高分辨率遥感影像中的城区与建筑物检测方法研究
|
摘要
在高分辨率遥感影像中，城区与建筑物作为两类重要的地物目标，其自动提取和解译在城市建设，GIS系统更新，数字化城市以及军事侦察等多个领域都有着重要应用。为此，本文在从高分辨率遥感影像上进行城区和建筑物提取方面进行了有意义的探索和尝试。概括而言，论文主要进行了如下四个方面的研究工作：
首先，对高分辨率遥感影像中常用的特征分析方法进行了较为全面的综述研究。根据特征类型不同将其抽象为三大类：光谱特征、纹理特征和局部关键点特征，然后阐述了每一类特征分析方法的基本思想，并对其中具有代表性的特征提取方法进行了较为深入的分析和讨论，在目标特征提取方面为后续的研究工作奠定了良好的基础。
其次，提出一种有监督的多特征融合的城区检测算法。该方法将最优的特征融合问题转化为多个特征核函数的组合问题，并采用多核学习的方法在进行多特征融合的同时，完成城区SVM分类器的学习。与传统的基于单种图像特征的城区检测方法相比，该算法通过有效地融合多种图像特征，大大提高了城区检测算法性能。
然后，传统的城区检测算法通常一次只能处理一幅输入影像，且为提高识别精度，需要提供大量人工标注的数据样本，过程十分繁琐。面对海量级的遥感影像数据，难以满足当前各种应用的自动化和实时化需求。针对这一问题，提出一种无监督的基于多幅高分辨率遥感影像的城区协同检测算法。首先对输入影像集中的各个影像进行独立处理：利用改进的Harris算子检测影像中的角点，然后根据影像中的角点分布提取其中的候选城区区域。在得到各个影像中的候选城区区域后，把它们合并起来得到一个候选城区区域集合，然后对它们进行纹理特征建模，最后利用谱聚类和graph cut算法从中筛选出真正的城区区域。与现有的无监督城区检测方法相比，由于该算法充分利用了多幅影像提供的信息，因此检测精度得到提高。与已有的有监督城区检测方法相比，该算法检测精度与它们相当，但由于无需任何训练样本，因此在效率上得到提高。
最后，在人类视觉认知理论的启发下，通过仿效人类认知复杂对象时通常采取的由易到难的策略，提出一种基于视觉认知启发的城区建筑物分级提取算法。该方法将面向对象的思想融入到基于邻域总变分的建筑物分割方法中，并通过分析分割后不同类型建筑物提取的难易程度，提出一种多特征融合的建筑物对象分级提取策略：首先通过形状分析检测一部分分割完整的矩形建筑物目标，然后采用新提出的多方向形态学道路滤波算法将建筑物与邻近光谱相似的道路目标分离，确保每一个候选建筑物目标都是独立的对象，最后利用初提取的建筑物对象和已剔除的非建筑物对象作为样本建立概率模型，根据贝叶斯准则进行复杂建筑物后提取。与传统的建筑物检测算法相比，该算法的优势在于能够直接从测试影像中提取训练样本用于建筑物目标模型学习，由于训练样本和测试对象在同一尺度和光照条件下获得（例如，它们均来自于同一幅影像），这充分保证了模型的置信度和稳健性；而且样本集的建立过程不需要人工参与，这也满足了算法自动化的需求。实验表明：该方法可以检测同一幅影像中具有不同形状结构和光谱特性的建筑物目标，准确率高、鲁棒性好，具有较高的实际应用价值。
In the past few years, urban area and building detection from high-resolutionremotely sensed image have become crucial for several applications. The main one is toupdate the geographic information databases, which are critical sources of information indiverse fields such as cartography, city planning and change detection. To this end, thisdissertation is trying to propose several algorithms for urban and building detection fromhigh resolution remotely sensed images. Concretely, main contents of this dissertationinclude the following four parts:
Firstly, we briefly review previous works on feature extraction approaches, anddivide them into three categories: spectral feature extraction, texture feature extraction andlocal feature extraction. Afterwards, we descible in detail the basic concept and principleof feature extraction approaches, which are commonly used in high-resolution imageinterpretation.
