User database hack.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "import keras \n",
    "from keras import *\n",
    "from keras.models import load_model\n",
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "\n",
    "import pickle\n",
    "import cv2\n",
    "import os.path\n",
    "from utility import img_to_encoding, resize_img\n",
    "from collections import defaultdict\n",
    "from webcam_utility import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# custom loss function for model\n",
    "def triplet_loss(y_true, y_pred, alpha = 0.2):\n",
    "    anchor, positive, negative = y_pred[0], y_pred[1], y_pred[2]\n",
    "    # triplet formula components\n",
    "    pos_dist = tf.reduce_sum( tf.square(tf.subtract(y_pred[0], y_pred[1])) )\n",
    "    neg_dist = tf.reduce_sum( tf.square(tf.subtract(y_pred[0], y_pred[2])) )\n",
    "    basic_loss = pos_dist - neg_dist + alpha\n",
    "    \n",
    "    loss = tf.maximum(basic_loss, 0.0)\n",
    "   \n",
    "    return loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# for loading the facenet trained model \n",
    "def load_FRmodel():\n",
    "    return load_model('models/model.h5', custom_objects={'triplet_loss': triplet_loss})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "# load the saved user database\n",
    "def ini_user_database():\n",
    "    global user_db\n",
    "    # check for existing database\n",
    "    if os.path.exists('database/user_dict.pickle'):\n",
    "        with open('database/user_dict.pickle', 'rb') as handle:\n",
    "            user_db = pickle.load(handle)\n",
    "    else:\n",
    "        # make a new one\n",
    "        # we use a dict for keeping track of mapping of each person with his/her face encoding\n",
    "        user_db = defaultdict(dict)\n",
    "    return user_db"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# deletes a registered user from database\n",
    "def delete_user(user_db, email):\n",
    "    popped = user_db.pop(email, None)\n",
    "\n",
    "    if popped is not None:\n",
    "        print('User ' + email + ' deleted successfully')\n",
    "        # save the database\n",
    "        with open('database/user_dict.pickle', 'wb') as handle:\n",
    "                pickle.dump(user_db, handle, protocol=pickle.HIGHEST_PROTOCOL)\n",
    "        return True\n",
    "    elif popped == None:\n",
    "        print('No such user !!')\n",
    "        return False"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# adds a new user face to the database using his/her image stored on disk using the image path\n",
    "def add_user_img_path(user_db, FRmodel, email, name, img_path):\n",
    "    if email not in user_db:\n",
    "        user_db[email]['encoding'] = img_to_encoding(img_path, FRmodel)\n",
    "        user_db[email]['name'] = name\n",
    "        # save the database\n",
    "        with open('database/user_dict.pickle', 'wb') as handle:\n",
    "                pickle.dump(user_db, handle, protocol=pickle.HIGHEST_PROTOCOL)\n",
    "        print('User ' + name + ' added successfully')\n",
    "        return True\n",
    "    else:\n",
    "        print('The name is already registered! Try a different name.........')\n",
    "        return False"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "# adds a new user using image taken from webcam\n",
    "def add_user_webcam(user_db, FRmodel, email, name):\n",
    "    # we can use the webcam to capture the user image then get it recognized\n",
    "    face_found = detect_face(user_db, FRmodel)\n",
    "\n",
    "    if face_found:\n",
    "        resize_img(\"saved_image/1.jpg\")\n",
    "        if email not in user_db:\n",
    "            return add_user_img_path(user_db, FRmodel, email, name, \"saved_image/1.jpg\")\n",
    "        else:\n",
    "            print('The name is already registered! Try a different name.........')\n",
    "            return True\n",
    "    else:\n",
    "        print('There was no face found in the visible frame. Try again...........')\n",
    "        return False"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "# for getting databse details\n",
    "def get_details(user_db, email = None, view_all = False):\n",
    "    print('Total users: ' + str(len(user_db)))\n",
    "    \n",
    "    user_db = ini_user_database()\n",
    "    if email is  not None and email in user_db:\n",
    "        print('Requested user')\n",
    "        print(user_db[email])\n",
    "    elif email:\n",
    "        print('User not registered !')\n",
    "    print()\n",
    "    if view_all:\n",
    "        for user in user_db.keys():\n",
    "            print(user)\n",
    "            print('Email: ' + str(user))\n",
    "            print('Name: ' + user_db[user]['name'])\n",
    "            print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "FRmodel = load_FRmodel()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "user_db = ini_user_database()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The name is already registered! Try a different name.........\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "add_user_img_path(user_db, FRmodel, 'susanta@gmail.com', 'Susanta', \"saved_image/1.jpg\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "User Susanta added successfully\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "add_user_webcam(user_db, FRmodel, 'susanta@gmail.com', 'Susanta')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "User susanta@gmail.com deleted successfully\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "delete_user(user_db, 'susanta@gmail.com')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total users: 1\n",
      "susanta@gmail.com\n",
      "Email: susanta@gmail.com\n",
      "Name: Susanta\n",
      "\n",
      "seeker@gmail.com\n",
      "Email: seeker@gmail.com\n",
      "Name: seeker\n",
      "\n"
     ]
    }
   ],
   "source": [
    "get_details(user_db, email = None, view_all = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "p = user_db['susanta@gmail.com'].pop('male', None)"
   ]
  }
 ],
 "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
}