基于改进SURF算法的双目视觉定位

随着机器人技术的发展,机器人视觉方面的研究也越来越受到人们的重视。在机器人视觉系统中,双目视觉应用最为广泛。在利用双目视觉对物体进行定位时,文中采用了各方面性能都较有优势的SURF算法,来对图像中的特征点进行提取与匹配。由于客观因素的影响,在SURF匹配过程中存在特征点误匹配现象,为了消除误匹配,文中对SURF算法做了改进,加入了剔除误匹配的RANSAC算法。实验结果表明,改进后的SURF算法,能够大大提高双目视觉定位的精准度。

Binocular Visual Positioning Based on Improved SURF Algorithm

With the development of robot technology,the research of robot vision get people's attention more and more.In robot vision system, the binocular vision applied most widely.When using the binocular vision to locate objects,this paper adopted SURF algorithm,it has more advantages on performance of many aspects,we use it to realize feature point extraction and matching of images.Because of the influence of the objective factors,there is false match in the process of SURF matching of feature points.In order to eliminate the false match,we improved the SURF algorithm,joined the RANSAC algorithm,it can eliminate false match.The experiment results show that the improved SURF algorithm can improve the binocular visual positioning accuracy greatly.