'
\nhave you ever felt like your brain was about to explode? like you just couldn't take one more day of reading dry, boring academic research papers? well, wonder no more! for i have the solution to your problem!
\npresenting the world-renowned collection of flower classification papers, guaranteed to make you smarter in under 60 seconds! (or 60 years if you actually read them all)
\nNo more struggling with tedious text on abstruse subjects! these papers have been selected by our team of experts consisting of only myself for their ability to easily teach you everything you need to know about the subject!
So sit back, relax, and let the papers do the hard work for you!
\nEnjoy <3
An Optimized Deep Learning Model For Flower Classification Using NAS-FPN And Faster R- CNN
\n\n\nTL;DR: In this paper, a deep convolutional neural network based on NAS-FPN and Faster R-CNN is proposed for flower object detection, localization and classification.
\n
In computer vision, object detection is widely used in many applications such as face detection, video surveillance, vehicle detection, plant leaf detection etc. Deep neural networks have greater capabilities for image pattern recognition and are widely used in Computer Vision algorithms. In this paper, a deep convolutional neural network based on NAS-FPN and Faster R-CNN is proposed for flower object detection, localization and classification. Using the method of transfer learning, different pre-trained models including ResNet 50, ResNet 101, Inception ResNet V2, Inception V2, NAS, and MobileNet V2 are trained and evaluated on flower 30 dataset and flower 102 dataset that contains 19679 flower images. Based on the experiment carried out, the result demonstrates that the performance of the proposed NAS-FPN with Faster R-CNN model using transfer learning approach gives optimum mAP score of 87.6% on 102 flower class and 96.2% on 30 flower class datasets. Also, the proposed model is able to detect, locate and classify flowers with other significant details that includes flower name, division, class, subclass, order, family, and herb flower using multiclass classification and multi-labeling techniques.
\nFlower Classification using Deep Learning models
\nDeep learning techniques are used widespread for image recognition and classification problems. Gradually, deep learning architectures have modified to comprise more layers and become more robust model for classification problems. In this paper, the base VGG16 model is fine-tuned for the classification flowers into five categories, namely, Daisy, Dandelion, Sunflower, Rose and Tulip flowers. The fine-tuned VGG16 model is trained using 3520 flower images. The model is achieved a classification accuracy of 97.67% for validation set and 95.00% for testing dataset. The Kaggle dataset is used for training, validation and testing of the proposed fine-tuned VGG16 model. The goal of this work is to show that a proper modified VGG16 deep model, which is, pre-trained on ImageNet for image classification can be used for other image data set using very small dataset without over fitting. The VGG16 model uses mall size 3x3 filters.
\nFlower Classification with Deep CNN and Machine Learning Algorithms
\n\n\nTL;DR: we used a deep convolutional neural network with data augmentation to classify flower images. the best accuracy we achieved was 98.5% for oxford 102-flowers dataset and 99.8% for oxford 17-flowers dataset.
\n
Development of the recognition of rare plant species will be advantageous in the fields such as the pharmaceutical industry, botany, agricultural, and trade activities. It was also very challenging that there is diversity of flower species and it is very hard to classify them when they can be very similar to each other indeed. Therefore, this subject has already become crucial. In this context, this paper presents a classification system for flower images by using Deep CNN and Data Augmentation. Recently, Deep CNN techniques have become the latest technology for such problems. However, the fact is that getting better performance for the flower classification is stuck due to the lack of labeled data. In the study, there are three primary contributions: First, we proposed a classification model to cultivate the performance of classifying of flower images by using Deep CNN for extracting the features and various machine learning algorithms for classifying purposes. Second, we demonstrated the use of image augmentation for achieving better performance results. Last, we compared the performances of the machine-learning classifiers such as SVM, Random Forest, KNN, and Multi-Layer Perceptron(MLP). In the study, we evaluated our classification system using two datasets: Oxford-17 Flowers, and Oxford-102 Flowers. We divided each dataset into the training and test sets by 0.8 and 0.2, respectively. As a result, we obtained the best accuracy for Oxford 102-Flowers Dataset as 98.5% using SVM Classifier. For Oxford 17-Flowers Dataset, we found the best accuracy as 99.8% with MLP Classifier. These results are better than others’ that classify the same datasets in the literature.
\nStudy of flower image classification using deep learning to support agricultural pollination
\nIn smart agriculture, research and development is advanced by robots performing agricultural works instead of humans. Agricultural works requires experience and the human sense of sight and touch. In our study, experience and the sense of sight are replaced by machine learning. We developed a deep learning classification method and implemented it for tomato flower pollination classification in the agricultural field.
\nFlora in Indonesia has about 25% of the species of flowering plant species present in the world. Roses are one type of flowering plants and are usually used as an ornamental plant that has a thorny stem. Roses have more than 150 species. In Indonesia there are several flower gardens that is larger than the others. One of the famous flower garden in Indonesia is located on Malang city, East Java. The flower garden in Malang has several varieties of many roses and has a large production of roses. To help the sales system of roses there, the researchers want to create a program that can classify the type of roses in order to help simplify the system of automatic sales of roses without through manual sorting. So that will accelerate the sale of roses with an automated system. Ordinary people with limited botanical knowledge usually don’t know how to classify the flowers just by looking at them. To classify the flowers properly, it is important to provide enough information, and one of them is the name of it. Convolutional Neural Network (CNN) is one method of deep learning that can be used for image classification process. The CNN design is motivated by the discovery of the visual mechanism, the visual cortex present in the brain. CNN has been widely used in many real-world applications, such as Face Recognition, Image Classification and Recognition, and Object Detection, because this is one of the most efficient method for extracting important features. In this research, the classification accuracy value obtained from the test data is 96.33% using 2-dimension Red Green Blue (RGB) input image, and the size of each image is 32 × 32 pixels that are trained with CNN algorithm and the network structure of four convolution layers and four layers pooling supported by dropout technique.
\nA Comprehensive Analysis of Deep Learning Techniques for Recognition of Flower Species
\nFlower Species Recognition may be a difficult issue due to the wide selection of features, like leaves and grass. The classification is done by the traditional method through color, shape, texture, petals, sepals etc. The image analysis and classification has been sharply developed by Deep Learning methods. This research work considered a dataset which contains 4242 images with 5 classes by using Convolutional Neural Network (CNN) to recognize flower species with high accuracy by using framework. Tensor Flow and Image Data Generator is used to augment the training set and avoid Overfitting.
\nFour-Dimension Deep Learning Method for Flower Quality Grading with Depth Information
\n\n\nTL;DR: Depth information was used to improve the accuracy of a convolutional neural network in classifying flower buds by their maturing status. The InceptionV3 network with RGBD input had the highest classification accuracy.
\n
Grading the quality of fresh cut flowers is an important practice in the flower industry. Based on the flower maturing status, a classification method based on deep learning and depth information was proposed for the grading of flower quality. Firstly, the RGB image and the depth image of a flower bud were collected and transformed into fused RGBD information. Then, the RGBD information of a flower was set as inputs of a convolutional neural network to determine the flower bud maturing status. Four convolutional neural network models (VGG16, ResNet18, MobileNetV2, and InceptionV3) were adjusted for a four-dimensional (4D) RGBD input to classify flowers, and their classification performances were compared with and without depth information. The experimental results show that the classification accuracy was improved with depth information, and the improved InceptionV3 network with RGBD achieved the highest classification accuracy (up to 98%), which means that the depth information can effectively reflect the characteristics of the flower bud and is helpful for the classification of the maturing status. These results have a certain significance for the intelligent classification and sorting of fresh flowers.
\nA Survey on Deep Learning Approaches For Flower Species Detection
\nIn modern’s world , the data is generated on multi-folded and cross platform which can be used to understand and analyse different domains such as Floriculture , Agriculture , Finance etc. In all real time analysis of data where time , money , man-power is playing important roles to measure, pass judgement and react/answer in terms of Floriculture domain because the floriculture cultivation in world are increasing day by day The data generated from analysis which will be in the form of shape , colour , petals , size etc. Flower species and recognition system can provide unique approach for providing flower species analysis which could be a type of clutter of flowers , an area of farm where more flowers having issues like disease , water problem etc. In this process , Lots of data analyse , processed in real tome , provides good level of accuracy , precision , entropy etc. We can use deep learning approaches for detection flower species. This paper provides an overview of several pattern classification and detection mythologies/algorithms in the literature. The objective of the paper having companion with the comparison between different algorithms along with different types of datasets. The main goal is to provide an idea for several methods with different data and to find the many different approaches of the methods used for the detection of flowers using different scenarios
\n\n\nTL;DR: We used a deep learning algorithm, a Convolutional Neural Network (CNN), to classify the quality of dried clove flowers. We found that the best model for this task was one that had five layers in its neural network, and that the best values for theCNN hyperparameters were those that produced a reading accuracy of 87.75%.
