论文信息 - A target recognition matching method with high reliability

A target recognition matching method with high reliability

Grayness correlation is a method with high reliability for target recognition. However, there is a lack of reliable guidelines for matching results. Basing on normalized correlation, this thesis uses the SURF (Speed up Robust Features) algorithm of local features of an image to verify matching results. By fully utilizing the global and local information of the image, this method ensures the reliability of the matching results. Emulation tests how that this method is hardly interfered by outside factors and can greatly save time by using parallel calculations. With a high accuracy of matching results, it is applicable for targets recognition.

Zhen Liu | Mingyu Xia | Junhui Wei | Shenpei Chen

[1] Matthijs C. Dorst. Distinctive Image Features from Scale-Invariant Keypoints , 2011 .

[2] L. Gool,et al. Interactive museum guide : fast and robust recognition of museum objects , 2006 .

[3] Paul A. Viola,et al. Rapid object detection using a boosted cascade of simple features , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[4] Herbert Bay,et al. From wide-baseline point and line correspondences to 3D , 2006 .

[5] Luc Van Gool,et al. SURF: Speeded Up Robust Features , 2006, ECCV.

[6] Niko Sünderhauf,et al. COMPARING SEVERAL IMPLEMENTATIONS OF TWO RECENTLY PUBLISHED FEATURE DETECTORS , 2007 .