Preprocessing.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# importing required modules \n",
    "from zipfile import ZipFile \n",
    "import datetime\n",
    "import os\n",
    "import shutil\n",
    "\n",
    "file_name='IDD_Segmentation'\n",
    "json_file_path='gtFine'\n",
    "image_path='leftImg8bit'\n",
    "temp_folder='All_data/New_data'\n",
    "\n",
    "curret_label_path= os.path.join(file_name,json_file_path,'train')\n",
    "curret_image_path= os.path.join(file_name,image_path,'train')\n",
    "files_name=[]\n",
    "\n",
    "\n",
    "for files in os.listdir(curret_image_path):\n",
    "    list_file=os.path.join(curret_image_path,files)\n",
    "    for file in os.listdir(list_file):\n",
    "        new_file_name=str(file.split('_')[0])\n",
    "        files_name.append(new_file_name)\n",
    "        json_name=new_file_name+'_gtFine_polygons.json'\n",
    "        #print(os.path.join(curret_label_path,files,json_name))\n",
    "        shutil.copy(os.path.join(curret_image_path,files,file),  temp_folder) \n",
    "        shutil.copy(os.path.join(curret_label_path,files,json_name), temp_folder)\n",
    "    \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open('file_names.txt', 'w') as filehandle:\n",
    "    for listitem in files_name:\n",
    "        filehandle.write('%s\\n' % listitem)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "files_name = []\n",
    "\n",
    "# open file and read the content in a list\n",
    "with open('file_names.txt', 'r') as filehandle:\n",
    "    for line in filehandle:\n",
    "        # remove linebreak which is the last character of the string\n",
    "        currentPlace = line[:-1]\n",
    "\n",
    "        # add item to the list\n",
    "        files_name.append(currentPlace)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "6993"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(files_name)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "train-70 --4895\n",
    "val-10 --700\n",
    "test-20 --1398"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import json\n",
    "import cv2\n",
    "from zipfile import ZipFile \n",
    "import datetime\n",
    "import os\n",
    "import shutil\n",
    "\n",
    "with open('label_txt.txt') as file:\n",
    "    label_to_color_map = json.load(file)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "source_path=os.path.join('All_data','New_data') \n",
    "\n",
    "def creat_lables(input_path,output_path,image_no):\n",
    "    get_json_file_path =os.path.join(source_path,image_no+'_gtFine_polygons.json') \n",
    "    get_image_file_path=os.path.join(source_path,image_no+'_leftImg8bit.png') \n",
    "    \n",
    "    \n",
    "    \n",
    "    with open(get_json_file_path) as json_file:\n",
    "        \n",
    "        json_data = json.load(json_file)\n",
    "       \n",
    "        cur_img = cv2.imread(get_image_file_path,cv2.IMREAD_UNCHANGED)\n",
    "        \n",
    "        for i, obj in enumerate(json_data['objects']):\n",
    "            obj_label = obj['label']\n",
    "            color_tuple = tuple(label_to_color_map[obj_label])\n",
    "            cv2.fillPoly(cur_img,[np.array(obj['polygon']).astype(np.int32)],color=color_tuple)\n",
    "            \n",
    " \n",
    "        cv2.imwrite(os.path.join(output_path,image_no+'_leftImg8bit.png'), cur_img)\n",
    "        shutil.copy(get_image_file_path,  input_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "for image_no in range(len(files_name)):\n",
    "    \n",
    "    if image_no >=0 and image_no<= 4895: #train\n",
    "        input_path =os.path.join('datasets_all','input','train') \n",
    "        output_path=os.path.join('datasets_all','output','train')\n",
    "        creat_lables(input_path,output_path,files_name[image_no])\n",
    "        \n",
    "    if image_no >=4896 and image_no<= 5595: #val\n",
    "        input_path =os.path.join('datasets_all','input','val') \n",
    "        output_path=os.path.join('datasets_all','output','val')\n",
    "        creat_lables(input_path,output_path,files_name[image_no])\n",
    "        \n",
    "    if image_no >=5596 and image_no<= 7000: #test\n",
    "        input_path =os.path.join('datasets_all','input','test') \n",
    "        output_path=os.path.join('datasets_all','output','test')\n",
    "        creat_lables(input_path,output_path,files_name[image_no])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "\n",
    "tt=np.array((1,2,3,3,4,6))[:3]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "def load_image(path):\n",
    "    image = cv2.cvtColor(cv2.imread(path,-1), cv2.COLOR_BGR2RGB)\n",
    "    return image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "image_paths_train = glob('new_dataset/leftImg8bit/train/*_leftImg8bit.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "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": [
    "image = cv2.imread('datasets_all/output/val/000300_leftImg8bit.png', 1)\n",
    "plt.imshow(image)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "\n",
    "label_values = np.array(\n",
    "[[128, 64,128] ,\n",
    "[250,170,160]  ,\n",
    "[ 81,  0, 81]  ,\n",
    " [244, 35,232]  ,\n",
    "[230,150,140]  ,\n",
    " [152,251,152]  ,\n",
    " [220, 20, 60]  ,\n",
    " [246, 198, 145],\n",
    " [255,  0,  0]  ,\n",
    " [  0,  0,230]  ,\n",
    " [119, 11, 32]  ,\n",
    " [255, 204, 54] ,\n",
    " [  0,  0,142]  ,\n",
    " [  0,  0, 70]  ,\n",
    " [  0, 60,100]  ,\n",
    " [  0,  0, 90]  ,\n",
    " [  0,  0,110]  ,\n",
    "[  0, 80,100]  ,\n",
    "[136, 143, 153],  \n",
    " [220, 190, 40],\n",
    " [102,102,156]  ,\n",
    "[190,153,153]  ,\n",
    " [180,165,180]  ,\n",
    " [174, 64, 67] ,\n",
    " [220,220,  0]  ,\n",
    " [250,170, 30]  ,\n",
    " [153,153,153]  ,\n",
    " [153,153,153]  ,\n",
    " [169, 187, 214] ,  \n",
    " [ 70, 70, 70]  ,\n",
    " [150,100,100]  ,\n",
    " [150,120, 90]  ,\n",
    " [107,142, 35]  ,\n",
    " [ 70,130,180]  ,\n",
    "  [169, 187, 214],\n",
    "  [  0,  0,  0]  ,\n",
    "  [  0,  0,  0]  ,\n",
    "  [  0,  0,  0]  ,\n",
    "  [  0,  0,  0]  ,\n",
    "  [  0,  0,142]  ], np.uint8)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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": "3.6.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}