feat: add ORB feature detection, BFMatcher, and homography computation #109
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Adds support for: - ORB_create() with full parameter support for feature detection - detectAndCompute() for keypoint + descriptor extraction - BFMatcher_create() for Hamming distance matching - matchBF() and knnMatchBF() for descriptor matching - findHomography() with RANSAC support New object types: - ORB (feature detector) - BFMatcher (descriptor matcher) - KeyPointVector (detected keypoints) - DMatchVector (descriptor matches) - DMatchVectorVector (for knnMatch results) New constants: - ORBScoreType (HARRIS_SCORE, FAST_SCORE) - NormTypes (NORM_HAMMING, NORM_L2, etc.) - HomographyMethod (RANSAC, LMEDS, RHO) These functions enable robust image registration for applications like photo scanning and panorama stitching. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
calib3d module is not included in FastOpenCV-iOS pod. Replaced findHomography with findHomographyFromMatches which uses getPerspectiveTransform from imgproc module. The caller must provide exactly 4 point pairs (pre-filtered best matches).
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary
This PR adds support for ORB (Oriented FAST and Rotated BRIEF) feature detection and matching, which enables robust image registration for applications like glare-free photo scanning.
New Functions
ORB_create- Creates an ORB feature detector with configurable parameters (nfeatures, scaleFactor, nlevels, etc.)detectAndCompute- Detects keypoints and computes descriptors using the ORB detectorBFMatcher_create- Creates a Brute-Force descriptor matcher (with NORM_HAMMING for ORB)matchBF- Finds the best match for each descriptorknnMatchBF- Finds k-nearest matches (useful for Lowe's ratio test)findHomographyFromMatches- Computes perspective transformation from 4 matched point pairsNew Types
ORB- ORB detector objectBFMatcher- Brute-Force matcher objectKeyPointVector- Vector of detected keypointsDMatchVector- Vector of descriptor matchesDMatchVectorVector- Vector of match vectors (for knnMatch)New Constants
ORBScoreType- HARRIS_SCORE, FAST_SCOREHomographyMethod- DEFAULT, LMEDS, RANSAC, RHO (for future use)Implementation Notes
cv::getPerspectiveTransforminstead ofcv::findHomographysince the calib3d module is not available in FastOpenCV-iOS podfindHomographyFromMatchesfunction expects exactly 4 point pairs (caller should pre-filter using ratio test and select well-distributed points)toJSValueconverters added for keypoint and match data structuresUse Case
This enables Google PhotoScan-style image alignment using feature matching + homography:
Testing
🤖 Generated with Claude Code