\n
The objective of this study is to classify the quality of dried clove flowers using deep learning method with Convolutional Neural Network (CNN) algorithm, and also to perform the sensitivity analysis of CNN hyperparameters to obtain best model for clove quality classification process. The quality of clove as raw material in this study was determined according to SNI 3392-1994 by PT. Perkebunan Nusantara XII Pancusari Plantation, Malang, East Java, Indonesia. In total 1,600 images of dried clove flower were divided into 4 qualities. Each clove quality has 225 training data, 75 validation data, and 100 test data. The first step of this study is to build CNN model architecture as first model. The result of that model gives 65.25% reading accuracy. The second step is to analyze CNN sensitivity or CNN hyperparameter on the first model. The best value of CNN hyperparameter in each step then to be used in the next stage. Finally, after CNN hyperparameter carried out the reading accuracy of the test data is improved to 87.75%.
\nFlower classification using CNN and transfer learning in CNN- Agriculture Perspective
\nClassiücation of flowers is a difficult task because of the huge number of flowering plant species, which are similar in shape, color and appearance. A flower classification can be used in various applications such as field monitoring, plant identification, medicinal plant, floriculture industry, research in plant taxonomy. In this study, the authors have demonstrated and analyzed the recent developments in deep learning methods such as CNN and transfer learning in CNN. Prototype CNN model architecture proposed and transfer learning approach as well examined on VGG16, MobileNet2 and Resnet50 architecture for flower classification on publicly available flower dataset.
\nLearning Salient Features for Flower Classification Using Convolutional Neural Network
\nImage features can be utilized for image classification. Convolutional Neural Network (CNN) is able to extract features automatically from images rather than collecting features by hand. There are more and more flower images on the internet, and it is necessary to develop a system for identification of flower type. In this paper, we collect flower images from the internet and label them according to the species, then by using a deep CNN, we learn salient features of the flower images and achieve a significant performance of 78% in term of classification accuracy.
\nFlower Identification and Classification applying CNN through Deep Learning Methodologies
\n\n\nTL;DR: Convolutional neural networks play a significant role in the identification of flora species. Deep learning methodologies support us in image identification based on properties such as color and shape. Every species is distinct concerning attributes like texture, the shape of petals, and sepals. In this paper, we classify five various categories of flora named as daisy, dandelion, rose, sunflower, tulip. The methodology utilized here is the convolutional neural network where for instinctive recognition based on flora pictures. We focused on a neural network which implies an approach called skip connections and is flexible enough to perform various classification methodologies
\n
Convolutional neural networks play a significant role in the identification of flora species. Deep learning methodologies support us in image identification based on properties such as color and shape. Every species is distinct concerning attributes like texture, the shape of petals, and sepals. In this paper, we classify five various categories of flora named as daisy, dandelion, rose, sunflower, tulip. The methodology utilized here is the convolutional neural network where for instinctive recognition based on flora pictures. We focused on a neural network which implies an approach called skip connections and is flexible enough to perform various classification methodologies. We concentrated related to image classification by training and validating the dataset. The methodology adopted here is a residual neural network with nine deep layers, where implementation and approaches are discussed related to the classification model. We observed different existing algorithms and discussed the disadvantages, based on the drawbacks. We designed an algorithm for the classification and identification of a flower. The Experimental methodologies adopted are based on PyTorch and datasets. Finally, we presented the experimental results along with the graphical analysis. Keywords—Convolutional Neural Networks, Deep learning, Classification, Data mining, Artificial Intelligence.
\nFlower Image Classification Using Deep Convolutional Neural Network
\nThese days deep learning methods play a pivotal role in complicated tasks, such as extracting useful features, segmentation, and semantic classification of images. These methods had significant effects on flower types classification during recent years. In this paper, we are trying to classify 102 flower species using a robust deep learning method. To this end, we used the transfer learning approach employing DenseNet121 architecture to categorize various species of oxford-102 flowers dataset. In this regard, we have tried to fine-tune our model to achieve higher accuracy respect to other methods. We performed preprocessing by normalizing and resizing of our images and then fed them to our fine-tuned pretrained model. We divided our dataset to three sets of train, validation, and test. We could achieve the accuracy of 98.6% for 50 epochs which is better than other deep-learning based methods for the same dataset in the study.
\nFlower classification using deep convolutional neural networks
\nFlower classification is a challenging task due to the wide range of flower species, which have a similar shape, appearance or surrounding objects such as leaves and grass. In this study, the authors propose a novel two-step deep learning classifier to distinguish flowers of a wide range of species. First, the flower region is automatically segmented to allow localisation of the minimum bounding box around it. The proposed flower segmentation approach is modelled as a binary classifier in a fully convolutional network framework. Second, they build a robust convolutional neural network classifier to distinguish the different flower types. They propose novel steps during the training stage to ensure robust, accurate and real-time classification. They evaluate their method on three well known flower datasets. Their classification results exceed 97% on all datasets, which are better than the state-of-the-art in this domain.
\nA deep learning approach for the classification of diseased plant leaf images
\n\n\nTL;DR: A multilayer convolutional neural network is proposed for the classification of diseased plant leaf images. The real-time images of four different plants in healthy and diseased condition are collected for validating the performance of the proposed model. Results, when compared with other methods, shows the higher classification accuracy of the proposed model.
\n
Plant growth monitoring and plant protection are the key elements in the plant production industry. These factors influence the quality and productivity of the plant and its yields. Diseases are the major factor that vitiates the plant health. More often they harm plant parts like fruit, flower, leaf or stem, but quite often the severity of diseases may even result in plant death. In recent years, computer vision techniques, machine learning algorithms, and deep learning models have gained importance due to their capability of dealing with complex data with precision. These techniques are well known for pattern recognition and classification problems. Therefore in this work, a multilayer convolutional neural network is proposed for the classification of diseased plant leaf images. The real-time images of four different plants in healthy and diseased condition are collected for validating the performance of the proposed model. Results, when compared with other methods, shows the higher classification accuracy of the proposed model.
\nFlower identification based on Deep Learning
\n\n\nTL;DR: builds a modified tiny darknet in flowers classification method to automatically extract the characteristics of flower images, then classifies and identifies flower test images.
\n
In the field of plant scientific research, agroforestry investigation and production and management, plant identification is crucial basic work, and flower identification is an important part of plant identification. Given the present artificial defects of labor cost, low efficiency and low accuracy in present artificial flower information query and traditional computer vision method, the study built a modified tiny darknet in flowers classification method. Seventeen types of flower datasets published by Oxford University are taken as the research objects and the input of the neural network model. The deep network classification model is trained to automatically extract the characteristics of flower images. Combined with softmax classifier, the flower test images are classified and identified. The experimental results show that the classification accuracy is 92% which is higher than the classification algorithm results of the original model and some current mainstream models. This model has a simple structure, few training parameters, and has achieved a good recognition effect. It is suitable for automatic classification and recognition in the field of flower planting and is convenient for the retrieval of agricultural plant information database.
\nWith the development of deep learning, flower classification has become a typical problem. The antagonistic generation network provides a method for data set expansion. At the same time, ResNet ensures that the depth of the neural network model can continue to deepen, while the training effect will not be worse than that of the shallow layer. In this paper, the two are combined. Firstly, the improved deep convolution antagonism generating network is used to expand the data set, and then the expanded data set is used in the ResNet model for training. In the experiment, by adjusting the parameters of DCGAN, performing multiple experiment comparisons to improve the performance of the model to generate pictures, finally, it is concluded that compared with the ResNet 101 trained on the original small data set, the classification accuracy of ResNet using the extended data set has been significantly improved.
\nFlower image classification based on generative adversarial network and transfer learning
\n\n\nTL;DR: This paper uses a generative adversarial network (GAN) with a residual network (ResNet-101) to improve the accuracy of flower classification.
\n
Aiming at the problem that the classification accuracy of the traditional flower classification method is low and the deep neural network requires a large amount of original data. This paper designs a flower classification model that combines generative adversarial network and ResNet-101 transfer learning algorithm, and uses stochastic gradient descent algorithm to optimize the training process of the model. The experimental results on the the international public flower recognition dataset, Oxford flower-102 dataset, show that by enhancing the original data, the accuracy of the network's recognition and classification of flowers is improved. At the same time, the model proposed in this paper is superior to other traditional network models, with higher recognition accuracy and robustness.
\nNowadays, deep learning techniques are increasingly growing in machine vision for object recognition, segmentation, classification, and so on, in a wide variety of applications. In this study, we apply the convolutional neural network (CNN) to flower classification. For this purpose, we firstly increase the data with the augmentation techniques and use them in the pre-trained CNN models in which classification part is removed and instead of it, we use global average pooling (GAP) in the last layer for extracting their features. The features obtained from these models are concatenated, and then we use a support vector machine (SVM) as classifier for the flower classification. We use the Oxford 102 flower and the Oxford 17 flower datasets in our experiments. By applying this method, we achieve 96.47% classification accuracy for the Oxford 102 flower and 97.64% classification accuracy for the Oxford 17 flower. The results show the effectiveness of the proposed strategy and perform more accurate classification than the traditional methods.
