draw_contours_and_merge_colors.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Reading image: input_images/vermeer.jpg\n",
      "Converting to grayscale\n",
      "Done\n",
      "Finding edges\n",
      "Drawing contours with thickness 1\n",
      "Saving image to: output_images/vermeer -- contours.png\n",
      "Reading image: input_images/unknown 1640.jpg\n",
      "Converting to grayscale\n",
      "Done\n",
      "Finding edges\n",
      "Drawing contours with thickness 1\n",
      "Saving image to: output_images/unknown 1640 -- contours.png\n",
      "Reading image: input_images/turner.webp\n",
      "Converting to grayscale\n",
      "Done\n",
      "Finding edges\n",
      "Drawing contours with thickness 1\n",
      "Saving image to: output_images/turner -- contours.png\n",
      "Reading image: input_images/leo putz_Halbindianerin mit Früchten.jpg\n",
      "Converting to grayscale\n",
      "Done\n",
      "Finding edges\n",
      "Drawing contours with thickness 1\n",
      "Saving image to: output_images/leo putz_Halbindianerin mit Früchten -- contours.png\n",
      "Reading image: input_images/cavalcanti_mulata against a green background.jpg\n",
      "Converting to grayscale\n",
      "Done\n",
      "Finding edges\n",
      "Drawing contours with thickness 1\n",
      "Saving image to: output_images/cavalcanti_mulata against a green background -- contours.png\n",
      "Reading image: input_images/bazille_family reunion.jpg\n",
      "Converting to grayscale\n",
      "Done\n",
      "Finding edges\n",
      "Drawing contours with thickness 1\n",
      "Saving image to: output_images/bazille_family reunion -- contours.png\n",
      "Reading image: input_images/alfred-sisley_autumn-banks-of-the-seine-near-bougival-1873.jpg\n",
      "Converting to grayscale\n",
      "Done\n",
      "Finding edges\n",
      "Drawing contours with thickness 1\n",
      "Saving image to: output_images/alfred-sisley_autumn-banks-of-the-seine-near-bougival-1873 -- contours.png\n",
      "Reading image: input_images/algernon-talmage_the-lake-district-for-holidays-honister-crag-london-midland-and-scottish-railway-poster-artwork.jpg\n",
      "Converting to grayscale\n",
      "Done\n",
      "Finding edges\n",
      "Drawing contours with thickness 1\n",
      "Saving image to: output_images/algernon-talmage_the-lake-district-for-holidays-honister-crag-london-midland-and-scottish-railway-poster-artwork -- contours.png\n",
      "Reading image: input_images/armand-guillaumin_echo-rock-1905.jpg\n",
      "Converting to grayscale\n",
      "Done\n",
      "Finding edges\n",
      "Drawing contours with thickness 1\n",
      "Saving image to: output_images/armand-guillaumin_echo-rock-1905 -- contours.png\n",
      "Reading image: input_images/benjamin-brown_eucalypti-near-arch-beach-california.jpg\n",
      "Converting to grayscale\n",
      "Done\n",
      "Finding edges\n",
      "Drawing contours with thickness 1\n",
      "Saving image to: output_images/benjamin-brown_eucalypti-near-arch-beach-california -- contours.png\n",
      "Reading image: input_images/benjamin-brown_grand-canyon.jpg\n",
      "Converting to grayscale\n",
      "Done\n",
      "Finding edges\n",
      "Drawing contours with thickness 1\n",
      "Saving image to: output_images/benjamin-brown_grand-canyon -- contours.png\n",
      "Reading image: input_images/benjamin-brown_poppies-and-eucalyptus.jpg\n",
      "Converting to grayscale\n",
      "Done\n",
      "Finding edges\n",
      "Drawing contours with thickness 1\n",
      "Saving image to: output_images/benjamin-brown_poppies-and-eucalyptus -- contours.png\n",
