SIFT characteristic reducing method facing close repeated image matching

Abstract

The invention discloses an SIFT characteristic reducing method facing close repeated image matching, comprising the following steps of carrying out gaussian kernel convolution processing on each image of an image library to obtain image key points; carrying out gaussian uniformization on the contrast ratio and the principal curvature ratio of the key points and linear weighting to obtain significance degree; sequencing from small to large according to the significance degree of the key points, and selecting a number of the key points specified by uses to realize reduction; generating descriptors for the reduced key points according to the positions, the dimensions and the directional information of the reduced key points to obtain SIFT characteristics; establishing an image library index for all SIFT characteristic sets by a partial sensitivity hash technique, and providing query function of the close repeated image matching. The invention utilizes the research and the realization results of an image partial characteristic technique and the partial sensitivity hash technique, can conveniently and fast provide the query ability of the close image matching and enable the users to regulate the weighting coefficient of a reduction algorithm and the upper limit threshold value of the number of the SIFT characteristics according to application requirements so as to provide the best property.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a SIFT feature trimming method for near-repetitive image matching. Background technique [0002] Near-duplicate image matching is an important problem in many application fields such as image copyright monitoring and image retrieval. Different domains have different definitions of near-duplicate images. The near-repetitive image targeted by this patent is generated by rotating, scaling, shearing, brightness and contrast changes, color adjustments, etc. of the source image. With the birth of various image editing tools, this has also become the main source of near-duplicate images on the Internet. [0003] In recent years, SIFT technology has achieved many achievements in the field of near-repetitive image matching, but there are still some problems. SIFT technology was first proposed in the field of object recognition. It is characterized by the ability to maintain a ...

AI Technical Summary

Problems solved by technology

[0004] However, in the previous image matching research, the importance of key point clipping is often ignored, usually based on experience, adjust the lower limit of contrast and the upper threshold of principal curvature ratio, roughly control the number of key points and filter unstable key points; Or simply by sorting the contrast of key points to obtain a controllable number of relatively effective key points

These oversimplified processing may lead to poor matching ability of the selected key points, causing unnecessary losses to the accuracy of the system

Images