CoALTP.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import glob\n",
    "import math\n",
    "from skimage import io,color,feature,exposure\n",
    "from matplotlib import pyplot as plt\n",
    "from matplotlib.gridspec import GridSpec\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "def LTP(image,scale,row,col,thres):\n",
    "    \n",
    "    ltpmat1 = np.zeros(shape=(r-2,c-2),dtype=np.uint8)\n",
    "    ltpmat2 = np.zeros(shape=(r-2,c-2),dtype=np.uint8)\n",
    "\n",
    "    for i in range(1,row-1):\n",
    "        for j in range(1,col-1):\n",
    "            window = image[i-1:i+2,j-1:j+2]\n",
    "            a=window[1,1]-thres\n",
    "            b=window[1,1]+thres\n",
    "            binary1=window>=b\n",
    "            binary2=window<=a\n",
    "            ltpmat1[i-1,j-1] = (sum(sum(binary1*scale)))\n",
    "            ltpmat2[i-1,j-1] = (sum(sum(binary2*scale)))\n",
    "    \n",
    "    return ltpmat1,ltpmat2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def GLCM(image,d,theta,levels):\n",
    "    return feature.greycomatrix(image,d,theta,levels=levels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def CoLTP_feature_vector(image):\n",
    "    scale = np.array([[2,0,1],[0,0,0],[4,0,8]],dtype=np.uint8)\n",
    "    (r,c) = image.shape\n",
    "    threshold=5\n",
    "    ltpmat1,ltpmat2 = LTP(image,scale,r,c,threshold)\n",
    "    \n",
    "    d=[4]\n",
    "    theta=[0,np.pi/4,np.pi/2,3*np.pi/4]\n",
    "    levels=16\n",
    "    ltpg1 = GLCM(ltpmat1,d,theta,levels).flatten()\n",
    "    ltpg2 = GLCM(ltpmat2,d,theta,levels).flatten()\n",
    "    return np.concatenate((ltpg1,ltpg2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x2304 with 10 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "image = io.imread('orl_faces\\\\s01\\\\1.pgm')\n",
    "scale = np.array([[2,0,1],[0,0,0],[4,0,8]],dtype=np.uint8)\n",
    "(r,c) = image.shape\n",
    "threshold=5\n",
    "ltp_image1,ltp_image2=LTP(image,scale,r,c,threshold)\n",
    "d=[4]\n",
    "theta=[0,np.pi/4,np.pi/2,3*np.pi/4]\n",
    "levels=16\n",
    "ltpg1 = GLCM(ltp_image1,d,theta,levels)\n",
    "ltpg2 = GLCM(ltp_image2,d,theta,levels)\n",
    "\n",
    "fig = plt.figure(figsize=(15,32))\n",
    "gs=GridSpec(5,2)\n",
    "\n",
    "ax1=fig.add_subplot(gs[0,0])\n",
    "ax2=fig.add_subplot(gs[0,1])\n",
    "ax3=fig.add_subplot(gs[1,0])\n",
    "ax4=fig.add_subplot(gs[1,1])\n",
    "ax5=fig.add_subplot(gs[2,0])\n",
    "ax6=fig.add_subplot(gs[2,1])\n",
    "ax7=fig.add_subplot(gs[3,0])\n",
    "ax8=fig.add_subplot(gs[3,1])\n",
    "ax9=fig.add_subplot(gs[4,0])\n",
    "ax10=fig.add_subplot(gs[4,1])\n",
    "\n",
    "ax1.set_title('LTP Image (Upper matrix)')\n",
    "ax1.imshow(ltp_image1,cmap='gray')\n",
    "ax1.set_yticklabels([])\n",
    "ax1.set_xticklabels([])\n",
    "\n",
    "ax2.set_title('LTP Image (Lower matrix)')\n",
    "ax2.imshow(ltp_image2,cmap='gray')\n",
    "ax2.set_yticklabels([])\n",
    "ax2.set_xticklabels([])\n",
    "\n",
    "ax3.matshow(ltpg1[:,:,0,0],cmap=plt.cm.gray)\n",
    "ax3.set_title('LTP GLCM (Upper matrix) d=4 theta=0 deg')\n",
    "ax3.set_xticks(np.arange(0,16))\n",
    "ax3.set_yticks(np.arange(0,16))\n",
    "\n",
    "ax4.matshow(ltpg2[:,:,0,0],cmap=plt.cm.gray)\n",
    "ax4.set_title('LTP GLCM (Lower matrix) d=4 theta=0 deg')\n",
    "ax4.set_xticks(np.arange(0,16))\n",
    "ax4.set_yticks(np.arange(0,16))\n",
    "\n",
    "ax5.matshow(ltpg1[:,:,0,1],cmap=plt.cm.gray)\n",
    "ax5.set_title('LTP GLCM (Upper matrix) d=4 theta=45 deg')\n",
    "ax5.set_xticks(np.arange(0,16))\n",
    "ax5.set_yticks(np.arange(0,16))\n",
    "\n",
    "ax6.matshow(ltpg2[:,:,0,1],cmap=plt.cm.gray)\n",
    "ax6.set_title('LTP GLCM (Lower matrix) d=4 theta=45 deg')\n",
    "ax6.set_xticks(np.arange(0,16))\n",
    "ax6.set_yticks(np.arange(0,16))\n",
    "\n",
    "ax7.matshow(ltpg1[:,:,0,2],cmap=plt.cm.gray)\n",
    "ax7.set_title('LTP GLCM (Upper matrix) d=4 theta=90 deg')\n",
    "ax7.set_xticks(np.arange(0,16))\n",
    "ax7.set_yticks(np.arange(0,16))\n",
    "\n",
    "ax8.matshow(ltpg2[:,:,0,2],cmap=plt.cm.gray)\n",
    "ax8.set_title('LTP GLCM (Lower matrix) d=4 theta=90 deg')\n",
    "ax8.set_xticks(np.arange(0,16))\n",
    "ax8.set_yticks(np.arange(0,16))\n",
    "\n",
    "ax9.matshow(ltpg1[:,:,0,3],cmap=plt.cm.gray)\n",
    "ax9.set_title('LTP GLCM (Upper matrix) d=4 theta=135 deg')\n",
    "ax9.set_xticks(np.arange(0,16))\n",
    "ax9.set_yticks(np.arange(0,16))\n",
    "\n",
    "ax10.matshow(ltpg2[:,:,0,3],cmap=plt.cm.gray)\n",
    "ax10.set_title('LTP GLCM (Lower matrix) d=4 theta=135 deg')\n",
    "ax10.set_xticks(np.arange(0,16))\n",
    "ax10.set_yticks(np.arange(0,16))\n",
    "\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "processing image :  1 :  orl_faces\\s01\\0.pgm\n",
      "processing image :  2 :  orl_faces\\s01\\1.pgm\n",
      "processing image :  3 :  orl_faces\\s01\\2.pgm\n",
      "processing image :  4 :  orl_faces\\s01\\3.pgm\n",
      "processing image :  5 :  orl_faces\\s01\\4.pgm\n",
      "processing image :  6 :  orl_faces\\s01\\5.pgm\n",
      "processing image :  7 :  orl_faces\\s01\\6.pgm\n",
      "processing image :  8 :  orl_faces\\s01\\7.pgm\n",
      "processing image :  9 :  orl_faces\\s01\\8.pgm\n",
      "processing image :  10 :  orl_faces\\s01\\9.pgm\n",
      "processing image :  11 :  orl_faces\\s02\\0.pgm\n",
      "processing image :  12 :  orl_faces\\s02\\1.pgm\n",
      "processing image :  13 :  orl_faces\\s02\\2.pgm\n",
