hand recogniztion opencv.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "import cv2\n",
    "import numpy as np\n",
    "import math "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "#hand = cv2.imread('/home/piyush/Downloads/hand.jpg',0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "#ret, the = cv2.threshold(hand,70,255,cv2.THRESH_BINARY)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "#_,contours,_ = cv2.findContours(the.copy(),cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "#hull = [cv2.convexHull(c) for c in contours]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "#final = cv2.drawContours(hand,hull,-1,(255,0,0))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "#cv2.imshow('Orignal Image',hand)\n",
    "#cv2.imshow('Threshold',the)\n",
    "#cv2.imshow('Convex Hull',final)\n",
    "#cv2.waitKey(0)\n",
    "#cv2.destroyAllWindows()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "capture = cv2.VideoCapture(0)\n",
    "while(True):\n",
    "    \n",
    "    #Capture frames from capture\n",
    "    ret,frame = capture.read()\n",
    "    \n",
    "    #Get hannd from rectangle sub window\n",
    "    cv2.rectangle(frame,(100,100),(300,300),(0,255,0),0)\n",
    "    crop_image = frame[100:300,100:300]\n",
    "    \n",
    "    #Apply gaussian blur\n",
    "    blur = cv2.GaussianBlur(crop_image,(3,3),0)\n",
    "    \n",
    "    #Change color from Bgr to Hsv\n",
    "    hsv = cv2.cvtColor(blur,cv2.COLOR_BGR2HSV)\n",
    "    \n",
    "    #mask image to make skin color to white and rest to black\n",
    "    mask = cv2.inRange(hsv , np.array([2,0,0]), np.array([20,255,255]))\n",
    "    \n",
    "    #kernel for morphological transformation\n",
    "    kernel = np.ones((5,5))\n",
    "    \n",
    "    #erosion and dilation to remove background noises\n",
    "    erosion = cv2.erode(mask, kernel, iterations=1)\n",
    "    dilation = cv2.dilate(erosion, kernel, iterations=1)\n",
    "    \n",
    "    #applying gausian blur and threshold\n",
    "    filtered = cv2.GaussianBlur(dilation, (3,3),0)\n",
    "    ret, the = cv2.threshold(filtered, 127, 255, 0)\n",
    "    \n",
    "    #Show Threshold Image\n",
    "    cv2.imshow('Thresholded',the)\n",
    "    \n",
    "    #Find Contours\n",
    "    image, contours, heirachy = cv2.findContours(the.copy(),cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)\n",
    "    \n",
    "    try:\n",
    "        #Find Contours with maximum area\n",
    "        contour = max(contours, key=lambda x:cv2.contourArea(x))\n",
    "        \n",
    "        #Create bounding rectangle around contour\n",
    "        x,y,w,h = cv2.boundingRect(contour)\n",
    "        cv2.rectangle(crop_image,(x,y),(x+w,y+h),(0,0,255),0)\n",
    "        \n",
    "        #Find convexHull\n",
    "        hull = cv2.convexHull(contour)\n",
    "        \n",
    "        #Draw Contours\n",
    "        drawing = np.zeros(crop_image.shape, np.uint8)\n",
    "        cv2.drawContours(drawing, [contour], -1, (255,0,0), 0)\n",
    "        cv2.drawContours(drawing, [hull], -1, (255,0,0), 0)\n",
    "        \n",
    "        #Find Convexity defects\n",
    "        hull = cv2.convexHull(contour, returnPoints=False)\n",
    "        defects = cv2.convexityDefects(contour, hull)\n",
    "        \n",
    "        for i in range(defects.shape[0]):\n",
    "            s, e, f, d = defects[i, 0]\n",
    "            start = tuple(contour[s][0])\n",
    "            end = tuple(contour[e][0])\n",
    "            far = tuple(contour[f][0])\n",
    "\n",
    "            a = math.sqrt((end[0] - start[0]) ** 2 + (end[1] - start[1]) ** 2)\n",
    "            b = math.sqrt((far[0] - start[0]) ** 2 + (far[1] - start[1]) ** 2)\n",
    "            c = math.sqrt((end[0] - far[0]) ** 2 + (end[1] - far[1]) ** 2)\n",
    "            angle = (math.acos((b ** 2 + c ** 2 - a ** 2) / (2 * b * c)) * 180) / 3.14\n",
    "\n",
    "            # if angle > 90 draw a circle at the far point\n",
    "            if angle <= 90:\n",
    "                count_defects += 1\n",
    "                cv2.circle(crop_image, far, 1, [0, 0, 255], -1)\n",
    "\n",
    "            cv2.line(crop_image, start, end, [0, 255, 0], 2)\n",
    "\n",
    "        # Print number of fingers\n",
    "        if count_defects == 0:\n",
    "            cv2.putText(frame, \"ONE\", (50, 50), cv2.FONT_HERSHEY_SIMPLEX, 2,(0,0,255),2)\n",
    "        elif count_defects == 1:\n",
    "            cv2.putText(frame, \"TWO\", (50, 50), cv2.FONT_HERSHEY_SIMPLEX, 2,(0,0,255), 2)\n",
    "        elif count_defects == 2:\n",
    "            cv2.putText(frame, \"THREE\", (5, 50), cv2.FONT_HERSHEY_SIMPLEX, 2,(0,0,255), 2)\n",
    "        elif count_defects == 3:\n",
    "            cv2.putText(frame, \"FOUR\", (50, 50), cv2.FONT_HERSHEY_SIMPLEX, 2,(0,0,255), 2)\n",
    "        elif count_defects == 4:\n",
    "            cv2.putText(frame, \"FIVE\", (50, 50), cv2.FONT_HERSHEY_SIMPLEX, 2,(0,0,255), 2)\n",
    "        else:\n",
    "            pass\n",
    "    except:\n",
    "        pass\n",
    "\n",
    "    # Show required images\n",
    "    cv2.imshow(\"Gesture\", frame)\n",
    "    all_image = np.hstack((drawing, crop_image))\n",
    "    cv2.imshow('Contours', all_image)\n",
    "\n",
    "    # Close the camera if 'q' is pressed\n",
    "    if cv2.waitKey(1) == ord('q'):\n",
    "        break\n",
    "        \n",
    "capture.release()\n",
    "cv2.destroyAllWindows()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "capture.release()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.6.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}