基于改进BRISK的图像拼接算法  

Image Mosaic Algorithm Based on Improved BRISK

作　　者：董强[1,2] 刘晶红[1] 王超 周前飞[1,2] DONG Qiang LIU Jinghong WANG Chao ZHOU Qianfei(Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, Changchun 130033, China University of Chinese Academy of Sciences, Beijing 100049, China Unit 77110 of People＇s Liberation Army, Shifang 618400, China)

机构地区：[1]中国科学院长春光学精密机械与物理研究所 [2]中国科学院大学 [3]解放军77110部队

出　　处：《电子与信息学报》2017年第2期444-450,共7页Journal of Electronics & Information Technology

基　　金：吉林省重大科技攻关项目(11ZDGG001);国家林业公益性行业科研专项(201204515)~~

摘　　要：为了获得精准的航空拼接图像,更好地解决图像拼接中经常出现的尺度变化、角度旋转、光照差异以及传统的BRISK(Binary Robust Invariant Scalable Keypoints)算法匹配正确率较低,图像拼接精度低等问题,该文提出一种全新的基于有向线段的BRISK特征的图像拼接模型。首先,使用BRISK算法进行图像匹配,得到粗匹配点对,再构造有向线段及其BRISK特征进行邻近线段匹配,通过概率统计模型进行特征点的精匹配,最后进行加权融合和亮度均衡化进行图像融合完成图像拼接。实验结果表明,该文算法针对图像的光照条件不同、角度旋转、分辨率低、尺度变化等均有良好的鲁棒性和稳定性,该文算法是一种耗时短、精确度高、拼接效果良好的图像拼接方法。In order to obtain accurate aeriM stitching images, this paper proposes a novel image mosaic method based on Binary Robust Invariant Scalable Keypoints （BRISK） feature of directed line segment, aiming to resolve incident scaling, rotation, changes in lighting condition, the low correct matching rate and low accuracy using conventional BRISK algorithm in image mosaic. This method firstly uses BRISK algorithm to match in order to acquire rough point matching. Secondly, it constructs directed line segments, describes them with BRISK feature, and matches those directed segments. The method is used to purified point matching based on statistical voting. Finally, weighted fusion and luminance equalization are used to image fusion to accomplish image mosaic. The experiment results show that the method has strong robustness and stability for lighting, rotation, resolution and scaling. The proposed method has high precision, and can achieve fine image mosaic results.

关 键 词：图像配准 图像拼接 BRISK特征 邻近线段 

分 类 号：TN911.73[电子电信—通信与信息系统]