      "processing image :  14 :  orl_faces\\s02\\3.pgm\n",
      "processing image :  15 :  orl_faces\\s02\\4.pgm\n",
      "processing image :  16 :  orl_faces\\s02\\5.pgm\n",
      "processing image :  17 :  orl_faces\\s02\\6.pgm\n",
      "processing image :  18 :  orl_faces\\s02\\7.pgm\n",
      "processing image :  19 :  orl_faces\\s02\\8.pgm\n",
      "processing image :  20 :  orl_faces\\s02\\9.pgm\n",
      "processing image :  21 :  orl_faces\\s03\\0.pgm\n",
      "processing image :  22 :  orl_faces\\s03\\1.pgm\n",
      "processing image :  23 :  orl_faces\\s03\\2.pgm\n",
      "processing image :  24 :  orl_faces\\s03\\3.pgm\n",
      "processing image :  25 :  orl_faces\\s03\\4.pgm\n",
      "processing image :  26 :  orl_faces\\s03\\5.pgm\n",
      "processing image :  27 :  orl_faces\\s03\\6.pgm\n",
      "processing image :  28 :  orl_faces\\s03\\7.pgm\n",
      "processing image :  29 :  orl_faces\\s03\\8.pgm\n",
      "processing image :  30 :  orl_faces\\s03\\9.pgm\n",
      "processing image :  31 :  orl_faces\\s04\\0.pgm\n",
      "processing image :  32 :  orl_faces\\s04\\1.pgm\n",
      "processing image :  33 :  orl_faces\\s04\\2.pgm\n",
      "processing image :  34 :  orl_faces\\s04\\3.pgm\n",
      "processing image :  35 :  orl_faces\\s04\\4.pgm\n",
      "processing image :  36 :  orl_faces\\s04\\5.pgm\n",
      "processing image :  37 :  orl_faces\\s04\\6.pgm\n",
      "processing image :  38 :  orl_faces\\s04\\7.pgm\n",
      "processing image :  39 :  orl_faces\\s04\\8.pgm\n",
      "processing image :  40 :  orl_faces\\s04\\9.pgm\n",
      "processing image :  41 :  orl_faces\\s05\\0.pgm\n",
      "processing image :  42 :  orl_faces\\s05\\1.pgm\n",
      "processing image :  43 :  orl_faces\\s05\\2.pgm\n",
      "processing image :  44 :  orl_faces\\s05\\3.pgm\n",
      "processing image :  45 :  orl_faces\\s05\\4.pgm\n",
      "processing image :  46 :  orl_faces\\s05\\5.pgm\n",
      "processing image :  47 :  orl_faces\\s05\\6.pgm\n",
      "processing image :  48 :  orl_faces\\s05\\7.pgm\n",
      "processing image :  49 :  orl_faces\\s05\\8.pgm\n",
      "processing image :  50 :  orl_faces\\s05\\9.pgm\n",
      "processing image :  51 :  orl_faces\\s06\\0.pgm\n",
      "processing image :  52 :  orl_faces\\s06\\1.pgm\n",
      "processing image :  53 :  orl_faces\\s06\\2.pgm\n",
      "processing image :  54 :  orl_faces\\s06\\3.pgm\n",
      "processing image :  55 :  orl_faces\\s06\\4.pgm\n",
      "processing image :  56 :  orl_faces\\s06\\5.pgm\n",
      "processing image :  57 :  orl_faces\\s06\\6.pgm\n",
      "processing image :  58 :  orl_faces\\s06\\7.pgm\n",
      "processing image :  59 :  orl_faces\\s06\\8.pgm\n",
      "processing image :  60 :  orl_faces\\s06\\9.pgm\n",
      "processing image :  61 :  orl_faces\\s07\\0.pgm\n",
      "processing image :  62 :  orl_faces\\s07\\1.pgm\n",
      "processing image :  63 :  orl_faces\\s07\\2.pgm\n",
      "processing image :  64 :  orl_faces\\s07\\3.pgm\n",
      "processing image :  65 :  orl_faces\\s07\\4.pgm\n",
      "processing image :  66 :  orl_faces\\s07\\5.pgm\n",
      "processing image :  67 :  orl_faces\\s07\\6.pgm\n",
      "processing image :  68 :  orl_faces\\s07\\7.pgm\n",
      "processing image :  69 :  orl_faces\\s07\\8.pgm\n",
      "processing image :  70 :  orl_faces\\s07\\9.pgm\n",
      "processing image :  71 :  orl_faces\\s08\\0.pgm\n",
      "processing image :  72 :  orl_faces\\s08\\1.pgm\n",
      "processing image :  73 :  orl_faces\\s08\\2.pgm\n",
      "processing image :  74 :  orl_faces\\s08\\3.pgm\n",
      "processing image :  75 :  orl_faces\\s08\\4.pgm\n",
      "processing image :  76 :  orl_faces\\s08\\5.pgm\n",
      "processing image :  77 :  orl_faces\\s08\\6.pgm\n",
      "processing image :  78 :  orl_faces\\s08\\7.pgm\n",
      "processing image :  79 :  orl_faces\\s08\\8.pgm\n",
      "processing image :  80 :  orl_faces\\s08\\9.pgm\n",
      "processing image :  81 :  orl_faces\\s09\\0.pgm\n",
      "processing image :  82 :  orl_faces\\s09\\1.pgm\n",
      "processing image :  83 :  orl_faces\\s09\\2.pgm\n",
      "processing image :  84 :  orl_faces\\s09\\3.pgm\n",
      "processing image :  85 :  orl_faces\\s09\\4.pgm\n",
      "processing image :  86 :  orl_faces\\s09\\5.pgm\n",
      "processing image :  87 :  orl_faces\\s09\\6.pgm\n",
      "processing image :  88 :  orl_faces\\s09\\7.pgm\n",
      "processing image :  89 :  orl_faces\\s09\\8.pgm\n",
      "processing image :  90 :  orl_faces\\s09\\9.pgm\n",
      "processing image :  91 :  orl_faces\\s10\\0.pgm\n",
      "processing image :  92 :  orl_faces\\s10\\1.pgm\n",
      "processing image :  93 :  orl_faces\\s10\\2.pgm\n",
      "processing image :  94 :  orl_faces\\s10\\3.pgm\n",
      "processing image :  95 :  orl_faces\\s10\\4.pgm\n",
      "processing image :  96 :  orl_faces\\s10\\5.pgm\n",
      "processing image :  97 :  orl_faces\\s10\\6.pgm\n",
      "processing image :  98 :  orl_faces\\s10\\7.pgm\n",
      "processing image :  99 :  orl_faces\\s10\\8.pgm\n",
      "processing image :  100 :  orl_faces\\s10\\9.pgm\n",
      "processing image :  101 :  orl_faces\\s11\\0.pgm\n",
      "processing image :  102 :  orl_faces\\s11\\1.pgm\n",
      "processing image :  103 :  orl_faces\\s11\\2.pgm\n",
      "processing image :  104 :  orl_faces\\s11\\3.pgm\n",
      "processing image :  105 :  orl_faces\\s11\\4.pgm\n",
      "processing image :  106 :  orl_faces\\s11\\5.pgm\n",
      "processing image :  107 :  orl_faces\\s11\\6.pgm\n",
      "processing image :  108 :  orl_faces\\s11\\7.pgm\n",
      "processing image :  109 :  orl_faces\\s11\\8.pgm\n",
      "processing image :  110 :  orl_faces\\s11\\9.pgm\n",
      "processing image :  111 :  orl_faces\\s12\\0.pgm\n",
      "processing image :  112 :  orl_faces\\s12\\1.pgm\n",
      "processing image :  113 :  orl_faces\\s12\\2.pgm\n",
      "processing image :  114 :  orl_faces\\s12\\3.pgm\n",
      "processing image :  115 :  orl_faces\\s12\\4.pgm\n",
      "processing image :  116 :  orl_faces\\s12\\5.pgm\n",
      "processing image :  117 :  orl_faces\\s12\\6.pgm\n",
      "processing image :  118 :  orl_faces\\s12\\7.pgm\n",
      "processing image :  119 :  orl_faces\\s12\\8.pgm\n",
      "processing image :  120 :  orl_faces\\s12\\9.pgm\n",
      "processing image :  121 :  orl_faces\\s13\\0.pgm\n",
      "processing image :  122 :  orl_faces\\s13\\1.pgm\n",
      "processing image :  123 :  orl_faces\\s13\\2.pgm\n",
      "processing image :  124 :  orl_faces\\s13\\3.pgm\n",
      "processing image :  125 :  orl_faces\\s13\\4.pgm\n",
      "processing image :  126 :  orl_faces\\s13\\5.pgm\n",
      "processing image :  127 :  orl_faces\\s13\\6.pgm\n",
      "processing image :  128 :  orl_faces\\s13\\7.pgm\n",
      "processing image :  129 :  orl_faces\\s13\\8.pgm\n",
      "processing image :  130 :  orl_faces\\s13\\9.pgm\n",
      "processing image :  131 :  orl_faces\\s14\\0.pgm\n",
      "processing image :  132 :  orl_faces\\s14\\1.pgm\n",
      "processing image :  133 :  orl_faces\\s14\\2.pgm\n",
      "processing image :  134 :  orl_faces\\s14\\3.pgm\n",
      "processing image :  135 :  orl_faces\\s14\\4.pgm\n",
      "processing image :  136 :  orl_faces\\s14\\5.pgm\n",
      "processing image :  137 :  orl_faces\\s14\\6.pgm\n",
      "processing image :  138 :  orl_faces\\s14\\7.pgm\n",
      "processing image :  139 :  orl_faces\\s14\\8.pgm\n",
      "processing image :  140 :  orl_faces\\s14\\9.pgm\n",
      "processing image :  141 :  orl_faces\\s15\\0.pgm\n",
      "processing image :  142 :  orl_faces\\s15\\1.pgm\n",
      "processing image :  143 :  orl_faces\\s15\\2.pgm\n",
      "processing image :  144 :  orl_faces\\s15\\3.pgm\n",
      "processing image :  145 :  orl_faces\\s15\\4.pgm\n",
      "processing image :  146 :  orl_faces\\s15\\5.pgm\n",
      "processing image :  147 :  orl_faces\\s15\\6.pgm\n",
      "processing image :  148 :  orl_faces\\s15\\7.pgm\n",
      "processing image :  149 :  orl_faces\\s15\\8.pgm\n",
      "processing image :  150 :  orl_faces\\s15\\9.pgm\n",
      "processing image :  151 :  orl_faces\\s16\\0.pgm\n",
      "processing image :  152 :  orl_faces\\s16\\1.pgm\n",
      "processing image :  153 :  orl_faces\\s16\\2.pgm\n",
      "processing image :  154 :  orl_faces\\s16\\3.pgm\n",
      "processing image :  155 :  orl_faces\\s16\\4.pgm\n",
      "processing image :  156 :  orl_faces\\s16\\5.pgm\n",
      "processing image :  157 :  orl_faces\\s16\\6.pgm\n",
      "processing image :  158 :  orl_faces\\s16\\7.pgm\n",
      "processing image :  159 :  orl_faces\\s16\\8.pgm\n",
      "processing image :  160 :  orl_faces\\s16\\9.pgm\n",
      "processing image :  161 :  orl_faces\\s17\\0.pgm\n",
      "processing image :  162 :  orl_faces\\s17\\1.pgm\n",
      "processing image :  163 :  orl_faces\\s17\\2.pgm\n",
      "processing image :  164 :  orl_faces\\s17\\3.pgm\n",
      "processing image :  165 :  orl_faces\\s17\\4.pgm\n",
      "processing image :  166 :  orl_faces\\s17\\5.pgm\n",
      "processing image :  167 :  orl_faces\\s17\\6.pgm\n",
      "processing image :  168 :  orl_faces\\s17\\7.pgm\n",
      "processing image :  169 :  orl_faces\\s17\\8.pgm\n",
      "processing image :  170 :  orl_faces\\s17\\9.pgm\n",
      "processing image :  171 :  orl_faces\\s18\\0.pgm\n",
      "processing image :  172 :  orl_faces\\s18\\1.pgm\n",
      "processing image :  173 :  orl_faces\\s18\\2.pgm\n",
      "processing image :  174 :  orl_faces\\s18\\3.pgm\n",
      "processing image :  175 :  orl_faces\\s18\\4.pgm\n",
      "processing image :  176 :  orl_faces\\s18\\5.pgm\n",
      "processing image :  177 :  orl_faces\\s18\\6.pgm\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "processing image :  178 :  orl_faces\\s18\\7.pgm\n",
      "processing image :  179 :  orl_faces\\s18\\8.pgm\n",
      "processing image :  180 :  orl_faces\\s18\\9.pgm\n",
      "processing image :  181 :  orl_faces\\s19\\0.pgm\n",
      "processing image :  182 :  orl_faces\\s19\\1.pgm\n",
      "processing image :  183 :  orl_faces\\s19\\2.pgm\n",
      "processing image :  184 :  orl_faces\\s19\\3.pgm\n",
      "processing image :  185 :  orl_faces\\s19\\4.pgm\n",
      "processing image :  186 :  orl_faces\\s19\\5.pgm\n",
      "processing image :  187 :  orl_faces\\s19\\6.pgm\n",
      "processing image :  188 :  orl_faces\\s19\\7.pgm\n",
      "processing image :  189 :  orl_faces\\s19\\8.pgm\n",
      "processing image :  190 :  orl_faces\\s19\\9.pgm\n",
      "processing image :  191 :  orl_faces\\s20\\0.pgm\n",
      "processing image :  192 :  orl_faces\\s20\\1.pgm\n",
      "processing image :  193 :  orl_faces\\s20\\2.pgm\n",
      "processing image :  194 :  orl_faces\\s20\\3.pgm\n",
      "processing image :  195 :  orl_faces\\s20\\4.pgm\n",
      "processing image :  196 :  orl_faces\\s20\\5.pgm\n",
      "processing image :  197 :  orl_faces\\s20\\6.pgm\n",
