{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.11.13","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"none","dataSources":[{"sourceId":113558,"databundleVersionId":14174843,"sourceType":"competition"}],"dockerImageVersionId":31153,"isInternetEnabled":false,"language":"python","sourceType":"notebook","isGpuEnabled":false}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"import os\nimport json\nimport numpy as np\nimport pandas as pd\nimport matplotlib.pyplot as plt\n\nfrom PIL import Image","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-01T02:11:43.251354Z","iopub.execute_input":"2025-11-01T02:11:43.251733Z","iopub.status.idle":"2025-11-01T02:11:43.636483Z","shell.execute_reply.started":"2025-11-01T02:11:43.251700Z","shell.execute_reply":"2025-11-01T02:11:43.635430Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"def rle_encode(mask, fg_val=1):\n \"\"\"\n Convert binary mask to RLE using the competition metric format\n \"\"\"\n dots = np.where(mask.T.flatten() == fg_val)[0]\n run_lengths = []\n prev = -2\n \n for b in dots:\n if b > prev + 1:\n run_lengths.extend((b + 1, 0))\n run_lengths[-1] += 1\n prev = b\n \n return run_lengths","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-01T02:11:45.596787Z","iopub.execute_input":"2025-11-01T02:11:45.597382Z","iopub.status.idle":"2025-11-01T02:11:45.604443Z","shell.execute_reply.started":"2025-11-01T02:11:45.597345Z","shell.execute_reply":"2025-11-01T02:11:45.603561Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"def visualize_mask(mask, title):\n \"\"\"Visualize mask\"\"\"\n plt.figure(figsize=(6, 6))\n plt.imshow(mask, cmap='gray', vmin=0, vmax=1)\n plt.title(title)\n plt.axis('off')\n \n # Add grid for clarity\n for i in range(mask.shape[0] + 1):\n plt.axhline(i - 0.5, color='red', alpha=0.3, linewidth=0.5)\n plt.axvline(i - 0.5, color='red', alpha=0.3, linewidth=0.5)\n \n # Show pixel values\n for i in range(mask.shape[0]):\n for j in range(mask.shape[1]):\n plt.text(j, i, str(mask[i, j]), ha='center', va='center', \n color='blue' if mask[i, j] == 0 else 'white', fontweight='bold')\n \n plt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-01T02:11:46.686592Z","iopub.execute_input":"2025-11-01T02:11:46.687202Z","iopub.status.idle":"2025-11-01T02:11:46.698054Z","shell.execute_reply.started":"2025-11-01T02:11:46.687141Z","shell.execute_reply":"2025-11-01T02:11:46.696063Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"our_example = np.array([\n [1, 0],\n [1, 1]\n])\n\nprint(f'Our example: {our_example}')\nprint(f\"\\nRLE encoding: {rle_encode(our_example)}\")\nvisualize_mask(our_example, \"Our mask\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-01T02:11:47.052254Z","iopub.execute_input":"2025-11-01T02:11:47.052619Z","iopub.status.idle":"2025-11-01T02:11:47.311998Z","shell.execute_reply.started":"2025-11-01T02:11:47.052593Z","shell.execute_reply":"2025-11-01T02:11:47.310929Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Create PLUS mask (9x9)\nplus_mask = np.zeros((9, 9), dtype=np.uint8)\n# Vertical line\nplus_mask[2:7, 4] = 1\n# Horizontal line \nplus_mask[4, 2:7] = 1\n\nprint(plus_mask)\nprint(f\"\\nRLE encoding: {rle_encode(plus_mask)}\")\nvisualize_mask(plus_mask, \"Plus - a mask for segmentation\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-01T02:11:47.435937Z","iopub.execute_input":"2025-11-01T02:11:47.436263Z","iopub.status.idle":"2025-11-01T02:11:47.709438Z","shell.execute_reply.started":"2025-11-01T02:11:47.436242Z","shell.execute_reply":"2025-11-01T02:11:47.707841Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Create MINUS mask (9x9)\nminus_mask = np.zeros((9, 9), dtype=np.uint8)\n# Horizontal line\nminus_mask[4, 2:7] = 1\n\nprint(minus_mask)\nprint(f\"\\nRLE encoding: {rle_encode(minus_mask)}\")\nvisualize_mask(minus_mask, \"Minus - segmentation mask\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-01T02:11:47.792599Z","iopub.execute_input":"2025-11-01T02:11:47.792931Z","iopub.status.idle":"2025-11-01T02:11:48.121369Z","shell.execute_reply.started":"2025-11-01T02:11:47.792907Z","shell.execute_reply":"2025-11-01T02:11:48.120420Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Detailed RLE explanation for plus\nprint(\"Plus mask (9x9):\")\nfor i in range(9):\n row = ''\n for j in range(9):\n row += f\"{plus_mask[i, j]} \"\n print(row)\n\nprint(f\"\\n1. Flatten to string:\")\nflat_plus = plus_mask.flatten()\nprint(' '.join(map(str, flat_plus)))\n\nprint(f\"\\n2. Split into sequences:\")\n# Add zeros at borders for correct boundary detection\npadded = np.concatenate([[0], flat_plus, [0]])\nchanges = np.where(padded[1:] != padded[:-1])[0] + 1\nruns = changes.copy()\nruns[1::2] -= runs[::2]\n\nprint(f\"Change positions: {changes}\")\nprint(f\"Sequence lengths: {runs}\")\n\nprint(f\"\\n3. Final RLE: '{rle_encode(plus_mask)}'\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-01T02:11:51.247716Z","iopub.execute_input":"2025-11-01T02:11:51.248081Z","iopub.status.idle":"2025-11-01T02:11:51.257743Z","shell.execute_reply.started":"2025-11-01T02:11:51.248054Z","shell.execute_reply":"2025-11-01T02:11:51.256485Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"
\n\n
\n\n
\n\n

\n Create sumission distributed by the most frequent position in the mask\n

","metadata":{}},{"cell_type":"code","source":"np.random.seed(71)\n\ndef mask_distribution():\n all_norm_positions = []\n heatmap_size = (100, 100)\n heatmap = np.zeros(heatmap_size, dtype=np.float32)\n \n train_masks_dir = '/kaggle/input/recodai-luc-scientific-image-forgery-detection/train_masks'\n \n if not os.path.exists(train_masks_dir):\n return (0.5, 0.5), None\n \n for mask_file in os.listdir(train_masks_dir):\n if mask_file.endswith('.npy'):\n mask_path = os.path.join(train_masks_dir, mask_file)\n try:\n mask = np.load(mask_path)\n \n if mask.ndim == 3:\n if mask.shape[0] == 1:\n mask = mask[0]\n elif mask.shape[2] == 1:\n mask = mask[:, :, 0]\n else:\n mask = (mask == 1).astype(np.uint8)\n if mask.ndim == 3:\n mask = mask[:, :, 0] if mask.shape[2] == 1 else mask[:, :, 0]\n \n if mask.ndim != 2:\n continue\n \n y_coords, x_coords = np.where(mask > 0)\n \n if len(y_coords) > 0:\n height, width = mask.shape\n \n for y, x in zip(y_coords, x_coords):\n norm_y = y / height\n norm_x = x / width\n \n heatmap_y = int(norm_y * heatmap_size[0])\n heatmap_x = int(norm_x * heatmap_size[1])\n \n heatmap_y = min(heatmap_y, heatmap_size[0] - 1)\n heatmap_x = min(heatmap_x, heatmap_size[1] - 1)\n \n heatmap[heatmap_y, heatmap_x] += 1\n all_norm_positions.append((norm_x, norm_y))\n \n except Exception as e:\n continue\n \n if all_norm_positions:\n max_heatmap_pos = np.unravel_index(np.argmax(heatmap), heatmap.shape)\n max_norm_y = max_heatmap_pos[0] / heatmap_size[0]\n max_norm_x = max_heatmap_pos[1] / heatmap_size[1]\n return (max_norm_x, max_norm_y), heatmap\n \n return (0.5, 0.5), heatmap\n\nhottest_norm_pos, heatmap = mask_distribution()\n\nif heatmap is not None:\n plt.figure(figsize=(10, 8))\n plt.imshow(heatmap, cmap='hot', interpolation='nearest')\n plt.colorbar()\n plt.title('Forgery Location Heatmap')\n plt.xlabel('Normalized X')\n plt.ylabel('Normalized Y')\n plt.show()\n\ntest_images_dir = '/kaggle/input/recodai-luc-scientific-image-forgery-detection/test_images'\nsample_submission = pd.read_csv('/kaggle/input/recodai-luc-scientific-image-forgery-detection/sample_submission.csv')","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-01T02:12:06.327883Z","iopub.execute_input":"2025-11-01T02:12:06.328233Z","iopub.status.idle":"2025-11-01T02:18:44.706337Z","shell.execute_reply.started":"2025-11-01T02:12:06.328202Z","shell.execute_reply":"2025-11-01T02:18:44.705167Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"submission_data = []\nfor case_id in sample_submission['case_id']:\n img_path = os.path.join(test_images_dir, f\"{case_id}.png\")\n \n with Image.open(img_path) as img:\n width, height = img.size\n \n if np.random.random() < 0.01:\n mask = np.zeros((height, width), dtype=np.uint8)\n \n offset_x = np.random.uniform(-0.3, 0.3) * width\n offset_y = np.random.uniform(-0.3, 0.3) * height\n \n center_x = int(hottest_norm_pos[0] * width + offset_x)\n center_y = int(hottest_norm_pos[1] * height + offset_y)\n \n h = 4\n w = 4\n \n y0 = max(0, center_y - h//2)\n x0 = max(0, center_x - w//2)\n y1 = min(height, y0 + h)\n x1 = min(width, x0 + w)\n \n actual_h = y1 - y0\n actual_w = x1 - x0\n \n if actual_h > 0 and actual_w > 0:\n mask[y0:y1, x0:x1] = 1\n \n RLE_res = rle_encode(mask)\n res = [int(x) for x in RLE_res]\n annotation = json.dumps(res)\n else:\n annotation = 'authentic'\n \n submission_data.append({\n 'case_id': case_id,\n 'annotation': annotation\n })\n\nsubmission = pd.DataFrame(submission_data)\nsubmission.to_csv('submission.csv', index=False)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-01T02:18:44.777655Z","iopub.execute_input":"2025-11-01T02:18:44.778024Z","iopub.status.idle":"2025-11-01T02:18:44.791840Z","shell.execute_reply.started":"2025-11-01T02:18:44.777994Z","shell.execute_reply":"2025-11-01T02:18:44.790869Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import os\nimport json\nimport numpy as np\nimport pandas as pd\nimport matplotlib.pyplot as plt\nfrom PIL import Image\n\n# Example setup\n# (Make sure you have these defined)\n# sample_submission = pd.read_csv(\"sample_submission.csv\")\n# test_images_dir = \"path/to/test/images\"\n# hottest_norm_pos = (0.5, 0.5) # Example center point (normalized 0–1)\n\ndef rle_encode(mask):\n pixels = mask.flatten()\n pixels = np.concatenate([[0], pixels, [0]])\n runs = np.where(pixels[1:] != pixels[:-1])[0] + 1\n runs[1::2] -= runs[::2]\n return runs\n\nsubmission_data = []\n\nfor case_id in sample_submission['case_id']:\n img_path = os.path.join(test_images_dir, f\"{case_id}.png\")\n\n with Image.open(img_path) as img:\n width, height = img.size\n img_array = np.array(img)\n\n if np.random.random() < 0.01:\n mask = np.zeros((height, width), dtype=np.uint8)\n offset_x = np.random.uniform(-0.3, 0.3) * width\n offset_y = np.random.uniform(-0.3, 0.3) * height\n center_x = int(hottest_norm_pos[0] * width + offset_x)\n center_y = int(hottest_norm_pos[1] * height + offset_y)\n h = 4\n w = 4\n\n y0 = max(0, center_y - h//2)\n x0 = max(0, center_x - w//2)\n y1 = min(height, y0 + h)\n x1 = min(width, x0 + w)\n actual_h = y1 - y0\n actual_w = x1 - x0\n\n if actual_h > 0 and actual_w > 0:\n mask[y0:y1, x0:x1] = 1\n RLE_res = rle_encode(mask)\n res = [int(x) for x in RLE_res]\n annotation = json.dumps(res)\n\n # 🔹 PLOT: visualize the image and mask overlay\n plt.figure(figsize=(5, 5))\n plt.imshow(img_array, cmap='gray')\n plt.imshow(mask, alpha=0.4, cmap='Reds') # overlay in red\n plt.title(f\"Mask overlay for {case_id}\")\n plt.axis('off')\n plt.show()\n\n else:\n annotation = 'authentic'\n else:\n annotation = 'authentic'\n\n submission_data.append({\n 'case_id': case_id,\n 