issue 1 solution i.e. align image of paper

phase1solution/.ipynb_checkpoints/DocScanner_Notebook_Representation-checkpoint.ipynb

@@ -0,0 +1,371 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Step by step DocScanner Algorithm Execution"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# By Abhinav Chauhan\n",
+    "# Refrence : Murtaza's Workshop (Youtube Channel)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Importing Libraries"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 27,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import cv2\n",
+    "import numpy as np\n",
+    "import matplotlib.pyplot as plt\n",
+    "%matplotlib inline"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 28,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Resizing image ( This step is optional)\n",
+    "widthImg = 500\n",
+    "heightImg = 540"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 29,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "-1"
+      ]
+     },
+     "execution_count": 29,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "path = \"1.jpg\"\n",
+    "img = cv2.imread(path)\n",
+    "img = cv2.resize(img, (widthImg, heightImg))\n",
+    "cv2.imshow('Real Image', img)\n",
+    "cv2.waitKey(0)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Image Preprocessing"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 31,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "-1"
+      ]
+     },
+     "execution_count": 31,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# Converting Image to Gray\n",
+    "\n",
+    "imgGray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)\n",
+    "cv2.imwrite('Gray.jpg', imgGray)\n",
+    "cv2.imshow('Gray Image', imgGray)\n",
+    "cv2.waitKey(0)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 32,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "-1"
+      ]
+     },
+     "execution_count": 32,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# Blurring Image\n",
+    "\n",
+    "imgBlur = cv2.GaussianBlur(imgGray,(5,5),1)\n",
+    "cv2.imwrite('Blur.jpg', imgBlur)\n",
+    "cv2.imshow('Blur Image', imgBlur)\n",
+    "cv2.waitKey(0)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 33,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "-1"
+      ]
+     },
+     "execution_count": 33,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# Detecting edges using canny edge detector\n",
+    "\n",
+    "imgCanny = cv2.Canny(imgBlur,200,200)\n",
+    "cv2.imwrite('Canny.jpg', imgCanny)\n",
+    "cv2.imshow('Canny Image', imgCanny)\n",
+    "cv2.waitKey(0)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 35,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "-1"
+      ]
+     },
+     "execution_count": 35,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# Dialation image\n",
+    "\n",
+    "kernel = np.ones((5,5))\n",
+    "imgDial = cv2.dilate(imgCanny,kernel, iterations = 2)\n",
+    "cv2.imwrite('Dilated.jpg', imgDial)\n",
+    "cv2.imshow('Dialated Image', imgDial)\n",
+    "cv2.waitKey(0)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 36,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "-1"
+      ]
+     },
+     "execution_count": 36,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# Eroded Image (Final image after Perprocessing)\n",
+    "\n",
+    "kernel = np.ones((5,5))\n",
+    "imgThres = cv2.erode(imgDial, kernel, iterations = 1)\n",
+    "cv2.imwrite('Threshold.jpg', imgThres)\n",
+    "cv2.imshow('Threshold Image', imgThres)\n",
+    "cv2.waitKey(0)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Function to get the desired countour and draw corner points around it"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 37,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[[[350  38]]\n",
+      "\n",
+      " [[105  50]]\n",
+      "\n",
+      " [[135 473]]\n",
+      "\n",
+      " [[410 433]]]\n"
+     ]
+    }
+   ],
+   "source": [
+    "imgContour = img.copy()\n",
+    "def getContours(img):\n",
+    "    biggest = np.array([])\n",
+    "    maxArea = 0\n",
+    "    #_, contours, _= cv2.findContours(img, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_NONE)\n",
+    "    contours, hierarchy = cv2.findContours(img, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_NONE)\n",
+    "    for cnt in contours:\n",
+    "        area = cv2.contourArea(cnt)\n",
+    "        if area> 5000:\n",
+    "            #cv2.drawContours(imgContour, cnt, -1, (255,0,0), 3)\n",
+    "            peri = cv2.arcLength(cnt, True)\n",
+    "            approx = cv2.approxPolyDP(cnt, 0.03*peri, True)\n",
+    "            if area > maxArea and len(approx) == 4:\n",
+    "                biggest = approx\n",
+    "                maxArea = area\n",
+    "\n",
+    "    cv2.drawContours(imgContour, biggest, -1, (255,0,0), 20)\n",
+    "    cv2.imwrite('Conttour.jpg', imgContour)\n",
+    "    cv2.imshow('Corner', imgContour)\n",
+    "    cv2.waitKey(0)\n",
+    "    return biggest\n",
+    "\n",
+    "biggest = getContours(imgThres)\n",
+    "print(biggest)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Now, the above printed biggest variable has the coordinates of the biggest contour detected\n",
+    "# But they are not in the correct order taken by the Wraping function.\n",
+    "# So we create another function to reorder them correctly according to the order required by Wraping function \n",
+    "# i.e (0,0) (width, 0) (0, height) and (width, height)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Function to get correct order of the corner points of the biggest contour detected."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 38,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def reorder(myPoints):\n",
+    "    myPoints = myPoints.reshape((4,2))\n",
+    "    myPointsNew = np.zeros((4,1,2), np.int32)\n",
+    "    add = myPoints.sum(1)\n",
+    "    myPointsNew[0] = myPoints[np.argmin(add)]\n",
+    "    myPointsNew[3] = myPoints[np.argmax(add)]\n",
+    "    diff = np.diff(myPoints, axis=1)\n",
+    "    myPointsNew[1] = myPoints[np.argmin(diff)]\n",
+    "    myPointsNew[2] = myPoints[np.argmax(diff)]\n",
+    "    return myPointsNew\n",
+    "\n",
+    "# We call this fucntion inside the Wraping fucntion which is in the very nect cell"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Function to wrap the image with the ordered points i.e. (0,0) (width, 0) (0, height) and (width, height)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 40,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "-1"
+      ]
+     },
+     "execution_count": 40,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "def getWarp(img, biggest):\n",
+    "    biggest = reorder(biggest)\n",
+    "    pts1 = np.float32(biggest)\n",
+    "    pts2 = np.float32([[0,0],[widthImg,0],[0,heightImg],[widthImg,heightImg]])\n",
+    "    matrix = cv2.getPerspectiveTransform(pts1,pts2)\n",
+    "    imgOutput = cv2.warpPerspective(img,matrix,(widthImg,heightImg))\n",
+    "\n",
+    "    return imgOutput\n",
+    "\n",
+    "imgWarped = getWarp(img,biggest)\n",
+    "cv2.imwrite('Scanned.jpg', imgWarped)\n",
+    "cv2.imshow('Scanned Image',imgWarped)\n",
+    "cv2.waitKey(0)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# All the output images at every step are saved and placed in the folder.\n",
+    "# Also the image will be saved at runtime!"
+   ]
+  }
+ ],
+ "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": "3.8.3"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}

phase1solution/.ipynb_checkpoints/Untitled-checkpoint.ipynb

@@ -0,0 +1,6 @@
+{
+ "cells": [],
+ "metadata": {},
+ "nbformat": 4,
+ "nbformat_minor": 4
+}