awoo

Signed-off-by: Balazs Horvath <[email protected]>

dataset_tools/done/Convert RGBA to RGB in PNGs.ipynb

@@ -1,96 +0,0 @@
-{
- "cells": [
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "# Convert RGBA to RGB in PNGs\n",
-    "\n",
-    "---\n",
-    "\n",
-    "This Python script automates the process of converting `.png` images from RGBA to RGB format in a specified directory, utilizing multiprocessing to enhance efficiency."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "import os\n",
-    "from PIL import Image\n",
-    "import glob\n",
-    "import multiprocessing\n",
-    "\n",
-    "# Set the maximum number of pixels allowed in an image to prevent DecompressionBombWarning.\n",
-    "Image.MAX_IMAGE_PIXELS = 139211472\n",
-    "\n",
-    "def convert_rgba_to_rgb(image_path):\n",
-    "    \"\"\"\n",
-    "    Convert an RGBA image to RGB format.\n",
-    "\n",
-    "    This function opens an image from a given path and checks if it's in RGBA mode.\n",
-    "    If it is, the image is converted to RGB mode and saved back to the same path.\n",
-    "    Any errors encountered during processing are caught and printed.\n",
-    "\n",
-    "    Parameters:\n",
-    "    - image_path (str): The file path of the image to be converted.\n",
-    "\n",
-    "    Returns:\n",
-    "    None\n",
-    "    \"\"\"\n",
-    "    try:\n",
-    "        with Image.open(image_path) as image:\n",
-    "            if image.mode == 'RGBA':\n",
-    "                rgb_image = image.convert('RGB')\n",
-    "                rgb_image.save(image_path)\n",
-    "                print(f\"Converted {image_path} to RGB.\")\n",
-    "            else:\n",
-    "                print(f\"{image_path} is not an RGBA image.\")\n",
-    "    except Exception as e:\n",
-    "        print(f\"Error processing {image_path}: {e}\")\n",
-    "\n",
-    "def main():\n",
-    "    \"\"\"\n",
-    "    Main function to convert all RGBA images to RGB in a directory.\n",
-    "\n",
-    "    This function searches for all .png files in a specified directory and its subdirectories.\n",
-    "    It then creates a pool of processes equal to the number of available CPUs and uses them\n",
-    "    to convert each RGBA image to RGB format concurrently.\n",
-    "\n",
-    "    Returns:\n",
-    "    None\n",
-    "    \"\"\"\n",
-    "    directory = r'E:\\training_dir'\n",
-    "    # Get all .png files in the directory recursively\n",
-    "    files = glob.glob(os.path.join(directory, '**', '*.png'), recursive=True)\n",
-    "    \n",
-    "    # Determine the number of processes based on the available CPUs\n",
-    "    num_processes = multiprocessing.cpu_count()\n",
-    "\n",
-    "    # Create a pool of processes\n",
-    "    with multiprocessing.Pool(num_processes) as pool:\n",
-    "        # Map the convert_rgba_to_rgb function to the files\n",
-    "        pool.map(convert_rgba_to_rgb, files)\n",
-    "\n",
-    "    print(\"Conversion complete.\")\n",
-    "\n",
-    "if __name__ == \"__main__\":\n",
-    "    main()"
-   ]
-  }
- ],
- "metadata": {
-  "kernelspec": {
-   "display_name": "base",
-   "language": "python",
-   "name": "python3"
-  },
-  "language_info": {
-   "name": "python",
-   "version": "3.12.2"
-  }
- },
- "nbformat": 4,
- "nbformat_minor": 2
-}

