基于改进蜘蛛网局部图像特征的图像配准算法  

Image registration algorithm based on improved spider local image feature

作　　者：杨真真[1] 匡楠 杨永鹏 许鹏飞 YANG Zhenzhen;KUANG Nan;YANG Yongpeng;XU Pengfei(National Engineering Research Center of Communication and Network Technology,Nanjing University of Posts and Telecommunications,Nanjing 210003,China;School of Network and Communication,Nanjing Vocational College of Information Technology,Nanjing 210023,China)

机构地区：[1]南京邮电大学通信与网络技术国家工程研究中心,江苏南京210003 [2]南京信息职业技术学院网络与通信学院,江苏南京210023

出　　处：《南京邮电大学学报（自然科学版）》2020年第2期27-34,共8页Journal of Nanjing University of Posts and Telecommunications：Natural Science Edition

基　　金：国家自然科学基金(61501251);中国博士后科学基金(2018M632326);通信与网络技术国家工程研究中心开放课题(TXKY17010);江苏省高校自然科学面上项目(19KJB510044);南京信息职业技术学院校级基金(YK20190402);江苏省高等学校大学生创新创业训练计划(201913112012Y)资助项目。

摘　　要：为了解决传统基于特征描述匹配算法鲁棒性和精度较差的问题,提出一种基于改进的蜘蛛网局部图像特征(Spider Local Image Feature,SLIF)描述子的图像配准算法。该算法对SLIF描述子进行改进,将特征点采样模型范围内的方形、径向、环形邻域像素信息纳入描述范围,增强SLIF描述子空域描述能力。此外,还对SLIF描述子描述方法进行改进,利用特征点采样模型范围内像素点之间自相似性构建特征描述方法。实验表明,该方法在平移旋转图像特征点配对的准确率和配准精度高,抗噪声能力强,具有较好的鲁棒性和图像变换适应性。To solve problems of the poor robustness and the accuracy of traditional feature-based matching algorithms,an image registration algorithm based on improved spider local image feature(SLIF)descriptor is proposed.The SLIF descriptor is improved by the algorithm,the square,radial,and circular neighborhood pixel information within the range of the feature point sampling model are incorporated into the description range,enhancing the spatial description capability of the SLIF descriptor.In addition,the SLIF description strategy is improved and the self-similarity between pixels within the range of the feature point sampling model is used to construct the feature description method.Experimental results show that the algorithm has high accuracy at the matching feature point and on the registering images,strong anti-noise ability,good robustness,and adaptability to image transformation,especially for translation and rotation images.

关 键 词：图像配准 特征匹配 特征描述子 邻域采样模型 

分 类 号：TN911.73[电子电信—通信与信息系统] TP391[电子电信—信息与通信工程]