多尺度边缘信息的遥感影像快速匹配
摘要
本文应用一种由改进的拉普拉斯金字塔算法提取的多尺度边缘信息作为特征模板,从而可将影像的灰度信息的匹配转化成特征的匹配,利用特征模板与待匹配影像建立间接的相似性度量,以达到提高速度、抗几何失真及灰度变化的目的。同时该方法的多尺度特性,可以满足对于匹配精度和匹配速度的多种需求。实验证明,该方法在保证了互相关影像匹配的精度优势条件下,提高了影像匹配的速度,具有很好的鲁棒性和抗干扰能力。
引文
[1]S.Sclaroff and A.P.Pentland,“Modal matching for correspondence and recognition,”IEEE Trans on PAMI,vo1.17(6),1995[J].
[2]J.P.Lewis,“Fast Template Matching,”Vision Interface 95,Canadian Image Processingand Pattern Recognition Society[M].Quebec City,Canada,pp.120-123.May 1995.
[3]J.P.W.Pluim,J.B.A.Maintz and M.A.Viergever,“Mutual-lnformation-based registration of medical images:a survey,”IEEE Trans[Z].on Medical Imaging,vol.22(8),pp.1986-1004,2003.
[4]S.Milan,Vaclav Hlavac and B.Roger,“Image processing,analysis,and machine vision,”Thomson learning publishing group,1999:52-60[Z].
[5]C.G.Rafael,E.W.Richard and L.E.Steven,Digital image processing using matlab,Publishing house of electronic industry,2005:285-320[Z].
[2]J.P.Lewis,“Fast Template Matching,”Vision Interface 95,Canadian Image Processingand Pattern Recognition Society[M].Quebec City,Canada,pp.120-123.May 1995.
[3]J.P.W.Pluim,J.B.A.Maintz and M.A.Viergever,“Mutual-lnformation-based registration of medical images:a survey,”IEEE Trans[Z].on Medical Imaging,vol.22(8),pp.1986-1004,2003.
[4]S.Milan,Vaclav Hlavac and B.Roger,“Image processing,analysis,and machine vision,”Thomson learning publishing group,1999:52-60[Z].
[5]C.G.Rafael,E.W.Richard and L.E.Steven,Digital image processing using matlab,Publishing house of electronic industry,2005:285-320[Z].