Secondly, a supervised urban detection approach based on mutil-feature fusion modelis proposed. In this method, we treat the problem of mutil-feature fusion as estimating aweighed linear combination of mutilple feature kernel functions. And the weight for eachfeature kernel function is automatically estimated in a multi-kernel support vector machine(SVM) learning framework during the training stage. In the classification stage, we firstdivide the test image into several non-overlapping image blocks, and then apply the SVMclassifier to determine whether each image block belongs to urban or not. Compared totraditional approaches using only texture information for urban detection, experimentalresults demonstrate that fusing multiple features can help improving urban detectionaccuracy rate.
Thirdly, given a set of high resolution satellite images covering different scenes, anunsupervised approach to simultaneously detect possible urban regions from them isproporsed. The motivation behind is that: the frequently recurring appearance patterns orrepeated textures corresponding to common objects of interest (e.g. urban area) in theinput image dataset can help us discriminate urban area from others. With this inspiration,our method consists of two steps. First, we extract a large set of local feature point by Harris corner detector. In order to achieve a reliable extraction of corners from urban areas,we further propose two criterions to validate and filter them. Afterwards, we incorporatethe extracted corners into a likelihood function to locate candidate regions in each inputimages. Given a set of candidate urban regions, in the second stage, we formulize theurban detection process as an unsupervised classification problem. The candidate regionsare modeled through their histogram representation of Gabor texture features, and theclassification problem is solved by spectrum clustering and graph cuts. The experimentalresults show that the proposed approach is capable of and efficient at simultaneouslydetecting urban regions from multiple high-resolution satellite images, and performscomparable or even better in comparison with the state-of-the-art supervised method.
In high-resolution satellite image, buildings can be considered as clustered objectsbelonging to the same category. Human perception of such objects consists of an initialidentification of simple instances followed by a recognition of more complicated ones bydeduction. Inspired by this theory, a novel hierarchical building extraction framework isproposed to simulate the process, which includes three major components. Firstly, a totalvariation based segmentation algorithm is presented to decompose the given image intoobject-level elements. Then, shape analysis is applied to extract some common and easilyidentified rectangular buildings. To ensure each candidate of building target is isolated, amultidirectional morphological road-filtering algorithm is designed to separate thebuildings from their neighboring roads with similar spectrum. Finally, the detection ofbuildings with complex structures is formulated as a deduction problem based onpreceding extracted information in terms of maximum a posteriori (MAP) estimation, anda Bayesian based approach is put forward to deal with it. Comparing to the conventionalway of detecting objects through the information learned from previously collectedtraining samples, our method has two advantages. First, our approach can learn buildingmodels directly from the original images. Therefore, it is highly automatic, for no manualaid is required in the collection of training data. More importantly, since the training dataare collected from the identical scale and illumination conditions (e.g., in the same image),our model is more discriminating. This enables that the proposed framework has theability to detect building with complex structures and varying spectral response,independent of pre-defined and limited building models.
