
Introduction to SURF (Speeded-Up Robust Features)
Learn the basics of SURF algorithm, a speeded-up version of SIFT that uses box filters and integral images to achieve 3x speedup while maintaining comparable performance.
Category
64 posts

Learn the basics of SURF algorithm, a speeded-up version of SIFT that uses box filters and integral images to achieve 3x speedup while maintaining comparable performance.

Learn about the concepts of SIFT algorithm. We cover scale-space extrema detection, keypoint localization, orientation assignment, and finding SIFT keypoints with OpenCV.

Learn about Shi-Tomasi Corner Detector, an improvement over Harris. We cover cv.goodFeaturesToTrack() to find the N strongest corners in an image.

Learn about Harris Corner Detection concepts and implementation. We cover cv.cornerHarris() and cv.cornerSubPix() for sub-pixel accuracy corner detection.

Learn what image features are, why they are important, and why corners are good features. Covers the basics of feature detection and description.

Learn to use marker-based image segmentation using watershed algorithm. We cover cv.watershed() with distance transform to segment touching objects.

Learn to use GrabCut algorithm to extract foreground in images. We cover cv2.grabCut() with both rectangle and mask initialization modes.

Learn about the hierarchy of contours, i.e. the parent-child relationship in Contours. We cover RETR_LIST, RETR_EXTERNAL, RETR_CCOMP and RETR_TREE flags with examples.

Learn about convexity defects, finding shortest distance from a point to a polygon, and matching different shapes. We cover cv2.convexityDefects(), cv2.pointPolygonTest() and cv2.matchShapes().