×

Registration algorithm based on fuzzy shape context and local vector similarity constraint. (Chinese. English summary) Zbl 1449.68086

Summary: Non-rigid point set registration is an essential research in pattern recognition. Two main contributions are presented based on the current popular non-rigid point set registration algorithm in this paper: (1) fuzzy shape context feature; and (2) local spatial vector similarity constraints based on local vector feature. Firstly, the correspondence of complementary of features is evaluated. At the same time, a fuzzy shape context (FSC) feature is defined and the Gaussian mixture model based on the fuzzy shape context distance and the global feature distance is designed. Secondly, the spatial transformation of complementary of constraint is updated. Meanwhile, a local vector feature is defined and the local spatial vector similarity constraint based on local vector feature is built. The proposed algorithm estimates the correspondence by using the Gaussian mixture model of complementary of feature. The proposed algorithm updates parameters of transformation by using the expectation maximization algorithm, and completes transformation updating by building energy function that has local spatial vector similarity constraints. Firstly, the retrieval rate of FSC is tested. Then the performance of point set registration and image registration of the proposed algorithm is tested by using public data sets. In the comparison experiments of currently popular 10 algorithms, accurate registration results of the proposed algorithm are acquired. It is proved that the proposed algorithm surpasses the popular algorithms in most of the registration precision.

MSC:

68T10 Pattern recognition, speech recognition
PDFBibTeX XMLCite
Full Text: DOI