\n\n\nTL;DR: We evaluated the impacts of large- , medium- and small-size deep residual-learning convolutional neural networks (DRL-CNN): ResNet50, ResNet35 and ResNet17 models on classifying Oxford-102 flowers image dataset with distinct number of 10, 50 and 102 flower classes. The Flowers image classification assessments rely on precision, recall, F1 scores and accuracy rates, averaged over 10-fold cross validation to ensure unbiased experimented results. Confusion matrix is also considered for more detail of results examination. The comparison results indicate the ResNet35 yields 0.201% and 0.706% few better
\n
This paper focuses on evaluating impacts of large -, medium - and small -size deep residual-learning convolutional neural network (DRL-CNN): ResNet50, ResNet35 and ResNet17 models on classifying Oxford-102 flowers image dataset with distinct number of 10, 50 and 102 flower classes. The Flowers image classification assessments rely on precision, recall, F1 scores and accuracy rates, averaged over 10-fold cross validation to ensure unbiased experimented results. Confusion matrix is also considered for more detail of results examination. The comparison results indicate the ResNet35 yields 0.201% and 0.706% few better recognition accuracy consecutively over ResNet50 and 17 according to 10-class dataset. For 50-class, 0.060% and 0.211% bits higher accuracy of ResNet35 than ResNet50 and ResNet17 are respectively generated. Whereas, 0.040% and 0.070% a few bits better performance of ResNet35 than ResNet50 and ResNet17 are sequentially attained on 100-class one. Decreasing rate regarding superiority of ResNet35 over ResNet50 and ResNet17 is indicated when increasing class size. However, less than 0.71% higher classification performance is indicated for ResNet35 than ResNet17 for all cases within the scope of this work. Thus, ResNet17 may be preferred to ResNet35 due to approximate 33% higher amount of parameters used in ResNet35 than ResNet17.
\n\n\nTL;DR: We have performed a comparative analysis of the performance of various object detection models. We have compared; SSD quantized model (8-bit) using MobileNet V1 and MobileNet V2, Atrous model using Faster R-CNN with Inception ResNet V2, low proposals model using Faster R-CNN with ResNet 50 and NAS, atrous and low proposals models using Faster R-CNN with ResNet 101 and Inception ResNet V2 with the proposed NAS-FPN with modified Faster RCNN model.
\n
Flower species identification refers to a process of comparing defined characteristics of a given flower to allocate a particular species to a known taxonomic group. Flowers can be identified and classified by observing certain distinguishing basic and morphological characteristics. Classifying flower species is challenging for people and needs in-depth specialist knowledge as some flower species look similar, whereas some look differently despite of being in the same species. Traditional computer vision methods remain inefficient and less accurate while considering environmental complexity and similarity and difference between flowers species. Deep CNN, an emerging field of machine learning and artificial intelligence, has grown rapidly and widely applied in computer vision applications with promising results, especially in the field of object detection from visual images. In this paper, we have performed a comparative analysis of the performance of various object detection models. We have compared; SSD quantized model (8-bit) using MobileNet V1 and MobileNet V2, Atrous model using Faster R-CNN with Inception ResNet V2, low proposals model using Faster R-CNN with ResNet 50 and NAS, atrous and low proposals models using Faster R-CNN with ResNet 101 and Inception ResNet V2 with the proposed NAS-FPN with modified Faster RCNN model. Based on the results obtained during experiment, the proposed NAS-FPN with modified Faster R-CNN model achieved good performance and highest mAP score of 87.6% on F102 flower class and 96.2% on J30 flower class datasets.
\nAutomated color detection in orchids using color labels and deep learning
\nThe color of particular parts of a flower is often employed as one of the features to differentiate between flower types. Thus, color is also used in flower-image classification. Color labels, such as ‘green’, ‘red’, and ‘yellow’, are used by taxonomists and lay people alike to describe the color of plants. Flower image datasets usually only consist of images and do not contain flower descriptions. In this research, we have built a flower-image dataset, especially regarding orchid species, which consists of human-friendly textual descriptions of features of specific flowers, on the one hand, and digital photographs indicating how a flower looks like, on the other hand. Using this dataset, a new automated color detection model was developed. It is the first research of its kind using color labels and deep learning for color detection in flower recognition. As deep learning often excels in pattern recognition in digital images, we applied transfer learning with various amounts of unfreezing of layers with five different neural network architectures (VGG16, Inception, Resnet50, Xception, Nasnet) to determine which architecture and which scheme of transfer learning performs best. In addition, various color scheme scenarios were tested, including the use of primary and secondary color together, and, in addition, the effectiveness of dealing with multi-class classification using multi-class, combined binary, and, finally, ensemble classifiers were studied. The best overall performance was achieved by the ensemble classifier. The results show that the proposed method can detect the color of flower and labellum very well without having to perform image segmentation. The result of this study can act as a foundation for the development of an image-based plant recognition system that is able to offer an explanation of a provided classification.
\n'
\nhave you ever felt like your brain was about to explode? like you just couldn't take one more day of reading dry, boring academic research papers? well, wonder no more! for i have the solution to your problem!
\npresenting the world-renowned collection of flower classification papers, guaranteed to make you smarter in under 60 seconds! (or 60 years if you actually read them all)
\nNo more struggling with tedious text on abstruse subjects! these papers have been selected by our team of experts consisting of only myself for their ability to easily teach you everything you need to know about the subject!
So sit back, relax, and let the papers do the hard work for you!
\nEnjoy <3
An Optimized Deep Learning Model For Flower Classification Using NAS-FPN And Faster R- CNN
\n\n\nTL;DR: In this paper, a deep convolutional neural network based on NAS-FPN and Faster R-CNN is proposed for flower object detection, localization and classification.
\n
In computer vision, object detection is widely used in many applications such as face detection, video surveillance, vehicle detection, plant leaf detection etc. Deep neural networks have greater capabilities for image pattern recognition and are widely used in Computer Vision algorithms. In this paper, a deep convolutional neural network based on NAS-FPN and Faster R-CNN is proposed for flower object detection, localization and classification. Using the method of transfer learning, different pre-trained models including ResNet 50, ResNet 101, Inception ResNet V2, Inception V2, NAS, and MobileNet V2 are trained and evaluated on flower 30 dataset and flower 102 dataset that contains 19679 flower images. Based on the experiment carried out, the result demonstrates that the performance of the proposed NAS-FPN with Faster R-CNN model using transfer learning approach gives optimum mAP score of 87.6% on 102 flower class and 96.2% on 30 flower class datasets. Also, the proposed model is able to detect, locate and classify flowers with other significant details that includes flower name, division, class, subclass, order, family, and herb flower using multiclass classification and multi-labeling techniques.
\nFlower Classification using Deep Learning models
\nDeep learning techniques are used widespread for image recognition and classification problems. Gradually, deep learning architectures have modified to comprise more layers and become more robust model for classification problems. In this paper, the base VGG16 model is fine-tuned for the classification flowers into five categories, namely, Daisy, Dandelion, Sunflower, Rose and Tulip flowers. The fine-tuned VGG16 model is trained using 3520 flower images. The model is achieved a classification accuracy of 97.67% for validation set and 95.00% for testing dataset. The Kaggle dataset is used for training, validation and testing of the proposed fine-tuned VGG16 model. The goal of this work is to show that a proper modified VGG16 deep model, which is, pre-trained on ImageNet for image classification can be used for other image data set using very small dataset without over fitting. The VGG16 model uses mall size 3x3 filters.
\nFlower Classification with Deep CNN and Machine Learning Algorithms
\n\n\nTL;DR: we used a deep convolutional neural network with data augmentation to classify flower images. the best accuracy we achieved was 98.5% for oxford 102-flowers dataset and 99.8% for oxford 17-flowers dataset.
\n
Development of the recognition of rare plant species will be advantageous in the fields such as the pharmaceutical industry, botany, agricultural, and trade activities. It was also very challenging that there is diversity of flower species and it is very hard to classify them when they can be very similar to each other indeed. Therefore, this subject has already become crucial. In this context, this paper presents a classification system for flower images by using Deep CNN and Data Augmentation. Recently, Deep CNN techniques have become the latest technology for such problems. However, the fact is that getting better performance for the flower classification is stuck due to the lack of labeled data. In the study, there are three primary contributions: First, we proposed a classification model to cultivate the performance of classifying of flower images by using Deep CNN for extracting the features and various machine learning algorithms for classifying purposes. Second, we demonstrated the use of image augmentation for achieving better performance results. Last, we compared the performances of the machine-learning classifiers such as SVM, Random Forest, KNN, and Multi-Layer Perceptron(MLP). In the study, we evaluated our classification system using two datasets: Oxford-17 Flowers, and Oxford-102 Flowers. We divided each dataset into the training and test sets by 0.8 and 0.2, respectively. As a result, we obtained the best accuracy for Oxford 102-Flowers Dataset as 98.5% using SVM Classifier. For Oxford 17-Flowers Dataset, we found the best accuracy as 99.8% with MLP Classifier. These results are better than others’ that classify the same datasets in the literature.
\nStudy of flower image classification using deep learning to support agricultural pollination
\nIn smart agriculture, research and development is advanced by robots performing agricultural works instead of humans. Agricultural works requires experience and the human sense of sight and touch. In our study, experience and the sense of sight are replaced by machine learning. We developed a deep learning classification method and implemented it for tomato flower pollination classification in the agricultural field.
