Image Thresholding in OpenCV | Binary, Adaptive, Otsu’s Method with Python

Chapter 2.2: Image Thresholding (Binary, Adaptive, Otsu’s Method)

Thresholding is a technique used to convert an image into a binary format (black & white) by comparing pixel values with a threshold. OpenCV provides several thresholding methods such as Binary, Adaptive, and Otsu’s thresholding.

⚫ Binary Thresholding

In binary thresholding, if the pixel value is greater than a threshold, it is set to white (255), otherwise black (0).

import cv2

img = cv2.imread("sample.jpg", cv2.IMREAD_GRAYSCALE)

# Apply binary thresholding
_, binary = cv2.threshold(img, 127, 255, cv2.THRESH_BINARY)

cv2.imshow("Binary Threshold", binary)
cv2.waitKey(0)
cv2.destroyAllWindows()

🟡 Adaptive Thresholding

Adaptive thresholding calculates thresholds for smaller regions of the image, making it useful for images with varying lighting conditions.

adaptive = cv2.adaptiveThreshold(img, 255, cv2.ADAPTIVE_THRESH_MEAN_C,
                                 cv2.THRESH_BINARY, 11, 2)

cv2.imshow("Adaptive Threshold", adaptive)
cv2.waitKey(0)
cv2.destroyAllWindows()

🔵 Otsu’s Thresholding

Otsu’s method automatically calculates the best threshold value by minimizing variance between foreground and background.

# Otsu’s thresholding
_, otsu = cv2.threshold(img, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)

cv2.imshow("Otsu Threshold", otsu)
cv2.waitKey(0)
cv2.destroyAllWindows()

✅ In this lesson, you learned about Binary, Adaptive, and Otsu’s thresholding. These techniques are essential for image segmentation, document processing, and object recognition.
👉 Next, we will explore image filtering techniques (blur, Gaussian, median, bilateral).