      "Reading image: input_images/cezanne_still life with apples_1895.jpg\n",
      "Converting to grayscale\n",
      "Done\n",
      "Finding edges\n",
      "Drawing contours with thickness 1\n",
      "Saving image to: output_images/cezanne_still life with apples_1895 -- contours.png\n",
      "Reading image: input_images/kahlo_viva la vida_1954.jpg\n",
      "Converting to grayscale\n",
      "Done\n",
      "Finding edges\n",
      "Drawing contours with thickness 1\n",
      "Saving image to: output_images/kahlo_viva la vida_1954 -- contours.png\n",
      "Reading image: input_images/Linda10 (1).jpg\n",
      "Converting to grayscale\n",
      "Done\n",
      "Finding edges\n",
      "Drawing contours with thickness 1\n",
      "Saving image to: output_images/Linda10 (1) -- contours.png\n",
      "Reading image: input_images/Ljus från lampa och omgivning.jpg\n",
      "Converting to grayscale\n",
      "Done\n",
      "Finding edges\n",
      "Drawing contours with thickness 1\n",
      "Saving image to: output_images/Ljus från lampa och omgivning -- contours.png\n",
      "Reading image: input_images/VAZQUEZ_Emmanuel_Honorato_01-scaled.jpg\n",
      "Converting to grayscale\n",
      "Done\n",
      "Finding edges\n",
      "Drawing contours with thickness 1\n",
      "Saving image to: output_images/VAZQUEZ_Emmanuel_Honorato_01-scaled -- contours.png\n",
      "Reading image: input_images/warhol_Capa do álbum Velvet Underground & Nico_1967.jpg\n",
      "Converting to grayscale\n",
      "Done\n",
      "Finding edges\n",
      "Drawing contours with thickness 1\n",
      "Saving image to: output_images/warhol_Capa do álbum Velvet Underground & Nico_1967 -- contours.png\n",
      "Reading image: input_images/warhol_flowers_1970.jpg\n",
      "Converting to grayscale\n",
      "Done\n",
      "Finding edges\n",
      "Drawing contours with thickness 1\n",
      "Saving image to: output_images/warhol_flowers_1970 -- contours.png\n",
      "Reading image: input_images/warhol_mao.jpg\n",
      "Converting to grayscale\n",
      "Done\n",
      "Finding edges\n",
      "Drawing contours with thickness 1\n",
      "Saving image to: output_images/warhol_mao -- contours.png\n",
      "Reading image: input_images/warhol_self portrait_1972.jpg\n",
      "Converting to grayscale\n",
      "Done\n",
      "Finding edges\n",
      "Drawing contours with thickness 1\n",
      "Saving image to: output_images/warhol_self portrait_1972 -- contours.png\n",
      "Reading image: input_images/warhol_sunset_1972.jpg\n",
      "Converting to grayscale\n",
      "Done\n",
      "Finding edges\n",
      "Drawing contours with thickness 1\n",
      "Saving image to: output_images/warhol_sunset_1972 -- contours.png\n",
      "Reading image: input_images/warhol_vesuvius_1985.jpg\n",
      "Converting to grayscale\n",
      "Done\n",
      "Finding edges\n",
      "Drawing contours with thickness 1\n",
      "Saving image to: output_images/warhol_vesuvius_1985 -- contours.png\n",
      "Reading image: input_images/kaggle dataset1.jpg\n",
      "Converting to grayscale\n",
      "Done\n",
      "Finding edges\n",
      "Drawing contours with thickness 1\n",
      "Saving image to: output_images/kaggle dataset1 -- contours.png\n",
      "Reading image: input_images/kaggle dataset2.jpg\n",
      "Converting to grayscale\n",
      "Done\n",
      "Finding edges\n",
      "Drawing contours with thickness 1\n",
      "Saving image to: output_images/kaggle dataset2 -- contours.png\n",
      "Reading image: input_images/kaggle dataset3.jpg\n",
      "Converting to grayscale\n",
      "Done\n",
