Applying Thresholding Techniques with OpenCV using Python

In the field of image processing, thresholding is a widely used technique to separate objects from the background. This technique involves converting a grayscale image into a binary image by dividing the pixels into two categories: foreground and background based on a threshold value. OpenCV, a popular library for computer vision tasks, provides several thresholding techniques to handle different scenarios. In this article, we will explore some of the common thresholding techniques and demonstrate how to apply them using Python and OpenCV.

1. Simple Thresholding

Simple thresholding is the most basic form of thresholding. It converts the grayscale image into a binary image based on a fixed threshold value. Pixels with intensity values above the threshold are set to a maximum value (typically white), while pixels below the threshold are set to a minimum value (typically black).

import cv2

# Read the grayscale image
image = cv2.imread('image.jpg', 0)

# Apply simple thresholding
_, thresholded = cv2.threshold(image, 127, 255, cv2.THRESH_BINARY)

# Display the thresholded image
cv2.imshow('Thresholded Image', thresholded)
cv2.waitKey(0)
cv2.destroyAllWindows()

In the code snippet above, we first read the grayscale image using cv2.imread('image.jpg', 0). Then, we apply simple thresholding using cv2.threshold() function. The second argument is the threshold value (127 in this case), the third argument is the maximum value assigned to pixels above the threshold (255 in this case), and the final argument specifies the thresholding method (cv2.THRESH_BINARY). Finally, we display the thresholded image using cv2.imshow() and cv2.waitKey() functions.

2. Adaptive Thresholding

Simple thresholding may not always produce satisfactory results, especially when the lighting conditions vary across different regions of the image. Adaptive thresholding addresses this issue by calculating a different threshold for each pixel based on the local neighborhood. OpenCV provides two types of adaptive thresholding techniques: ADAPTIVE_THRESH_MEAN_C and ADAPTIVE_THRESH_GAUSSIAN_C.

import cv2

# Read the grayscale image
image = cv2.imread('image.jpg', 0)

# Apply adaptive thresholding
thresholded_mean = cv2.adaptiveThreshold(image, 255, cv2.ADAPTIVE_THRESH_MEAN_C, cv2.THRESH_BINARY, 11, 2)
thresholded_gaussian = cv2.adaptiveThreshold(image, 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C, cv2.THRESH_BINARY, 11, 2)

# Display the thresholded images
cv2.imshow('Mean Adaptive Thresholding', thresholded_mean)
cv2.imshow('Gaussian Adaptive Thresholding', thresholded_gaussian)
cv2.waitKey(0)
cv2.destroyAllWindows()

In the code snippet above, we read the grayscale image as before. Then, we apply adaptive thresholding using cv2.adaptiveThreshold() function. The second argument is the maximum value assigned to pixels above the threshold, and the thresholding method is specified as either cv2.ADAPTIVE_THRESH_MEAN_C or cv2.ADAPTIVE_THRESH_GAUSSIAN_C. The next two arguments are the block size (11 in this case) and the constant to subtract from the mean or weighted mean (2 in this case). Finally, we display the thresholded images using cv2.imshow() and cv2.waitKey() functions.

Conclusion

Thresholding is a powerful technique for segmenting objects from images based on their intensity values. In this article, we have discussed simple thresholding and adaptive thresholding techniques using OpenCV and Python. These techniques can be valuable in various computer vision applications such as image segmentation, object detection, and image enhancement. Experiment with different thresholding methods and parameters to achieve the desired results for your specific use case.