The syntax for applying simple thresholding in OpenCV is as follows:<\/strong><\/p>\n Code with Explanation:<\/strong><\/p>\n Let’s illustrate simple thresholding using binary thresholding as an example:<\/strong><\/p>\n Original Image :<\/strong><\/p>\n Binary<\/strong><\/p>\n Inverse Binary<\/strong><\/p>\n Truncated<\/strong><\/p>\n Tozero<\/strong><\/p>\n Tozero inverse<\/strong><\/p>\n <\/p>\n In this code, we read a grayscale image and then apply all five types of simple thresholding using the cv2.threshold() function. We create a separate window for each thresholding output and display them using the cv2.imshow() function. The images will be displayed one after the other, and you can close the window by pressing any key.<\/p>\n In the ever-evolving world of image processing, adaptive thresholding stands as a versatile technique for tackling varying lighting conditions and image complexities. Unlike simple thresholding, where a fixed threshold is applied globally to the entire image, adaptive thresholding employs dynamic thresholds at each pixel’s local neighborhood. This adaptive nature empowers the algorithm to adapt to local changes, making it highly effective in scenarios with uneven illumination and varying contrasts.<\/p>\n Types of Adaptive Thresholding and Their Functions in OpenCV:<\/strong><\/p>\n Calculates the mean of the pixel intensities in a local neighborhood and uses it as the threshold value. Each pixel is compared to its local mean, which helps in handling varying illumination conditions.<\/p>\n Identical to adaptive mean thresholding, but uses a Gaussian window to derive the threshold value based on the weighted sum of the pixel intensities in the immediate area rather than the mean.<\/p>\n Syntax and Parameters:<\/strong><\/p>\n The syntax for applying adaptive thresholding in OpenCV is as follows:<\/strong><\/p>\n Code with Explanation:<\/strong><\/p>\n Original Image :<\/strong><\/p>\n Adaptive Mean Thresholding :<\/strong><\/p>\n Adaptive Gaussian Thresholding :<\/strong><\/p>\n In this code, we read a grayscale image and then apply both types of adaptive thresholding using the cv2.adaptiveThreshold() function. We create separate windows for each adaptive thresholding output and display them using the cv2.imshow() function. The images will be displayed one after the other, and you can close the window by pressing any key.<\/p>\n In the realm of image processing, adding borders to an image plays a crucial role in various applications. OpenCV offers a powerful function called copyMakeBorder(), which allows you to expand the image canvas and create a border of desired size and style. This function acts as a shield, safeguarding the image from unintended distortions and enhancing its visual appearance.<\/p>\n Syntax and Parameters:<\/strong><\/p>\n The syntax for using copyMakeBorder() in OpenCV is as follows:<\/strong><\/p>\n Code with Explanation :<\/strong><\/p>\n Let’s demonstrate copyMakeBorder() by adding a red border to an image:<\/strong><\/p>\n Original image :<\/strong><\/p>\nretval, threshold_output = cv2.threshold(input_image, threshold_value, max_value, threshold_type)<\/pre>\n
\n
import cv2\r\nimport numpy as np\r\n\r\n# Read the grayscale image\r\nimage = cv2.imread(r\"C:\\Users\\sanji\\Desktop\\DataFlair\\Thresholding in Opencv\\nature-img.jpg\", cv2.IMREAD_GRAYSCALE)\r\n\r\n# Threshold values\r\nthreshold_value = 150\r\nmax_value = 255\r\n\r\n\r\n\r\n# Binary Thresholding\r\n_, binary_threshold_output = cv2.threshold(image, threshold_value, max_value, cv2.THRESH_BINARY)\r\n\r\n# Inverse Binary Thresholding\r\n_, inverse_binary_threshold_output = cv2.threshold(image, threshold_value, max_value, cv2.THRESH_BINARY_INV)\r\n\r\n# Truncated Thresholding\r\n_, truncated_threshold_output = cv2.threshold(image, threshold_value, max_value, cv2.THRESH_TRUNC)\r\n\r\n# Tozero