\nFlora in Indonesia has about 25% of the species of flowering plant species present in the world. Roses are one type of flowering plants and are usually used as an ornamental plant that has a thorny stem. Roses have more than 150 species. In Indonesia there are several flower gardens that is larger than the others. One of the famous flower garden in Indonesia is located on Malang city, East Java. The flower garden in Malang has several varieties of many roses and has a large production of roses. To help the sales system of roses there, the researchers want to create a program that can classify the type of roses in order to help simplify the system of automatic sales of roses without through manual sorting. So that will accelerate the sale of roses with an automated system. Ordinary people with limited botanical knowledge usually don’t know how to classify the flowers just by looking at them. To classify the flowers properly, it is important to provide enough information, and one of them is the name of it. Convolutional Neural Network (CNN) is one method of deep learning that can be used for image classification process. The CNN design is motivated by the discovery of the visual mechanism, the visual cortex present in the brain. CNN has been widely used in many real-world applications, such as Face Recognition, Image Classification and Recognition, and Object Detection, because this is one of the most efficient method for extracting important features. In this research, the classification accuracy value obtained from the test data is 96.33% using 2-dimension Red Green Blue (RGB) input image, and the size of each image is 32 × 32 pixels that are trained with CNN algorithm and the network structure of four convolution layers and four layers pooling supported by dropout technique.
\nA Comprehensive Analysis of Deep Learning Techniques for Recognition of Flower Species
\nFlower Species Recognition may be a difficult issue due to the wide selection of features, like leaves and grass. The classification is done by the traditional method through color, shape, texture, petals, sepals etc. The image analysis and classification has been sharply developed by Deep Learning methods. This research work considered a dataset which contains 4242 images with 5 classes by using Convolutional Neural Network (CNN) to recognize flower species with high accuracy by using framework. Tensor Flow and Image Data Generator is used to augment the training set and avoid Overfitting.
\nFour-Dimension Deep Learning Method for Flower Quality Grading with Depth Information
\n\n\nTL;DR: Depth information was used to improve the accuracy of a convolutional neural network in classifying flower buds by their maturing status. The InceptionV3 network with RGBD input had the highest classification accuracy.
\n
Grading the quality of fresh cut flowers is an important practice in the flower industry. Based on the flower maturing status, a classification method based on deep learning and depth information was proposed for the grading of flower quality. Firstly, the RGB image and the depth image of a flower bud were collected and transformed into fused RGBD information. Then, the RGBD information of a flower was set as inputs of a convolutional neural network to determine the flower bud maturing status. Four convolutional neural network models (VGG16, ResNet18, MobileNetV2, and InceptionV3) were adjusted for a four-dimensional (4D) RGBD input to classify flowers, and their classification performances were compared with and without depth information. The experimental results show that the classification accuracy was improved with depth information, and the improved InceptionV3 network with RGBD achieved the highest classification accuracy (up to 98%), which means that the depth information can effectively reflect the characteristics of the flower bud and is helpful for the classification of the maturing status. These results have a certain significance for the intelligent classification and sorting of fresh flowers.
\nA Survey on Deep Learning Approaches For Flower Species Detection
\nIn modern’s world , the data is generated on multi-folded and cross platform which can be used to understand and analyse different domains such as Floriculture , Agriculture , Finance etc. In all real time analysis of data where time , money , man-power is playing important roles to measure, pass judgement and react/answer in terms of Floriculture domain because the floriculture cultivation in world are increasing day by day The data generated from analysis which will be in the form of shape , colour , petals , size etc. Flower species and recognition system can provide unique approach for providing flower species analysis which could be a type of clutter of flowers , an area of farm where more flowers having issues like disease , water problem etc. In this process , Lots of data analyse , processed in real tome , provides good level of accuracy , precision , entropy etc. We can use deep learning approaches for detection flower species. This paper provides an overview of several pattern classification and detection mythologies/algorithms in the literature. The objective of the paper having companion with the comparison between different algorithms along with different types of datasets. The main goal is to provide an idea for several methods with different data and to find the many different approaches of the methods used for the detection of flowers using different scenarios
\n\n\nTL;DR: We used a deep learning algorithm, a Convolutional Neural Network (CNN), to classify the quality of dried clove flowers. We found that the best model for this task was one that had five layers in its neural network, and that the best values for theCNN hyperparameters were those that produced a reading accuracy of 87.75%.
\n
The objective of this study is to classify the quality of dried clove flowers using deep learning method with Convolutional Neural Network (CNN) algorithm, and also to perform the sensitivity analysis of CNN hyperparameters to obtain best model for clove quality classification process. The quality of clove as raw material in this study was determined according to SNI 3392-1994 by PT. Perkebunan Nusantara XII Pancusari Plantation, Malang, East Java, Indonesia. In total 1,600 images of dried clove flower were divided into 4 qualities. Each clove quality has 225 training data, 75 validation data, and 100 test data. The first step of this study is to build CNN model architecture as first model. The result of that model gives 65.25% reading accuracy. The second step is to analyze CNN sensitivity or CNN hyperparameter on the first model. The best value of CNN hyperparameter in each step then to be used in the next stage. Finally, after CNN hyperparameter carried out the reading accuracy of the test data is improved to 87.75%.
\nFlower classification using CNN and transfer learning in CNN- Agriculture Perspective
\nClassiücation of flowers is a difficult task because of the huge number of flowering plant species, which are similar in shape, color and appearance. A flower classification can be used in various applications such as field monitoring, plant identification, medicinal plant, floriculture industry, research in plant taxonomy. In this study, the authors have demonstrated and analyzed the recent developments in deep learning methods such as CNN and transfer learning in CNN. Prototype CNN model architecture proposed and transfer learning approach as well examined on VGG16, MobileNet2 and Resnet50 architecture for flower classification on publicly available flower dataset.
\nLearning Salient Features for Flower Classification Using Convolutional Neural Network
\nImage features can be utilized for image classification. Convolutional Neural Network (CNN) is able to extract features automatically from images rather than collecting features by hand. There are more and more flower images on the internet, and it is necessary to develop a system for identification of flower type. In this paper, we collect flower images from the internet and label them according to the species, then by using a deep CNN, we learn salient features of the flower images and achieve a significant performance of 78% in term of classification accuracy.
\nFlower Identification and Classification applying CNN through Deep Learning Methodologies
\n\n\nTL;DR: Convolutional neural networks play a significant role in the identification of flora species. Deep learning methodologies support us in image identification based on properties such as color and shape. Every species is distinct concerning attributes like texture, the shape of petals, and sepals. In this paper, we classify five various categories of flora named as daisy, dandelion, rose, sunflower, tulip. The methodology utilized here is the convolutional neural network where for instinctive recognition based on flora pictures. We focused on a neural network which implies an approach called skip connections and is flexible enough to perform various classification methodologies
\n
Convolutional neural networks play a significant role in the identification of flora species. Deep learning methodologies support us in image identification based on properties such as color and shape. Every species is distinct concerning attributes like texture, the shape of petals, and sepals. In this paper, we classify five various categories of flora named as daisy, dandelion, rose, sunflower, tulip. The methodology utilized here is the convolutional neural network where for instinctive recognition based on flora pictures. We focused on a neural network which implies an approach called skip connections and is flexible enough to perform various classification methodologies. We concentrated related to image classification by training and validating the dataset. The methodology adopted here is a residual neural network with nine deep layers, where implementation and approaches are discussed related to the classification model. We observed different existing algorithms and discussed the disadvantages, based on the drawbacks. We designed an algorithm for the classification and identification of a flower. The Experimental methodologies adopted are based on PyTorch and datasets. Finally, we presented the experimental results along with the graphical analysis. Keywords—Convolutional Neural Networks, Deep learning, Classification, Data mining, Artificial Intelligence.
\nFlower Image Classification Using Deep Convolutional Neural Network
\nThese days deep learning methods play a pivotal role in complicated tasks, such as extracting useful features, segmentation, and semantic classification of images. These methods had significant effects on flower types classification during recent years. In this paper, we are trying to classify 102 flower species using a robust deep learning method. To this end, we used the transfer learning approach employing DenseNet121 architecture to categorize various species of oxford-102 flowers dataset. In this regard, we have tried to fine-tune our model to achieve higher accuracy respect to other methods. We performed preprocessing by normalizing and resizing of our images and then fed them to our fine-tuned pretrained model. We divided our dataset to three sets of train, validation, and test. We could achieve the accuracy of 98.6% for 50 epochs which is better than other deep-learning based methods for the same dataset in the study.
\nFlower classification using deep convolutional neural networks
\nFlower classification is a challenging task due to the wide range of flower species, which have a similar shape, appearance or surrounding objects such as leaves and grass. In this study, the authors propose a novel two-step deep learning classifier to distinguish flowers of a wide range of species. First, the flower region is automatically segmented to allow localisation of the minimum bounding box around it. The proposed flower segmentation approach is modelled as a binary classifier in a fully convolutional network framework. Second, they build a robust convolutional neural network classifier to distinguish the different flower types. They propose novel steps during the training stage to ensure robust, accurate and real-time classification. They evaluate their method on three well known flower datasets. Their classification results exceed 97% on all datasets, which are better than the state-of-the-art in this domain.