      "processing image :  198 :  orl_faces\\s20\\7.pgm\n",
      "processing image :  199 :  orl_faces\\s20\\8.pgm\n",
      "processing image :  200 :  orl_faces\\s20\\9.pgm\n",
      "processing image :  201 :  orl_faces\\s21\\0.pgm\n",
      "processing image :  202 :  orl_faces\\s21\\1.pgm\n",
      "processing image :  203 :  orl_faces\\s21\\2.pgm\n",
      "processing image :  204 :  orl_faces\\s21\\3.pgm\n",
      "processing image :  205 :  orl_faces\\s21\\4.pgm\n",
      "processing image :  206 :  orl_faces\\s21\\5.pgm\n",
      "processing image :  207 :  orl_faces\\s21\\6.pgm\n",
      "processing image :  208 :  orl_faces\\s21\\7.pgm\n",
      "processing image :  209 :  orl_faces\\s21\\8.pgm\n",
      "processing image :  210 :  orl_faces\\s21\\9.pgm\n",
      "processing image :  211 :  orl_faces\\s22\\0.pgm\n",
      "processing image :  212 :  orl_faces\\s22\\1.pgm\n",
      "processing image :  213 :  orl_faces\\s22\\2.pgm\n",
      "processing image :  214 :  orl_faces\\s22\\3.pgm\n",
      "processing image :  215 :  orl_faces\\s22\\4.pgm\n",
      "processing image :  216 :  orl_faces\\s22\\5.pgm\n",
      "processing image :  217 :  orl_faces\\s22\\6.pgm\n",
      "processing image :  218 :  orl_faces\\s22\\7.pgm\n",
      "processing image :  219 :  orl_faces\\s22\\8.pgm\n",
      "processing image :  220 :  orl_faces\\s22\\9.pgm\n",
      "processing image :  221 :  orl_faces\\s23\\0.pgm\n",
      "processing image :  222 :  orl_faces\\s23\\1.pgm\n",
      "processing image :  223 :  orl_faces\\s23\\2.pgm\n",
      "processing image :  224 :  orl_faces\\s23\\3.pgm\n",
      "processing image :  225 :  orl_faces\\s23\\4.pgm\n",
      "processing image :  226 :  orl_faces\\s23\\5.pgm\n",
      "processing image :  227 :  orl_faces\\s23\\6.pgm\n",
      "processing image :  228 :  orl_faces\\s23\\7.pgm\n",
      "processing image :  229 :  orl_faces\\s23\\8.pgm\n",
      "processing image :  230 :  orl_faces\\s23\\9.pgm\n",
      "processing image :  231 :  orl_faces\\s24\\0.pgm\n",
      "processing image :  232 :  orl_faces\\s24\\1.pgm\n",
      "processing image :  233 :  orl_faces\\s24\\2.pgm\n",
      "processing image :  234 :  orl_faces\\s24\\3.pgm\n",
      "processing image :  235 :  orl_faces\\s24\\4.pgm\n",
      "processing image :  236 :  orl_faces\\s24\\5.pgm\n",
      "processing image :  237 :  orl_faces\\s24\\6.pgm\n",
      "processing image :  238 :  orl_faces\\s24\\7.pgm\n",
      "processing image :  239 :  orl_faces\\s24\\8.pgm\n",
      "processing image :  240 :  orl_faces\\s24\\9.pgm\n",
      "processing image :  241 :  orl_faces\\s25\\0.pgm\n",
      "processing image :  242 :  orl_faces\\s25\\1.pgm\n",
      "processing image :  243 :  orl_faces\\s25\\2.pgm\n",
      "processing image :  244 :  orl_faces\\s25\\3.pgm\n",
      "processing image :  245 :  orl_faces\\s25\\4.pgm\n",
      "processing image :  246 :  orl_faces\\s25\\5.pgm\n",
      "processing image :  247 :  orl_faces\\s25\\6.pgm\n",
      "processing image :  248 :  orl_faces\\s25\\7.pgm\n",
      "processing image :  249 :  orl_faces\\s25\\8.pgm\n",
      "processing image :  250 :  orl_faces\\s25\\9.pgm\n",
      "processing image :  251 :  orl_faces\\s26\\0.pgm\n",
      "processing image :  252 :  orl_faces\\s26\\1.pgm\n",
      "processing image :  253 :  orl_faces\\s26\\2.pgm\n",
      "processing image :  254 :  orl_faces\\s26\\3.pgm\n",
      "processing image :  255 :  orl_faces\\s26\\4.pgm\n",
      "processing image :  256 :  orl_faces\\s26\\5.pgm\n",
      "processing image :  257 :  orl_faces\\s26\\6.pgm\n",
      "processing image :  258 :  orl_faces\\s26\\7.pgm\n",
      "processing image :  259 :  orl_faces\\s26\\8.pgm\n",
      "processing image :  260 :  orl_faces\\s26\\9.pgm\n",
      "processing image :  261 :  orl_faces\\s27\\0.pgm\n",
      "processing image :  262 :  orl_faces\\s27\\1.pgm\n",
      "processing image :  263 :  orl_faces\\s27\\2.pgm\n",
      "processing image :  264 :  orl_faces\\s27\\3.pgm\n",
      "processing image :  265 :  orl_faces\\s27\\4.pgm\n",
      "processing image :  266 :  orl_faces\\s27\\5.pgm\n",
      "processing image :  267 :  orl_faces\\s27\\6.pgm\n",
      "processing image :  268 :  orl_faces\\s27\\7.pgm\n",
      "processing image :  269 :  orl_faces\\s27\\8.pgm\n",
      "processing image :  270 :  orl_faces\\s27\\9.pgm\n",
      "processing image :  271 :  orl_faces\\s28\\0.pgm\n",
      "processing image :  272 :  orl_faces\\s28\\1.pgm\n",
      "processing image :  273 :  orl_faces\\s28\\2.pgm\n",
      "processing image :  274 :  orl_faces\\s28\\3.pgm\n",
      "processing image :  275 :  orl_faces\\s28\\4.pgm\n",
      "processing image :  276 :  orl_faces\\s28\\5.pgm\n",
      "processing image :  277 :  orl_faces\\s28\\6.pgm\n",
      "processing image :  278 :  orl_faces\\s28\\7.pgm\n",
      "processing image :  279 :  orl_faces\\s28\\8.pgm\n",
      "processing image :  280 :  orl_faces\\s28\\9.pgm\n",
      "processing image :  281 :  orl_faces\\s29\\0.pgm\n",
      "processing image :  282 :  orl_faces\\s29\\1.pgm\n",
      "processing image :  283 :  orl_faces\\s29\\2.pgm\n",
      "processing image :  284 :  orl_faces\\s29\\3.pgm\n",
      "processing image :  285 :  orl_faces\\s29\\4.pgm\n",
      "processing image :  286 :  orl_faces\\s29\\5.pgm\n",
      "processing image :  287 :  orl_faces\\s29\\6.pgm\n",
      "processing image :  288 :  orl_faces\\s29\\7.pgm\n",
      "processing image :  289 :  orl_faces\\s29\\8.pgm\n",
      "processing image :  290 :  orl_faces\\s29\\9.pgm\n",
      "processing image :  291 :  orl_faces\\s30\\0.pgm\n",
      "processing image :  292 :  orl_faces\\s30\\1.pgm\n",
      "processing image :  293 :  orl_faces\\s30\\2.pgm\n",
      "processing image :  294 :  orl_faces\\s30\\3.pgm\n",
      "processing image :  295 :  orl_faces\\s30\\4.pgm\n",
      "processing image :  296 :  orl_faces\\s30\\5.pgm\n",
      "processing image :  297 :  orl_faces\\s30\\6.pgm\n",
      "processing image :  298 :  orl_faces\\s30\\7.pgm\n",
      "processing image :  299 :  orl_faces\\s30\\8.pgm\n",
      "processing image :  300 :  orl_faces\\s30\\9.pgm\n",
      "processing image :  301 :  orl_faces\\s31\\0.pgm\n",
      "processing image :  302 :  orl_faces\\s31\\1.pgm\n",
      "processing image :  303 :  orl_faces\\s31\\2.pgm\n",
      "processing image :  304 :  orl_faces\\s31\\3.pgm\n",
      "processing image :  305 :  orl_faces\\s31\\4.pgm\n",
      "processing image :  306 :  orl_faces\\s31\\5.pgm\n",
      "processing image :  307 :  orl_faces\\s31\\6.pgm\n",
      "processing image :  308 :  orl_faces\\s31\\7.pgm\n",
      "processing image :  309 :  orl_faces\\s31\\8.pgm\n",
      "processing image :  310 :  orl_faces\\s31\\9.pgm\n",
      "processing image :  311 :  orl_faces\\s32\\0.pgm\n",
      "processing image :  312 :  orl_faces\\s32\\1.pgm\n",
      "processing image :  313 :  orl_faces\\s32\\2.pgm\n",
      "processing image :  314 :  orl_faces\\s32\\3.pgm\n",
      "processing image :  315 :  orl_faces\\s32\\4.pgm\n",
      "processing image :  316 :  orl_faces\\s32\\5.pgm\n",
      "processing image :  317 :  orl_faces\\s32\\6.pgm\n",
      "processing image :  318 :  orl_faces\\s32\\7.pgm\n",
      "processing image :  319 :  orl_faces\\s32\\8.pgm\n",
      "processing image :  320 :  orl_faces\\s32\\9.pgm\n",
      "processing image :  321 :  orl_faces\\s33\\0.pgm\n",
      "processing image :  322 :  orl_faces\\s33\\1.pgm\n",
      "processing image :  323 :  orl_faces\\s33\\2.pgm\n",
      "processing image :  324 :  orl_faces\\s33\\3.pgm\n",
      "processing image :  325 :  orl_faces\\s33\\4.pgm\n",
      "processing image :  326 :  orl_faces\\s33\\5.pgm\n",
      "processing image :  327 :  orl_faces\\s33\\6.pgm\n",
      "processing image :  328 :  orl_faces\\s33\\7.pgm\n",
      "processing image :  329 :  orl_faces\\s33\\8.pgm\n",
      "processing image :  330 :  orl_faces\\s33\\9.pgm\n",
      "processing image :  331 :  orl_faces\\s34\\0.pgm\n",
      "processing image :  332 :  orl_faces\\s34\\1.pgm\n",
      "processing image :  333 :  orl_faces\\s34\\2.pgm\n",
      "processing image :  334 :  orl_faces\\s34\\3.pgm\n",
      "processing image :  335 :  orl_faces\\s34\\4.pgm\n",
      "processing image :  336 :  orl_faces\\s34\\5.pgm\n",
      "processing image :  337 :  orl_faces\\s34\\6.pgm\n",
      "processing image :  338 :  orl_faces\\s34\\7.pgm\n",
      "processing image :  339 :  orl_faces\\s34\\8.pgm\n",
      "processing image :  340 :  orl_faces\\s34\\9.pgm\n",
      "processing image :  341 :  orl_faces\\s35\\0.pgm\n",
      "processing image :  342 :  orl_faces\\s35\\1.pgm\n",
      "processing image :  343 :  orl_faces\\s35\\2.pgm\n",
      "processing image :  344 :  orl_faces\\s35\\3.pgm\n",
      "processing image :  345 :  orl_faces\\s35\\4.pgm\n",
      "processing image :  346 :  orl_faces\\s35\\5.pgm\n",
      "processing image :  347 :  orl_faces\\s35\\6.pgm\n",
      "processing image :  348 :  orl_faces\\s35\\7.pgm\n",
      "processing image :  349 :  orl_faces\\s35\\8.pgm\n",
      "processing image :  350 :  orl_faces\\s35\\9.pgm\n",
      "processing image :  351 :  orl_faces\\s36\\0.pgm\n",
      "processing image :  352 :  orl_faces\\s36\\1.pgm\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "processing image :  353 :  orl_faces\\s36\\2.pgm\n",
      "processing image :  354 :  orl_faces\\s36\\3.pgm\n",
      "processing image :  355 :  orl_faces\\s36\\4.pgm\n",
      "processing image :  356 :  orl_faces\\s36\\5.pgm\n",
      "processing image :  357 :  orl_faces\\s36\\6.pgm\n",
      "processing image :  358 :  orl_faces\\s36\\7.pgm\n",
      "processing image :  359 :  orl_faces\\s36\\8.pgm\n",
      "processing image :  360 :  orl_faces\\s36\\9.pgm\n",
      "processing image :  361 :  orl_faces\\s37\\0.pgm\n",
      "processing image :  362 :  orl_faces\\s37\\1.pgm\n",
      "processing image :  363 :  orl_faces\\s37\\2.pgm\n",
      "processing image :  364 :  orl_faces\\s37\\3.pgm\n",
      "processing image :  365 :  orl_faces\\s37\\4.pgm\n",
      "processing image :  366 :  orl_faces\\s37\\5.pgm\n",
      "processing image :  367 :  orl_faces\\s37\\6.pgm\n",
      "processing image :  368 :  orl_faces\\s37\\7.pgm\n",
      "processing image :  369 :  orl_faces\\s37\\8.pgm\n",
      "processing image :  370 :  orl_faces\\s37\\9.pgm\n",
      "processing image :  371 :  orl_faces\\s38\\0.pgm\n",
      "processing image :  372 :  orl_faces\\s38\\1.pgm\n",
      "processing image :  373 :  orl_faces\\s38\\2.pgm\n",
      "processing image :  374 :  orl_faces\\s38\\3.pgm\n",
      "processing image :  375 :  orl_faces\\s38\\4.pgm\n",
      "processing image :  376 :  orl_faces\\s38\\5.pgm\n",
      "processing image :  377 :  orl_faces\\s38\\6.pgm\n",
      "processing image :  378 :  orl_faces\\s38\\7.pgm\n",
      "processing image :  379 :  orl_faces\\s38\\8.pgm\n",
      "processing image :  380 :  orl_faces\\s38\\9.pgm\n",