首先，对高分辨率遥感影像中常用的特征分析方法进行了较为全面的综述研究。根据特征类型不同将其抽象为三大类：光谱特征、纹理特征和局部关键点特征，然后阐述了每一类特征分析方法的基本思想，并对其中具有代表性的特征提取方法进行了较为深入的分析和讨论，在目标特征提取方面为后续的研究工作奠定了良好的基础。
其次，提出一种有监督的多特征融合的城区检测算法。该方法将最优的特征融合问题转化为多个特征核函数的组合问题，并采用多核学习的方法在进行多特征融合的同时，完成城区SVM分类器的学习。与传统的基于单种图像特征的城区检测方法相比，该算法通过有效地融合多种图像特征，大大提高了城区检测算法性能。
然后，传统的城区检测算法通常一次只能处理一幅输入影像，且为提高识别精度，需要提供大量人工标注的数据样本，过程十分繁琐。面对海量级的遥感影像数据，难以满足当前各种应用的自动化和实时化需求。针对这一问题，提出一种无监督的基于多幅高分辨率遥感影像的城区协同检测算法。首先对输入影像集中的各个影像进行独立处理：利用改进的Harris算子检测影像中的角点，然后根据影像中的角点分布提取其中的候选城区区域。在得到各个影像中的候选城区区域后，把它们合并起来得到一个候选城区区域集合，然后对它们进行纹理特征建模，最后利用谱聚类和graph cut算法从中筛选出真正的城区区域。与现有的无监督城区检测方法相比，由于该算法充分利用了多幅影像提供的信息，因此检测精度得到提高。与已有的有监督城区检测方法相比，该算法检测精度与它们相当，但由于无需任何训练样本，因此在效率上得到提高。
最后，在人类视觉认知理论的启发下，通过仿效人类认知复杂对象时通常采取的由易到难的策略，提出一种基于视觉认知启发的城区建筑物分级提取算法。该方法将面向对象的思想融入到基于邻域总变分的建筑物分割方法中，并通过分析分割后不同类型建筑物提取的难易程度，提出一种多特征融合的建筑物对象分级提取策略：首先通过形状分析检测一部分分割完整的矩形建筑物目标，然后采用新提出的多方向形态学道路滤波算法将建筑物与邻近光谱相似的道路目标分离，确保每一个候选建筑物目标都是独立的对象，最后利用初提取的建筑物对象和已剔除的非建筑物对象作为样本建立概率模型，根据贝叶斯准则进行复杂建筑物后提取。与传统的建筑物检测算法相比，该算法的优势在于能够直接从测试影像中提取训练样本用于建筑物目标模型学习，由于训练样本和测试对象在同一尺度和光照条件下获得（例如，它们均来自于同一幅影像），这充分保证了模型的置信度和稳健性；而且样本集的建立过程不需要人工参与，这也满足了算法自动化的需求。实验表明：该方法可以检测同一幅影像中具有不同形状结构和光谱特性的建筑物目标，准确率高、鲁棒性好，具有较高的实际应用价值。
In the past few years, urban area and building detection from high-resolutionremotely sensed image have become crucial for several applications. The main one is toupdate the geographic information databases, which are critical sources of information indiverse fields such as cartography, city planning and change detection. To this end, thisdissertation is trying to propose several algorithms for urban and building detection fromhigh resolution remotely sensed images. Concretely, main contents of this dissertationinclude the following four parts:
Firstly, we briefly review previous works on feature extraction approaches, anddivide them into three categories: spectral feature extraction, texture feature extraction andlocal feature extraction. Afterwards, we descible in detail the basic concept and principleof feature extraction approaches, which are commonly used in high-resolution imageinterpretation.
Secondly, a supervised urban detection approach based on mutil-feature fusion modelis proposed. In this method, we treat the problem of mutil-feature fusion as estimating aweighed linear combination of mutilple feature kernel functions. And the weight for eachfeature kernel function is automatically estimated in a multi-kernel support vector machine(SVM) learning framework during the training stage. In the classification stage, we firstdivide the test image into several non-overlapping image blocks, and then apply the SVMclassifier to determine whether each image block belongs to urban or not. Compared totraditional approaches using only texture information for urban detection, experimentalresults demonstrate that fusing multiple features can help improving urban detectionaccuracy rate.
