基于改进SIFT算法的多源遥感影像特征匹配
|
摘要
影像匹配是影像处理及应用的基础。异源遥感影像在灰度信息、比例尺及旋转角度方面都存在较大差异,采用传统的匹配算法难以对其进行匹配。SIFT算法在影像匹配方面有着广泛的应用,本文以传统SIFT算法为基础,对其结构和度量方面做出了改进,即将SIFT算子由局部向全局结构化转变,且用准欧氏距离代替欧氏距离作为相似性判定测度,从而实现了异源影像的高精度配准。以天绘一号及高分二号卫星影像进行实验,结果表明,改进后的SIFT算法在稳定性、可靠性及精度方面都有较大的提升,且能较好地匹配不同分辨率及光照变化下的异源遥感影像。
Image matching is the basis of image processing and application. Because of significant differences in the gray information,scale and rotation of multi-source remote sensing images,it is difficult to match them with traditional algorithms. The algorithm of SIFT is widely used in image matching. Based on SIFT,improvements have been made in its structure and measurement in this paper.The structure of SIFT is transformed from local to global and Euclidean distance is utilized to replace Euclidean distance as the similarity measure. In this way,high precision of matching multi-source remote sensing images could be realized. Experiments were conducted on the Mapping Satellite-1 and Gaofen Satellite-2 images. The results show that the improved SIFT algorithm has a great improvement in stability,reliability and precision and could match multi-source remote sensing images with various resolutions and illumination.
Image matching is the basis of image processing and application. Because of significant differences in the gray information,scale and rotation of multi-source remote sensing images,it is difficult to match them with traditional algorithms. The algorithm of SIFT is widely used in image matching. Based on SIFT,improvements have been made in its structure and measurement in this paper.The structure of SIFT is transformed from local to global and Euclidean distance is utilized to replace Euclidean distance as the similarity measure. In this way,high precision of matching multi-source remote sensing images could be realized. Experiments were conducted on the Mapping Satellite-1 and Gaofen Satellite-2 images. The results show that the improved SIFT algorithm has a great improvement in stability,reliability and precision and could match multi-source remote sensing images with various resolutions and illumination.
引文
[1]禄丰年．多源遥感影像配准技术分析[J]．测绘科学技术学报，2007,24(4):251-254.
[2] Ke Y,Sukthankar R. PCA-SIFT:a more distinctive representation for local image descriptors[C]//IEEE Computer Society,2004:506-513.
[3]张羽，朱丹，王玉良．一种改进的快速SIFT特征匹配算法[J]．微计算机信息，2008,24(33):220-222.
[4] Yang Z L,Guo B L. Image Mosaic Based on SIFT[C]//International Conference on Intelligent Information Hiding and Multimedia Signal Processing. IEEE Computer Society,2008:1 422-1 425.
[5] Brackle D V,Rangarajan K,Shah M. Optimal corner detector[J]. Comput Vis,1989,48(2):230-245.
[6]程德志，李言俊，余瑞星．基于改进SIFT算法的图像匹配方法[J]．计算机仿真，2011,28(7):285-289.
[7]倪希亮．基于尺度不变特征的多源遥感影像配准[D]．青岛:山东科技大学，2010.
[8]刘松涛，杨绍清．图像配准技术研究进展[J]．电光与控制，2007,14(6):99-105.
[9]肖健．SIFT特征匹配算法研究与改进[D]．重庆:重庆大学，2012.
[10]勒中鑫．数字图像信息处理[M]．北京:国防工业出版社，2003.
[11] Lowe D G. Distinctive Image Features from Scale-Invariant Keypoints[J]. International Journal of Computer Vision,2004,60(2):91-110．
[2] Ke Y,Sukthankar R. PCA-SIFT:a more distinctive representation for local image descriptors[C]//IEEE Computer Society,2004:506-513.
[3]张羽，朱丹，王玉良．一种改进的快速SIFT特征匹配算法[J]．微计算机信息，2008,24(33):220-222.
[4] Yang Z L,Guo B L. Image Mosaic Based on SIFT[C]//International Conference on Intelligent Information Hiding and Multimedia Signal Processing. IEEE Computer Society,2008:1 422-1 425.
[5] Brackle D V,Rangarajan K,Shah M. Optimal corner detector[J]. Comput Vis,1989,48(2):230-245.
[6]程德志，李言俊，余瑞星．基于改进SIFT算法的图像匹配方法[J]．计算机仿真，2011,28(7):285-289.
[7]倪希亮．基于尺度不变特征的多源遥感影像配准[D]．青岛:山东科技大学，2010.
[8]刘松涛，杨绍清．图像配准技术研究进展[J]．电光与控制，2007,14(6):99-105.
[9]肖健．SIFT特征匹配算法研究与改进[D]．重庆:重庆大学，2012.
[10]勒中鑫．数字图像信息处理[M]．北京:国防工业出版社，2003.
[11] Lowe D G. Distinctive Image Features from Scale-Invariant Keypoints[J]. International Journal of Computer Vision,2004,60(2):91-110．
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |