期刊名称：International Journal of Signal Processing, Image Processing and Pattern Recognition
印刷版ISSN：2005-4254
出版年度：2014
卷号：7
期号：3
页码：83-90
DOI：10.14257/ijsip.2014.7.3.08
出版社：SERSC
摘要：The key problem of computer vision reconstruction technology is to improve accuracy of shape reconstruction and authenticity of color reconstruction. First, color edge Gray code is presented, which encodes with red and blue Gray code stripes, and decodes with sub-pixel located stripe edges. So quantization error and decoding error were eliminated and accuracy of shape reconstruction was improved. Second, color correction method is presented. The coupling phenomena of RGB primary colours existed in the encode and decode process was analyzed, and then Color coupler Correction scheme based on Caspi model Hardware Calibration was designed. Accordingly, a color imbalance caused by the surface curvature in the reconstruction process was analyzed, and color imbalance correction scheme used surface geometric information was also proposed. Shape reconstruction experimental results show that relative error of reconstructed plane is 0.05%, and reconstructed complex surface has same visual effects as real surface. Color reconstruction experimental results show that color of reconstructed surface is uniform after correction, which has same color as real surface. And moreover, This method has strong anti-interference ability and can help to improve the accuracy of shape and color reconstruction.
关键词：Structured light; Shape reconstruction; Color reconstruction; Gray code
