融合颜色与光照信息的改进AKAZE图像匹配算法

doi:10.19678/j.issn.1000-3428.0057709

摘要/Abstract

摘要： 传统AKAZE算法主要以图像的灰度信息作为输入，忽略了颜色与光照信息，且存在过多的冗余特征点。提出一种融合颜色与光照信息的改进图像匹配算法。在图像原始灰度信息中增加颜色补偿量和光照补偿量，克服传统AKAZE算法未利用颜色和光照信息来区分同名点的缺点。在特征点邻域内计算欧式距离和区域重复率以确定重叠程度较高的相邻特征点，比较相邻特征点的响应强度，将响应强度值较小的特征点作为冗余点并删除。在图像匹配后利用随机一致性算法剔除误匹配点对，通过并行运算对算法进行加速。实验结果表明，该算法可以有效识别同名点并剔除图像中的冗余点，运行效率优于SIFT和SURF等算法。

关键词: 图像匹配, AKAZE算法, 颜色补偿, 光照补偿, 冗余特征点

Abstract: The traditional AKAZE algorithm mainly takes the grayscale information of the image as the input and ignores the color and illumination information of the image, making it hard to distinguish identical points.In addition, it leaves too many redundant feature points unsolved.This paper proposes an improved image matching algorithm that combines color and illumination information.The algorithm introduces color compensation and illumination compensation into the original grayscale information of the image, to distinguish the identical points.To deal with the redundant feature points, the Euclidean distance and regional repetition rate in the neighborhood of feature points are calculated to determine the highly overlapped adjacent feature points.Then the response intensity values of neighboring feature points are compared, and the feature points with a low response intensity are deleted as redundant points.After image matching is done, the random consistency algorithm is used to eliminate the false matching point pairs, and the whole algorithm is accelerated through parallel operations.Experimental results show that the proposed algorithm can effectively identify the identical points and eliminate the redundant points in the image.Its operation efficiency is also higher than that of SIFT, SURF and other algorithms.

Key words: image matching, AKAZE algorithm, color compensation, illumination compensation, redundant feature point

中图分类号:

TP391.4

张宇, 胡立坤, 黄彬. 融合颜色与光照信息的改进AKAZE图像匹配算法[J]. 计算机工程, 2021, 47(6): 225-233.

ZHANG Yu, HU Likun, HUANG Bin. Improved AKAZE Image Matching Algorithm Combining Color and Illumination Information[J]. Computer Engineering, 2021, 47(6): 225-233.