\nA deep learning approach for the classification of diseased plant leaf images
\n\n\nTL;DR: A multilayer convolutional neural network is proposed for the classification of diseased plant leaf images. The real-time images of four different plants in healthy and diseased condition are collected for validating the performance of the proposed model. Results, when compared with other methods, shows the higher classification accuracy of the proposed model.
\n
Plant growth monitoring and plant protection are the key elements in the plant production industry. These factors influence the quality and productivity of the plant and its yields. Diseases are the major factor that vitiates the plant health. More often they harm plant parts like fruit, flower, leaf or stem, but quite often the severity of diseases may even result in plant death. In recent years, computer vision techniques, machine learning algorithms, and deep learning models have gained importance due to their capability of dealing with complex data with precision. These techniques are well known for pattern recognition and classification problems. Therefore in this work, a multilayer convolutional neural network is proposed for the classification of diseased plant leaf images. The real-time images of four different plants in healthy and diseased condition are collected for validating the performance of the proposed model. Results, when compared with other methods, shows the higher classification accuracy of the proposed model.
\nFlower identification based on Deep Learning
\n\n\nTL;DR: builds a modified tiny darknet in flowers classification method to automatically extract the characteristics of flower images, then classifies and identifies flower test images.
\n
In the field of plant scientific research, agroforestry investigation and production and management, plant identification is crucial basic work, and flower identification is an important part of plant identification. Given the present artificial defects of labor cost, low efficiency and low accuracy in present artificial flower information query and traditional computer vision method, the study built a modified tiny darknet in flowers classification method. Seventeen types of flower datasets published by Oxford University are taken as the research objects and the input of the neural network model. The deep network classification model is trained to automatically extract the characteristics of flower images. Combined with softmax classifier, the flower test images are classified and identified. The experimental results show that the classification accuracy is 92% which is higher than the classification algorithm results of the original model and some current mainstream models. This model has a simple structure, few training parameters, and has achieved a good recognition effect. It is suitable for automatic classification and recognition in the field of flower planting and is convenient for the retrieval of agricultural plant information database.
\nWith the development of deep learning, flower classification has become a typical problem. The antagonistic generation network provides a method for data set expansion. At the same time, ResNet ensures that the depth of the neural network model can continue to deepen, while the training effect will not be worse than that of the shallow layer. In this paper, the two are combined. Firstly, the improved deep convolution antagonism generating network is used to expand the data set, and then the expanded data set is used in the ResNet model for training. In the experiment, by adjusting the parameters of DCGAN, performing multiple experiment comparisons to improve the performance of the model to generate pictures, finally, it is concluded that compared with the ResNet 101 trained on the original small data set, the classification accuracy of ResNet using the extended data set has been significantly improved.
\nFlower image classification based on generative adversarial network and transfer learning
\n\n\nTL;DR: This paper uses a generative adversarial network (GAN) with a residual network (ResNet-101) to improve the accuracy of flower classification.
\n
Aiming at the problem that the classification accuracy of the traditional flower classification method is low and the deep neural network requires a large amount of original data. This paper designs a flower classification model that combines generative adversarial network and ResNet-101 transfer learning algorithm, and uses stochastic gradient descent algorithm to optimize the training process of the model. The experimental results on the the international public flower recognition dataset, Oxford flower-102 dataset, show that by enhancing the original data, the accuracy of the network's recognition and classification of flowers is improved. At the same time, the model proposed in this paper is superior to other traditional network models, with higher recognition accuracy and robustness.
\nNowadays, deep learning techniques are increasingly growing in machine vision for object recognition, segmentation, classification, and so on, in a wide variety of applications. In this study, we apply the convolutional neural network (CNN) to flower classification. For this purpose, we firstly increase the data with the augmentation techniques and use them in the pre-trained CNN models in which classification part is removed and instead of it, we use global average pooling (GAP) in the last layer for extracting their features. The features obtained from these models are concatenated, and then we use a support vector machine (SVM) as classifier for the flower classification. We use the Oxford 102 flower and the Oxford 17 flower datasets in our experiments. By applying this method, we achieve 96.47% classification accuracy for the Oxford 102 flower and 97.64% classification accuracy for the Oxford 17 flower. The results show the effectiveness of the proposed strategy and perform more accurate classification than the traditional methods.
\n\n\nTL;DR: We evaluated the impacts of large- , medium- and small-size deep residual-learning convolutional neural networks (DRL-CNN): ResNet50, ResNet35 and ResNet17 models on classifying Oxford-102 flowers image dataset with distinct number of 10, 50 and 102 flower classes. The Flowers image classification assessments rely on precision, recall, F1 scores and accuracy rates, averaged over 10-fold cross validation to ensure unbiased experimented results. Confusion matrix is also considered for more detail of results examination. The comparison results indicate the ResNet35 yields 0.201% and 0.706% few better
\n
This paper focuses on evaluating impacts of large -, medium - and small -size deep residual-learning convolutional neural network (DRL-CNN): ResNet50, ResNet35 and ResNet17 models on classifying Oxford-102 flowers image dataset with distinct number of 10, 50 and 102 flower classes. The Flowers image classification assessments rely on precision, recall, F1 scores and accuracy rates, averaged over 10-fold cross validation to ensure unbiased experimented results. Confusion matrix is also considered for more detail of results examination. The comparison results indicate the ResNet35 yields 0.201% and 0.706% few better recognition accuracy consecutively over ResNet50 and 17 according to 10-class dataset. For 50-class, 0.060% and 0.211% bits higher accuracy of ResNet35 than ResNet50 and ResNet17 are respectively generated. Whereas, 0.040% and 0.070% a few bits better performance of ResNet35 than ResNet50 and ResNet17 are sequentially attained on 100-class one. Decreasing rate regarding superiority of ResNet35 over ResNet50 and ResNet17 is indicated when increasing class size. However, less than 0.71% higher classification performance is indicated for ResNet35 than ResNet17 for all cases within the scope of this work. Thus, ResNet17 may be preferred to ResNet35 due to approximate 33% higher amount of parameters used in ResNet35 than ResNet17.
\n\n\nTL;DR: We have performed a comparative analysis of the performance of various object detection models. We have compared; SSD quantized model (8-bit) using MobileNet V1 and MobileNet V2, Atrous model using Faster R-CNN with Inception ResNet V2, low proposals model using Faster R-CNN with ResNet 50 and NAS, atrous and low proposals models using Faster R-CNN with ResNet 101 and Inception ResNet V2 with the proposed NAS-FPN with modified Faster RCNN model.
\n
Flower species identification refers to a process of comparing defined characteristics of a given flower to allocate a particular species to a known taxonomic group. Flowers can be identified and classified by observing certain distinguishing basic and morphological characteristics. Classifying flower species is challenging for people and needs in-depth specialist knowledge as some flower species look similar, whereas some look differently despite of being in the same species. Traditional computer vision methods remain inefficient and less accurate while considering environmental complexity and similarity and difference between flowers species. Deep CNN, an emerging field of machine learning and artificial intelligence, has grown rapidly and widely applied in computer vision applications with promising results, especially in the field of object detection from visual images. In this paper, we have performed a comparative analysis of the performance of various object detection models. We have compared; SSD quantized model (8-bit) using MobileNet V1 and MobileNet V2, Atrous model using Faster R-CNN with Inception ResNet V2, low proposals model using Faster R-CNN with ResNet 50 and NAS, atrous and low proposals models using Faster R-CNN with ResNet 101 and Inception ResNet V2 with the proposed NAS-FPN with modified Faster RCNN model. Based on the results obtained during experiment, the proposed NAS-FPN with modified Faster R-CNN model achieved good performance and highest mAP score of 87.6% on F102 flower class and 96.2% on J30 flower class datasets.
\nAutomated color detection in orchids using color labels and deep learning
\nThe color of particular parts of a flower is often employed as one of the features to differentiate between flower types. Thus, color is also used in flower-image classification. Color labels, such as ‘green’, ‘red’, and ‘yellow’, are used by taxonomists and lay people alike to describe the color of plants. Flower image datasets usually only consist of images and do not contain flower descriptions. In this research, we have built a flower-image dataset, especially regarding orchid species, which consists of human-friendly textual descriptions of features of specific flowers, on the one hand, and digital photographs indicating how a flower looks like, on the other hand. Using this dataset, a new automated color detection model was developed. It is the first research of its kind using color labels and deep learning for color detection in flower recognition. As deep learning often excels in pattern recognition in digital images, we applied transfer learning with various amounts of unfreezing of layers with five different neural network architectures (VGG16, Inception, Resnet50, Xception, Nasnet) to determine which architecture and which scheme of transfer learning performs best. In addition, various color scheme scenarios were tested, including the use of primary and secondary color together, and, in addition, the effectiveness of dealing with multi-class classification using multi-class, combined binary, and, finally, ensemble classifiers were studied. The best overall performance was achieved by the ensemble classifier. The results show that the proposed method can detect the color of flower and labellum very well without having to perform image segmentation. The result of this study can act as a foundation for the development of an image-based plant recognition system that is able to offer an explanation of a provided classification.
