利用贝叶斯网络融合空间上下文的高分辨遥感图像分类
收稿日期: 2010-08-02
修回日期: 2010-10-28
网络出版日期: 2011-01-25
基金资助
国家自然科学基金资助项目（60373000）
Integrating Contexts into the Classification of HighResolution Remote Sensing Images Using the Bayesian Networks
Received date: 2010-08-02
Revised date: 2010-10-28
Online published: 2011-01-25
针对高分辨遥感图像，本文提出了一种基于贝叶斯网络的上下文模型，以及基于该模型的面向对象的遥感图像分类方法。首先，利用支持向量机（SVM）实现分割区域的初始分类，获得各个类别的候选区域。然后，利用提出的上下文模型融合候选区域及其周围区域的上下文信息，通过贝叶斯网络推理，将候选区域分类到各类地物类型中。基于贝叶斯网络的上下文模型由候选区域节点、相关区域节点和上下文节点三部分组成。对于不同类型的地物，通过贝叶斯网络的结构学习算法学习得到不同的空间关系作为上下文节点。因此，该模型能够针对不同的地物类别利用不同的空间上下文信息，使得分类过程更智能和有效。实验结果表明，本文提出的算法能够很好地利用上下文信息，对高分辨遥感图像中的各种地物进行有效的分类和检测。
程环环，王润生 . 利用贝叶斯网络融合空间上下文的高分辨遥感图像分类[J]. 计算机工程与科学, 2011 , 33(1) : 70 -76 . DOI: 10.3969/j.issn.1007130X.2011.
In this paper, a Bayesian networkbased context model (RCBN) is presented to classify high resolution remote sensing (HRRS) images. First of all, image regions are classified by SVMs and candidate regions for each ground cover types are obtained. Then, the hybrid streams of candidate regions and spatial context information are piped into the context model, which will produce the category labels of regions by performing inference through the network. The RCBN consists of three kinds of nodes, which are nodes for candidate regions, related regions and contexts. The context nodes vary with different ground cover types, which are learned by the structure learning algorithm from training samples. Therefore, our RCBN model is capable of using the specific context information for each ground cover type, which makes the classification process more intelligent and efficient. The performance of the approach is evaluated qualitatively and quantitatively with comparative experiments, and the results show that the proposed methods are able to automatically classify and detect segments belonging to the same object classes.
［1］Jensen J R. Introductory Digital Image Processing: A Remote Sensing Perspective［M］.Third Edition. Beijing: China Machine Press, 2007.
［2］黄昕, 李平湘, 张良培. 基于多层形状特征提取与融合的城市高光谱影像解译 ［J］.测绘科学, 2009, 34（6）:6264.
［3］Fauvel M, et al. Spectral and Spatial Classification of Hyperspectral Data Using SVMs and Morphological Profiles ［J］. IEEE Trans on Geosience and Remote Sensing, 2008, 1(12)：49344837.
［4］Porway J, et al. A Hierarchical and Contextual Model for Aerial Image Understanding ［C］∥Proc of IEEE Conf on Computer Vision and Pattern Recognition,2008.
［5］Zhong P,Wang R. A Multiple Conditional Random Fields Ensemble Model for Urban Area Detection in Remote Sensing Optical Images ［J］. IEEE Trans on Geoscience and Remote Sensing, 2007, 45(12):39783988.
［6］Bhagavathy S,Manjunath B S. Modeling and Detection of Geospatial Objects Using Texture Motifs ［J］. IEEE Trans on Geoscience Remote Sensing, 2006, 44(12):37063715.
［7］Li Y,Bretschneider T R. SemanticSensitive Satellite Image Retrieval ［J］. IEEE Trans on Geoscience and Remote Sensing, 2007,45(4):853860.
［8］Heitz G.Graphical Models for HighLevel Computer Vision ［D］. Stanford University, 2009.
［9］Akcay H G,Aksoy S. Automatic Detection of Geospatial Objects Using Multiple Hierarchical Segmentations ［J］. IEEE Trans on Geoscience and Remote Sensing, 2008, 46(7):20972111.
［10］Zhou H,Wang R,Wang C. A Novel Extended LocalBinaryPattern Operator for Texture Analysis ［J］. Information Sciences, 2008, 178(22):43144325.
［11］Chang C C,Lin C J. LIBSVM: A Library for Support Vector Machines ［EB/OL］.［20080807］.http://www.csie.ntu.edu.tw/~cjlin/libsvm.
［12］Friedman N. Learning Belief Networks in the Presence of Missing Values and Hidden Variables ［C］∥Proc of Int’l Conf on Machine Learning,1997:125133.
［13］Geman S，Geman D. Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images［J］.Journal of Applied Statistics,1993,20(5):2562.
［14］Yu Q,Gong P,Clinton N,et al.ObjectBased Detailed Begatation Classification with Airborne High Spatial Resolution Remote Sensing Imagery［J］.Photogrammetric Engineering and Remote Sensing,2006,72(7):799811.
/
| 〈 |
|
〉 |