[1]姬晓飞,秦宁丽,刘洋.多特征的光学遥感图像多目标识别算法[J].智能系统学报,2016,11(5):655-662.[doi:10.11992/tis.201511011]
 JI Xiaofei,QIN Ningli,LIU Yang.Research on multi-feature based multi-target recognition algorithm for optical remote sensing image[J].CAAI Transactions on Intelligent Systems,2016,11(5):655-662.[doi:10.11992/tis.201511011]
点击复制

多特征的光学遥感图像多目标识别算法(/HTML)
分享到:

《智能系统学报》[ISSN:1673-4785/CN:23-1538/TP]

卷:
第11卷
期数:
2016年5期
页码:
655-662
栏目:
出版日期:
2016-11-01

文章信息/Info

Title:
Research on multi-feature based multi-target recognition algorithm for optical remote sensing image
作者:
姬晓飞1 秦宁丽2 刘洋1
1. 沈阳航空航天大学 自动化学院, 辽宁 沈阳 110136;
2. 北京国电通网络技术有限公司, 北京 100070
Author(s):
JI Xiaofei1 QIN Ningli2 LIU Yang1
1. School of Automation, Shenyang Aerospace University, Shenyang 110136, China;
2. Beijing GuoDianTong Network Technology Co. Ltd, Beijing 100070, China
关键词:
光学遥感图像多特征的决策级融合分层的BoF-SIFT特征SC形状特征Hu不变矩特征支持向量机
Keywords:
optical remote sensing imagemulti-features decision level fusionhierarchical BoF-SIFT featureshape context featureHu moment invariantssupport vector machine
分类号:
TP751.1
DOI:
10.11992/tis.201511011
摘要:
基于单一特征的光学遥感图像多目标分类识别存在准确性较差的问题,提出一种新的基于多特征决策级融合的多目标分类识别算法。首先对光学遥感图像目标提取3种能够同时满足平移、旋转和尺度不变性的特征:可以描述局部和全局分布特性的分层BoF-SIFT特征,描述目标边缘轮廓点信息的改进后的SC形状特征,对图像中较大目标识别较好的Hu不变矩特征;其次采用基于径向基核函数的一对一支持向量机算法分别获得3种特征的目标识别概率,并设计了一种多特征决策级加权融合的策略实现对多目标的分类。经多次实验验证该算法对光学遥感图像多目标具有较好的分类识别性能,达到了93.52%的正确识别率。
Abstract:
A novel multi-feature decision level fusion recognition algorithm is proposed to solve the problem of poor levels of accuracy with single feature based multi-target classification of optical remote sensing images. Firstly, three kinds of features which can not only meet translation, rotation, and scale invariance are extracted. One is the hierarchical BoF-SIFT feature which can simultaneously describe local and global distributions. Another is the improved shape context feature which is used to describe the target edge contour point information. The other one is Hu moment invariants which gives better levels of recognition performance for large targets. Secondly, the recognition probabilities of these features are obtained using a one versus one support vector machine based on a radial basis function. Thirdly a strategy for multi-feature decision level fusion is designed. A large number of experiments show that the algorithm for multi-target classification of optical remote sensing images performs better with the recognition rate of targets reaching 93.52%.

参考文献/References:

