Python Tutorial

Morphological operations can be used for extracting image components that are helpful for the description and representation of the shape of a region. Morphological operations are the fundamental tasks that are dependent on the image shape. It typically takes place on binary images. It requires two sources of data, the first of which is an input image, and the second is known as the structuring component.

Morphological operators are able to take an input image as well as the structuring component as input. The two elements are then combined with set operators. The objects of the image input are treated based on the characteristics of the shape of the image. These can be encoded by the structuring part.

The Opening is similar to erosion, as it removes bright foreground pixels from the edges of the foreground pixels. The result of this operation is to secure the foreground region, which is similar to the structuring component or even completely contains the structuring component while taking away all other regions in the foreground of pixels. The opening operation is utilized for the removal of internal noise within an image.

The Opening is an erosion operation that is followed by dilation.

Morphological Operations in Image Processing in Python

Required modules and libraries:

To perform the morphological operation in the image processing, we will need the following module:

1. cv2

Install: !pip3 install cv2

2. NumPy

Install: !pip3 install numpy

Syntax:

Syntax of cv2morphology is:

Parameters:

image: It takes input Image array.
MORPH_OPEN: It is used for applying the Morphological Opening operation.
kernel: It is used for structuring element.

Here below is the Python code for explaining Opening Morphological Operation -

Code:

# First, we will import the required libraries:
import cv2 as CV
import numpy as nmp 
 
# here, it will return video from the first webcam on our computer. 
screenRead1 = CV.VideoCapture(0)
# now, it will loop runs if capturing has been initialized.
while(1):
    # now, it is reading frames from a camera
    _, image = screenRead1.read()
     
    # Converting to HSV color space because OCV only reads colors as BGR 
    # and the frame is converted to hsv
    hsv = CV.cvtColor(image, CV.COLOR_BGR2HSV)
     
    # here, we will define the range of masking
    blue_1 = nmp.array([111, 52, 52])
    blue_2 = n,p.array([131, 257, 257])
     
    # Now, we will initialize the mask forconvoluting over input image
    mask = CV.inRange(hsv, blue_1, blue_2)
 
    # then, we will pass the bitwise_and over each pixel convoluted
    res = CV.bitwise_and(image, image, mask = mask)
     
    # Here, we will define the kernel that is Structuring element
    kernel = nmp.ones((5, 5), nmp.uint8)
     
    # Now, we will define the opening function over the image and structuring   # element
    opening = CV.morphologyEx(mask, CV.MORPH_OPEN, kernel)
    
    # The mask and opening operation is shown in the window
    CV.imshow('Mask', mask)
    CV.imshow('Opening', opening)
     
    # Now, it will wait for 'a' key to stop the program
    if CV.waitKey(1) & 0xFF == ord('a'):
        break
 
# De-allocate any associated memory usage 
CV.destroyAllWindows()
 
# At last, we will close the window and Release webcam
screenRead.release()

Output

Input image

Mask:

Output Frame

The system recognizes the designated blue book as the input and removes and eliminates the noise inside the area of interest with the aid of the Opening function.

Next TopicRole of Python in Artificial Intelligence

← prev next →