基于角点检测的遥感图像几何质量评价方法.pdf

文档介绍：第 40 卷
第 2 期测
绘
学
报 Vol. 40, No. 2
20
11 年 4 月 Acta Geoda etica et Ca rtographica Sinica Apr. , 2011
文章编号: 1001
1595( 2011) 02
0175
05
基于角点检测的遥感图像几何质量评价方法
王明富1, 2 , 杨世洪1 , 吴钦章1
1. 中国科学院光电技术研究所, 四川成都 610209; 2. 中国科学院研究生院, 北京 100049
A Geometric Quality Assessment Algorithm of Remote Sensing Image Based on Corner Detec
tion
WA N G M ing fu1, 2, Y AN G Shiho ng 1 , WU Qinzhang 1
1. Institute of Opti cs a nd Electronics, Chinese Aca demy of Sciences, Chengdu 610209, China ; 2. Gra dua te Uni versity, Chi nese
Academy of Sciences, Beijing 100049, China
AbstractTo study geometric quality influence of remote sensing ima ge brought by pressing system, a new
image geometric quality assessment algorithm ba sed on ima ge corner is proposed. First, classic Harris corner de
tecting algorithm is introduced, then it is improved to solve its disa dva nta ge of imprecise point location: fitting the
surfa ce of corner response function and evalua ting pea k location of the fitted surfa ce to check the sub
pixel corner
loca tion. Finally, this algorithm is a pplied to geometric qua lity assessment system of remote sensing ima ge, a nd is
compared to traditional surface fitting extremum algorithm. The experimenta l results show that this a lgorithm can
accurately express the a berration of geometric qua lity between original image pressed one. So, it ca n be
used into quality