      "processing image :  381 :  orl_faces\\s39\\0.pgm\n",
      "processing image :  382 :  orl_faces\\s39\\1.pgm\n",
      "processing image :  383 :  orl_faces\\s39\\2.pgm\n",
      "processing image :  384 :  orl_faces\\s39\\3.pgm\n",
      "processing image :  385 :  orl_faces\\s39\\4.pgm\n",
      "processing image :  386 :  orl_faces\\s39\\5.pgm\n",
      "processing image :  387 :  orl_faces\\s39\\6.pgm\n",
      "processing image :  388 :  orl_faces\\s39\\7.pgm\n",
      "processing image :  389 :  orl_faces\\s39\\8.pgm\n",
      "processing image :  390 :  orl_faces\\s39\\9.pgm\n",
      "processing image :  391 :  orl_faces\\s40\\0.pgm\n",
      "processing image :  392 :  orl_faces\\s40\\1.pgm\n",
      "processing image :  393 :  orl_faces\\s40\\2.pgm\n",
      "processing image :  394 :  orl_faces\\s40\\3.pgm\n",
      "processing image :  395 :  orl_faces\\s40\\4.pgm\n",
      "processing image :  396 :  orl_faces\\s40\\5.pgm\n",
      "processing image :  397 :  orl_faces\\s40\\6.pgm\n",
      "processing image :  398 :  orl_faces\\s40\\7.pgm\n",
      "processing image :  399 :  orl_faces\\s40\\8.pgm\n",
      "processing image :  400 :  orl_faces\\s40\\9.pgm\n"
     ]
    }
   ],
   "source": [
    "coltp_feature_matrix = []\n",
    "k=1\n",
    "for img_file in glob.glob(\"orl_faces\\\\**\\\\*.pgm\"):\n",
    "    print(\"processing image : \",k,\": \",img_file)\n",
    "    image = io.imread(img_file)\n",
    "    coltp_feature_matrix.append(CoLTP_feature_vector(image))\n",
    "    k+=1\n",
    "    \n",
    "np.savetxt(\"CSV\\\\orl_faces_CoLTP_FeatureMatrix.csv\",coltp_feature_matrix,fmt='%f',delimiter=\",\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "def cityBlockDistance(feature_matrix):\n",
    "    (row,col)=np.shape(feature_matrix)\n",
    "    city_block_distance_matrix = np.zeros(shape=(row,row),dtype=np.double)\n",
    "    for i in range(row):\n",
    "        for j in range(i+1,row):\n",
    "            city_block_distance_matrix[i,j] = city_block_distance_matrix[j,i] = sum(np.abs(feature_matrix[i]-feature_matrix[j]))\n",
    "    return city_block_distance_matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "def euclideanDistance(feature_matrix):\n",
    "    (row,col)=np.shape(feature_matrix)\n",
    "    euclidean_distance_matrix = np.zeros(shape=(row,row),dtype=np.double)\n",
    "    for i in range(row):\n",
    "        for j in range(i+1,row):\n",
    "            euclidean_distance_matrix[i,j] = euclidean_distance_matrix[j,i] = math.sqrt(sum(np.square(feature_matrix[i]-feature_matrix[j])))\n",
    "    return euclidean_distance_matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "def d1Distance(feature_matrix):\n",
    "    (row,col)=np.shape(feature_matrix)\n",
    "    d1_distance_matrix = np.zeros(shape=(row,row),dtype=np.double)\n",
    "    for i in range(row):\n",
    "        for j in range(i+1,row):\n",
    "            d1_distance_matrix[i,j] = d1_distance_matrix[j,i] = sum(abs((feature_matrix[i]-feature_matrix[j])/(1+feature_matrix[i]+feature_matrix[j])))\n",
    "    return d1_distance_matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "city_block_distance_matrix = cityBlockDistance(np.array(coltp_feature_matrix,dtype=np.double))\n",
    "euclidean_distance_matrix = euclideanDistance(np.array(coltp_feature_matrix,dtype=np.double))\n",
    "d1_distance_matrix = d1Distance(np.array(coltp_feature_matrix,dtype=np.double))\n",
    "\n",
    "np.savetxt(\"CSV\\\\coltp_distance\\\\orl_faces_coltp_CityBlockDistanceMatrix.csv\",city_block_distance_matrix,fmt='%f',delimiter=\",\")\n",
    "np.savetxt(\"CSV\\\\coltp_distance\\\\orl_faces_coltp_EuclideanDistanceMatrix.csv\",euclidean_distance_matrix,fmt='%f',delimiter=\",\")\n",
    "np.savetxt(\"CSV\\\\coltp_distance\\\\orl_faces_coltp_D1DistanceMatrix.csv\",d1_distance_matrix,fmt='%f',delimiter=\",\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "def averagePrecision(distance_matrix,classes,retrive_images,relevant_images_DB):\n",
    "    \n",
    "    class_precision = [0]*classes\n",
    "    \n",
    "    i=0\n",
    "    for distmat in distance_matrix:\n",
    "        relevant_images=sum(np.floor(np.argsort(distmat)[:retrive_images]/relevant_images_DB)==math.floor(i/relevant_images_DB))\n",
    "        class_precision[math.floor(i/relevant_images_DB)]+=relevant_images/retrive_images\n",
    "        l=i\n",
    "        i=i+1;\n",
    "        if math.floor(i/relevant_images_DB)!=math.floor(l/relevant_images_DB):\n",
    "            class_precision[math.floor(l/relevant_images_DB)]/=relevant_images_DB\n",
    "            \n",
    "    class_precision*=100\n",
    "    average_precision = sum(class_precision)/classes\n",
    "    \n",
    "    return class_precision,average_precision"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "def averageRecall(distance_matrix,classes,retrive_images,relevant_images_DB):\n",
    "    \n",
    "    class_recall = [0]*classes\n",
    "    \n",
    "    i=0\n",
    "    for distmat in distance_matrix:\n",
    "        relevant_images=sum(np.floor(np.argsort(distmat)[:retrive_images]/relevant_images_DB)==math.floor(i/relevant_images_DB))\n",
    "        class_recall[math.floor(i/relevant_images_DB)]+=relevant_images/relevant_images_DB\n",
    "        l=i\n",
    "        i=i+1;\n",
    "        if math.floor(i/relevant_images_DB)!=math.floor(l/relevant_images_DB):\n",
    "            class_recall[math.floor(l/relevant_images_DB)]/=relevant_images_DB\n",
