Hand-Biometrics.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Step1: Biometric using Hand Geometry"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from PIL import Image\n",
    "from scipy import ndimage as ndi\n",
    "from skimage.feature import peak_local_max\n",
    "from skimage import data, img_as_float\n",
    "import binascii\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.image as mpimg\n",
    "import cv2\n",
    "import numpy as np\n",
    "import imutils\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def displayImage(name,data):\n",
    " \n",
    "    '''\n",
    "     Input:- Image-name, image-path to be display\n",
    "     Output:- Display image on separate window\n",
    " \n",
    "    '''\n",
    "    cv2.imshow(name, data)\n",
    "    cv2.waitKey(0)\n",
    "    cv2.destroyAllWindows()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Step 1a): Image Preprocessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "'''\n",
    " --Acquired the coloured hand image\n",
    " --resize the image \n",
    " --rotate the image by -90-degree\n",
    " \n",
    "'''\n",
    "originalImage = cv2.imread('hand.jpg')\n",
    "resizedImage1 = cv2.resize(originalImage, (200,200))\n",
    "resizedImage = imutils.rotate(resizedImage1, -90)\n",
    "displayImage(\"resized\",resizedImage)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "'''\n",
    "  convert the hand-image into Gray image \n",
    "  \n",
    "'''\n",
    "#convert the image into GrayImage\n",
    "grayImage = cv2.cvtColor(resizedImage, cv2.COLOR_BGR2GRAY)\n",
    "\n",
    "#convert the image into black and white\n",
    "(thresh, blackAndWhiteImage) = cv2.threshold(grayImage, 127, 255, cv2.THRESH_BINARY)\n",
    " \n",
    "displayImage('Black white image', blackAndWhiteImage)\n",
    "displayImage('Original image',resizedImage)\n",
    "displayImage('Gray image', grayImage)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "'''\n",
    "Gaussian filter had used to remove the background noise of the image\n",
    "\n",
    "'''\n",
    "#median filter \n",
    "median = cv2.medianBlur(resizedImage, 5)\n",
    "compare = np.concatenate((resizedImage, median), axis=1) #side by side comparison\n",
    "displayImage('img', compare)\n",
    "\n",
    "#gaussian filtering\n",
    "blur = cv2.GaussianBlur(grayImage,(5,5),0)\n",
    "displayImage(\"Gaussianblur\",blur)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "'''\n",
    "hand image is changed over to the “binary image” by thresholding\n",
    "# applying Otsu thresholding \n",
    "# as an extra flag in binary  \n",
    "# thresholding\n",
    "'''\n",
    "ret, thresh1 = cv2.threshold(blur, 0, 255, cv2.THRESH_BINARY +  cv2.THRESH_OTSU) \n",
    "displayImage('Otsu Threshold', thresh1)          \n",
    "\n",
    "ret3,th3 = cv2.threshold(blur,0,255,cv2.THRESH_BINARY+cv2.THRESH_OTSU)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Step 1b): Feature Extraction\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "'''\n",
    "Boundary Tracing of the preprocessed hand image\n",
    "-trace contours of image used to plot all the contours point\n",
    "'''\n",
    "\n",
    "cnts = cv2.findContours(thresh1, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)\n",
    "#print(cnts)\n",
    "\n",
    "cnts = imutils.grab_contours(cnts)\n",
    "c = max(cnts, key=cv2.contourArea)\n",
    "\n",
    "\n",
    "# determine the most extreme points along the contour\n",
    "extLeft = tuple(c[c[:, :, 0].argmin()][0])\n",
    "extRight = tuple(c[c[:, :, 0].argmax()][0])\n",
    "extTop = tuple(c[c[:, :, 1].argmin()][0])\n",
    "extBot = tuple(c[c[:, :, 1].argmax()][0])\n",
    "\n",
    "# draw the outline of the object, then draw each of the\n",
    "# extreme points, where the left-most is red, right-most\n",
    "# is green, top-most is blue, and bottom-most is teal\n",