Thirdly, given a set of high resolution satellite images covering different scenes, anunsupervised approach to simultaneously detect possible urban regions from them isproporsed. The motivation behind is that: the frequently recurring appearance patterns orrepeated textures corresponding to common objects of interest (e.g. urban area) in theinput image dataset can help us discriminate urban area from others. With this inspiration,our method consists of two steps. First, we extract a large set of local feature point by Harris corner detector. In order to achieve a reliable extraction of corners from urban areas,we further propose two criterions to validate and filter them. Afterwards, we incorporatethe extracted corners into a likelihood function to locate candidate regions in each inputimages. Given a set of candidate urban regions, in the second stage, we formulize theurban detection process as an unsupervised classification problem. The candidate regionsare modeled through their histogram representation of Gabor texture features, and theclassification problem is solved by spectrum clustering and graph cuts. The experimentalresults show that the proposed approach is capable of and efficient at simultaneouslydetecting urban regions from multiple high-resolution satellite images, and performscomparable or even better in comparison with the state-of-the-art supervised method.
In high-resolution satellite image, buildings can be considered as clustered objectsbelonging to the same category. Human perception of such objects consists of an initialidentification of simple instances followed by a recognition of more complicated ones bydeduction. Inspired by this theory, a novel hierarchical building extraction framework isproposed to simulate the process, which includes three major components. Firstly, a totalvariation based segmentation algorithm is presented to decompose the given image intoobject-level elements. Then, shape analysis is applied to extract some common and easilyidentified rectangular buildings. To ensure each candidate of building target is isolated, amultidirectional morphological road-filtering algorithm is designed to separate thebuildings from their neighboring roads with similar spectrum. Finally, the detection ofbuildings with complex structures is formulated as a deduction problem based onpreceding extracted information in terms of maximum a posteriori (MAP) estimation, anda Bayesian based approach is put forward to deal with it. Comparing to the conventionalway of detecting objects through the information learned from previously collectedtraining samples, our method has two advantages. First, our approach can learn buildingmodels directly from the original images. Therefore, it is highly automatic, for no manualaid is required in the collection of training data. More importantly, since the training dataare collected from the identical scale and illumination conditions (e.g., in the same image),our model is more discriminating. This enables that the proposed framework has theability to detect building with complex structures and varying spectral response,independent of pre-defined and limited building models.
引文
[1]李德仁.论21世纪遥感与GIS的发展.武汉大学学报(信息科学版),2003,28(2):127:130
[2]黄昕.高分辨率遥感影像多尺度纹理、形状、特征提取与面向对象分类方法研究:[博士学位论文].武汉:武汉大学,2009
[3]江万寿.航空影像多视匹配与规则建筑物自动提取方法研究:[博士学位论文].武汉:武汉大学,2001
[4] B. Sirmacek, C. Unsalan. Urban Area Detection Using Local Feature Points andSpatial Voting. IEEE Geoscience and Remote Sensing Letter.2010,7(1):146:150
