Image processing involves using mathematical algorithms to perform operations on digital images to enhance their quality, extract useful information, or separate objects or regions of interest.
Here are some common algorithms used for these tasks:
Image Enhancement:
- Histogram Equalization: enhances the contrast of the image by spreading out the intensity values.
- Filtering: such as median, Gaussian, or bilateral filter, which can remove noise or sharpen the image.
Image Segmentation:
- Thresholding: separates an image into its foreground and background by defining a threshold value.
- Watershed: a marker-based algorithm that segments an image by treating it as a topographic map and finding the watershed lines.
- Clustering: such as K-means or Expectation Maximization, which groups similar pixels together to form segments.
Feature Extraction:
- Edge Detection: detects the boundaries of objects in an image by finding the points of rapid intensity change.
- SIFT (Scale-Invariant Feature Transform): detects and describes local features in an image, which can be used for object recognition or matching.
- HOG (Histogram of Oriented Gradients): extracts features by counting the occurrences of gradient orientation in localized portions of an image.
Code Matlab:
% Load an image
I = imread('image.jpg');
% Image Enhancement
% Convert the image to grayscale
I_gray = rgb2gray(I);
% Adjust the image contrast using histogram equalization
I_eq = histeq(I_gray);
% Image Segmentation
% Apply a threshold to segment the image
level = graythresh(I_gray);
I_bw = im2bw(I_gray, level);
% Feature Extraction
% Detect corners in the image using the Harris corner detector
corners = detectHarrisFeatures(I_gray);
% Display the results
figure;
subplot(2,2,1), imshow(I), title('Original Image');
subplot(2,2,2), imshow(I_eq), title('Enhanced Image');
subplot(2,2,3), imshow(I_bw), title('Segmented Image');
subplot(2,2,4), imshow(I_gray), title('Grayscale Image'); hold on;
plot(corners.selectStrongest(50));
title('Feature Extraction: Corners');