    "    class_recall*=100\n",
    "    average_recall = sum(class_recall)/classes\n",
    "    return class_recall,average_recall"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "City Block Average Precision: 25.74999999999997%\n",
      "City Block Average Recall: 77.24999999999875%\n",
      "Euclidean Average Precision: 21.174999999999716%\n",
      "Euclidean Average Recall: 63.524999999999785%\n",
      "D1 Average Precision: 26.258333333333233%\n",
      "D1 Average Recall: 78.77500000000019%\n"
     ]
    }
   ],
   "source": [
    "classes = 40\n",
    "retrive_images = 30\n",
    "relevant_images_DB = 10\n",
    "\n",
    "city_block_class_precision,city_block_average_precision = averagePrecision(city_block_distance_matrix,classes,retrive_images,relevant_images_DB)\n",
    "city_block_class_recall,city_block_average_recall = averageRecall(city_block_distance_matrix,classes,retrive_images,relevant_images_DB)\n",
    "\n",
    "euclidean_class_precision,euclidean_average_precision = averagePrecision(euclidean_distance_matrix,classes,retrive_images,relevant_images_DB)\n",
    "euclidean_class_recall,euclidean_average_recall = averageRecall(euclidean_distance_matrix,classes,retrive_images,relevant_images_DB)\n",
    "\n",
    "d1_class_precision,d1_average_precision = averagePrecision(d1_distance_matrix,classes,retrive_images,relevant_images_DB)\n",
    "d1_class_recall,d1_average_recall = averageRecall(d1_distance_matrix,classes,retrive_images,relevant_images_DB)\n",
    "print('City Block Average Precision: '+str(city_block_average_precision)+'%')\n",
    "print('City Block Average Recall: '+str(city_block_average_recall)+'%')\n",
    "print('Euclidean Average Precision: '+str(euclidean_average_precision)+'%')\n",
    "print('Euclidean Average Recall: '+str(euclidean_average_recall)+'%')\n",
    "print('D1 Average Precision: '+str(d1_average_precision)+'%')\n",
    "print('D1 Average Recall: '+str(d1_average_recall)+'%')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "n_groups = 40\n",
    " \n",
    "fig, ax = plt.subplots(figsize=(15, 5))\n",
    "\n",
    "index = np.arange(n_groups)\n",
    "bar_width = 0.25\n",
    "opacity = 0.8\n",
    " \n",
    "rects1 = plt.bar(index, city_block_class_precision[0:40], bar_width, alpha=opacity, color='b', label='City Block Distance')\n",
    "rects2 = plt.bar(index + bar_width, euclidean_class_precision[0:40], bar_width, alpha=opacity, color='y', label='Euclidean Distance')\n",
    "rects3 = plt.bar(index + 2*bar_width, d1_class_precision[0:40], bar_width, alpha=opacity, color='r', label='D1 Distance')\n",
    " \n",
    "plt.xlabel('Classes')\n",
    "plt.ylabel('Precision')\n",
    "plt.title('Class wise precisions for different distances')\n",
    "plt.xticks(index + bar_width, range(1,41))\n",
    "plt.legend()\n",
    " \n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "n_groups = 40\n",
    " \n",
    "fig, ax = plt.subplots(figsize=(15, 5))\n",
    "\n",
    "index = np.arange(n_groups)\n",
    "bar_width = 0.25\n",
    "opacity = 0.8\n",
    " \n",
    "rects1 = plt.bar(index, city_block_class_recall[0:40], bar_width, alpha=opacity, color='b', label='City Block Distance')\n",
    "rects2 = plt.bar(index + bar_width, euclidean_class_recall[0:40], bar_width, alpha=opacity, color='y', label='Euclidean Distance')\n",
    "rects3 = plt.bar(index + 2*bar_width, d1_class_recall[0:40], bar_width, alpha=opacity, color='r', label='D1 Distance')\n",
    " \n",
    "plt.xlabel('Classes')\n",
    "plt.ylabel('Recall')\n",
    "plt.title('Class wise recall rate for different distances')\n",
    "plt.xticks(index + bar_width, range(1,41))\n",
    "plt.legend()\n",
    " \n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "distances = ('City Block','Euclidean','D1')\n",
    "y_pos = np.arange(len(distances))\n",
    "average_precisions = [city_block_average_precision,euclidean_average_precision,d1_average_precision]\n",
    "plt.figure(figsize=(10,5))\n",
    "plt.bar(y_pos, average_precisions,0.2 ,align='center')\n",
    "plt.xticks(y_pos, distances)\n",
    "plt.ylabel('Average Precision')\n",
    "plt.title('Comparision of average precisions using different distance formulae')\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "average_recalls = [city_block_average_recall,euclidean_average_recall,d1_average_recall]\n",
    "plt.figure(figsize=(10,5))\n",
    "plt.bar(y_pos, average_recalls,0.2 ,align='center')\n",
    "plt.xticks(y_pos, distances)\n",
    "plt.ylabel('Average Recall Rate')\n",
    "plt.title('Comparision of average recall rate using different distance formulae')\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "city_block_average_precisions = []\n",
    "city_block_average_recalls = []\n",
    "euclidean_average_precisions = []\n",
    "euclidean_average_recalls = []\n",
    "d1_average_precisions = []\n",
    "d1_average_recalls = []\n",
    "for retrive_images in range(10,101,10):\n",
    "    city_block_average_precisions.append(averagePrecision(city_block_distance_matrix,classes,retrive_images,relevant_images_DB)[1])\n",
    "    city_block_average_recalls.append(averageRecall(city_block_distance_matrix,classes,retrive_images,relevant_images_DB)[1])\n",
    "    euclidean_average_precisions.append(averagePrecision(euclidean_distance_matrix,classes,retrive_images,relevant_images_DB)[1])\n",
    "    euclidean_average_recalls.append(averageRecall(euclidean_distance_matrix,classes,retrive_images,relevant_images_DB)[1])\n",