http://www.ecice06.com/CN/Y2021/V47/I6/225

图/表 18

20210618184323

20210618184327

20210618184330

20210618184333

20210618184340

20210618184344

20210618184347

20210618184350

20210618184353

20210618184357

20210618184401

20210618184404

20210618184409

20210618184413

20210618184417

20210618184421

20210618184425

20210618184429

参考文献

[1] NASROLLAHI K,MOESLUND T B.Super-resolution:a comprehensive survey[J].Machine Vision and Applications,2014,25(6):1423-1468.
[2] LI Yali,WANG Shengjin,TIAN Qi,et al.A survey of recent advances in visual feature detection[J].Neurocomputing,2015,149(5):736-751.
[3] GHOSH D,KAABOUCH N.A survey on image mosaicing techniques[J].Journal of Visual Communication and Image Representation,2016,34:1-11.
[4] WANG Zhicheng,CHEN Yufei,ZHU Zewei,et al.An automatic panoramic image mosaic method based on graph model[J].Multimedia Tools and Applications,2016,75(5):2725-2740.
[5] WU Hanqian,LI Chengchao,XIE Jue.Application of improved A-KAZE algorithm in image registration[J].Journal of Southeast University(Natural Science),2017,47(4):667-672.(in Chinese)吴含前,李程超,谢珏.一种改进的A-KAZE算法在图像配准中的应用[J].东南大学学报(自然科学版),2017,47(4):667-672.
[6] KWON O S,HA Y H.Panoramic video using scale-invariant feature transform with embedded color-invariant values[J].IEEE Transactions on Consumer Electronics,2010,56(2):792-798.
[7] LOWE D G.Object recognition from local scale-invariant features[C]//Proceedings of the 7th IEEE International Conference on Computer Vision.Washington D.C.,USA:IEEE Press,1999:1150-1157.
[8] LOWE D G.Distinctive image features from scale-invariant keypoints[J].International Journal of Computer Vision,2004,60(2):91-110.
[9] BAY H,TUYTELAARS T,VAN GOOL L.SURF:speeded up robust features[C]//Proceedings of the 9th European Conference on Computer Vision.Berlin,Germany:Springer,2006:404-417.
[10] RUBLEE E,RABAUD V,KONOLIGE K,et al.ORB:an efficient alternative to SIFT or SURF[C]//Proceedings of International Conference on Computer Vision.Washington D.C.,USA:IEEE Press,2011:2564-2571.
[11] ROSTEN E,DRUMMOND T.Machine learning for high-speed corner detection[C]//Proceedings of the 9th European Conference on Computer Vision.Berlin,Germany:Springer,2006:430-443.
[12] CALONDER M,LEPETIT V,STRECHA C,et al.BRIEF:binary robust independent elementary features[C]//Proceedings of the 11th European Conference on Computer Vision.Berlin,Germany:Springer,2010:778-792.
[13] ALCANTARILLA P F,BARTOLI A,DAVISON A J.KAZE features[C]//Proceedings of the 12th European Conference on Computer Vision.Berlin,Germany:Springer,2012:214-227.
[14] WEICKERT J,GEWENIG S,SCHROERS C,et al.Cyclic schemes for PDE-based image analysis[J].International Journal of Computer Vision,2016,118(3):275-299.
[15] HAN Min,YAN Kuo,QIN Guoshuai.A mosaic alorithm for UAV aerial image based on improved KAZE[J].Acta Automatica Sinica,2019,45(2):305-314.(in Chinese)韩敏,闫阔,秦国帅.基于改进KAZE的无人机航拍图像拼接算法[J].自动化学报,2019,45(2):305-314.
[16] LING Huanqing,XIE Yi,FU Sizhou.UAV image registration algorithm using color and AKAZE feature[J].Acta Geodaetica et Cartographica Sinca,2017,46(7):900-909.(in Chinese)梁焕青,谢意,付四洲.颜色不变量与AKAZE特征相结合的无人机影像匹配算法[J].测绘学报,2017,46(7):900-909.
[17] ZHAO Y,ZHAI Y,ERIC D,et al.Image matching algorithm based on SIFT using color and exposure information[J].Journal of Systems Engineering and Electronics,2016,27(3):691-699.
[18] WANG Wei,WANG Zhanqing.Image of multi-scale feature points clustering[J].Journal of Chinese Computer Systems,2017,38(11):2597-2603.(in Chinese)王薇,王展青.多尺度特征点聚类的图像配准算法[J].小型微型计算机系统,2017,38(11):2597-2603.
[19] GAO Fei,WANG Cong,JIAO Donghang.An image registration algorithm based on block information entropy characteristic scale[J].Transactions of Beijing Institute of Technology,2016,36(11):1194-1199.(in Chinese)高飞,王聪,矫东航.基于分块信息熵和特征尺度的图像配准算法[J].北京理工大学学报,2016,36(11):1194-1199.
[20] HOSSEIN-NEJAD Z,NASRI M.RKEM:redundant keypoint elimination method in image registration[J].IET Image Processing,2017,11(5):273-284.
[21] ZOU Guoliang,ZHANG Shiming,CHEN Changji.Design of image matching parallel algorithm based on improved minimum distance method[J].Computer Engineering,2018,44(9):218-223.(in Chinese)邹国良,张石明,陈长吉.基于改进最小距离法的图像匹配并行算法设计[J].计算机工程,2018,44(9):218-223.

选择文件类型/文献管理软件名称

选择包含的内容