Image Processing
Algorithms that transform or analyze images.
Image processing algorithms work directly with pixels to detect edges, smooth noise, segment regions, or fill connected areas.
Sobel Edge Detection
Finds edges by measuring horizontal and vertical brightness changes.
Gaussian Blur
Smooths an image by convolving it with a Gaussian kernel.
Thresholding
Separates pixels into groups based on whether they cross a chosen value.
Flood Fill
Fills a connected region starting from one seed point.
Bilateral Filter
Smooths images while preserving edges by considering distance and color similarity.
Canny Edge Detection
Finds thin edges by smoothing, measuring gradients, suppressing nonmaxima, and linking thresholds.
Histogram Equalization
Improves contrast by spreading image intensities across the available range.
Hough Transform
Detects shapes by voting in a parameter space.
Laplacian Edge Detection
Finds edges by measuring second-order brightness changes.
Morphological Erosion
Shrinks bright or foreground regions using a local structuring element.
Difference of Gaussians
Highlights image structures by subtracting two Gaussian-blurred versions at different scales.
Connected Component Labeling
Assigns labels to contiguous foreground regions in a binary or segmented image.
Unsharp Masking
Sharpens an image by adding back high-frequency detail separated by blur.
Template Matching
Finds image regions that resemble a smaller template patch.
GrabCut
Segments foreground from background using graph cuts and color models.
Lucas-Kanade Optical Flow
Estimates local image motion by solving small least-squares systems over patches.
Horn-Schunck Optical Flow
Estimates dense optical flow by combining brightness constancy with smoothness.
Image Pyramid
Stores progressively downsampled versions of an image for multi-scale processing.
Integral Image
Precomputes cumulative pixel sums so rectangular sums are answered in constant time.
Morphological Dilation
Expands bright or foreground regions according to a structuring element.
Watershed Segmentation
Segments regions by treating image intensity as a terrain flooded from markers.
FAST Corner Detection
FAST Corner Detection detects repeatable local image features.
Harris Corner Detection
Harris Corner Detection detects repeatable local image features.
Shi-Tomasi Corner Detection
Shi-Tomasi Corner Detection detects repeatable local image features.
SIFT Feature Detection
SIFT Feature Detection detects repeatable local image features.
SURF Feature Detection
SURF Feature Detection detects repeatable local image features.
ORB Feature Detection
ORB Feature Detection detects repeatable local image features.
RANSAC Image Homography
RANSAC Image Homography transforms image pixels into features, masks, or restored output.
Otsu Thresholding
Otsu Thresholding separates pixels by intensity or local contrast.
Adaptive Thresholding
Adaptive Thresholding separates pixels by intensity or local contrast.
K-Means Image Segmentation
K-Means Image Segmentation partitions an image into meaningful regions.
Mean Shift Segmentation
Mean Shift Segmentation partitions an image into meaningful regions.
Region Growing
Region Growing transforms image pixels into features, masks, or restored output.
Active Contour Model
Active Contour Model partitions an image into meaningful regions.
Anisotropic Diffusion
Anisotropic Diffusion transforms image pixels into features, masks, or restored output.
Richardson-Lucy Deconvolution
Richardson-Lucy Deconvolution transforms image pixels into features, masks, or restored output.
Retinex Algorithm
Retinex Algorithm transforms image pixels into features, masks, or restored output.
HOG Feature Descriptor
Describes local image shape by accumulating gradient-orientation histograms over cells.
SIFT
Detects and describes scale-invariant image keypoints from local gradient structure.
SURF
SURF addresses fast feature detection by approximating Hessian responses with integral-image box filters.
ORB
Detects oriented FAST keypoints and describes them with rotated binary BRIEF-style tests.
RANSAC Homography Estimation
Estimates a planar homography while rejecting mismatched feature correspondences.