\nThis is helpful, thanks for putting it together.
", "rawMarkdown": "This is helpful, thanks for putting it together." }, { "id": 2634966, "postDate": "2024-02-04T05:56:48.843Z", "content": "Thanks for your work
", "rawMarkdown": "Thanks for your work" } ], "comments": [ { "id": 2439662, "author_name": "Yun Xing", "author_url": "", "post_date": "2023-09-15T03:50:49.410000", "content": "This is helpful, thanks for putting it together.
", "votes": 0, "replies": [] }, { "id": 2634966, "author_name": "charlesyao1568", "author_url": "", "post_date": "2024-02-04T05:56:48.843000", "content": "Thanks for your work
", "votes": 0, "replies": [] } ], "raw_markdown_by_id": { "1801409": "'\n\nhave you ever felt like your brain was about to explode? like you just couldn't take one more day of reading dry, boring academic research papers? well, wonder no more! for i have the solution to your problem!\npresenting the world-renowned collection of flower classification papers, guaranteed to make you smarter in under 60 seconds! (or 60 years if you actually read them all)\nNo more struggling with tedious text on abstruse subjects! these papers have been selected by our team of experts consisting of only myself for their ability to easily teach you everything you need to know about the subject!\n\nSo sit back, relax, and let the papers do the hard work for you!\nEnjoy <3\n\n\n_____\n**[An Optimized Deep Learning Model For Flower Classification Using NAS-FPN And Faster R- CNN](https://scholar.google.com/scholar?q=An%20Optimized%20Deep%20Learning%20Model%20For%20Flower%20Classification%20Using%20NAS-FPN%20And%20Faster%20R-%20CNN)**\n\n\n> **TL;DR:** In this paper, a deep convolutional neural network based on NAS-FPN and Faster R-CNN is proposed for flower object detection, localization and classification.\n\nIn computer vision, object detection is widely used in many applications such as face detection, video surveillance, vehicle detection, plant leaf detection etc. Deep neural networks have greater capabilities for image pattern recognition and are widely used in Computer Vision algorithms. In this paper, a deep convolutional neural network based on NAS-FPN and Faster R-CNN is proposed for flower object detection, localization and classification. Using the method of transfer learning, different pre-trained models including ResNet 50, ResNet 101, Inception ResNet V2, Inception V2, NAS, and MobileNet V2 are trained and evaluated on flower 30 dataset and flower 102 dataset that contains 19679 flower images. Based on the experiment carried out, the result demonstrates that the performance of the proposed NAS-FPN with Faster R-CNN model using transfer learning approach gives optimum mAP score of 87.6% on 102 flower class and 96.2% on 30 flower class datasets. Also, the proposed model is able to detect, locate and classify flowers with other significant details that includes flower name, division, class, subclass, order, family, and herb flower using multiclass classification and multi-labeling techniques.\n_____\n\n\n_____\n**[Flower Classification using Deep Learning models](https://doi.org/10.1109/ICSTCEE49637.2020.9277041)**\n\nDeep learning techniques are used widespread for image recognition and classification problems. Gradually, deep learning architectures have modified to comprise more layers and become more robust model for classification problems. In this paper, the base VGG16 model is fine-tuned for the classification flowers into five categories, namely, Daisy, Dandelion, Sunflower, Rose and Tulip flowers. The fine-tuned VGG16 model is trained using 3520 flower images. The model is achieved a classification accuracy of 97.67% for validation set and 95.00% for testing dataset. The Kaggle dataset is used for training, validation and testing of the proposed fine-tuned VGG16 model. The goal of this work is to show that a proper modified VGG16 deep model, which is, pre-trained on ImageNet for image classification can be used for other image data set using very small dataset without over fitting. The VGG16 model uses mall size 3x3 filters.\n_____\n\n\n_____\n**[Flower Classification with Deep CNN and Machine Learning Algorithms](https://doi.org/10.1109/ISMSIT.2019.8932908)**\n\n\n> **TL;DR:** we used a deep convolutional neural network with data augmentation to classify flower images. the best accuracy we achieved was 98.5% for oxford 102-flowers dataset and 99.8% for oxford 17-flowers dataset.\n\nDevelopment of the recognition of rare plant species will be advantageous in the fields such as the pharmaceutical industry, botany, agricultural, and trade activities. It was also very challenging that there is diversity of flower species and it is very hard to classify them when they can be very similar to each other indeed. Therefore, this subject has already become crucial. In this context, this paper presents a classification system for flower images by using Deep CNN and Data Augmentation. Recently, Deep CNN techniques have become the latest technology for such problems. However, the fact is that getting better performance for the flower classification is stuck due to the lack of labeled data. In the study, there are three primary contributions: First, we proposed a classification model to cultivate the performance of classifying of flower images by using Deep CNN for extracting the features and various machine learning algorithms for classifying purposes. Second, we demonstrated the use of image augmentation for achieving better performance results. Last, we compared the performances of the machine-learning classifiers such as SVM, Random Forest, KNN, and Multi-Layer Perceptron(MLP). In the study, we evaluated our classification system using two datasets: Oxford-17 Flowers, and Oxford-102 Flowers. We divided each dataset into the training and test sets by 0.8 and 0.2, respectively. As a result, we obtained the best accuracy for Oxford 102-Flowers Dataset as 98.5% using SVM Classifier. For Oxford 17-Flowers Dataset, we found the best accuracy as 99.8% with MLP Classifier. These results are better than others’ that classify the same datasets in the literature.\n_____\n\n\n_____\n**[Study of flower image classification using deep learning to support agricultural pollination](https://doi.org/10.1109/ICCE-TW52618.2021.9603106)**\n\nIn smart agriculture, research and development is advanced by robots performing agricultural works instead of humans. Agricultural works requires experience and the human sense of sight and touch. In our study, experience and the sense of sight are replaced by machine learning. We developed a deep learning classification method and implemented it for tomato flower pollination classification in the agricultural field.\n_____\n\n\n_____\n**[Implementation of Deep Learning Using Convolutional Neural Network Algorithm for Classification Rose Flower](https://doi.org/10.1088/1742-6596/1842/1/012002)**\n\nFlora in Indonesia has about 25% of the species of flowering plant species present in the world. Roses are one type of flowering plants and are usually used as an ornamental plant that has a thorny stem. Roses have more than 150 species. In Indonesia there are several flower gardens that is larger than the others. One of the famous flower garden in Indonesia is located on Malang city, East Java. The flower garden in Malang has several varieties of many roses and has a large production of roses. To help the sales system of roses there, the researchers want to create a program that can classify the type of roses in order to help simplify the system of automatic sales of roses without through manual sorting. So that will accelerate the sale of roses with an automated system. Ordinary people with limited botanical knowledge usually don’t know how to classify the flowers just by looking at them. To classify the flowers properly, it is important to provide enough information, and one of them is the name of it. Convolutional Neural Network (CNN) is one method of deep learning that can be used for image classification process. The CNN design is motivated by the discovery of the visual mechanism, the visual cortex present in the brain. CNN has been widely used in many real-world applications, such as Face Recognition, Image Classification and Recognition, and Object Detection, because this is one of the most efficient method for extracting important features. In this research, the classification accuracy value obtained from the test data is 96.33% using 2-dimension Red Green Blue (RGB) input image, and the size of each image is 32 × 32 pixels that are trained with CNN algorithm and the network structure of four convolution layers and four layers pooling supported by dropout technique.\n_____\n\n\n_____\n**[A Comprehensive Analysis of Deep Learning Techniques for Recognition of Flower Species](https://doi.org/10.1109/ICICV50876.2021.9388503)**\n\n\nFlower Species Recognition may be a difficult issue due to the wide selection of features, like leaves and grass. The classification is done by the traditional method through color, shape, texture, petals, sepals etc. The image analysis and classification has been sharply developed by Deep Learning methods. This research work considered a dataset which contains 4242 images with 5 classes by using Convolutional Neural Network (CNN) to recognize flower species with high accuracy by using framework. Tensor Flow and Image Data Generator is used to augment the training set and avoid Overfitting.\n_____\n\n\n_____\n**[Four-Dimension Deep Learning Method for Flower Quality Grading with Depth Information](https://doi.org/10.3390/electronics10192353)**\n\n\n> **TL;DR:** Depth information was used to improve the accuracy of a convolutional neural network in classifying flower buds by their maturing status. The InceptionV3 network with RGBD input had the highest classification accuracy.\n\nGrading the quality of fresh cut flowers is an important practice in the flower industry. Based on the flower maturing status, a classification method based on deep learning and depth information was proposed for the grading of flower quality. Firstly, the RGB image and the depth image of a flower bud were collected and transformed into fused RGBD information. Then, the RGBD information of a flower was set as inputs of a convolutional neural network to determine the flower bud maturing status. Four convolutional neural network models (VGG16, ResNet18, MobileNetV2, and InceptionV3) were adjusted for a four-dimensional (4D) RGBD input to classify flowers, and their classification performances were compared with and without depth information. The experimental results show that the classification accuracy was improved with depth information, and the improved InceptionV3 network with RGBD achieved the highest classification accuracy (up to 98%), which means that the depth information can effectively reflect the characteristics of the flower bud and is helpful for the classification of the maturing status. These results have a certain significance for the intelligent classification and sorting of fresh flowers.