objsplit.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Object splitter and measurement - Notebook version\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import time\n",
    "import os\n",
    "import sys\n",
    "import cv2\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import warnings\n",
    "\n",
    "dirstr=str(round(time.time()))\n",
    "os.mkdir(\"out_\"+dirstr)\n",
    "os.mkdir(\"out_\"+dirstr+\"/split\")\n",
    "\n",
    "f2 = open(\"out_\"+dirstr+\"/\"+dirstr+\"_ind_data.txt\", 'w')\n",
    "f2.write('ID,L(km),W(km),Area(km2),perimeter(km),xc,yc\\n')\n",
    "\n",
    "fname_sb='terrasp.jpg'\n",
    "img = cv2.imread(fname_sb)\n",
    "\n",
    "scale= 3   # pix/km  \n",
    "tresh_8b=150  \n",
    "tresh_bin=int(round(tresh_8b/100*255))#84 #84 (33%), 102 (40%) , 114 (45%)\n",
    "font = cv2.FONT_HERSHEY_SIMPLEX\n",
    "filtareamin=5000/(scale*scale)  # in km2\n",
    "filtareamax=np.size(img)/(3)\n",
    "        \n",
    "gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)\n",
    "ret,patt = cv2.threshold(gray,tresh_8b,255,cv2.THRESH_BINARY)\n",
    "contours, hierarchy = cv2.findContours(patt, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_NONE)\n",
    "contpic=cv2.imread(fname_sb)# cv2.drawContours(bwcont, contours, -1, (255, 0, 0), 1)\n",
    "\n",
    "areatot=0\n",
    "areafilt=0\n",
    "nfilt=0\n",
    "\n",
    "areafstats=[]\n",
    "perfstats=[]\n",
    "areastats=[]\n",
    "boxc=[]\n",
    "Lboxst=[]\n",
    "Wboxst=[]\n",
    "\n",
    "for j in range(0,len(contours)-1,1):\n",
    "    area = cv2.contourArea(contours[j])*scale**2\n",
    "    perimeter = cv2.arcLength(contours[j],True)*scale\n",
    "    areatot=areatot+area\n",
    "    areastats.append(area)\n",
    "    if (area > filtareamin) & (area < filtareamax):\n",
    "        nfilt=nfilt+1\n",
    "        areafstats.append(area)\n",
    "        perfstats.append(perimeter)\n",
    "        areafilt=areafilt+area\n",
    "        \n",
    "        rect = cv2.minAreaRect(contours[j])\n",
    "        box = cv2.boxPoints(rect)\n",
    "        box = np.int0(box)\n",
    "        boxc.append(box)\n",
    "        \n",
    "        x0=box[0][0]\n",
    "        y0=box[0][1]\n",
    "        x1=box[1][0]\n",
    "        y1=box[1][1]\n",
    "        x2=box[2][0]\n",
    "        y2=box[2][1]\n",
    "        x3=box[3][0]\n",
    "        y3=box[3][1]\n",
    "        \n",
    "        pm0x=int(round((box[0][0]+box[1][0])/2))\n",
    "        pm0y=int(round((box[0][1]+box[1][1])/2))\n",
    "        pm1x=int(round((box[1][0]+box[2][0])/2))\n",
    "        pm1y=int(round((box[1][1]+box[2][1])/2))\n",
    "        pm2x=int(round((box[2][0]+box[3][0])/2))\n",
    "        pm2y=int(round((box[2][1]+box[3][1])/2))\n",
    "        pm3x=int(round((box[3][0]+box[0][0])/2))\n",
    "        pm3y=int(round((box[3][1]+box[0][1])/2)) \n",
    "       \n",
    "        d01box=((box[0][0]-box[1][0])**2+(box[0][1]-box[1][1])**2)**0.5\n",
    "        d12box=((box[1][0]-box[2][0])**2+(box[1][1]-box[2][1])**2)**0.5\n",
    "        \n",