dataset_tools/done/Replace Transparency with Black.ipynb

@@ -1,110 +0,0 @@
-{
- "cells": [
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "# Replace Transparency with Black\n",
-    "\n",
-    "----\n",
-    "\n",
-    "This Python script processes all `.png` images in a specified directory by adding a black layer to each, utilizing multiprocessing to handle the images in parallel for efficiency."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "import os\n",
-    "from PIL import Image\n",
-    "import glob\n",
-    "from multiprocessing import Pool\n",
-    "\n",
-    "def add_black_layer(image_path):\n",
-    "    \"\"\"\n",
-    "    Adds a black layer to the image at the given path and saves the modified image.\n",
-    "\n",
-    "    This function opens an image, converts it to 'RGBA' mode, creates a new black layer,\n",
-    "    pastes the original image onto the black layer, and saves the result back to the disk.\n",
-    "\n",
-    "    Parameters:\n",
-    "    image_path (str): The file path to the image to be processed.\n",
-    "\n",
-    "    Raises:\n",
-    "    Exception: If there is an error opening or processing the image.\n",
-    "    \"\"\"\n",
-    "    try:\n",
-    "        with Image.open(image_path) as img:\n",
-    "            # Ensure the image is in 'RGBA' mode to handle transparency\n",
-    "            img = img.convert('RGBA')\n",
-    "            black_layer = Image.new('RGBA', img.size, (0, 0, 0, 255))  # The fourth value is the alpha channel\n",
-    "            black_layer.paste(img, (0, 0), img)\n",
-    "            black_layer.save(image_path)\n",
-    "            print(f\"Black layer added to and overwritten {image_path}\")\n",
-    "    except Exception as e:\n",
-    "        print(f\"Error processing {image_path}: {e}\")\n",
-    "\n",
-    "def process_image(image_path):\n",
-    "    \"\"\"\n",
-    "    Processes a single image by adding a black layer.\n",
-    "\n",
-    "    This function is designed to be used with multiprocessing. It calls the 'add_black_layer'\n",
-    "    function and handles any exceptions that occur.\n",
-    "\n",
-    "    Parameters:\n",
-    "    image_path (str): The file path to the image to be processed.\n",
-    "    \"\"\"\n",
-    "    try:\n",
-    "        add_black_layer(image_path)\n",
-    "        print(f\"Black layer added to and overwritten {image_path}\")\n",
-    "    except Exception as e:\n",
-    "        print(f\"Error processing {image_path}: {e}\")\n",
-    "\n",
-    "def process_directory(directory):\n",
-    "    \"\"\"\n",
-    "    Processes all .png images in a directory by adding a black layer to each.\n",
-    "\n",
-    "    This function finds all .png images within the specified directory (including subdirectories),\n",
-    "    then creates a pool of worker processes to process each image in parallel.\n",
-    "\n",
-    "    Parameters:\n",
-    "    directory (str): The directory path where the .png images are located.\n",
-    "    \"\"\"\n",
-    "    # Get a list of all .png images in the directory recursively\n",
-    "    image_paths = glob.glob(os.path.join(directory, '**', '*.png'), recursive=True)\n",
-    "    \n",
-    "    # Create a pool of workers equal to the number of CPU cores\n",
-    "    with Pool() as pool:\n",
-    "        # Map the process_image function to the list of image paths\n",
-    "        pool.map(process_image, image_paths)\n",
-    "\n",
-    "if __name__ == \"__main__\":\n",
-    "    directory = r'E:\\training_dir'\n",
-    "    process_directory(directory)"
-   ]
-  }
- ],
- "metadata": {
-  "kernelspec": {
-   "display_name": "base",
-   "language": "python",
-   "name": "python3"
-  },
-  "language_info": {
-   "codemirror_mode": {
-    "name": "ipython",
-    "version": 3
-   },
-   "file_extension": ".py",
-   "mimetype": "text/x-python",
-   "name": "python",
-   "nbconvert_exporter": "python",
-   "pygments_lexer": "ipython3",
-   "version": "3.12.2"
-  }
- },
- "nbformat": 4,
- "nbformat_minor": 2
-}

