Implementing Shi-Tomasi Corner Detector in OpenCV Python

When it comes to image processing, detecting corners in an image can be a crucial step. Corners are, in general, regions in an image where two or more edges meet. The Shi-Tomasi corner detector is an algorithm widely used to detect corners in images. In this article, we will explore how to implement the Shi-Tomasi corner detector in OpenCV Python.

What is Shi-Tomasi Corner Detector?

The Shi-Tomasi corner detector is a modification of the Harris corner detector algorithm. The algorithm calculates the minimum eigenvalue for each pixel in the image, which corresponds to the corner’s strength. The algorithm works by computing the gradient of the image, and then calculating the sum of the squares of the gradient in a given window.

The Shi-Tomasi corner detector selects only the corners with the highest quality, which are the ones with the lowest eigenvalue. This approach drastically improves the detection of key features in an image, especially in low light conditions or images with low contrast.

Implementing Shi-Tomasi Corner Detector in OpenCV Python

To implement Shi-Tomasi corner detector in OpenCV Python, follow these steps:

1. Import the OpenCV library:

import cv2

2. Load the image:

img = cv2.imread('image.jpg')

3. Convert the image to grayscale:

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

4. Define the Shi-Tomasi parameters:

max_corners = 100
quality_level = 0.3
min_distance = 7

5. Apply corner detection using the cv2.goodFeaturesToTrack() function:

corners = cv2.goodFeaturesToTrack(gray, max_corners, quality_level, min_distance)

6. Convert the corners to integers:

corners = np.int0(corners)

7. Iterate through each corner and draw a circle:

for corner in corners:
    x, y = corner.ravel()
    cv2.circle(img, (x, y), 3, (0, 0, 255), -1)

8. Display the image:

cv2.imshow('Shi-Tomasi Corner Detector',img)
cv2.waitKey(0)

Sample Code

import cv2
import numpy as np

# Load image
img = cv2.imread('image.jpg')

# Convert to grayscale
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

# Define Shi-Tomasi parameters
max_corners = 100
quality_level = 0.3
min_distance = 7

# Apply corner detection
corners = cv2.goodFeaturesToTrack(gray, max_corners, quality_level, min_distance)

# Convert corners to integers
corners = np.int0(corners)

# Iterate through corners and draw a circle
for corner in corners:
    x, y = corner.ravel()
    cv2.circle(img, (x, y), 3, (0, 0, 255), -1)

# Display image
cv2.imshow('Shi-Tomasi Corner Detector',img)
cv2.waitKey(0)

Conclusion

In this article, we have explored the Shi-Tomasi Corner Detector algorithm and learned how to implement it in OpenCV Python. Implementing this algorithm can significantly impact our image processing pipelines, especially in applications where we want to detect key features in an image with low contrast or low lighting conditions. With the help of OpenCV Python, we can efficiently detect corners and extract the features we need from an image with ease.