    "cv2.drawContours(thresh1, [c], -1, (125, 125, 125), 2)\n",
    "cv2.circle(thresh1, extLeft, 8, (50, 100, 200), -1)\n",
    "cv2.circle(thresh1, extRight, 8, (100, 50, 80), -1)\n",
    "cv2.circle(thresh1, extTop, 8, (50, 50, 50), -1)\n",
    "cv2.circle(thresh1, extBot, 8, (50, 50, 50), -1)\n",
    "\n",
    "displayImage(\"Image\", thresh1)\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x25eae314860>]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#print all the contours point\n",
    "#remove\n",
    "#print(cnts)\n",
    "items = [tuple(item[0]) for npar in cnts for item in npar]\n",
    "\n",
    "#print(items)\n",
    "\n",
    "xRes = [lis[0] for lis in items]\n",
    "yRes = [lis[1] for lis in items]\n",
    "\n",
    "plt.plot(yRes,xRes)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x25eadf69f60>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "'''\n",
    "- Detect the convex contour\n",
    "- find the X-Y coordinates of hand image\n",
    "- find convex hull\n",
    "'''\n",
    "\n",
    "hull = cv2.convexHull(cnts[0])\n",
    "\n",
    "#print(hull)\n",
    "\n",
    "xhull = [item[0] for sublist in hull for item in sublist]\n",
    "yhull = [item[1] for sublist in hull for item in sublist]\n",
    "#print('convexHull-X-coord:= ',xhull)\n",
    "#print('convexHull-Y-coord:= ',yhull)\n",
    "\n",
    "img_copy = thresh1.copy()\n",
    "img_hull = cv2.drawContours(img_copy, contours = [hull], \n",
    "                            contourIdx = 0, \n",
    "                            color = (125, 125, 125), thickness = 2)\n",
    "plt.imshow(img_hull)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "'''\n",
    "- find the coordinates of center point for hand image\n",
    "- highlight the center point\n",
    "'''\n",
    "# calculate moments of binary image\n",
    "M = cv2.moments(th3)\n",
    " \n",
    "# calculate x,y coordinate of center\n",
    "cX = int(M[\"m10\"] / M[\"m00\"])\n",
    "cY = int(M[\"m01\"] / M[\"m00\"])\n",
    "#print(\"cX\",cX) \n",
    "#print(\"cY\",cY) \n",
    "\n",
    "# put text and highlight the center\n",
    "cv2.circle(th3, (cX, cY), 5, (125, 125, 255), -1)\n",
    "cv2.putText(th3, \"centroid\", (cX - 25, cY - 25),cv2.FONT_HERSHEY_SIMPLEX, 0.5, (255, 255, 255), 2)\n",
    " \n",
    "# display the image\n",
    "displayImage(\"Image\", th3)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0,0.5,'xhull')"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "'''\n",
    "- plot the X-Y coordinates of contours point and center point\n",
    "'''\n",
    "plt.plot(yRes,xRes)\n",
    "plt.plot(yhull,xhull,'ro')\n",
    "plt.plot(cY,cX,'go')\n",
    "plt.xlabel(\"yhull\")\n",
    "plt.ylabel(\"xhull\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0,0.5,'xhullupdated')"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "'''\n",
    " To find the tips point from hand geometry\n",
    " Consider the points above the center point which indicate the tip points\n",
    "'''\n",
    "\n",
    "#only take the coordinates of tip points\n",
    "xhullUpdated = []\n",
    "yhullUpdated = []\n",
    "t=0;\n",
    "for i in yhull:\n",
    "    if(i<cY):\n",
    "        yhullUpdated.append(i)\n",
    "        xhullUpdated.append(xhull[t])\n",
    "    t=t+1\n",
    "\n",
    "plt.plot(yRes,xRes)\n",
    "plt.plot(yhullUpdated,xhullUpdated,'ro')\n",
    "plt.plot(cY,cX,'go')\n",
    "plt.xlabel(\"yhullUpdated\")\n",
    "plt.ylabel(\"xhullupdated\")\n",
    "\n",
    "#print('yhullUpdated',yhullUpdated)\n",
    "#print('xhullUpdated',xhullUpdated)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Step 1c): Key Generation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "'''\n",
    "Euclidean distance between center point and all other points of convex hull(tips point)\n",
    "\n",
    "'''\n",
    "distVector = []\n",
    "length=int(len(yhullUpdated))\n",
