Live Sketch Through WebCam

Computer Vision is a field in Computer science that focuses on enabling computers to see, identify, and process images in the same way as the human mind does. The importance of computer vision is in the problems it can solve. It is one of the main technologies that enable the digital world to interact with the physical world.

OpenCV is a very powerful computer vision library. There are many other applications of computer vision using OpenCV, For eg: Face Detection, Face Recognition, License plate reading, etc. In, this blog article we will see how we can use this library to implement a Computer Vision model that creates a live video black and white sketch of real-time webcam video.

Before we start make sure you have installed “OpenCV” and “NumPy” libraries in your system. By using the following commands:

!pip install opencv
!pip install numpy

Importing Libraries
The most important step, importing our OpenCV and NumPy libraries.

import cv2
import numpy as np

Making a sketch generating function
We will make a function in which we will generate our sketch:

def sketch(img):

Converting an image into grayscale
As we are making our black and white live sketcher we should convert an image into grayscale to remove colors that are present in our image.

img_gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

Clean up the image using Gaussian Blur
Blurring our image using the Gaussian Blur function provided by OpenCV. There are several other functions that are provided by OpenCV to blur images for eg: cv2.blur, cv2.medianBlur, and cv2.bilateralFilter. Here, I have used cv2.GaussianBlur as it saves the edges to an extent and blurs the rest of the image whereas also there is more emphasis or weighting on points around the center.

img_gray_blur = cv2.GaussianBlur(img_gray, (5,5), 0)

Extracting edges
Edges can be defined as sudden changes in an image and they can encode just as much information as pixels. OpenCV again provides us with different functions to detect edges like- Sobel, Laplacian, and Canny. Here, we are using Canny as it optimal due to its low error rate, well-defined edges, and accurate detection.

canny_edges = cv2.Canny(img_gray_blur, 10, 70)

Do an invert binarize the image
Thresholding is the act of converting an image to a binary form. There are different function s provided by OpenCV to apply a threshold on the image for eg: cv2.THRESH_BINARY, cv2.THRESH_BINARY_INV, cv2.THRESH_BINARY_INV, cv2.THRESH_TRUNC, cv2.THRESH_TOZERO, cv2.THRESH_TOZERO_INV. Here I have used cv2.THRESH_BINARY_INV as a threshold calculated individually for each pixel.

ret, mask = cv2.threshold(canny_edges, 70,   255,cv2.THRESH_BINARY_INV)
return mask

Capturing the video
Getting real-time video input through webcam.

cap=cv2.VideoCapture(0)

Displaying our live sketch
Your live black and white sketch will be displayed on your screen. This code will keep running unless you terminate it. Press ‘ESC’ to terminate and exit the window.

while True:
 ret, frame = cap.read()
 cv2.imshow(‘Our Live Sketcher’, sketch(frame))
 
 k= cv2.waitKey(13)
 if k == 27:
 break
 
cap.release()
cv2.destroyAllWindows()

Complete Code

Here, is the complete code:

#Importing Libraries
import cv2
import numpy as np#Making a sketch generating function
def sketch(img): #Converting image into grayscale
 img_gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
 
 #Clean up image using Gausian Blur
 img_gray_blur = cv2.GaussianBlur(img_gray, (5,5), 0)
 
 #Extracting edges
 canny_edges = cv2.Canny(img_gray_blur, 10, 70)
 
 #Do an invert binarize the image
 ret, mask = cv2.threshold(canny_edges, 70, 255,  cv2.THRESH_BINARY_INV)
 return mask
cap=cv2.VideoCapture(0)while True:
 ret, frame = cap.read()
 cv2.imshow(‘Our Live Sketcher’, sketch(frame))
 
 k= cv2.waitKey(13)
 if k == 27:
 break
 
cap.release()
cv2.destroyAllWindows()

Results
Voila! Your Live Sketch!!!

Live Sketch through WebCam

Implementation
You can also implement to convert an image to add some filter or make your own live filter example:

1) Sepia filter
2) Cartoonization filter
3) Duo-Tone Image
4) Embossing Image
5) Colourful Live Sketcher
6) Oldify Image
7) Low Poly Image
8) Grayscale Filter
9) Water Colour Filter

Thanks for reading this!!!

Connect with me on:

Twitter: https://twitter.com/aree_yarr_sharu

Linkedin: https://t.co/g0A8rcvcYo?amp=1