[1] 姬晓飞,秦宁丽. 基于光学遥感图像的目标检测与分类识别方法[J]. 沈阳航空航天大学学报, 2015, 32(1):23-31. JI Xiaofei, QIN Ningli. Target detection and classified recognition method based on optical remote sensing image[J]. Journal of Shenyang aerospace university, 2015, 32(1):23-31.
[2] 杨斌, 赵红漫, 赵宗涛, 等. 一个改进的遥感图像目标纹理分类识别算法[J]. 微电子学与计算机, 2004, 21(9):111-113. YANG Bin, ZHAO Hongman, ZHAO Zongtao, et al. A removed texture classification and distinction algorithm of remoted sensing martial object[J]. Microelectronics & computer, 2004, 21(9):111-113.
[3] CHEN Zhong, LIU Jianguo, WANG Guoyou. A new circle targets extraction method from high resolution remote sensing imagery[C]//Proceedings of the 2011 Fourth International Workshop on Advanced Computational Intelligence. Wuhan, 2011:529-533.
[4] LIU Ge, SUN Xian, FU Kun, et al. Aircraft recognition in high-resolution satellite images using coarse-to-fine shape prior[J]. IEEE geoscience and remote sensing letters, 2013, 10(3):573-577.
[5] 刘帅, 李士进, 冯钧. 多特征融合的遥感图像分类[J]. 数据采集与处理, 2014, 29(1):108-115. LIU Shuai, LI Shijin, FENG Jun. Remote sensing image classification based on adaptive fusion of multiple features[J]. Journal of data acquisition and processing, 2014, 29(1):108-115.
[6] LI Xiangjuan, SUN Xian, SUN Hao, et al. Generalized multiple kernel framework for multiclass geospatial objects detection in high-resolution remote sensing images[J]. Optical engineering, 2012, 51(1):1-10.
[7] SHARMA G, CHAUDHURY S, SRIVASTAVA J B. Bag-of-features kernel eigen spaces for classification[C]//Proceedings of the 19th International Conference on Pattern Recognition. Tampa, FL, 2008:1-4.
[8] ARTHUR D, VASSILVITSKⅡ S. K-means++:the advantages of careful seeding[C]//Proceedings of the eighteenth annual ACM-SIAM symposium on Discrete algorithms. Philadelphia, PA, 2007:1027-1035.
[9] LAZEBNIK S, SCHMID C, PONCE J. Beyond bags of features:spatial pyramid matching for recognizing natural scene categories[C]//Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition. New York, NY, USA, 2006:2169-2178.
[10] BELONGIE S, MALIK J, PUZICHA J. Matching shapes[C]//Proceedings of the Eighth International Conference on Computer Vision. Vancouver, BC, 2001, 1:454-461.
[11] 张儒良, 席泓, 王林. 一种基于Hu不变矩的匹配演化算法[J]. 西南师范大学学报:自然科学版, 2012, 37(5):11-15. ZHANG Ruliang, XI Hong, WANG Lin. An image matching evolutionary algorithm based on Hu invariant moments[J]. Journal of southwest China normal university:natural science edition, 2012, 37(5):11-15.
[12] HUANG Qirui, WU Guangmin, CHEN Jianming, et al. Automated remote sensing image classification method based on FCM and SVM[C]//Proceedings of the 20122nd International Conference on Remote Sensing, Environment and Transportation Engineering. Nanjing, 2012:1-4.
[13] 薛宁静. 多类支持向量机分类器对比研究[J]. 计算机工程与设计, 2011, 32(5):1792-1795. XUE Ningjing. Comparison of multi-class support vector machines[J]. Computer engineering and design, 2011, 32(5):1792-1795.

备注/Memo

备注/Memo:
收稿日期:2015-11-10。
基金项目:国家自然科学基金项目(61103123);辽宁省高等学校优秀人才支持计划项目(LJQ214018);辽宁省自然科学基金项目(2015020101).
作者简介:姬晓飞,女,1978年生,副教授,博士,主要研究方向为视频分析与处理、模式识别。承担国家自然科学基金、教育部留学回国启动基金等多项课题研究。发表学术论文40余篇,被SCI、EI检索20余篇。参与编著英文专著1部;秦宁丽,女,1991年生,硕士研究生,主要研究方向为遥感图像处理与分析。发表学术论文1篇;刘洋,1977年生,副教授,博士,主要研究方向为图像处理、模式识别。
通讯作者:姬晓飞.E-mail:jixiaofei7804@126.com
更新日期/Last Update: 1900-01-01