      "Finding edges\n",
      "Drawing contours with thickness 1\n",
      "Saving image to: output_images/kaggle dataset3 -- contours.png\n",
      "Reading image: input_images/kaggle dataset4.jpg\n",
      "Converting to grayscale\n",
      "Done\n",
      "Finding edges\n",
      "Drawing contours with thickness 1\n",
      "Saving image to: output_images/kaggle dataset4 -- contours.png\n",
      "Reading image: input_images/kaggle dataset5.jpg\n",
      "Converting to grayscale\n",
      "Done\n",
      "Finding edges\n",
      "Drawing contours with thickness 1\n",
      "Saving image to: output_images/kaggle dataset5 -- contours.png\n"
     ]
    }
   ],
   "source": [
    "from skimage import data, io, segmentation, color, util, img_as_ubyte\n",
    "from skimage import graph\n",
    "import numpy as np\n",
    "import cv2 as cv\n",
    "import os\n",
    "from scipy.spatial import KDTree\n",
    "\n",
    "compactness = 5 #Higher values make the shape of the segments more square/cubic. Lower values make the shape of the segments more circular/spherical.\n",
    "n_segments = 2000 #Desired number of segments. Lower values result in larger segments, and higher values create more, smaller segments. Segments might be fewer than set value, but not more.\n",
    "thresh = 50 # Higher value results in fewer and larger segments. Lower values result in more and smaller segments.\n",
    "\n",
    "\n",
    "# parameters for contour detection\n",
    "contour_color = (255, 0, 0) # color of the contour lines\n",
    "t_lower = 50 # Lower Threshold\n",
    "t_upper = 130  # Upper threshold\n",
    "cont_thickness = 1 #Contour Thickness\n",
    "\n",
    "images = [{'image_name': 'vermeer', 'image_file_format': '.jpg', 'compactness': 5, 'n_segments': 2000, 'thresh': 50, 't_lower': 50, 't_upper': 130, 'cont_thickness': 1},\n",
    "          {'image_name': 'unknown 1640', 'image_file_format': '.jpg', 'compactness': 5, 'n_segments': 2000, 'thresh': 50, 't_lower': 50, 't_upper': 130, 'cont_thickness': 1},\n",
    "          {'image_name': 'turner', 'image_file_format': '.webp', 'compactness': 5, 'n_segments': 2000, 'thresh': 50, 't_lower': 50, 't_upper': 130, 'cont_thickness': 1},\n",
    "          {'image_name': 'leo putz_Halbindianerin mit Früchten', 'image_file_format': '.jpg', 'compactness': 5, 'n_segments': 2000, 'thresh': 50, 't_lower': 50, 't_upper': 130, 'cont_thickness': 1},\n",
    "          {'image_name': 'cavalcanti_mulata against a green background', 'image_file_format': '.jpg', 'compactness': 5, 'n_segments': 2000, 'thresh': 50, 't_lower': 50, 't_upper': 130, 'cont_thickness': 1},\n",
    "          {'image_name': 'bazille_family reunion', 'image_file_format': '.jpg', 'compactness': 5, 'n_segments': 2000, 'thresh': 50, 't_lower': 50, 't_upper': 130, 'cont_thickness': 1},\n",
    "          {'image_name': 'alfred-sisley_autumn-banks-of-the-seine-near-bougival-1873', 'image_file_format': '.jpg', 'compactness': 5, 'n_segments': 2000, 'thresh': 50, 't_lower': 50, 't_upper': 130, 'cont_thickness': 1},\n",
    "          {'image_name': 'algernon-talmage_the-lake-district-for-holidays-honister-crag-london-midland-and-scottish-railway-poster-artwork', 'image_file_format': '.jpg', 'compactness': 5, 'n_segments': 2000, 'thresh': 50, 't_lower': 50, 't_upper': 130, 'cont_thickness': 1},\n",
    "          {'image_name': 'armand-guillaumin_echo-rock-1905', 'image_file_format': '.jpg', 'compactness': 5, 'n_segments': 2000, 'thresh': 50, 't_lower': 50, 't_upper': 130, 'cont_thickness': 1},\n",
