Local Descriptor for Object Recognition in High Resolution Remote Sensing Images

摘要：

A Local feature description algorithm towards object recognition in high-resolution remote images, called the Normalized Pixel Distribution Histogram Local Descriptor (NPDHLD), is proposed. First we extract object edges, the log-polar coordinate is established by using every edge points as the coordinate origin; Normalize every pixel value, the descriptor at the origin points is constructed by capturing the distribution of the remaining points relative to it. Describe sensitive objects with the proposed local describe algorithm to building a sensitive object feature database. The local features of to-be-matched objects to which the local features in the sensitive object database is corresponding, are matched under a two-step matching strategy. Object recognition is implemented after the matching.

关键词： Local descriptor Image matching Hungarian method Object recognition

作者: Tian Hao Li Guohui Jia Li Xu Shukui

作者单位: Dept. of Information System & Management, National University of Defense Technology Changsha, China

会议类型: 国际会议

会议名称: 2010 International Conference on Information Security and Artificial Intelligence(2010年信息安全与人工智能国际会议 ISAI 2010)

会议地点: 成都

会议语种:英文

页码: 915-919

在线出版日期: 2010-12-17（万方平台首次上网日期，不代表论文的发表时间）