    "for i in range(length):\n",
    "    distVector.append(np.sqrt( (xhullUpdated[i]-cX)**2 + (yhullUpdated[i]-cY)**2 ))\n",
    "    #distVector.append(dist)\n",
    "\n",
    "#print(distVector)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Variance of sample set is 59.81944190387565\n",
      "Standard Deviation of sample is 7.734302935874418 \n"
     ]
    }
   ],
   "source": [
    "'''\n",
    "Variance and standard deviation is calculated for pairwise Euclidean distance\n",
    "'''\n",
    "import statistics \n",
    "\n",
    "variance = statistics.variance(distVector)\n",
    "std_dev = statistics.stdev(distVector)\n",
    "print(\"Variance of sample set is % s\"  %(variance))\n",
    "print(\"Standard Deviation of sample is % s \" %(std_dev)) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Key Generated:-  58\n"
     ]
    }
   ],
   "source": [
    "'''\n",
    "Unique 3 digit key generation\n",
    "Key= (Most Significant 1 Digits of p1) * 10 + (Least Significant 1 Digits p2)\n",
    "Where\n",
    "P1 = variance of Hand points pairs wise distance\n",
    "P2 = standard deviation of Hand points pairs wise distance\n",
    "\n",
    "'''\n",
    "MSB = int(str(variance)[:1])\n",
    "LSB = int(str(std_dev)[-1:])\n",
    "#print(MSB,LSB)\n",
    "key = (MSB)* 10 + (LSB)\n",
    "print(\"Key Generated:- \",key)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Step2:- Encrypt Msg using Ceaser Cipher\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "hand_key = key\n",
    "Number = hand_key\n",
    "Sum = 0\n",
    "while(Number > 0):\n",
    "    r = Number % 10\n",
    "    Sum = Sum + r\n",
    "    Number = Number//10\n",
    "#print(Sum)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Enter data to be encoded(Without Space) : Steganography\n",
      "Cipher-Msg: Fgrtnabtencul\n"
     ]
    }
   ],
   "source": [
    "def encrypt(text,s):\n",
    "    result = \"\"\n",
    "    #print(\"lrngth\" , len(text))\n",
    "    for i in range(len(text)):\n",
    "        char = text[i]\n",
    "        # Encrypt uppercase characters in plain text\n",
    "        if (char.isupper()):\n",
    "            result += chr((ord(char) + s-65) % 26 + 65)\n",
    "        # Encrypt lowercase characters in plain text\n",
    "        else:\n",
    "            result += chr((ord(char) + s - 97) % 26 + 97)\n",
    "    return result\n",
    "\n",
    "text = input(\"Enter data to be encoded(Without Space) : \") \n",
    "#print (\"Plain Text : \" + text)\n",
    "#print (\"Shift pattern : \" + str(Sum))\n",
    "print (\"Cipher-Msg: \" + encrypt(text,Sum))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Step 3: Image Steganography using LSB\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "def to_bin(data):\n",
    "    \"\"\"Convert `data` to binary format as string\"\"\"\n",
    "    if isinstance(data, str):\n",
    "        return ''.join([ format(ord(i), \"08b\") for i in data ])\n",
    "    elif isinstance(data, bytes) or isinstance(data, np.ndarray):\n",
    "        return [ format(i, \"08b\") for i in data ]\n",
    "    elif isinstance(data, int) or isinstance(data, np.uint8):\n",
    "        return format(data, \"08b\")\n",
    "    else:\n",
    "        raise TypeError(\"Type not supported.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "'''\n",
    "- First carrier image or cover image has been read and converted in to array of bits\n",
    "- the secret message which is in the form of characters are converted in to the “ASCII” values\n",
    "- ASCII values are converted in to array of bits\n",
    "- key will gives unique pixel position of cover image to embed the data LSB of particular image pixel of cover object\n",
    "\n",
    "Output: Stego-image has been generated which contains secret message embedded within cover image.\n",
    "\n",
    "'''\n",
    "#res = input(\"Enter data to be encoded(Without Space) : \") \n",
