计算机科学 ›› 2017, Vol. 44 ›› Issue (7): 315-317.doi: 10.11896/j.issn.1002-137X.2017.07.057

• 图形图像与模式识别 • 上一篇    下一篇

基于图论的无监督区域遥感图像检索算法研究

李丽萍,赵传荣,孔德仁,王芳   

  1. 南京理工大学机械工程学院 南京210094,安徽工业大学电气与信息工程学院 马鞍山243032,南京理工大学机械工程学院 南京210094,南京理工大学机械工程学院 南京210094
  • 出版日期:2018-11-13 发布日期:2018-11-13
  • 基金资助:
    本文受国家自然科学基金(11372143),国防科工局基础计量项目(J092013B003)资助

Research on Unsupervised Regional Remote Sensing Image Retrieval Algorithm Based on Graph Theory

LI Li-ping, ZHAO Chuan-rong, KONG De-ren and WANG Fang   

  • Online:2018-11-13 Published:2018-11-13

摘要: 为了改善基于内容的遥感图像检索技术,以遥感图像区域检索为框架,提出了一种新的基于图论的无监督学习遥感图像检索算法。首先,提出的方法 用图表为每一幅图像建模,同时结合局部信息和相关的空间结构,提供基于区域的图像代表。将每一幅图像初步划分为不同的区域,再通过属性关系图建模,节点和边界分别代表区域特征和它们之间存在的空间关系。然后,通过评估基于图像的相似点实现最相似于查询图像的图像检索。为匹配相应的图像以及按照图像相似点实现图像检索,采用了结合子图同构算法和光谱图嵌入技术的新型非精确图像匹配策略。实验结果显示,与其他两种无监督遥感图像检索方法相比,所提方法的检索性能明显改善。

关键词: 图像检索,图论,无监督学习,属性关系图(ARG),子图同构

Abstract: In order to improve the content based remote sensing image retrieval technology,a new image retrieval algorithm based on graph theory was proposed.First,the proposed method models each image by a graph and combines local information and related spatial structures,which provides the region based image representation.Each image is initially divided into different regions.The nodes and boundaries of the attribute relation graph respectively represent the regionalfeatures and the spatial relations between them.Then,image retrieval is achieved based on image similarity.To match the corresponding image and realize image retrieval according to image similarities,a new type of non-accurate image matching strategy is used based on sub-graph isomorphism algorithm and spectral graph embedding technology.The experimental results show that compared with the most advanced unsupervised remote sensing image retrieval methods,the retrieval performance of the proposed method is significantly improved.

Key words: Image retrieval,Graph theory,Unsupervised learning,Attribute relation graph(ARG),Sub-graph isomorphism

[1] NIU L,NI L,MIAO Y,et al.The compressed domain based on ROI multi spectral remote sensing image retrieval Chinese [J].Journal of Image and Graphics,2005,10(10):1212-1217.(in Chinese) 牛蕾,倪林,MIAO Y,等.基于ROI的压缩域多谱段遥感图像的检索[J].中国图象图形学报,2005,10(10):1212-1217.
[2] ZHU J L,LI S J,WAN D S,et al.Remote sensing image retrie-val,feature selection and semi supervised learning based on China [J].Journal of Image and Graphics,2011,16(8):1474-1482.(in Chinese) 朱佳丽,李士进,万定生,等.基于特征选择和半监督学习的遥感图像检索[J].中国图象图形学报,2011,16(8):1474-1482.
[3] YANG J,LIU J B,DAI Q.An Improved Remote Sensing Imgae Retrieval Method Based on Bag of Word Framework[J].Geomatics and Information Science of Wuhan University, 2014,38(9):1109-1113.(in Chinese) 杨进,刘建波,戴芹.一种改进包模型的遥感图像检索方法[J].武汉大学学报(信息科学版),2014,39(9):1109-1113.
[4] TANG X H,QIN K,MENG L K.Qualitative description model of directional relations based on topological constraints [J].Journal of Surveying and Mapping,2014,43(4):396-403.(in Chinese) 唐雪华,秦昆,孟令奎.基于拓扑约束的方向关系定性描述模型[J].测绘学报,2014,43(4):396-403.
[5] XU Y X,CHEN F.Recent research progress of local image de-scriptor [J].China Journal of Image and Graphics,2015,20(9):1133-1150.(in Chinese) 许允喜,陈方.局部图像描述符最新研究进展[J].中国图象图形学报,2015,20(9):1133-1150.
[6] DEMIR B,BRUZZONE L.A Novel Active Learning Method in Relevance Feedback for Content-Based Remote Sensing Image Retrieval[J].IEEE Transactions on Geoscience & Remote Sen-sing,2015,53(5):2323-2334.
[7] DU Z,LI X,LU X.Local structure learning in high resolution remote sensing image retrieval[J].Neurocomputing,2016,207:813-822.
[8] SEVILLA J,JIMENEZ L I,PLAZA A.Sparse Unmixing-Based Content Retrieval of Hyperspectral Images on Graphics Proces-sing Units[J].IEEE Geoscience & Remote Sensing Letters,2015,12(12):2443-2447.
[9] CHAUDHURI B,DEMIR B,BRUZZONE L,et al.Region-Based Retrieval of Remote Sensing Images Using an Unsupervised Graph-Theoretic Approach[J].IEEE Geoscience & Remote Sensing Letters,2016,13(7):987-991.
[10] JIAO L,TANG X,HOU B,et al.SAR Images Retrieval Based on Semantic Classification and Region-Based Similarity Measure for Earth Observation[J].IEEE Journal of Selected Topics in Applied Earth Observations & Remote Sensing,2015,8(8):1-16.
[11] JIANG J,MA X,CAI Z,et al.Sparse Support Regression forImage Super-Resolution[J].IEEE Photonics Journal,2015,7(5):1.
[12] CHEN X,HAN G,ZHANG Z.Real-time video super-resolution restruction based on GPU acceleration[J].Journal of Computer Applications,2013,3(12):3540-3543.
[13] SEDAGHAT A,EBADI H.Remote Sensing Image MatchingBased on Adaptive Binning SIFT Descriptor[J].IEEE Transactions on Geoscience & Remote Sensing,2015,53(10):5283-5293.
[14] DEMIR B,BRUZZONE L.Hashing-Based Scalable Remote Sen-sing Image Search and Retrieval in Large Archives[J].IEEE Transactions on Geoscience & Remote Sensing,2015,54:1-13.
[15] YANG Y,NEWSAM S.Geographic Image Retrieval Using Local Invariant Features[J].IEEE Transactions on Geoscience & Remote Sensing,2013,1(2):818-832.
[16] ZHOU Y G,WANG P,GAO Y H.Remote Sensing Image Classification with Bag-of-Visval-Words Model[J].Journal of Chongqing University Technology(Natural Science),2015,29(5):71-77.(in Chinese) 周宇谷,王平,高颖慧.基于视觉词袋模型的遥感图像分类方法[J].重庆理工大学学报(自然科学),2015,29(5):71-77.

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!