    "    d1_average_precisions.append(averagePrecision(d1_distance_matrix,classes,retrive_images,relevant_images_DB)[1])\n",
    "    d1_average_recalls.append(averageRecall(d1_distance_matrix,classes,retrive_images,relevant_images_DB)[1])\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(15,8))\n",
    "gs=GridSpec(1,2)\n",
    "index = range(10,101,10)\n",
    "ax1=fig.add_subplot(gs[0,0])\n",
    "ax2=fig.add_subplot(gs[0,1])\n",
    "\n",
    "ax1.plot(index,city_block_average_precisions,'-g.',label='City Block')\n",
    "ax1.plot(index,euclidean_average_precisions,'-r.',label='Euclidean')\n",
    "ax1.plot(index,d1_average_precisions,'-b.',label='D1')\n",
    "ax1.set_xlabel('Number of retrive images')\n",
    "ax1.set_ylabel('Precision')\n",
    "ax1.set_title('Average Precision for different numbers of retrival images')\n",
    "ax1.set_xticks(range(10,101,10))\n",
    "ax1.set_yticks(np.arange(math.floor(min(min(d1_average_precisions),min(euclidean_average_precisions),min(city_block_average_precisions))), max(max(d1_average_precisions),max(euclidean_average_precisions),max(city_block_average_precisions))+1, 5.0))\n",
    "ax1.grid()\n",
    "ax1.legend()\n",
    "\n",
    "ax2.plot(index,city_block_average_recalls,'-g.',label='City Block')\n",
    "ax2.plot(index,euclidean_average_recalls,'-r.',label='Euclidean')\n",
    "ax2.plot(index,d1_average_recalls,'-b.',label='D1')\n",
    "ax2.set_xlabel('Number of retrive images')\n",
    "ax2.set_ylabel('Recall Rate')\n",
    "ax2.set_title('Average Recall Rate for different numbers of retrival images')\n",
    "ax2.set_xticks(range(10,101,10))\n",
    "ax2.set_yticks(np.arange(math.floor(min(min(d1_average_recalls),min(euclidean_average_recalls),min(city_block_average_recalls))), max(max(d1_average_recalls),max(euclidean_average_recalls),max(city_block_average_recalls))+1, 5.0))\n",
    "ax2.grid()\n",
    "ax2.legend()\n",
    "\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'math' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-2-b946ee35d503>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m()\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0mqueryclass\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mmath\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mfloor\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mquery\u001b[0m\u001b[1;33m/\u001b[0m\u001b[1;36m10\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m      2\u001b[0m \u001b[0mqueryimg\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mquery\u001b[0m\u001b[1;33m%\u001b[0m\u001b[1;36m10\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      3\u001b[0m \u001b[0mqueryimage\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mio\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mimread\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m\"orl_faces\\\\s\"\u001b[0m\u001b[1;33m+\u001b[0m\u001b[0mstr\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mqueryclass\u001b[0m\u001b[1;33m+\u001b[0m\u001b[1;36m1\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mzfill\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;36m2\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m+\u001b[0m\u001b[1;34m\"\\\\\"\u001b[0m\u001b[1;33m+\u001b[0m\u001b[0mstr\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mqueryimg\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m+\u001b[0m\u001b[1;34m\".pgm\"\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      4\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      5\u001b[0m \u001b[0mquery_coltp_feature\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mCoLTP_feature_vector\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mqueryimage\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mNameError\u001b[0m: name 'math' is not defined"
     ]
    }
   ],
   "source": [
    "queryclass=math.floor(query/10)\n",
    "queryimg=query%10\n",
    "queryimage=io.imread(\"orl_faces\\\\s\"+str(queryclass+1).zfill(2)+\"\\\\\"+str(queryimg)+\".pgm\")\n",
    "\n",
    "query_coltp_feature=CoLTP_feature_vector(queryimage)\n",
    "\n",
    "coltp_feature_matrix=np.array(coltp_feature_matrix,dtype=np.double)\n",
    "distance_matrix_city_block=np.zeros(shape=(coltp_feature_matrix.shape[0]),dtype=np.double)\n",
    "for i in range(coltp_feature_matrix.shape[0]):\n",
    "    distance_matrix_city_block[i]=sum(np.abs(query_coltp_feature-coltp_feature_matrix[i]))\n",
    "\n",
    "best=8\n",
    "bestimgs=np.argsort(distance_matrix_city_block)[:best]\n",
    "classes=np.floor(bestimgs/10).astype(np.intc)+1\n",
    "imgs=bestimgs%10\n",
    "\n",
    "retrieve_images=[]\n",
    "for i in range(best):\n",
    "    image=io.imread(\"orl_faces\\\\s\"+str(classes[i]).zfill(2)+\"\\\\\"+str(imgs[i])+\".pgm\")\n",
    "    retrieve_images.append(image)\n",
    "    \n",
    "print(\"User Given Image:-\")\n",
    "\n",
    "fig = plt.figure(figsize=(8,8))\n",
    "gs=GridSpec(int(math.floor(best/4))+1,4)\n",
    "\n",
    "ax1=fig.add_subplot(gs[0,:])\n",
    "ax1.imshow(queryimage,cmap='gray')\n",
    "ax1.set_yticklabels([])\n",
    "ax1.set_xticklabels([])\n",
    "\n",
    "for i in range(best):\n",
    "    ax=fig.add_subplot(gs[math.floor(i/4)+1,i%4])\n",
    "    ax.imshow(retrieve_images[i],cmap='gray')\n",
    "    ax.set_yticklabels([])\n",
    "    ax.set_xticklabels([])\n",
    "\n",
    "fig.show()"
   ]
  },
  {
   "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}