    "res= encrypt(text,Sum)\n",
    "\n",
    "#data = ''.join(format(ord(i), 'b') for i in res) \n",
    "#print(\"The string after binary conversion : \" + data) \n",
    "data = to_bin(res)\n",
    "msg_len = len(data)\n",
    "#print(\"msg length\",msg_len)\n",
    "            \n",
    "    \n",
    "i=0\n",
    "\n",
    "with Image.open(\"source_img.png\") as img:\n",
    "   \n",
    "    width, height = img.size\n",
    "    #print(width)\n",
    "    #print(height)\n",
    "    #img.show()\n",
    "    for x in range(key, width):\n",
    "        for y in range(key, height):\n",
    "            pixel = list(img.getpixel((x, y)))\n",
    "            for n in range(0,3):\n",
    "                if(i < len(data)):\n",
    "                    pixel[n] = pixel[n] & ~1 | int(data[i])\n",
    "                    i+=1\n",
    "                    \n",
    "            img.putpixel((x,y), tuple(pixel))\n",
    "            \n",
    "    \n",
    "    img.save(\"source_secret.png\", \"PNG\")\n",
    "        \n",
    "        "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Step 4: AES Cryptography (CBC-mode) Encryption"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "from PIL import Image\n",
    "from Crypto.Cipher import AES\n",
    "\n",
    "filename = \"source_secret.png\"\n",
    "filename_out = \"AES_encrypted_image\"\n",
    "format = \"PNG\"\n",
    "\n",
    "#hand_key=5980\n",
    "key = str(hand_key)*8\n",
    "#print('key', key)\n",
    "\n",
    "\n",
    "# AES requires that plaintexts be a multiple of 16, so we have to pad the data\n",
    "def pad(data):\n",
    "    return data + b\"\\x00\"*(16-len(data)%16) \n",
    " \n",
    "# Maps the RGB \n",
    "def convert_to_RGB(data):\n",
    "    r, g, b = tuple(map(lambda d: [data[i] for i in range(0,len(data)) if i % 3 == d], [0, 1, 2]))\n",
    "    pixels = tuple(zip(r,g,b))\n",
    "    return pixels\n",
    "    \n",
    "def process_image(filename):\n",
    "    # Opens image and converts it to RGB format for PIL\n",
    "    im = Image.open(filename)\n",
    "    data = im.convert(\"RGB\").tobytes()\n",
    " \n",
    "    # Since we will pad the data to satisfy AES's multiple-of-16 requirement, we will store the original data length and \"unpad\" it later.\n",
    "    original = len(data) \n",
    " \n",
    "    # Encrypts using desired AES mode (we'll set it to ECB by default)\n",
    "    new = convert_to_RGB(aes_cbc_encrypt(key, pad(data))[:original]) \n",
    "    \n",
    "    # Create a new PIL Image object and save the old image data into the new image.\n",
    "    im2 = Image.new(im.mode, im.size)\n",
    "    im2.putdata(new)\n",
    "    \n",
    "    #Save image\n",
    "    im2.save(filename_out+\".\"+format, format)\n",
    " \n",
    "\n",
    "# CBC\n",
    "def aes_cbc_encrypt(key, data, mode=AES.MODE_CBC):\n",
    "    IV = \"A\"*16  #We'll manually set the initialization vector to simplify things\n",
    "    aes = AES.new(key, mode, IV)\n",
    "    new_data = aes.encrypt(data)\n",
    "    return new_data\n",
    "# ECB\n",
    "def aes_ecb_encrypt(key, data, mode=AES.MODE_ECB):\n",
    "    aes = AES.new(key, mode)\n",
    "    new_data = aes.encrypt(data)\n",
    "    return new_data\n",
    " \n",
    "process_image(filename)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Step 5: AES Cryptography (CBC-mode) Decryption\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "def aes_cbc_decrypt(key, data, mode=AES.MODE_CBC):\n",
    "    IV = \"A\"*16 \n",
    "    aes = AES.new(key, mode, IV)\n",
    "    decd = aes.decrypt(data)\n",
    "    return decd\n",
    "\n",
    "\n",
    "def decrypt_image(filename):\n",
    "    # Opens image and converts it to RGB format for PIL\n",
    "    im = Image.open(filename)\n",
    "    data = im.convert(\"RGB\").tobytes()\n",
    " \n",
