Exploration of Virtual Reality System of Dimensionality Reconstruction Technique

Abstract

In order to improve the accuracy of the target matching and recognition in digital image processing field and reduce the time of the target matching and recognition, then use the stability and robustness of SIFT descriptor in image matching and identification, and couple with some data dimension reduction techniques to remove the redundant data in the SIFT descriptor, so as to solve the defect of SIFT descriptor in the matching speed and the accuracy, so the multilayer dimension reduction plan of the PCA algorithm and the CAIM algorithm is put forward in this paper, which is based on SIFT feature matching algorithm, and after the experiment analysis and comparison, it can be found that our method can effectively remove redundant data, improve the matching speed, and also enhance the accuracy of the matching to a certain degree.