contour_detection.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "import cv2\n",
    "import numpy as np\n",
    "\n",
    "image = cv2.imread('IMG_5050.JPG',-1)\n",
    "# paper = cv2.resize(image, (1000, 500))\n",
    "paper = image\n",
    "# ret, thresh_gray = cv2.threshold(cv2.cvtColor(paper, cv2.COLOR_BGR2GRAY),\n",
    "#                         200, 255, cv2.THRESH_BINARY)\n",
    "ret, thresh_gray = cv2.threshold(cv2.cvtColor(paper, cv2.COLOR_BGR2GRAY),\n",
    "                                 200, 255, cv2.THRESH_BINARY)\n",
    "\n",
    "plt.imshow(thresh_gray);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(4447, 1608077.0), (4283, 420448.0), (4515, 384080.0)]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sorted(list(map(lambda x: (x[0], cv2.contourArea(x[1])), enumerate(contours))),\n",
    "                   key=lambda x: x[1], reverse=True)[:3]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "contours, hier = cv2.findContours(thresh_gray, cv2.RETR_EXTERNAL,cv2.CHAIN_APPROX_NONE)\n",
    "for i, _ in sorted(list(map(lambda x: (x[0], cv2.contourArea(x[1])), enumerate(contours))),\n",
    "                   key=lambda x: x[1], reverse=True)[:3]:\n",
    "    rect = cv2.minAreaRect(contours[i])\n",
    "    box = cv2.boxPoints(rect)\n",
    "    box = np.int0(box)\n",
    "    paper = image.copy()\n",
    "    cv2.drawContours(paper, [box], 0, (0, 255, 0), 10)\n",
    "    cv2.imwrite('paper_{}.jpg'.format(i),paper)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "contours, hier = cv2.findContours(thresh_gray, cv2.RETR_EXTERNAL,cv2.CHAIN_APPROX_NONE)\n",
    "for c in contours:\n",
    "    rect = cv2.minAreaRect(c)\n",
    "    box = cv2.boxPoints(rect)\n",
    "    # convert all coordinates floating point values to int\n",
    "    box = np.int0(box)\n",
    "    # draw a green 'nghien' rectangle\n",
    "    cv2.drawContours(paper, [box], 0, (0, 255, 0),1)\n",
    "\n",
    "cv2.imwrite('paper.jpg',paper)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "contours, hier = cv2.findContours(thresh_gray, cv2.RETR_EXTERNAL,cv2.CHAIN_APPROX_NONE)\n",
    "\n",
    "rect = cv2.minAreaRect(max(contours, key=lambda x: cv2.contourArea(x)))\n",
    "box = cv2.boxPoints(rect)\n",
    "box = np.int0(box)\n",
    "cv2.drawContours(paper, [box], 0, (0, 255, 0),1)\n",
    "\n",
    "cv2.imwrite('paper.jpg',paper)"
   ]
  }
 ],
 "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": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}