character recognition -Hog+ANN.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import cv2\n",
    "import glob\n",
    "import numpy as np\n",
    "from PIL import Image\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn import svm\n",
    "from sklearn.neural_network import MLPClassifier\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn import decomposition\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from skimage.feature import hog\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "import tensorflow as tf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "X=np.zeros((2860,50*50))\n",
    "Y=np.zeros(2860)\n",
    "for i in range(11,63):\n",
    "    temp=glob.glob(\"Img/Sample0\"+str(i)+\"/\"+\"*.png\")\n",
    "    for j in range(55):\n",
    "        a=Image.open(temp[j])\n",
    "        a=a.resize((50,50), Image.ANTIALIAS)\n",
    "        a=np.asarray(a, dtype=np.uint8)\n",
    "        a=cv2.cvtColor(a, cv2.COLOR_BGR2GRAY)\n",
    "        X[55*(i-11)+j,:]=a.reshape(1,50*50)\n",
    "        Y[55*(i-11)+j]=i-11"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "X=X/255"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Nabeel\\Anaconda3\\lib\\site-packages\\skimage\\feature\\_hog.py:119: skimage_deprecation: Default value of `block_norm`==`L1` is deprecated and will be changed to `L2-Hys` in v0.15\n",
      "  'be changed to `L2-Hys` in v0.15', skimage_deprecation)\n"
     ]
    }
   ],
   "source": [
    "fd = hog(X[1,:].reshape(50,50), orientations=8, pixels_per_cell=(6, 6),cells_per_block=(1, 1))\n",
    "\n",
    "train_hog=np.zeros((2860,512))\n",
    "for i in range(2860):\n",
    "    fd = hog(X[i,:].reshape(50,50), orientations=8, pixels_per_cell=(6, 6),cells_per_block=(1, 1))\n",
    "    train_hog[i,:]=fd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(train_hog, Y, test_size=0.20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Nabeel\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\multilayer_perceptron.py:564: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.58216783216783219"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classifier = MLPClassifier(hidden_layer_sizes=(100,100))\n",
    "classifier.fit(X_train,y_train)\n",
    "classifier.score(X_test,y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Nabeel\\Anaconda3\\lib\\site-packages\\skimage\\feature\\_hog.py:119: skimage_deprecation: Default value of `block_norm`==`L1` is deprecated and will be changed to `L2-Hys` in v0.15\n",
      "  'be changed to `L2-Hys` in v0.15', skimage_deprecation)\n"
     ]
    }
   ],
   "source": [
    "a=Image.open('pic5.png')\n",
    "A=np.zeros((1,512))\n",
    "a=a.resize((50,50), Image.ANTIALIAS)\n",
    "a=np.asarray(a, dtype=np.uint8)\n",
    "a=a.reshape(1,50*50)\n",
    "a = hog(a.reshape(50,50), orientations=8, pixels_per_cell=(6, 6),cells_per_block=(1, 1))\n",
    "A[:,:]=a"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 38.])"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classifier.predict(A)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "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": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}