diff --git a/opencv.ipynb b/opencv.ipynb
new file mode 100644
index 0000000..4173cb9
--- /dev/null
+++ b/opencv.ipynb
@@ -0,0 +1,87 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import cv2\n",
+    "\n",
+    "#load some pretrained data on face frontals opencv supplied\n",
+    "trained_face_data= cv2.CascadeClassifier('haarcascade_frontalface_default.xml')\n",
+    "\n",
+    "#to read an image to detect faces\n",
+    "#img=cv2.imread('MSD.jpg')\n",
+    "webcam=cv2.VideoCapture(0)\n",
+    "\n",
+    "\n",
+    "\n",
+    "while True:\n",
+    "    \n",
+    "    success_frame_read, frame=webcam.read()\n",
+    "#to show img\n",
+    "#cv2.imshow('face detec',img)\n",
+    "\n",
+    "#got to make image black as we are using harcascade algo , it uses gray scale \n",
+    "\n",
+    "    grayscaled_img=cv2.cvtColor(frame,cv2.COLOR_BGR2GRAY)\n",
+    "\n",
+    "    cv2.imshow('images',grayscaled_img)\n",
+    "\n",
+    "#now we need to plug our image to the algo, so we use trained data\n",
+    "#detect faces\n",
+    "#face_coordinates=trained_face_data.detectMultiScale(grayscaled_img)\n",
+    "\n",
+    "#print(face_coordinates)\n",
+    "#detectmultiscale detects objects of diff sizes in the input img. and returned as rectangles\n",
+    "\n",
+    "\n",
+    "#draw rectangles around the faces (x,y,w,h) coordinates\n",
+    "#cv2.rectangle(img,(x,y),(x+w,y+h),(r,g,b), rect thickness)\n",
+    "\n",
+    "#cv2.rectangle(img,(660,62),(660+349,62+349),(0,255,0),3)\n",
+    "\n",
+    "\n",
+    "#cv2.imshow('show',img) to show img with rec\n",
+    "#waits for image to open\n",
+    "    key=cv2.waitKey(2)\n",
+    "    \n",
+    "    if key==81 or key==113:\n",
+    "        break\n",
+    "        \n",
+    "webcam.release()\n",
+    "\n",
+    "#if you have images next to next use for loop \n",
+    "#for (x,y,w,h) in face_coordinates:\n",
+    "   # cv2.rectangle(img,(x,y),(x+w,y+h),(r,g,b), rect thickness)\n",
+    "\n",
+    "#to capture video from webcam\n",
+    "\n",
+    "    \n",
+    "    \n",
+    "    \n",
+    "\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "q\n"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}