Constructor

obj = cv.FeaturesMatcher(matcherType)
obj = cv.FeaturesMatcher(matcherType, 'OptionName',optionValue, ...)

Input

• matcherType One of:
  • BestOf2NearestMatcher A "best of 2 nearest" matcher. Features matcher which finds two best matches for each feature and leaves the best one only if the ratio between descriptor distances is greater than the threshold MatchConf.
  • BestOf2NearestRangeMatcher
  • AffineBestOf2NearestMatcher A "best of 2 nearest" matcher that expects affine trasformation between images. Features matcher similar to BestOf2NearestMatcher which finds two best matches for each feature and leaves the best one only if the ratio between descriptor distances is greater than the threshold MatchConf. Unlike BestOf2NearestMatcher this matcher uses affine transformation (affine trasformation estimate will be placed in matches_info).

Options

The following are options accepted by all matchers:

• TryUseGPU Should try to use GPU or not. default false
• MatchConf Match distances ration threshold. Confidence for feature matching step. default 0.3
• NumMatchesThresh1 Minimum number of matches required for the 2D projective transform estimation used in the inliers classification step. default 6
• NumMatchesThresh2 Minimum number of matches required for the 2D projective transform re-estimation on inliers. default 6

The following are options for the various algorithms:

BestOf2NearestRangeMatcher

• RangeWidth Range width. default 5

AffineBestOf2NearestMatcher

• FullAffine whether to use full affine transformation with 6 degress of freedom (cv.estimateAffine2D) or reduced transformation with 4 degrees of freedom (cv.estimateAffinePartial2D) using only rotation, translation and uniform scaling. default false