[5] P. Zhon, R. Wang. A multiple conditional random fields ensemble model for urbanarea detection in remote sensing optical images. IEEE Transaction on Geoscienceand Remote Sensing,2007,45(12):3978-3988
[6] W. Sun, G. Xu, P. Gong. Textual and Local Spatial Statistics for the Object-orientedClassification of Urban Areas Using High Resolution Imagery. International.Journal of Remote Sensing,2006,27(22):4963-4990
[7] M. Pesaresi, A. Gerhardinger, F. Kayitakire. A robust built-up area presence indexby anisotropic rotation-invariant textural measure. IEEE Journal of Selected Topicsin Applied Earth Observations and Remote Sensing,2008,1(3):180-192
[8] J. A. Benediktsson, J. A. Palmason, J. R. Sveinsson, Classification of hyperspectraldata from urban areas based on extended morphological profiles. IEEE Transactionon Geoscience and Remote Sensing,2005,43(3):480-491
[9] M. Pesaresi, J. A. Benediktsson. A new approach for the morphologicalsegmentation of high-resolution satellite imagery. IEEE Transaction on Geoscienceand Remote Sensing,2001,39(2):309-320
[10] C. ünsalan, K. L. Boyer. Classifying land development in high-resolutionpanchromatic satellite images using straight-line statistics. IEEE Transaction onGeoscience and Remote Sensing,2004,42(4):907-919
[11] X. Huang, L. P. Zhang, P. Li. Classification of very high spatial resolution imagerybased on the fusion of edge and multispectral information. PhotogrammetriEngineering and Remote Sensing,2008,74(12):1585-1596
[12] L. Weizman, J. Goldberger. Detection of Urban Zones in Satellite Images UsingVisual Words. IEEE conference on Geoscience and Remote Sensing Symposium(IGARSS),2008:160-163
[13] L. Fei-Fei, P. Perona. A Bayesian hierarchical model for learning natural scenecategories. IEEE Conference on Computer Vision and Pattern Recognition,2005,2:524-531
[14] R. Fergus, P. Perona, A. Zisserman. Object class recognition by unsupervisedscale-invariant learning. IEEE Conference on Computer Vision and PatternRecognition,2003,2:264-271
[15] B. Sirmacek, C. Unsalan. Urban-Area and Building Detection Using SIFTKeypoints and Graph Theory. IEEE Transaction on Geoscience and. RemoteSensing,2009,47(4):1156-1167
[16] B. Sirmacek, C. Unsalan. Using Local Features to Measure Land Development inUrban Regions. Pattern Recognition Letters,2010,31(10):1155:1159
[17]朱江洪,李江风,叶菁.利用决策树工具的土地利用类型遥感识别方法研究.武汉大学学报,2011,36(3):301-305
[18] V. Vapnik. The Nature of Statistical Learning Theory. New York: Springer-Verlag,1995
[19] M. Pal, P. M. Mather. Support Vector Machines for Classification in Remote Sensing.International Journal of Remote Sensing,2005,26(5):1007-1011
[20]骆剑承,周成虎,杨艳.人工神经网络遥感影像分类模型及其与知识集成方法研究.遥感学报,2001,5(2):121-129
[21] Y. Freund, R. E. Schapire. A Decision-theoretic generalization of online learningand an application to boosting. Journal of Computer and System Sciences,1997,55(1):119-120
[22] M. J. Barnsley, S. L. Rarr. Inferring Urban Land Use From Satellite Sensor UsingKernel-based Spatial Reclassification. Photogrammetric Engineering and RemoteSensing,1996,62(7):949-958
[23] M. Fauvel, J. Chanussot, J. A. Benediktsson. Decision fusion for the classificationof urban remote sensing images. IEEE Transaction on Geoscience and RemoteSensing,2006,44(10):2828-2838
[24]黄昕,张良培,李平湘.高空间分辨率遥感图像分类的SSMC方法.中国图象图形学报,2006,11(4):529-534
[25]潘欣,张树清,李晓峰等.利用α-torrent粗集的遥感影像分类方法研究.武汉大学学报:信息科学版,2009,34(10):1240-1244
[26]陶超,谭毅华,彭碧发等.一种基于概率潜在语义模型的高分辨率遥感影像分类方法.测绘学报,2011,40(2):156-162
[27]唐亮,谢维信,黄建军等.从航空影像中自动提取高层建筑物.计算机学报.2005,28(7):1199-1204