    "    # Since we will pad the data to satisfy AES's multiple-of-16 requirement, we will store the original data length and \"unpad\" it later.\n",
    "    original = len(data) \n",
    "\n",
    "    # Decrypts using desired AES mode (we'll set it to ECB by default)\n",
    "    new = convert_to_RGB(aes_cbc_decrypt(key, pad(data))[:original]) \n",
    "    \n",
    "    # Create a new PIL Image object and save the old image data into the new image.\n",
    "    im3 = Image.new(im.mode, im.size)\n",
    "    im3.putdata(new)\n",
    "    \n",
    "    #Save image\n",
    "    im3.save(filename_in+\".\"+format, format)\n",
    "    \n",
    "\n",
    "filenames = filename_out+\".\"+format\n",
    "filename_in = \"AES_decrypted_image\"\n",
    "decrypt_image(filenames)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Step 6: Data Extraction using LSB"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "decoded_msg:=  Fgrtnabtencul\n"
     ]
    }
   ],
   "source": [
    "'''\n",
    "- Embedded secret data has extracted from stego-image\n",
    "- Extracting secret data the hand key is used\n",
    "\n",
    "'''\n",
    "extracted_bin = []\n",
    "with Image.open(\"AES_decrypted_image.png\") as img:\n",
    "    width, height = img.size\n",
    "    byte = []\n",
    "    for x in range(hand_key, width):\n",
    "        for y in range(hand_key, height):\n",
    "            pixel = list(img.getpixel((x, y)))\n",
    "            for n in range(0,3):\n",
    "                extracted_bin.append(pixel[n]&1)\n",
    "\n",
    "data = \"\".join([str(x) for x in extracted_bin])\n",
    "msg =''\n",
    "\n",
    "for i in range(msg_len):\n",
    "    msg+=str(extracted_bin[i])\n",
    "#print(msg)\n",
    "\n",
    "decoded_msg = ''.join(chr(int(msg[i*8:i*8+8],2)) for i in range(len(msg)//8))\n",
    "print(\"decoded_msg:= \",decoded_msg)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Step 7: Decrypt Msg using Ceaser Cipher"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Received_msg : Fgrtnabtencul\n",
      "Decipher: Steganography\n"
     ]
    }
   ],
   "source": [
    "def decrypt(text,s):\n",
    "    result = \"\"\n",
    "    #print(\"lrngth\" , len(text))\n",
    "    for i in range(len(text)):\n",
    "        char = text[i]\n",
    "        # Encrypt uppercase characters in plain text\n",
    "        if (char.isupper()):\n",
    "            result += chr((ord(char) - s-65) % 26 + 65)\n",
    "        # Encrypt lowercase characters in plain text\n",
    "        else:\n",
    "            result += chr((ord(char) - s - 97) % 26 + 97)\n",
    "    return result\n",
    "\n",
    "print (\"Received_msg : \" + decoded_msg)\n",
    "print (\"Decipher: \" + decrypt(decoded_msg,Sum))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Step 8: PSNR"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "105.36680886253231\n"
     ]
    }
   ],
   "source": [
    "#calculate PSNR\n",
    "import math\n",
    "original = cv2.imread('source_img.png')\n",
    "updated = cv2.imread('source_secret.png',1)\n",
    "\n",
    "def psnr(img1, img2):\n",
    "    mse = np.mean( (img1 - img2) ** 2 )\n",
    "    if mse == 0:\n",
    "        return 100\n",
    "    PIXEL_MAX = 255.0\n",
    "    return 20 * math.log10(PIXEL_MAX / math.sqrt(mse))\n",
    "\n",
    "d=psnr(original,updated)\n",
    "print(d)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Step 9: MSE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5.669186305024455e-06\n"
     ]
    }
   ],
   "source": [
    "#calculate MSE\n",
    "def mse(original1, contrast1):\n",
    "    err = np.sum((original1.astype(\"float\") - contrast1.astype(\"float\")) ** 2)\n",
    "    err /= float(original1.shape[0] * original1.shape[1])\n",
    "    return err\n",
    "\n",
    "MSE=mse(original,updated)\n",
    "print(MSE)"
   ]
  }
 ],
 "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
}