遥感图像几何校正的支持向量机算法研究

doi:10.3969/j.issn.1001-2400.2011.05.020

J4 ›› 2011, Vol. 38 ›› Issue (5): 121-128+146.doi: 10.3969/j.issn.1001-2400.2011.05.020

遥感图像几何校正的支持向量机算法研究

厍向阳¹;李崇贵²;姚顽强²

(1. 西安科技大学计算机科学与技术学院，陕西西安 710054;
2. 西安科技大学测绘科学与技术学院，陕西西安 710054)

收稿日期:2010-08-20 出版日期:2011-10-20 发布日期:2012-01-14
通讯作者: 厍向阳
作者简介:厍向阳(1968-)，男，副教授，博士，E-mail: xiangyangshe@sohu.com
基金资助:
国家自然科学基金资助项目(30872023);陕西省自然科学基金资助项目(2009JM7007);教育部科学技术研究重点资助项目(2009122);陕西省教育厅专项科研计划资助项目(08JK354)

Research on the geometric correction algorithm for the remote sensing image by a support vector machine

SHE Xiangyang¹;LI Chonggui²;YAO Wanqiang²

(1. Computer Sci. & Tech. College, Xi'an Univ. of Sci. and Tech., Xi'an 710054， China;
2. Surveying and Mapping Sci. & Tech. College，Xi'an Univ. of Sci. and Tech., Xi'an 710054， China)

Received:2010-08-20 Online:2011-10-20 Published:2012-01-14
Contact: SHE Xiangyang

摘要/Abstract

摘要：

针对目前遥感图像几何校正算法存在的不足，提出了一种新的遥感图像几何校正算法．引入支持向量机理论和方法，结合遥感图像近似几何校正基本原理，提出遥感图像几何校正的支持向量机算法和实现步骤|选择实验区，使用差分GPS实测地面控制点坐标，使用遥感图像处理软件量测地面控制点对应的影像坐标|使用聚类算法分别选择不同数量的控制点作为遥感图像几何校正的控制点，其余控制点作为检查点|分别使用近似几何校正算法、神经网络和支持向量机算法进行遥感图像的几何校正，并进行校正误差比较分析．算法测试表明：遥感图像几何校正的支持向量机算法具有校正误差小、泛化能力强等特点．

关键词: 遥感图像, 图像几何校正, 支持向量机(SVM), 最小二乘法(LSM), 误差比较

Abstract:

Facing the disadvantages of geometric correction algorithm for the remote sensing image at present,a new algorithm is studied and we put forward the geometric correction algorithm and solving steps for the remote sensing image based on SVM, introduce the SVM theory and approach and adopt the essence theory of the remote sensing image approximate geometric correction. One testing region is selected, the ground control points coordinates and altitudes are surveyed by differential GPS, and the coordinates of the ground control points in the remote sensing image are measured with image processing software. We select a varying number of control points to correct the remote sensing image in the geometrical plain, and use other control points as testing points by the cluster algorithm in all the ground control points. We carry out remote sensing image geometric correction by the approximate geometric correction , the neural network model and the SVM algorithms, analyze and compare correction precision. Algorithm testing show that the algorithm for the SVM has good correction precision and generalizing ability. The algorithm for the SVM supports remote sensing application to develop the quantitative and accurate technique, and enlarges the geometric correction approach.

Key words: remote sensing image, image geometric correction, support vector machine(SVM), least squares method(LSM), correction precision

厍向阳;李崇贵;姚顽强. 遥感图像几何校正的支持向量机算法研究[J]. J4, 2011, 38(5): 121-128+146.

SHE Xiangyang;LI Chonggui;YAO Wanqiang. Research on the geometric correction algorithm for the remote sensing image by a support vector machine[J]. J4, 2011, 38(5): 121-128+146.

