Multiclass semantic segmentation annotation.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "c10c9d86",
   "metadata": {},
   "outputs": [],
   "source": [
    "from PIL import Image\n",
    "import numpy as np\n",
    "import os\n",
    "\n",
    "root_dir = 'E:\\Job\\IUB RA\\Aquatic plants Dataset\\Selected Frames\\garden pics\\garden_frame15'\n",
    "image_1 = Image.open(os.path.join(root_dir,\"bg.tif\"))   ##img1 is for Background\n",
    "image_2 = Image.open(os.path.join(root_dir,\"1.tif\"))\n",
    "image_3 = Image.open(os.path.join(root_dir,\"2.tif\"))\n",
    "image_4 = Image.open(os.path.join(root_dir,\"3.tif\"))\n",
    "image_5 = Image.open(os.path.join(root_dir,\"4.tif\"))\n",
    "\n",
    "# Convert images to NumPy arrays\n",
    "array_1 = np.array(image_1)\n",
    "array_2 = np.array(image_2)\n",
    "array_3 = np.array(image_3)\n",
    "array_4 = np.array(image_4)\n",
    "array_5 = np.array(image_5)\n",
    "\n",
    "#################################################################\n",
    "####### NEED TO SET IT MANUALLY FROM OUR DOCUMENTED LABEL #######\n",
    "#################################################################\n",
    "class_1_label = 1    ####label 1 is for Background ###\n",
    "class_2_label = 2\n",
    "class_3_label = 3\n",
    "class_4_label = 4\n",
    "class_5_label = 5\n",
    "\n",
    "multiclass_mask = np.ones_like(array_1)\n",
    "\n",
    "# Assign labels to different classes \n",
    "multiclass_mask[array_1 == 1] = class_1_label  # Change 1 to the value representing class 1 in image_1\n",
    "multiclass_mask[array_2 == 1] = class_2_label  \n",
    "multiclass_mask[array_3 == 1] = class_3_label  \n",
    "multiclass_mask[array_4 == 1] = class_4_label \n",
    "multiclass_mask[array_5 == 1] = class_5_label "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "d3ca8163",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "location = 'museum'  #garden\n",
    "num_of_frame = 15\n",
    "\n",
    "# Display the NumPy array as a grayscale image using Matplotlib\n",
    "plt.imshow(multiclass_mask, cmap='gray')\n",
    "plt.axis('off') \n",
    "plt.show()\n",
    "plt.imsave(f'{location}_frame{num_of_frame}.png', multiclass_mask, cmap='gray')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "fcc5ca76",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1, 2, 3, 4, 5], dtype=uint8)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.unique(multiclass_mask)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ec77bbe0",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}