Edge-reinforced color space maximally stable extremal region detection method

Abstract

The invention discloses an edge-reinforced color space maximally stable extremal region detection method which comprises the steps of using a multiscale probability of boundary (mPb) edge detection method to detect edges of a color image to obtain edge information and obtain distance a weighting coefficient according to the edge information and a distance conversion formula, calculating the surface characteristics and the distance value of difference between adjacent pixels in a color space, using the distance weighting coefficient to weight the distance value to obtain a final distance set, obtaining a distance threshold set according to the distribution situation of the final distance set, combining the adjacent pixels with the distance value smaller than the threshold to the same region by continuously increasing a distance threshold, and extracting regions with the size change rate reaching to the local minimum along with the change of the threshold as the maximally stable region characteristics. The edge-reinforced color space maximally stable extremal region detection method comprehensively uses edge information and color information of the image to further improve fuzzy mapping invariance in keeping the self invariance of the maximally stable extremal region to the greatest degree.

Description

technical field [0001] The invention belongs to the field of image local feature extraction, in particular to an edge-enhanced method for detecting the most stable extremum region in color space. Background technique [0002] Image local invariant features have become a research hotspot in the field of image processing and computer vision in recent years. After artificial intelligence and neural networks, it has once again ignited people's enthusiasm for machine intelligence research. The core of image local invariant feature research is "invariance", that is, when recognizing an object, no matter how far or near the object is, and whether the object is rotated or observed from different angles, it can Correctly identify the object. Common invariances include perspective invariance, scale invariance, rotation invariance, illumination invariance, affine invariance, etc., but so far there is no local feature that has all the above invariances, and general local features can o...

AI Technical Summary

Problems solved by technology

[0004] In order to solve the above problems, the object of the present invention is to provide an edge-enhanced method for detecting the most stable extremum region in color space, which comprehensively utilizes the edge information and color information of the image, and retains the invariance of the most stable extremum region itself as much as possible. At the same time, it further improves its invariance in fuzzy transformation, and overcomes the disadvantage of the existing most stable extremum region MSER with poor invariance to fuzzy changes.

Images