    "          {'image_name': 'benjamin-brown_eucalypti-near-arch-beach-california', 'image_file_format': '.jpg', 'compactness': 5, 'n_segments': 2000, 'thresh': 50, 't_lower': 50, 't_upper': 130, 'cont_thickness': 1},\n",
    "          {'image_name': 'benjamin-brown_grand-canyon', 'image_file_format': '.jpg', 'compactness': 5, 'n_segments': 2000, 'thresh': 50, 't_lower': 50, 't_upper': 130, 'cont_thickness': 1},\n",
    "          {'image_name': 'benjamin-brown_poppies-and-eucalyptus', 'image_file_format': '.jpg', 'compactness': 5, 'n_segments': 2000, 'thresh': 50, 't_lower': 50, 't_upper': 130, 'cont_thickness': 1},\n",
    "          {'image_name': 'cezanne_still life with apples_1895', 'image_file_format': '.jpg', 'compactness': 5, 'n_segments': 2000, 'thresh': 50, 't_lower': 50, 't_upper': 130, 'cont_thickness': 1},\n",
    "          {'image_name': 'kahlo_viva la vida_1954', 'image_file_format': '.jpg', 'compactness': 5, 'n_segments': 2000, 'thresh': 50, 't_lower': 50, 't_upper': 130, 'cont_thickness': 1},\n",
    "          {'image_name': 'Linda10 (1)', 'image_file_format': '.jpg', 'compactness': 5, 'n_segments': 2000, 'thresh': 50, 't_lower': 50, 't_upper': 130, 'cont_thickness': 1},\n",
    "          {'image_name': 'Ljus från lampa och omgivning', 'image_file_format': '.jpg', 'compactness': 5, 'n_segments': 2000, 'thresh': 50, 't_lower': 50, 't_upper': 130, 'cont_thickness': 1},\n",
    "          {'image_name': 'VAZQUEZ_Emmanuel_Honorato_01-scaled', 'image_file_format': '.jpg', 'compactness': 5, 'n_segments': 2000, 'thresh': 50, 't_lower': 50, 't_upper': 130, 'cont_thickness': 1},\n",
    "          {'image_name': 'warhol_Capa do álbum Velvet Underground & Nico_1967', 'image_file_format': '.jpg', 'compactness': 5, 'n_segments': 2000, 'thresh': 50, 't_lower': 50, 't_upper': 130, 'cont_thickness': 1},\n",
    "          {'image_name': 'warhol_flowers_1970', 'image_file_format': '.jpg', 'compactness': 5, 'n_segments': 2000, 'thresh': 50, 't_lower': 50, 't_upper': 130, 'cont_thickness': 1},\n",
    "          {'image_name': 'warhol_mao', 'image_file_format': '.jpg', 'compactness': 5, 'n_segments': 2000, 'thresh': 50, 't_lower': 50, 't_upper': 130, 'cont_thickness': 1},\n",
    "          {'image_name': 'warhol_self portrait_1972', 'image_file_format': '.jpg', 'compactness': 5, 'n_segments': 2000, 'thresh': 50, 't_lower': 50, 't_upper': 130, 'cont_thickness': 1},\n",
    "          {'image_name': 'warhol_sunset_1972', 'image_file_format': '.jpg', 'compactness': 5, 'n_segments': 2000, 'thresh': 50, 't_lower': 50, 't_upper': 130, 'cont_thickness': 1},\n",
    "          {'image_name': 'warhol_vesuvius_1985', 'image_file_format': '.jpg', 'compactness': 5, 'n_segments': 2000, 'thresh': 50, 't_lower': 50, 't_upper': 130, 'cont_thickness': 1},\n",
    "          {'image_name': 'kaggle dataset1', 'image_file_format': '.jpg', 'compactness': 5, 'n_segments': 2000, 'thresh': 50, 't_lower': 50, 't_upper': 130, 'cont_thickness': 1},\n",
    "          {'image_name': 'kaggle dataset2', 'image_file_format': '.jpg', 'compactness': 5, 'n_segments': 2000, 'thresh': 50, 't_lower': 50, 't_upper': 130, 'cont_thickness': 1},\n",
    "          {'image_name': 'kaggle dataset3', 'image_file_format': '.jpg', 'compactness': 5, 'n_segments': 2000, 'thresh': 50, 't_lower': 50, 't_upper': 130, 'cont_thickness': 1},\n",
    "          {'image_name': 'kaggle dataset4', 'image_file_format': '.jpg', 'compactness': 5, 'n_segments': 2000, 'thresh': 50, 't_lower': 50, 't_upper': 130, 'cont_thickness': 1},\n",
    "          {'image_name': 'kaggle dataset5', 'image_file_format': '.jpg', 'compactness': 5, 'n_segments': 2000, 'thresh': 50, 't_lower': 50, 't_upper': 130, 'cont_thickness': 1},   \n",
    "          ]\n",
    "\n",
    "predefined_colors = np.array([\n",
    "    [208, 72, 154], [236, 24, 121], [236, 28, 35], [250, 164, 23], [241, 235, 32], [154, 201, 59], [113, 193, 82], [111, 197, 164], [59, 184, 235], [67, 111, 182], [90, 80, 162], [134, 80, 160],\n",
    "    [214, 111, 171], [237, 95, 141], [240, 90, 65], [251, 181, 75], [244, 235, 98], [173, 211, 96], [141, 199, 116], [136, 209, 181], [109, 192, 236], [101, 133, 193], [117, 105, 174], [154, 108, 177],\n",
    "    [223, 144, 189], [243, 135, 166], [243, 134, 99], [254, 197, 119], [248, 239, 139], [191, 222, 135], [170, 213, 150], [171, 219, 196], [149, 207, 242], [134, 153, 205], [143, 134, 190], [171, 137, 190],\n",
    "    [230, 179, 209], [247, 174, 193], [248, 168, 141], [255, 215, 161], [249, 244, 176], [212, 228, 174], [196, 225, 181], [196, 228, 213], [183, 221, 244], [169, 182, 220], [175, 167, 210], [196, 171, 211],\n",
    "    [241, 214, 232], [250, 213, 221], [253, 210, 191], [254, 234, 204], [252, 249, 215], [231, 241, 212], [225, 239, 216], [226, 240, 232], [220, 236, 250], [209, 214, 236], [209, 206, 233], [222, 210, 229]\n",
    "  ])\n",
    "\n",
    "\n",
    "def closest_predefined_color(mean_color, predefined_colors):\n",
    "    tree = KDTree(predefined_colors)\n",
    "    _, idx = tree.query(mean_color)\n",
    "    return predefined_colors[idx]\n",
    "\n",
    "def _weight_mean_color(graph, src, dst, n):\n",
    "    \"\"\"Callback to handle merging nodes by recomputing mean color.\n",
    "\n",
    "    The method expects that the mean color of `dst` is already computed.\n",
    "\n",
    "    Parameters\n",
    "    ----------\n",
    "    graph : RAG\n",
    "        The graph under consideration.\n",
    "    src, dst : int\n",
    "        The vertices in `graph` to be merged.\n",
    "    n : int\n",
    "        A neighbor of `src` or `dst` or both.\n",
    "\n",
    "    Returns\n",
    "    -------\n",
    "    data : dict\n",
    "        A dictionary with the `\"weight\"` attribute set as the absolute\n",
    "        difference of the mean color between node `dst` and `n`.\n",
    "    \"\"\"\n",
    "\n",
    "    diff = graph.nodes[dst]['mean color'] - graph.nodes[n]['mean color']\n",
    "    diff = np.linalg.norm(diff)\n",
    "    return {'weight': diff}\n",
    "\n",
    "\n",
    "def merge_mean_color(graph, src, dst):\n",
    "    \"\"\"Callback called before merging two nodes of a mean color distance graph.\n",
    "\n",
    "    This method computes the mean color of `dst`.\n",
    "\n",
    "    Parameters\n",
    "    ----------\n",
    "    graph : RAG\n",
    "        The graph under consideration.\n",
    "    src, dst : int\n",
    "        The vertices in `graph` to be merged.\n",
    "    \"\"\"\n",
    "    graph.nodes[dst]['total color'] += graph.nodes[src]['total color']\n",
    "    graph.nodes[dst]['pixel count'] += graph.nodes[src]['pixel count']\n",
    "    graph.nodes[dst]['mean color'] = (graph.nodes[dst]['total color'] /\n",
    "                                      graph.nodes[dst]['pixel count'])\n",
    "\n",
    "\n",
    "# Define a function to make nearly black parts transparent\n",
    "def make_black_transparent(image_data, threshold=50):\n",
    "    # Extract the RGB channels\n",
    "    r, g, b, a = np.rollaxis(image_data, axis=-1)\n",
    "    # Create a mask for nearly black pixels\n",
    "    mask = (r < threshold) & (g < threshold) & (b < threshold)\n",
    "    # Set the alpha channel to 0 for nearly black pixels\n",
    "    image_data[mask, 3] = 0\n",
    "    return image_data\n",
    "\n",
    "\n",