Thresholding\r\n_, tozero_threshold_output = cv2.threshold(image, threshold_value, max_value, cv2.THRESH_TOZERO)\r\n\r\n# Tozero Inverse Thresholding\r\n_, tozero_inverse_threshold_output = cv2.threshold(image, threshold_value, max_value, cv2.THRESH_TOZERO_INV)\r\n\r\n# Create a window for each type of thresholding\r\ncv2.namedWindow('Binary Thresholding', cv2.WINDOW_NORMAL)\r\ncv2.namedWindow('Inverse Binary Thresholding', cv2.WINDOW_NORMAL)\r\ncv2.namedWindow('Truncated Thresholding', cv2.WINDOW_NORMAL)\r\ncv2.namedWindow('Tozero Thresholding', cv2.WINDOW_NORMAL)\r\ncv2.namedWindow('Tozero Inverse Thresholding', cv2.WINDOW_NORMAL)\r\n\r\n# Display the thresholded images\r\ncv2.imshow('Binary Thresholding', binary_threshold_output)\r\ncv2.imshow('Inverse Binary Thresholding', inverse_binary_threshold_output)\r\ncv2.imshow('Truncated Thresholding', truncated_threshold_output)\r\ncv2.imshow('Tozero Thresholding', tozero_threshold_output)\r\ncv2.imshow('Tozero Inverse Thresholding', tozero_inverse_threshold_output)\r\n\r\ncv2.waitKey(0)\r\ncv2.destroyAllWindows()<\/pre>\n![\"Thresholding](\"https:\/\/data-flair.training\/blogs\/wp-content\/uploads\/sites\/2\/2023\/11\/Thresholding-original-image-.webp\")
<\/a><\/p>\n<\/a><\/p>\n<\/a><\/p>\n<\/a><\/p>\n<\/a><\/p>\n<\/a><\/p>\nAdaptive Thresholding in OpenCV: Mastering Image Segmentation with Dynamic Thresholds<\/h3>\n
1. Adaptive Mean Thresholding: cv2.ADAPTIVE_THRESH_MEAN_C<\/h4>\n
2. Adaptive Gaussian Thresholding: cv2.ADAPTIVE_THRESH_GAUSSIAN_C<\/h4>\n
threshold_output = cv2.adaptiveThreshold(input_image, max_value, adaptive_method, threshold_type, block_size, constant_value)<\/pre>\n
\n
import cv2\r\n\r\n# Read the grayscale image\r\nimage = cv2.imread(r\"C:\\Users\\sanji\\Desktop\\DataFlair\\Thresholding in Opencv\\nature-img.jpg\", cv2.IMREAD_GRAYSCALE)\r\n\r\n# Maximum pixel value for the thresholded image\r\nmax_value = 255\r\n\r\n# Adaptive Mean Thresholding\r\nblock_size_mean = 11\r\nconstant_value_mean = 2\r\nadaptive_mean_output = cv2.adaptiveThreshold(image, max_value, cv2.ADAPTIVE_THRESH_MEAN_C, cv2.THRESH_BINARY, block_size_mean, constant_value_mean)\r\n\r\n# Adaptive Gaussian Thresholding\r\nblock_size_gaussian = 11\r\nconstant_value_gaussian = 2\r\nadaptive_gaussian_output = cv2.adaptiveThreshold(image, max_value, cv2.ADAPTIVE_THRESH_GAUSSIAN_C, cv2.THRESH_BINARY, block_size_gaussian, constant_value_gaussian)\r\n\r\n# Create windows for each type of adaptive thresholding\r\ncv2.namedWindow('Adaptive Mean Thresholding', cv2.WINDOW_NORMAL)\r\ncv2.namedWindow('Adaptive Gaussian Thresholding', cv2.WINDOW_NORMAL)\r\n\r\n# Display the adaptive thresholded images\r\ncv2.imshow('Adaptive Mean Thresholding', adaptive_mean_output)\r\ncv2.imshow('Adaptive Gaussian Thresholding', adaptive_gaussian_output)\r\n\r\ncv2.waitKey(0)\r\ncv2.destroyAllWindows()<\/pre>\n<\/a><\/p>\n<\/a><\/p>\n<\/a><\/p>\nAdding Borders in OpenCV: Amplifying Image Boundaries with copyMakeBorder() :<\/h3>\n
border_image = cv2.copyMakeBorder(src, top, bottom, left, right, border_type, value)<\/pre>\n
\n
import cv2\r\nimport numpy as np\r\n\r\n# Read the image\r\nimage = cv2.imread(r\"C:\\Users\\sanji\\Desktop\\DataFlair\\Thresholding in Opencv\\nature-img.jpg\")\r\n\r\n# Define the border size in pixels\r\nborder_size = 20\r\n\r\n# Define the border color (in BGR format)\r\nborder_color = (0, 0, 255) # Red color\r\n\r\n# Add the border using copyMakeBorder()\r\nborder_image = cv2.copyMakeBorder(image, border_size, border_size, border_size, border_size, cv2.BORDER_CONSTANT, value=border_color)\r\n\r\n# Display the original image\r\ncv2.imshow('Original Image', image)\r\ncv2.waitKey(0)\r\n\r\n# Display the bordered image\r\ncv2.imshow('Bordered Image', border_image)\r\ncv2.waitKey(0)\r\n\r\ncv2.destroyAllWindows()<\/pre>\n