'annotation': annotation\n })\n\nsubmission = pd.DataFrame(submission_data)\nsubmission.to_csv('submission.csv', index=False)\n\nprint(\"✅ Submission file saved as 'submission.csv'\")\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-01T02:18:44.708343Z","iopub.execute_input":"2025-11-01T02:18:44.708680Z","iopub.status.idle":"2025-11-01T02:18:44.776660Z","shell.execute_reply.started":"2025-11-01T02:18:44.708659Z","shell.execute_reply":"2025-11-01T02:18:44.775432Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import os\nimport json\nimport numpy as np\nimport pandas as pd\nimport matplotlib.pyplot as plt\nfrom PIL import Image\nimport matplotlib.patches as patches\n\n# --------------------------\n# SETUP: define your paths\n# --------------------------\n# sample_submission = pd.read_csv(\"sample_submission.csv\") # your submission template\n# test_images_dir = \"path/to/test/images\" # folder with test images\n# hottest_norm_pos = (0.5, 0.5) # normalized center (example)\n\n# Make folder to save plots\nos.makedirs(\"plots\", exist_ok=True)\n\n# --------------------------\n# RLE Encode function\n# --------------------------\ndef rle_encode(mask):\n pixels = mask.flatten()\n pixels = np.concatenate([[0], pixels, [0]])\n runs = np.where(pixels[1:] != pixels[:-1])[0] + 1\n runs[1::2] -= runs[::2]\n return runs\n\n# --------------------------\n# MAIN LOOP\n# --------------------------\nsubmission_data = []\n\nfor case_id in sample_submission['case_id']:\n img_path = os.path.join(test_images_dir, f\"{case_id}.png\")\n\n with Image.open(img_path) as img:\n width, height = img.size\n img_array = np.array(img)\n\n if np.random.random() < 0.01:\n mask = np.zeros((height, width), dtype=np.uint8)\n offset_x = np.random.uniform(-0.3, 0.3) * width\n offset_y = np.random.uniform(-0.3, 0.3) * height\n center_x = int(hottest_norm_pos[0] * width + offset_x)\n center_y = int(hottest_norm_pos[1] * height + offset_y)\n h = 4\n w = 4\n\n y0 = max(0, center_y - h//2)\n x0 = max(0, center_x - w//2)\n y1 = min(height, y0 + h)\n x1 = min(width, x0 + w)\n actual_h = y1 - y0\n actual_w = x1 - x0\n\n if actual_h > 0 and actual_w > 0:\n mask[y0:y1, x0:x1] = 1\n RLE_res = rle_encode(mask)\n res = [int(x) for x in RLE_res]\n annotation = json.dumps(res)\n\n # --------------------------\n # PLOT: Image + Mask + Bounding Box\n # --------------------------\n fig, ax = plt.subplots(figsize=(6, 6))\n ax.imshow(img_array, cmap='gray')\n ax.imshow(mask, alpha=0.4, cmap='Reds')\n rect = patches.Rectangle((x0, y0), actual_w, actual_h,\n linewidth=2, edgecolor='cyan', facecolor='none')\n ax.add_patch(rect)\n ax.set_title(f\"Mask Overlay for {case_id}\")\n ax.axis('off')\n\n # Save plot\n plt.savefig(f\"plots/{case_id}_overlay.png\", bbox_inches='tight')\n plt.close()\n\n else:\n annotation = 'authentic'\n else:\n annotation = 'authentic'\n\n submission_data.append({\n 'case_id': case_id,\n 'annotation': annotation\n })\n\n# --------------------------\n# SAVE FINAL SUBMISSION\n# --------------------------\nsubmission = pd.DataFrame(submission_data)\nsubmission.to_csv('submission.csv', index=False)\nprint(\"✅ Submission file saved as 'submission.csv'\")\nprint(\"✅ Mask overlay plots saved in 'plots/' folder\")\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-01T02:18:44.793833Z","iopub.execute_input":"2025-11-01T02:18:44.794137Z","iopub.status.idle":"2025-11-01T02:18:44.852600Z","shell.execute_reply.started":"2025-11-01T02:18:44.794114Z","shell.execute_reply":"2025-11-01T02:18:44.851167Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"","metadata":{"trusted":true},"outputs":[],"execution_count":null}]}