Add files via upload

PixelPerMetric of leaf.ipynb

@@ -0,0 +1,207 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "\"\"\"importing the necessary libraries, numpy for handling the mathematical complexities\n",
+    "cv2 - OpenCV, for contouring the image and classifying the data points based on colour differences\"\"\"\n",
+    "\n",
+    "import cv2\n",
+    "import numpy as np\n",
+    "import matplotlib.pyplot as plt"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#loading the test image to an image variable\n",
+    "#this part of the code will be changed dynamically to accept images loaded on the RPi\n",
+    "\n",
+    "img = cv2.imread(\"/home/siddhant/Pictures/leaf.jpg\", cv2.IMREAD_COLOR)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#function to display the image since the process is tedious with multiple lines of code for compatibility in Jupyter Notebooks\n",
+    "# the kernel will stop working if you simply close an image, you have to press a random key on the keyboard to successfully close the image file\n",
+    "# opened after this function is executed\n",
+    "\n",
+    "def show(img, title=\"img\"):\n",
+    "    cv2.imshow(title, img)\n",
+    "    cv2.waitKey()\n",
+    "    cv2.destroyAllWindows()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# converting image to gray scale for easy colour segregation\n",
+    "gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)\n",
+    "\n",
+    "# blur the image to reduce noise as it makes the contouring more accurate. \n",
+    "blur = cv2.blur(gray, (3, 3))\n",
+    "\n",
+    "# binary thresholding of the image. Changes the image into all black or all white and no gray-spots in between.\n",
+    "# colours closer to black turn complete black and likewise with white\n",
+    "\n",
+    "THRESH_VAL = 220 # variable\n",
+    "ret, thresh = cv2.threshold(blur, THRESH_VAL, 255, cv2.THRESH_BINARY)\n",
+    "\n",
+    "show(thresh, title=\"Thresholded Image\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# find contours\n",
+    "# finds a set of joined points around the black and white image borders. Helps in segmentation and auto-selection of parts of the image\n",
+    "\n",
+    "contours, hierarchy = cv2.findContours(thresh, \n",
+    "                                       cv2.RETR_TREE, \n",
+    "                                        cv2.CHAIN_APPROX_SIMPLE)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Length of contours: 17\n"
+     ]
+    }
+   ],
+   "source": [
+    "# number of contour arrays, there are 17 arrays in this particular test image\n",
+    "\n",
+    "print(\"Length of contours: {}\".format(len(contours)))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# here we use numpy to draw an image. This is a part I don't quite understand myself enough to document correctly, but it creates a bitmap image.\n",
+    "# this created image is later filled with the contours we stored in the variable earlier to draw a blueprint of the image parts we're interested in\n",
+    "\n",
+    "drawing = np.zeros((thresh.shape[0], thresh.shape[1], 3), np.uint8)\n",
+    "color_contours = (0, 255, 0) # color for contours\n",
+    "for i in range(len(contours)):\n",
+    "    cv2.drawContours(drawing, contours, i, color_contours, 2, 8, hierarchy)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "show(drawing, title='Contours')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Finally: max_x = 1220, max_y = 766\n",
+      "Finally: min_x = 48, min_y = 92\n"
+     ]
+    }
+   ],
+   "source": [
+    "# The important part, here we run two simple loops to find the maximum and minimum distance from the length and the width axes respectively\n",
+    "# The max and min are recorded after being compared to every single element of the contour array\n",
+    "\n",
+    "max_x = 0\n",
+    "max_y = 0\n",
+    "min_x = thresh.shape[1]\n",
+    "min_y = thresh.shape[0]\n",
+    "\n",
+    "for i in range(len(contours)):\n",
+    "    # print(\"Finding for: \", contours[i])\n",
+    "    for j in contours[i]:\n",
+    "        if(j[0][0] > max_x and j[0][0] < 1250):\n",
+    "            max_x = j[0][0]\n",
+    "        if(j[0][1] > max_y and j[0][1] < 850):\n",
+    "            max_y = j[0][1]\n",
+    "        if(j[0][0] < min_x and j[0][0] > 10):\n",
+    "            min_x = j[0][0]\n",
+    "        if(j[0][1] < min_y and j[0][1] > 10):\n",
+    "            min_y = j[0][1]\n",
+    "            \n",
+    "print(\"Finally: max_x = {}, max_y = {}\".format(max_x, max_y))\n",
+    "print(\"Finally: min_x = {}, min_y = {}\".format(min_x, min_y))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Length 1172\n",
+      "Width: 674\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Naturally the length and width of the leaf are the difference between the points of min and max.\n",
+    "\n",
+    "print(f'Length {max_x - min_x}')\n",
+    "print(\"Width: {}\".format(max_y - min_y))\n",
+    "\n",
+    "# These values will be sent to a code further to calculate actual value in metric units."
+   ]
+  }
+ ],
+ "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": "3.7.0"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}