基于低空摄影测量影像的特征点提取算子比较研究  

Study on Feature Point Extraction Operators Based on Low Altitude Photogrammetry Image

作　　者：张凯南[1] 杨志强[1] 张学伟 石震[1] ZHANG Kai-nan;YANG Zhi-qiang;ZHANG Xue-wei;SHI Zhen(Institute of Surveying and Spatial Information of Chang'an University,Xi'an 710054,China)

机构地区：[1]长安大学测绘与空间信息研究所,陕西西安710054

出　　处：《地理与地理信息科学》2017年第3期30-35,共6页Geography and Geo-Information Science

基　　金：国家青年科学基金项目(41504001);中央高校基本科研业务费(310826175027)

摘　　要：影像匹配是低空遥感数据处理的核心步骤,而特征提取是影像匹配的基础。该文从兴趣算子的角度分析了摄影测量中几种主流特征点提取算法:Moravec算子、Forstner算子、SUSAN算子、Harris和SIFT算子,以角点类型较多的普通几何图形、卫星遥感影像和动力三角翼拍摄的低空遥感影像为数据,通过实验得到各算法的速度、精度、局限性和适应性。针对低空摄影测量影像,以重复率为指标定量比较和分析了各算法在抗噪、抗对比对变化、抗光照变化和抗旋转变化等方面的性能。实验结果表明,针对灰度信息丰富的低空遥感影像,SIFT算子具备尺度不变性,抗噪性最好,Harris算子提取速度最快,Forstner算子精度最高。实验结论为低空摄影测量影像处理提供了一种可行性方法。Low altitude photogr ammetry technology has become research focus in the field of remote sensing,since it has the advantages of simple and easy,low cost,high speed,and precision.Image matching is the key of low altituder emote sensing dataprocessing,and feature extraction is the basis of image matching.This paper studied sever al mainstream feature extr action algor ithms in the v iew of inter est o perat ors:Moravec operator,Forstner operator,SUSAN operator,Harr is operator and SIFT operator,took the data of geometric figure with many corner types,images of satellite and low altitude photo grammetry image gottenby dynamic delta wing,and got the speed,precision,limitation and adaptability of the algo rit hms thr ough the exper iments.Thealgor ithms were analyzed and compared from the resistance performance in no ise,contrast change,illumination change and rotatechange with the repetition rate as the index.The research showed that SIFT operator had scale inv ariance and the best noise resistance,Harr is was the fastest operator,and Forstner operator had the highest precision for the rich gray low altitude photogrammetry.The experimental conclusion offers the method to the low space photo grammetry image processing.

关 键 词：低空摄影测量 特征点提取 FORSTNER算子 HARRIS算子 SIFT算子 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]