基于注意力选择的局部特征匹配方法  

Matching method of local feature based on attention selection

作　　者：杨燕 胡小鹏[1] 吴思宁 王凡 刘培启 YANG Yan;HU Xiaopeng;WU Sining;WANG Fan;LIU Peiqi(School of Computer Science and Technology,Faculty of Electronic Information and Electrical Engineering,Dalian University of Technology,Dalian 116024,China;School of Computer and Information Technology,Liaoning Normal University,Dalian 116029,China)

机构地区：[1]大连理工大学电子信息与电气工程学部计算机科学与技术学院,辽宁大连116024 [2]辽宁师范大学计算机与信息技术学院,辽宁大连116029

出　　处：《大连理工大学学报》2019年第2期186-193,共8页Journal of Dalian University of Technology

基　　金：辽宁省教育厅科技研究资助项目(L201683682);国家自然科学基金资助项目(21676048)

摘　　要：为提高局部特征匹配性能,提出了基于人类注意力选择机制的特征匹配方法.通过局部特征的显著度分析,优先选择区域显著的特征进行匹配,利用可靠的匹配点对进行三角区域约束,实现区域内重复特征的精确匹配.实验结果表明,通过逐步缩小对应匹配搜索区域的方法抑制由相似特征引起的匹配歧义性,兼顾了高匹配精度和查全率.To improve the matching performance of local features,a feature matching method based on human attention selection mechanism is proposed.Based on the saliency analysis for local features,to achieve the accurate matching of repetitive features in region,the most salient local features are prefered to be matched and the robust correspondences are found to perform the following triangle constraint.The experimental results show that the reduction of the matching region is effective for reducing the matching ambiguities of local feature.The matching performance including both the matching precision and the recall rate is improved,especially for the scenes with repetitive structures.

关 键 词：特征匹配 注意力选择 重复结构 三角约束 

分 类 号：TP39[自动化与计算机技术—计算机应用技术]