激光与光电子学进展, 2018, 55 (3): 031005, 网络出版: 2018-09-10
结合区域分块的快速BRISK图像拼接算法 下载: 1087次
Fast Image Mosaic Algorithm Based on Area Blocking and BRISK
图像处理 图像拼接 区域分块 有向线段 投影误差 效率提升 image processing image mosaic area blocking directed line segment projection error efficiency improvement
摘要
为了解决传统图像拼接算法匹配率低以及拼接速度慢的问题,提出一种结合区域分块的快速BRISK图像拼接算法。采用频域相位相关算法寻找两幅图像间的相似部分,将其均匀划分成N×N '的图像块,计算各图像块的标准差,选择标准差较大且分布在不同位置的两个图像块,用BRISK算法进行特征点粗匹配。获得匹配点对后,构造有向线段并进行邻近线段匹配。通过随机选择的匹配点对计算投影变换矩阵,进行投影误差校正,最后进行加权融合和亮度均衡化完成图像拼接。实验结果表明:该算法能够保证图像拼接的正确率大于90%,并且拼接速度提升63.7%。
Abstract
In order to solve the problems of the low matching rate and the low mosaic efficiency of the traditional image mosaic algorithms, we propose a fast image mosaic algorithm based on area blocking and binary robust invariant scalable keypoints (BRISK). Firstly, we use the frequency domain phase correlation algorithm to find the similar areas between the two images. Secondly, we divide the similar areas into N×N' image blocks evenly and calculate the standard deviation of each image block. Thirdly, selecting two image blocks with large standard deviations in different positions, we coarsely match the feature points by BRISK algorithm. Fourthly, we construct the directed line segments and match the adjacent line segments after obtaining the matching points. Finally, we calculate the projection transformation matrix by the randomly matching point pairs to correct the projection error, and accomplish the image mosaic by weighted fusion and luminance equalization. The experimental results show that the matching correctness of the proposed algorithm is more than 90%, and the mosaic speed is increased by 63.7%.
赵婷, 康海林, 张正平. 结合区域分块的快速BRISK图像拼接算法[J]. 激光与光电子学进展, 2018, 55(3): 031005. Ting Zhao, Hailin Kang, Zhengping Zhang. Fast Image Mosaic Algorithm Based on Area Blocking and BRISK[J]. Laser & Optoelectronics Progress, 2018, 55(3): 031005.