参考文献

［1］徐青，张艳，耿则勋，等．遥感图像影像融合与分辨率增强技术［M］. 北京：科学出版社，2007.
［2］朱述龙，朱宝山，王卫红．遥感图像处理与应用［M］. 北京：科学出版社，2006.
［3］孙家抦．遥感原理与应用［M］. 2版. 武汉：武汉大学出版社，2009.
［4］ David P S, Saumeul F A. Accuracy Rectification Using Topographic Map Versus GPS Ground Control Points［J］ . Photogrametric Engineering & Remote Sensing, 2001, 67(5): 565-570.
［5］ Jay G. Non-Differential GPS as an Alternative Source of Planimetric Control for Rectifying Satellite Image［J］ . Photogrametric Engineering & Remote Sensing, 2001, 67(1): 49-55.
［6］ Baltsavias E S. Metric Information Extraction form SPOT Images and the Role of Polynomial Mapping Function［J］ . International Archives of Photogrammetry and Remote Sensing, 1992, 29(B4): 358-364.
［7］ Yasser E M, Kurt N. Precision Rectification of SPOT Imagery Using the Direct Linear Transformation Model［J］ . Photogrammetric Engineering & Remote Sensing, 1996, 62(1): 67-72.
［8］ Savopol F, Armenakis C. Modeling of IRS-1C Satellite Pan Imagery Using the DLT Approach［J］ . International Archives of Photogrammetry and Remote Sensing,1998, 32(4): 511-514.
［9］ Vincent C T, Yong H A. Comprehensive Study of the Rational Function Model for Photogrammetic Processing［J］ . Photogrametric Engineering & Remote Sensing, 2001, 67(12): 1347-1357.
［10］ Vincent C T, Yong H A. Updating Solutions of Rational Function Model Using Additional Control Information［J］ . Photogrametric Engineering & Remote Sensing, 2002, 68(7): 715-724.
［11］ Vincent C T, Yong H A. 3D Reconstruction Methods Based on Rational Function Model［J］ . Photogrametric Engineering & Remote Sensing, 2002, 68(7): 705-716.
［12］ Dowman J, Dolloff J T. An Evaluation of Rational Functions for Photogrammetric Restitution［J］ International Archives of Photogrammetry and Remote Sensing, 2000, 33(B3/1): 252-266.
［13］ Fraser C S, Baltsavias E, Gruen A. Processing of Imagery for Submitter 3D Positioning and Building Extraction［J］ . ISPRS Journal of Photogrammery & Remote Sensing, 2002, 56(2): 177-194.
［14］ Clive S F, Happy B, Hanleny T Y. Three-dimensional Geopositioning Accuracy of IKONOS Imagery［J］ . Photogrammetric Record, 2002, 17(99): 465-479.
［15］朱述龙，史文中，张艳，等．线阵推扫式影像近似几何校正算法的精度比较［J］．遥感学报，2004, 8(3): 221-226.
Zhu Shulong, Shi Wenzhong, Zhang Yan, et al. Precision Comparison of Several Algorithms for Approximate Rectification of Linear Array Push-broom Imagery［J］ . Journal of Remote Rensing, 2004, 8(3): 221-226.
［16］边肇祺，张学工．模式识别［M］. 2版．北京：清华大学出版社，2000.
［17］ (美)Vapnik V N．统计学习理论［M］. 许建华，张学工，译．北京：电子工业出版社，2004.
［18］邓乃扬，田英杰．支持向量机——理论、算法与拓展［M］. 北京：科学出版社，2009.
［19］张光澄，王文娟，韩会磊，等．非线性优化计算算法［M］. 北京：高等教育出版社，2005.
［20］陈宝林．最优化理论与算法［M］. 2版．北京：清华大学出版社，2005.
［21］ Mingmin Chi, Rui Feng , Lorenzo Bruzzone. Classification of Hyperspectral Remote-sensing Data with Primal SVM for Small-sized Training Dataset Problem［J］ . Advances in Space Research, 2008(41): 1793-1799.
［22］ Valls G C, Bruzzone L. Kernel Methods for Remote Sensing Data Analysis［M］. Chichester: John Wiley & Sons Ltd, 2009.
［23］杨晓元，王志刚，王育民．支持向量机在图像隐秘检测中的应用［J］．西安电子科技大学学报，2005, 32(3): 457-459.
Yang Xiaoyuan, Wang Zhigang, Wang Yumin. The Application of Support Vector Machines in Detection of Images Steganography［J］ . Journal of Xidian University, 2005, 32(3): 457-459.

[1]	董如婵,焦李成,赵进,沈维燕. 一种深度融合机制的遥感图像目标检测技术[J]. 西安电子科技大学学报, 2021, 48(5): 128-138.
[2]	王军军,孙岳,李颖. 一种生成对抗网络的遥感图像去云方法[J]. 西安电子科技大学学报, 2021, 48(5): 23-29.
[3]	李红;刘芳;张凯. 稀疏非负矩阵分解下的遥感图像融合[J]. 西安电子科技大学学报, 2016, 43(2): 193-198.

遥感图像几何校正的支持向量机算法研究

Research on the geometric correction algorithm for the remote sensing image by a support vector machine

PDF (PC)

赞

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 3

Metrics

本文评价

推荐阅读 10