    "# Convert an RGB color from 0-255 range to 0-1 range\n",
    "def convert_color_range(color):\n",
    "    return tuple([c / 255.0 for c in color])\n",
    "\n",
    "\n",
    "def get_contours(image): \n",
    "\n",
    "    cont_color = (0, 0, 255)  # BGR color for contour lines, red in this case\n",
    "\n",
    "    print(\"Reading image: \" + image)\n",
    "    im = cv.imread(image)\n",
    "    assert im is not None, \"File could not be read, check with os.path.exists()\"\n",
    "    print(\"Converting to grayscale\")\n",
    "    imgray = cv.cvtColor(im, cv.COLOR_BGR2GRAY)\n",
    "    print(\"Done\")\n",
    "    blur = cv.GaussianBlur(im, (3, 3), 0)\n",
    "\n",
    "    # Applying the Canny Edge filter\n",
    "    print(\"Finding edges\")\n",
    "    edge = cv.Canny(blur, t_lower, t_upper)\n",
    "    contours, hierarchy = cv.findContours(edge, cv.RETR_TREE, cv.CHAIN_APPROX_SIMPLE)\n",
    "\n",
    "    print(f\"Drawing contours with thickness {cont_thickness}\")\n",
    "    # Create a new black image of the same size as the original\n",
    "    black_image = np.zeros_like(im)\n",
    "\n",
    "    # Drawing the contours on the black image\n",
    "    cv.drawContours(black_image, contours, -1, cont_color, cont_thickness)\n",
    "\n",
    "    # Ensure the output directory exists\n",
    "    output_dir = 'output_images'\n",
    "    if not os.path.exists(output_dir):\n",
    "        os.makedirs(output_dir)\n",
    "    \n",
    "    # Saving the image with contours on black background\n",
    "    image_name = os.path.splitext(os.path.basename(image))[0]\n",
    "    full_save_path = os.path.join(output_dir, image_name + ' -- contours.png')\n",
    "    print(f\"Saving image to: {full_save_path}\")\n",
    "    cv.imwrite(full_save_path, black_image)\n",
    "\n",
    "    # Displaying the image for verification\n",
    "    # cv.imshow('Contours', black_image)\n",
    "    # print(\"Press any key to close the image window.\")\n",
    "    # cv.waitKey(0)  # Waits indefinitely for a key press\n",
    "    # cv.destroyAllWindows()  # Closes all the OpenCV windows\n",
    "\n",
    "for image in images:\n",
    "    image_path = 'input_images/'\n",
    "    #image_name = 'vermeer'\n",
    "    #image_file_format = '.jpg'\n",
    "    image_name = image['image_name']\n",
    "    image_file_format = image['image_file_format']\n",
    "    compactness = image['compactness']\n",
    "    n_segments = image['n_segments']\n",
    "    thresh = image['thresh']\n",
    "    t_lower = image['t_lower']\n",
    "    t_upper = image['t_upper']\n",
    "    cont_thickness = image['cont_thickness']\n",
    "\n",
    "    file_name = image_name + image_file_format\n",
    "    img = io.imread(image_path + file_name)\n",
    "    #img = data.coffee()\n",
    "    labels = segmentation.slic(img, compactness=compactness, n_segments=n_segments, start_label=1)\n",
    "\n",
    "    # Create the RAG\n",
    "    g = graph.rag_mean_color(img, labels)\n",
    "\n",
    "    # Perform hierarchical merging\n",
    "    labels2 = graph.merge_hierarchical(labels, g, thresh=thresh, rag_copy=False,\n",
    "                                    in_place_merge=True,\n",
    "                                    merge_func=merge_mean_color,\n",
    "                                    weight_func=_weight_mean_color)\n",
    "\n",
    "    # Map the final labels to colors\n",
    "    out_original_colors = color.label2rgb(labels2, img, kind='avg', bg_label=0)\n",
    "    io.imsave('output_images/' + image_name + ' -- color_merge.png', out_original_colors)\n",
    "\n",
    "    output_image = np.zeros_like(img, dtype=np.uint8)\n",
    "    for region in np.unique(labels2):\n",
    "        mask = labels2 == region\n",
    "        mean_color = img[mask].mean(axis=0)\n",
    "        closest_color = closest_predefined_color(mean_color, predefined_colors)\n",
    "        output_image[mask] = closest_color\n",
    "\n",
    "    # Save and display the output image\n",
    "    io.imsave('output_images/' + image_name + ' -- color_merge_forced_colors.png', output_image)\n",
    "    f = open('output_images/' + image_name + '_parameters.txt', \"w\")\n",
    "    f.write(\"compactness = \" + str(compactness) + \"\\n\" + \"n_segments = \" + str(n_segments) + \"\\n\" + \"thresh = \" + str(thresh) + \"\\n\" + \"t_lower = \" + str(t_lower) + \"\\n\" + \"t_upper = \" + str(t_upper) + \"\\n\" + \"cont_thickness = \" + str(cont_thickness) + \"\\n\")\n",
    "    f.close()\n",
    "\n",
    "    #io.imshow(output_image)\n",
    "    #io.show()\n",
    "\n",
    "\n",
    "    get_contours(image_path + file_name)\n",
    "\n",
    "    # BELOW IS THE OLD CONTOUR CODE\n",
    "    # Convert to 0-1 range for use with mark_boundaries\n",
    "    # contour_color_01 = convert_color_range(contour_color)\n",
    "\n",
    "    # # Creating an image with only the contours against a black background\n",
    "    # contours_only = np.zeros_like(img)\n",
    "    # contours_only = segmentation.mark_boundaries(contours_only, labels2, color=contour_color_01, outline_color=contour_color_01, mode='thick')\n",
    "    # contours_only = util.img_as_ubyte(contours_only)  # Convert the image to byte format\n",
    "\n",
    "    # # Display the contours-only image\n",
    "    # io.imshow(contours_only)\n",
    "    # io.show()\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "    from PIL import Image\n",
    "\n",
    "    # Load the images\n",
    "    edge_image_path = 'output_images/' + image_name + ' -- contours.png'\n",
    "    color_image_path = 'output_images/' + image_name + ' -- color_merge.png'\n",
    "    forced_colors_image_path = 'output_images/' + image_name + ' -- color_merge_forced_colors.png'\n",
    "\n",
    "    edge_image = Image.open(edge_image_path).convert(\"RGBA\")\n",
    "    color_image = Image.open(color_image_path).convert(\"RGBA\")\n",
    "    forced_colors_image = Image.open(forced_colors_image_path).convert(\"RGBA\")\n",
    "\n",
    "    # Convert the edge image to a numpy array\n",
    "    edge_data = np.array(edge_image)\n",
    "\n",
    "\n",
    "\n",
    "    # Make nearly black parts transparent\n",
    "    edge_data = make_black_transparent(edge_data)\n",
    "\n",
    "    # Convert back to an image\n",
    "    edge_image_transparent = Image.fromarray(edge_data)\n",
    "\n",
    "    # Combine the images\n",
    "    combined_image = Image.alpha_composite(color_image, edge_image_transparent)\n",
    "\n",
    "    # Save the combined image\n",
    "    combined_image_path = 'output_images/' + image_name + ' -- combined all colors.png'\n",
    "    combined_image.save(combined_image_path)\n",
    "\n",
    "    forced_combined_image = Image.alpha_composite(forced_colors_image, edge_image_transparent)\n",
    "\n",
    "    forced_colors_combined_image_path = 'output_images/' + image_name + ' -- combined forced colors.png'\n",
    "    forced_combined_image.save(forced_colors_combined_image_path)\n",
    "\n",
    "    # Display the combined image\n",
    "    #combined_image.show()\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "hapto",
   "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": "3.10.14"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}