Image Thresholding Techniques with OpenCV

Overview

Thresholding is a fundamental technique in image processing that allows you to segment images by setting pixel values to either a maximum or minimum value based on a threshold. This is useful for separating objects from the background, document scanning, and more. In this tutorial, you'll learn how to use both basic and adaptive thresholding methods in OpenCV with Python.

Prerequisites

You'll need the following Python libraries:

• opencv-python: For image processing operations
• matplotlib: For displaying images
• numpy: For numerical operations

Install them with:

pip install opencv-python matplotlib numpy

Step 1: Reading and Displaying the Image

Let's start by reading an image and converting it to grayscale:

import cv2
import numpy as np
import matplotlib.pyplot as plt

# Read the image in grayscale mode
gray_img = cv2.imread('Assets/kerem.jpeg', 0)
plt.imshow(gray_img, cmap='gray')
plt.show()

Step 2: Basic Thresholding

Thresholding assigns pixel values based on a threshold value. Any value below the threshold is set to 0, and values above the threshold are set to a maximum value.

# Apply a basic threshold
ret, thresh1 = cv2.threshold(gray_img, 127, 255, cv2.THRESH_TOZERO)
print(ret)
print(thresh1)
plt.imshow(thresh1, cmap='gray')
plt.show()

Common Thresholding Types in OpenCV

• THRESH_TRUNC: If pixel intensity is greater than the threshold, it is set to the threshlod value.
• THRESH_BINARY: If pixel intensity is greater than the threshold, it is set to 255, else set to 0.
• THRESH_TOZERO: Pixel intensity is set to 0 for all pixels less than the threshold value.

Step 3: Thresholding a Crossword Image

Let's try thresholding on a different image:

# Read a crossword image
gray_crossword = cv2.imread('Assets/crossword.jpg', 0)

def show(img):
    fig = plt.figure(figsize=(6,6))
    ax = fig.add_subplot(111)
    ax.imshow(img, cmap='gray')
    plt.show()

show(gray_crossword)

# Apply binary thresholding
ret, th1 = cv2.threshold(gray_crossword, 180, 255, cv2.THRESH_BINARY)
show(th1)

Step 4: Adaptive Thresholding

Adaptive thresholding calculates the threshold for a small region of the image, allowing for different threshold values in different areas. This is useful for images with varying lighting conditions.

# Adaptive thresholding
th2 = cv2.adaptiveThreshold(
    gray_crossword, 255, cv2.ADAPTIVE_THRESH_MEAN_C, cv2.THRESH_BINARY, 11, 8)
show(th2)

Parameters:

• src: Source image
• maxValue: Maximum value to use with the THRESH_BINARY and THRESH_BINARY_INV thresholding types
• adaptiveMethod: Adaptive thresholding algorithm to use (e.g., ADAPTIVE_THRESH_MEAN_C or ADAPTIVE_THRESH_GAUSSIAN_C)
• thresholdType: Type of thresholding (e.g., THRESH_BINARY or THRESH_BINARY_INV)
• blockSize: Size of a pixel neighborhood used to calculate the threshold value (must be odd)
• C: Constant subtracted from the mean or weighted mean

Step 5: Blending Thresholded Images

You can blend two thresholded images to combine their effects:

# Blend two images with different weights
blended = cv2.addWeighted(src1=th1, alpha=0.6, src2=th2, beta=0.4, gamma=0)
show(blended)