dataset_tools/done/convert_rgba_to_rgb_in_pngs.py

@@ -0,0 +1,53 @@
+import os
+import glob
+import multiprocessing
+from PIL import Image
+
+# Set the maximum number of pixels allowed in an image to prevent DecompressionBombWarning.
+Image.MAX_IMAGE_PIXELS = 139211472
+
+
+def convert_rgba_to_rgb(image_path):
+    """
+    Convert an RGBA image to RGB format.
+    """
+    try:
+        print(f"Opening image: {image_path}")
+        with Image.open(image_path) as image:
+            print(f"Image mode: {image.mode}")
+            if image.mode == "RGBA":
+                rgb_image = image.convert("RGB")
+                rgb_image.save(image_path)
+                print(f"Converted {image_path} to RGB.")
+            else:
+                print(f"{image_path} is not an RGBA image.")
+    except Exception as e:
+        print(f"Error processing {image_path}: {e}")
+
+
+def main():
+    """
+    Main function to convert all RGBA images to RGB in a directory.
+    """
+    directory = r"E:\training_dir"
+    print(f"Directory set to: {directory}")
+
+    # Get all .png files in the directory recursively
+    files = glob.glob(os.path.join(directory, "**", "*.png"), recursive=True)
+    print(f"Found {len(files)} .png files in directory and subdirectories.")
+
+    # Determine the number of processes based on the available CPUs
+    num_processes = multiprocessing.cpu_count()
+    print(f"Number of processes set to the number of available CPUs: {num_processes}")
+
+    # Create a pool of processes
+    with multiprocessing.Pool(num_processes) as pool:
+        print("Pool of processes created.")
+        # Map the convert_rgba_to_rgb function to the files
+        pool.map(convert_rgba_to_rgb, files)
+
+    print("Conversion complete.")
+
+
+if __name__ == "__main__":
+    main()

dataset_tools/done/replace_transparency_with_black.py

@@ -0,0 +1,69 @@
+import os
+import glob
+from multiprocessing import Pool
+from PIL import Image
+
+
+def add_black_layer(image_path):
+    """
+    Adds a black layer to the image at the given path and saves the modified image.
+
+    This function opens an image, converts it to 'RGBA' mode, creates a new black layer,
+    pastes the original image onto the black layer, and saves the result back to the disk.
+
+    Parameters:
+    image_path (str): The file path to the image to be processed.
+
+    Raises:
+    Exception: If there is an error opening or processing the image.
+    """
+    print(f"Processing {image_path}...")
+    try:
+        with Image.open(image_path) as img:
+            img = img.convert("RGBA")
+            black_layer = Image.new("RGBA", img.size, (0, 0, 0, 255))
+            black_layer.paste(img, (0, 0), img)
+            black_layer.save(image_path)
+            print(f"Black layer added to {image_path}")
+    except Exception as e:
+        print(f"Error processing {image_path}: {e}")
+        raise
+
+
+def process_image(image_path):
+    """
+    Processes a single image by adding a black layer.
+
+    This function is designed to be used with multiprocessing. It calls the 'add_black_layer'
+    function and handles any exceptions that occur.
+
+    Parameters:
+    image_path (str): The file path to the image to be processed.
+    """
+    try:
+        add_black_layer(image_path)
+        print(f"Black layer added to and overwritten {image_path}")
+    except Exception as e:
+        print(f"Error processing {image_path}: {e}")
+
+
+def process_directory(directory):
+    """
+    Processes all .png images in a directory by adding a black layer to each.
+
+    This function finds all .png images within the specified directory (including subdirectories),
+    then creates a pool of worker processes to process each image in parallel.
+
+    Parameters:
+    directory (str): The directory path where the .png images are located.
+    """
+    image_paths = glob.glob(os.path.join(directory, "**", "*.png"), recursive=True)
+    print(f"Found {len(image_paths)} images to process.")
+    with Pool() as pool:
+        pool.map(add_black_layer, image_paths)
+
+
+if __name__ == "__main__":
+    directory = r"E:\training_dir"
+    print(f"Starting processing of images in {directory}")
+    process_directory(directory)