计算机工程与应用 ›› 2017, Vol. 53 ›› Issue (12): 208-213.DOI: 10.3778/j.issn.1002-8331.1601-0287

• 图形图像处理 • 上一篇    下一篇

基于MSCR与CS-LBP的鱼眼图像特征区域匹配方法

万  丽1,李晓芬2,冯为嘉3,龙帮强4,朱均超1   

  1. 1.天津理工大学 自动化学院,天津 300384
    2.天津国际工程咨询公司 低碳发展部,天津 300074
    3.天津师范大学 计算机信息与工程学院,天津 300387
    4.天津大学 精密测试技术及仪器国家重点实验室,天津 300072
  • 出版日期:2017-06-15 发布日期:2017-07-04

Matching method of fish-eye image based on CS-LBP and MSCR regions

WAN Li1, LI Xiaofen2, FENG Weijia3, LONG Bangqiang4, ZHU Junchao1   

  1. 1.School of Electrical Engineering, Tianjin University of Technology, Tianjin 300384, China
    2.Low Carbon Development, Tianjin International Engineering Consultants Corporation, Tianjin 300074, China
    3.College of Computer and Information Engineering, Tianjin Normal University, Tianjin 300387, China
    4.State Key Laboratory of Precision Measuring Technology and Instruments, Tianjin University, Tianjin 300072, China
  • Online:2017-06-15 Published:2017-07-04

摘要: 相比普通镜头,鱼眼镜头拥有更大的视场角,甚至可以直接获取半球域的图像信息,在立体视觉领域,应用鱼眼镜头来采集全景图像可减少镜头及图像采集模块数目,简化系统、提高运算速度、降低成本。但同时鱼眼镜头图像也存在一定程度的畸变,越靠近边缘畸变越严重。因此,在光轴正交或是角度更大的立体视觉系统中,进行相关图像的特征点匹配存在困难,直接影响立体视觉系统的应用效果。然而采用一种具有仿射不变性的图像匹配算法即可解决这个问题,首先提取原始图像的MSCR特征区域,其次引进CS-LBP算子对各个MSCR区域进行特征描述,应用特征权重的卡方距离比较法进行唯一匹配,最后进行椭圆拟合及连线标记使得匹配结果可视化。且通过实验验证了此方法的稳定一致性,可应用于大旋转角度的鱼眼图像的特征匹配。

关键词: 鱼眼图像, 特征区域匹配, 最大稳定色彩区域, 中心对称局部二值模式, 卡方距离

Abstract: Compared with ordinary lens, fish-eye lens has larger field of view, and even direct access to the image information of the hemisphere domain. In the field of stereoscopic vision, using fish-eye lens to get a panoramic image can reduce the number of lens and image acquisition modules, simplifying the system, increase speed, and reduce costs. But images taken with a fisheye lens will produce a certain amount of distortion, and the more closer to the edge of image the more larger to the distortion. Therefore, it is difficult to match the feature points of the image in the stereo vision system with the orthogonal or the angle of the optical axis, which directly affects the application effect of the stereo vision system. However, the problem can be solved by an image matching algorithm based on affine invariants. This paper first extracts the characteristics of MSCR regions from the original image, and then introduces the CS-LBP operator to describe the characteristics of each MSCR regions and finds the only matching by the application of the chi square distance comparison method, which based on feature weights. Finally, the ellipse fitting and connection mark makes the matching result visualization. The stability and consistency of this method can be verified by the experiments, so this method can be applied to the matching of the fish-eye images with large rotation angle.

Key words: fish-eye image, matching of character regions, Maximally Stable Colour Regions(MSCR), Center Symmet-Local Binary Pattern(CS-LBP), chi-square measure