    "        Abox=d01box*d12box*scale**2\n",
    "        Lbox=np.maximum(d01box,d12box)*scale\n",
    "        Wbox=np.minimum(d01box,d12box)*scale    \n",
    "    \n",
    "        Lboxst.append(Lbox)\n",
    "        Wboxst.append(Wbox)\n",
    "        \n",
    "        xmaxb=np.max((x0,x1,x2,x3))\n",
    "        ymaxb=np.max((y0,y1,y2,y3))\n",
    "        xminb=np.min((x0,x1,x2,x3))\n",
    "        yminb=np.min((y0,y1,y2,y3))\n",
    "        \n",
    "        if(xminb < 1):\n",
    "            xminb=1\n",
    "        if(yminb < 1): \n",
    "            yminb=1\n",
    "        if(xmaxb < 1): \n",
    "            xmaxb=1\n",
    "        if(ymaxb < 1): \n",
    "            ymaxb=1\n",
    "            \n",
    "        #cv2.putText(contpic, str(round(area,2)), (int(round((pm0x+pm2x)*0.5)+5),int(round((pm2y+pm0y)*0.5))), font, 0.3, (255, 0, 0), 1, cv2.LINE_AA)\n",
    "        cv2.putText(contpic, str(nfilt), (int(round((pm0x+pm2x)*0.5)+5),int(round((pm2y+pm0y)*0.5))), font, 0.3, (255, 0, 0), 1, cv2.LINE_AA)\n",
    "        cv2.drawContours(contpic, contours[j], -1, (255, 0, 0), 1)\n",
    "        cv2.drawContours(contpic,[box],0,(0,255,255),1) \n",
    "        cv2.imwrite(\"out_\"+dirstr+\"/split/filtobj_\"+str(nfilt)+\".tif\", contpic[yminb:ymaxb,xminb:xmaxb,:])\n",
    "        f2.write(str(nfilt)+','+str(round(Lbox,2))+','+str(round(Wbox,2))+','+str(round(area,2))+','+str(round((pm0x+pm2x)*0.5,2))+','+str(round((pm2y+pm0y)*0.5,2))+'\\n')\n",
    "\n",
    "cv2.imwrite(\"out_\"+dirstr+\"/\"+dirstr+\"_allcont.tif\", contpic)\n",
    "f2.close()\n",
    "\n",
    "# Statistical analysis plots\n",
    "# create a figure \n",
    "fig = plt.figure() \n",
    "# define subplots \n",
    "plt1 = fig.add_subplot(221) \n",
    "plt2 = fig.add_subplot(222) \n",
    "plt3 = fig.add_subplot(223) \n",
    "plt4 = fig.add_subplot(224) \n",
    "plt1.hist(areafstats, bins=10) \n",
    "plt1.set_title('Area') \n",
    "plt2.hist(Lboxst, bins=10) \n",
    "plt2.set_title('Lenght') \n",
    "plt3.hist(Wboxst, bins=10)  \n",
    "plt3.set_title('Width') \n",
    "plt4.hist(perfstats, bins=10) \n",
    "plt4.set_title('Perimeter') \n",
    "# Space between subplots \n",
    "fig.subplots_adjust(hspace=.5,wspace=0.5) \n",
    "plt.savefig(\"out_\"+dirstr+\"/\"+dirstr+\"_histplot.png\",dpi=200)\n",
    "\n",
    "# create a figure \n",
    "fig2 = plt.figure() \n",
    "# define subplots \n",
    "plt5 = fig2.add_subplot(221) \n",
    "plt6 = fig2.add_subplot(222) \n",
    "plt7 = fig2.add_subplot(223) \n",
    "plt8 = fig2.add_subplot(224) \n",
    "plt5.boxplot(areafstats) \n",
    "plt5.set_title('Area') \n",
    "plt6.boxplot(Lboxst)  \n",
    "plt6.set_title('Lenght') \n",
    "plt7.boxplot(Wboxst)    \n",
    "plt7.set_title('Width') \n",
    "plt8.boxplot(perfstats)  \n",
    "plt8.set_title('Perimeter') \n",
    "# Space between subplots \n",
    "fig2.subplots_adjust(hspace=.5,wspace=0.5) \n",
    "plt.savefig(\"out_\"+dirstr+\"/\"+dirstr+\"_boxsplot.png\",dpi=200)\n",
    "\n",
    "\n"
   ]
  }
 ],
 "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
}