\n_____\n\n\n_____\n**[A Survey on Deep Learning Approaches For Flower Species Detection](https://scholar.google.com/scholar?q=A%20Survey%20on%20Deep%20Learning%20Approaches%20For%20Flower%20Species%20Detection)**\n\nIn modern’s world , the data is generated on multi-folded and cross platform which can be used to understand and analyse different domains such as Floriculture , Agriculture , Finance etc. In all real time analysis of data where time , money , man-power is playing important roles to measure, pass judgement and react/answer in terms of Floriculture domain because the floriculture cultivation in world are increasing day by day The data generated from analysis which will be in the form of shape , colour , petals , size etc. Flower species and recognition system can provide unique approach for providing flower species analysis which could be a type of clutter of flowers , an area of farm where more flowers having issues like disease , water problem etc. In this process , Lots of data analyse , processed in real tome , provides good level of accuracy , precision , entropy etc. We can use deep learning approaches for detection flower species. This paper provides an overview of several pattern classification and detection mythologies/algorithms in the literature. The objective of the paper having companion with the comparison between different algorithms along with different types of datasets. The main goal is to provide an idea for several methods with different data and to find the many different approaches of the methods used for the detection of flowers using different scenarios\n_____\n\n\n_____\n**[Image classification of different clove (Syzygium aromaticum) quality using deep learning method with convolutional neural network algorithm](https://doi.org/10.1088/1755-1315/905/1/012018)**\n\n\n> **TL;DR:** We used a deep learning algorithm, a Convolutional Neural Network (CNN), to classify the quality of dried clove flowers. We found that the best model for this task was one that had five layers in its neural network, and that the best values for theCNN hyperparameters were those that produced a reading accuracy of 87.75%.\n\nThe objective of this study is to classify the quality of dried clove flowers using deep learning method with Convolutional Neural Network (CNN) algorithm, and also to perform the sensitivity analysis of CNN hyperparameters to obtain best model for clove quality classification process. The quality of clove as raw material in this study was determined according to SNI 3392-1994 by PT. Perkebunan Nusantara XII Pancusari Plantation, Malang, East Java, Indonesia. In total 1,600 images of dried clove flower were divided into 4 qualities. Each clove quality has 225 training data, 75 validation data, and 100 test data. The first step of this study is to build CNN model architecture as first model. The result of that model gives 65.25% reading accuracy. The second step is to analyze CNN sensitivity or CNN hyperparameter on the first model. The best value of CNN hyperparameter in each step then to be used in the next stage. Finally, after CNN hyperparameter carried out the reading accuracy of the test data is improved to 87.75%.\n_____\n\n\n_____\n**[Flower classification using CNN and transfer learning in CNN- Agriculture Perspective](https://doi.org/10.1109/ICISS49785.2020.9316030)**\n\nClassiücation of flowers is a difficult task because of the huge number of flowering plant species, which are similar in shape, color and appearance. A flower classification can be used in various applications such as field monitoring, plant identification, medicinal plant, floriculture industry, research in plant taxonomy. In this study, the authors have demonstrated and analyzed the recent developments in deep learning methods such as CNN and transfer learning in CNN. Prototype CNN model architecture proposed and transfer learning approach as well examined on VGG16, MobileNet2 and Resnet50 architecture for flower classification on publicly available flower dataset.\n_____\n\n\n_____\n**[Learning Salient Features for Flower Classification Using Convolutional Neural Network](https://doi.org/10.1109/ICAIIS49377.2020.9194931)**\n\nImage features can be utilized for image classification. Convolutional Neural Network (CNN) is able to extract features automatically from images rather than collecting features by hand. There are more and more flower images on the internet, and it is necessary to develop a system for identification of flower type. In this paper, we collect flower images from the internet and label them according to the species, then by using a deep CNN, we learn salient features of the flower images and achieve a significant performance of 78% in term of classification accuracy.\n_____\n\n\n_____\n**[Flower Identification and Classification applying CNN through Deep Learning Methodologies](https://doi.org/10.1109/MECON53876.2022.9752231)**\n\n\n> **TL;DR:** Convolutional neural networks play a significant role in the identification of flora species. Deep learning methodologies support us in image identification based on properties such as color and shape. Every species is distinct concerning attributes like texture, the shape of petals, and sepals. In this paper, we classify five various categories of flora named as daisy, dandelion, rose, sunflower, tulip. The methodology utilized here is the convolutional neural network where for instinctive recognition based on flora pictures. We focused on a neural network which implies an approach called skip connections and is flexible enough to perform various classification methodologies\n\nConvolutional neural networks play a significant role in the identification of flora species. Deep learning methodologies support us in image identification based on properties such as color and shape. Every species is distinct concerning attributes like texture, the shape of petals, and sepals. In this paper, we classify five various categories of flora named as daisy, dandelion, rose, sunflower, tulip. The methodology utilized here is the convolutional neural network where for instinctive recognition based on flora pictures. We focused on a neural network which implies an approach called skip connections and is flexible enough to perform various classification methodologies. We concentrated related to image classification by training and validating the dataset. The methodology adopted here is a residual neural network with nine deep layers, where implementation and approaches are discussed related to the classification model. We observed different existing algorithms and discussed the disadvantages, based on the drawbacks. We designed an algorithm for the classification and identification of a flower. The Experimental methodologies adopted are based on PyTorch and datasets. Finally, we presented the experimental results along with the graphical analysis. Keywords—Convolutional Neural Networks, Deep learning, Classification, Data mining, Artificial Intelligence.\n_____\n\n\n_____\n**[Flower Image Classification Using Deep Convolutional Neural Network](https://doi.org/10.1109/ICWR51868.2021.9443129)**\n\nThese days deep learning methods play a pivotal role in complicated tasks, such as extracting useful features, segmentation, and semantic classification of images. These methods had significant effects on flower types classification during recent years. In this paper, we are trying to classify 102 flower species using a robust deep learning method. To this end, we used the transfer learning approach employing DenseNet121 architecture to categorize various species of oxford-102 flowers dataset. In this regard, we have tried to fine-tune our model to achieve higher accuracy respect to other methods. We performed preprocessing by normalizing and resizing of our images and then fed them to our fine-tuned pretrained model. We divided our dataset to three sets of train, validation, and test. We could achieve the accuracy of 98.6% for 50 epochs which is better than other deep-learning based methods for the same dataset in the study.\n_____\n\n\n_____\n**[Flower classification using deep convolutional neural networks](https://doi.org/10.1049/iet-cvi.2017.0155)**\n\nFlower classification is a challenging task due to the wide range of flower species, which have a similar shape, appearance or surrounding objects such as leaves and grass. In this study, the authors propose a novel two-step deep learning classifier to distinguish flowers of a wide range of species. First, the flower region is automatically segmented to allow localisation of the minimum bounding box around it. The proposed flower segmentation approach is modelled as a binary classifier in a fully convolutional network framework. Second, they build a robust convolutional neural network classifier to distinguish the different flower types. They propose novel steps during the training stage to ensure robust, accurate and real-time classification. They evaluate their method on three well known flower datasets. Their classification results exceed 97% on all datasets, which are better than the state-of-the-art in this domain.\n_____\n\n\n_____\n**[A deep learning approach for the classification of diseased plant leaf images](https://doi.org/10.1109/ICCES45898.2019.9002201)**\n\n\n> **TL;DR:** A multilayer convolutional neural network is proposed for the classification of diseased plant leaf images. The real-time images of four different plants in healthy and diseased condition are collected for validating the performance of the proposed model. Results, when compared with other methods, shows the higher classification accuracy of the proposed model.\n\nPlant growth monitoring and plant protection are the key elements in the plant production industry. These factors influence the quality and productivity of the plant and its yields. Diseases are the major factor that vitiates the plant health. More often they harm plant parts like fruit, flower, leaf or stem, but quite often the severity of diseases may even result in plant death. In recent years, computer vision techniques, machine learning algorithms, and deep learning models have gained importance due to their capability of dealing with complex data with precision. These techniques are well known for pattern recognition and classification problems. Therefore in this work, a multilayer convolutional neural network is proposed for the classification of diseased plant leaf images. The real-time images of four different plants in healthy and diseased condition are collected for validating the performance of the proposed model. Results, when compared with other methods, shows the higher classification accuracy of the proposed model.\n_____\n\n\n_____\n**[Flower identification based on Deep Learning](https://doi.org/10.1088/1742-6596/1237/2/022060)**\n\n\n> **TL;DR:** builds a modified tiny darknet in flowers classification method to automatically extract the characteristics of flower images, then classifies and identifies flower test images.\n\nIn the field of plant scientific research, agroforestry investigation and production and management, plant identification is crucial basic work, and flower identification is an important part of plant identification. Given the present artificial defects of labor cost, low efficiency and low accuracy in present artificial flower information query and traditional computer vision method, the study built a modified tiny darknet in flowers classification method. Seventeen types of flower datasets published by Oxford University are taken as the research objects and the input of the neural network model. The deep network classification model is trained to automatically extract the characteristics of flower images. Combined with softmax classifier, the flower test images are classified and identified. The experimental results show that the classification accuracy is 92% which is higher than the classification algorithm results of the original model and some current mainstream models. This model has a simple structure, few training parameters, and has achieved a good recognition effect. It is suitable for automatic classification and recognition in the field of flower planting and is convenient for the retrieval of agricultural plant information database.\n_____\n\n\n_____\n**[Flower Data set Expansion Based on DCGAN and ResNet Classification Algorithm Based on Transfer Learning](https://doi.org/10.1109/ISCIPT53667.2021.00043)**\n\nWith the development of deep learning, flower classification has become a typical problem. The antagonistic generation network provides a method for data set expansion. At the same time, ResNet ensures that the depth of the neural network model can continue to deepen, while the training effect will not be worse than that of the shallow layer. In this paper, the two are combined. Firstly, the improved deep convolution antagonism generating network is used to expand the data set, and then the expanded data set is used in the ResNet model for training. In the experiment, by adjusting the parameters of DCGAN, performing multiple experiment comparisons to improve the performance of the model to generate pictures, finally, it is concluded that compared with the ResNet 101 trained on the original small data set, the classification accuracy of ResNet using the extended data set has been significantly improved.\n_____\n\n\n_____\n**[Flower image classification based on generative adversarial network and transfer learning](https://doi.org/10.1088/1755-1315/647/1/012180)**\n\n\n> **TL;DR:** This paper uses a generative adversarial network (GAN) with a residual network (ResNet-101) to improve the accuracy of flower classification.\n\nAiming at the problem that the classification accuracy of the traditional flower classification method is low and the deep neural network requires a large amount of original data. This paper designs a flower classification model that combines generative adversarial network and ResNet-101 transfer learning algorithm, and uses stochastic gradient descent algorithm to optimize the training process of the model. The experimental results on the the international public flower recognition dataset, Oxford flower-102 dataset, show that by enhancing the original data, the accuracy of the network's recognition and classification of flowers is improved. At the same time, the model proposed in this paper is superior to other traditional network models, with higher recognition accuracy and robustness.\n_____\n\n\n_____\n**[An Improved Image Classification Based In Feature Extraction From Convolutional Neural Network: Application To Flower Classification](https://doi.org/10.1109/IKT54664.2021.9685994)**\n\nNowadays, deep learning techniques are increasingly growing in machine vision for object recognition, segmentation, classification, and so on, in a wide variety of applications. In this study, we apply the convolutional neural network (CNN) to flower classification. For this purpose, we firstly increase the data with the augmentation techniques and use them in the pre-trained CNN models in which classification part is removed and instead of it, we use global average pooling (GAP) in the last layer for extracting their features. The features obtained from these models are concatenated, and then we use a support vector machine (SVM) as classifier for the flower classification. We use the Oxford 102 flower and the Oxford 17 flower datasets in our experiments. By applying this method, we achieve 96.47% classification accuracy for the Oxford 102 flower and 97.64% classification accuracy for the Oxford 17 flower. The results show the effectiveness of the proposed strategy and perform more accurate classification than the traditional methods.\n_____\n\n\n_____\n**[Impacts of Layer Sizes in Deep Residual-Learning Convolutional Neural Network on Flower Image Classification with Different class sizes](https://doi.org/10.1109/iEECON53204.2022.9741662)**\n\n\n> **TL;DR:** We evaluated the impacts of large- , medium- and small-size deep residual-learning convolutional neural networks (DRL-CNN): ResNet50, ResNet35 and ResNet17 models on classifying Oxford-102 flowers image dataset with distinct number of 10, 50 and 102 flower classes. The Flowers image classification assessments rely on precision, recall, F1 scores and accuracy rates, averaged over 10-fold cross validation to ensure unbiased experimented results. Confusion matrix is also considered for more detail of results examination. The comparison results indicate the ResNet35 yields 0.201% and 0.706% few better\n\nThis paper focuses on evaluating impacts of large -, medium - and small -size deep residual-learning convolutional neural network (DRL-CNN): ResNet50, ResNet35 and ResNet17 models on classifying Oxford-102 flowers image dataset with distinct number of 10, 50 and 102 flower classes. The Flowers image classification assessments rely on precision, recall, F1 scores and accuracy rates, averaged over 10-fold cross validation to ensure unbiased experimented results. Confusion matrix is also considered for more detail of results examination. The comparison results indicate the ResNet35 yields 0.201% and 0.706% few better recognition accuracy consecutively over ResNet50 and 17 according to 10-class dataset. For 50-class, 0.060% and 0.211% bits higher accuracy of ResNet35 than ResNet50 and ResNet17 are respectively generated. Whereas, 0.040% and 0.070% a few bits better performance of ResNet35 than ResNet50 and ResNet17 are sequentially attained on 100-class one. Decreasing rate regarding superiority of ResNet35 over ResNet50 and ResNet17 is indicated when increasing class size. However, less than 0.71% higher classification performance is indicated for ResNet35 than ResNet17 for all cases within the scope of this work. Thus, ResNet17 may be preferred to ResNet35 due to approximate 33% higher amount of parameters used in ResNet35 than ResNet17.\n_____\n\n\n\n_____\n**[A Comparative Analysis of Applying Object Detection Models with Transfer Learning for Flower Species Detection and Classification](https://scholar.google.com/scholar?q=A%20Comparative%20Analysis%20of%20Applying%20Object%20Detection%20Models%20with%20Transfer%20Learning%20for%20Flower%20Species%20Detection%20and%20Classification)**\n\n\n> **TL;DR:** We have performed a comparative analysis of the performance of various object detection models. We have compared; SSD quantized model (8-bit) using MobileNet V1 and MobileNet V2, Atrous model using Faster R-CNN with Inception ResNet V2, low proposals model using Faster R-CNN with ResNet 50 and NAS, atrous and low proposals models using Faster R-CNN with ResNet 101 and Inception ResNet V2 with the proposed NAS-FPN with modified Faster RCNN model.\n\nFlower species identification refers to a process of comparing defined characteristics of a given flower to allocate a particular species to a known taxonomic group. Flowers can be identified and classified by observing certain distinguishing basic and morphological characteristics. Classifying flower species is challenging for people and needs in-depth specialist knowledge as some flower species look similar, whereas some look differently despite of being in the same species. Traditional computer vision methods remain inefficient and less accurate while considering environmental complexity and similarity and difference between flowers species. Deep CNN, an emerging field of machine learning and artificial intelligence, has grown rapidly and widely applied in computer vision applications with promising results, especially in the field of object detection from visual images. In this paper, we have performed a comparative analysis of the performance of various object detection models. We have compared; SSD quantized model (8-bit) using MobileNet V1 and MobileNet V2, Atrous model using Faster R-CNN with Inception ResNet V2, low proposals model using Faster R-CNN with ResNet 50 and NAS, atrous and low proposals models using Faster R-CNN with ResNet 101 and Inception ResNet V2 with the proposed NAS-FPN with modified Faster RCNN model. Based on the results obtained during experiment, the proposed NAS-FPN with modified Faster R-CNN model achieved good performance and highest mAP score of 87.6% on F102 flower class and 96.2% on J30 flower class datasets.\n_____\n\n\n_____\n**[Automated color detection in orchids using color labels and deep learning](https://doi.org/10.1371/journal.pone.0259036)**\n\nThe color of particular parts of a flower is often employed as one of the features to differentiate between flower types. Thus, color is also used in flower-image classification. Color labels, such as ‘green’, ‘red’, and ‘yellow’, are used by taxonomists and lay people alike to describe the color of plants. Flower image datasets usually only consist of images and do not contain flower descriptions. In this research, we have built a flower-image dataset, especially regarding orchid species, which consists of human-friendly textual descriptions of features of specific flowers, on the one hand, and digital photographs indicating how a flower looks like, on the other hand. Using this dataset, a new automated color detection model was developed. It is the first research of its kind using color labels and deep learning for color detection in flower recognition. As deep learning often excels in pattern recognition in digital images, we applied transfer learning with various amounts of unfreezing of layers with five different neural network architectures (VGG16, Inception, Resnet50, Xception, Nasnet) to determine which architecture and which scheme of transfer learning performs best. In addition, various color scheme scenarios were tested, including the use of primary and secondary color together, and, in addition, the effectiveness of dealing with multi-class classification using multi-class, combined binary, and, finally, ensemble classifiers were studied. The best overall performance was achieved by the ensemble classifier. The results show that the proposed method can detect the color of flower and labellum very well without having to perform image segmentation. The result of this study can act as a foundation for the development of an image-based plant recognition system that is able to offer an explanation of a provided classification.\n_____\n\n", "2439662": "This is helpful, thanks for putting it together.", "2634966": "Thanks for your work" } }