[28] Z. W. Kim, R. Nevatia. Automatic description of complex buildings from multipleimages. Computer Vision and Image Understanding.2004,96:60-95
[29] E. Simonton, H. Oriot, R. Garello. Rectangular Building Extraction fromStereoscopic Airborne Radar Images. IEEE Transactions on Geoscience and RemoteSensing.2005,43(10):2386-2395
[30]孙显,王宏琦,张正.基于对象的Boosting方法自动提取高分辨率遥感图像中建筑物目标.电子与信息学报.2009,31(1):177-181
[31] X. Jin, C. Davis. Automated Building Extraction from High-Resolution SatelliteImagery in Urban Areas Using Structural, Contextual, and Spectral Information.EURASIP Journal on Applied Signal Processing,2005,14(5):2196-2206
[32]陶超,谭毅华,蔡华杰等.面向对象的高分辨率遥感影像城区建筑物分级提取方法.测绘学报.2010,39(1):39-45
[33]周亚男,沈占锋,骆剑承等.阴影辅助下的面向对象城市建筑物提取.地理与地理信息科学.2010,26(3):37-40
[34] K. Karantzalos, N. Paragios. Recognition-Driven Two-Dimensional CompetingPriors Toward Automatic and Accurate Building Detection. IEEE Transaction onGeoscience and Remote Sensing,2009,47(1):133-144
[35] G. Gao. An Improved Scheme for Target Discrimination in High-Resolution SARImages. IEEE Transaction on Geoscience and Remote Sensing,2011,49(1):277-294
[36] L. Cheng, J. Y. Gong. Building Boundary Extraction Using Very High ResolutionImages and LiDAR. Acta Geodaetica et Cartographica Sinica,2008,37(3):391-393
[37] A. Sampath, J. Shan. Building Boundary Tracing and Regularization from AirborneLidar Point Clouds. Photogrammetric Engineering&Remote Sensing,2007,73(7):805-812
[38] C. Lin. Perception of3-D objects from an intensity image using simple geometricmodels[[PhD thesis]. America: the University of Southern California,1996
[39]强永刚,殷建平,陈涛等,利用阴影信息检测高分辨率遥感图像中的建筑物.计算机科学,2006,33(8):338-340
[40] K. Y. Ren, H. X. Sun, Q. X. Jia, et al. Building Recognition from Aerial ImagesCombining Segmentation and Shadow, IEEE conference on Intelligent Computingand Intelligent System,2009:578-582
[41] F. Rottensteiner, A. C. Briese. A New Method for Building Extraction in UrbanAreas from High-Resolution LIDAR data. ISPRS Commission III Symposium,Austria,2002
[42] S. Vinson, L. D. Cohen. Multiple Rectangle Model for Buildings Segmentation and3D Scene Reconstruction. International Conference on Pattern Recognition,Quebec-City,2002,3:623-626
[43] S. Noronha, R. Nevatia. Detection and Modeling of Building from Multiple AerialImages. IEEE Transaction on Pattern Analysis and Machine Intelligence,2001,23(5):501-518
[44] J. Hu, S. You, U. Neumann. Approaches to Large-scale Urban Modeling. IEEEComputer Graphics and Applications,2003,23(6):62-69
[45]王继阳,文贡坚,吕金建等.建筑物三维重建方法综述.遥感技术与应用,2009,24(6):832-840
[46]柳嫁航.利用遥感技术进行城市建筑物震害的自动识别与分类方法研究:[博士学位论文].北京:中国地震局地质研究所,2008
[47] A. Huertas, R. Nevatia. Detecting changes in aerial view form man-made structures.Image Vision and Computing,2000,18:583-596
[48]刘亚文,叶晓新.城区人工地物变化检测方法的研究.测绘通报,2001,7:9-11
[49]刘直芳,张继平,张剑清.基于DSM和影像持征的城市变化检测.遥感技术与应用,2002,17(5):240-244
[50] H. Wersing, E. Koerner. Learning Optimized Features for Hierarchical Models forInvariant Recognition. Neural Computation,2003,15(7):1559-1588
[51] T. J. Palmeri, I. Gauthier. Visual object understanding. Nature ReviewsNeuroscience,2004,5(1):291-303
[52]章毓晋.图像工程（上册）:图像处理和分析.北京:清华大学出版社,1993:218-240
[53] J. W. Rouse, R. H. Haas, J. A. Schell, et al. Monitoring the vernal advancement ofnatural vegetation, Final report, NASA/GCSFC, Greenbelt, MD,1974
[54]陈君颖,田庆久.高分辨率遥感植被分类研究.遥感学报,2007,11(2):221-2
[55] L. Bruzzone, L. Carlin. A multilevel context-based system for classification of veryhigh spatial resolution images. IEEE Transactions on Geoscience and RemoteSensing,2006,44(9):2587-2600
[56]余鹏,张震龙,侯至群.基于高斯马尔可夫随机场混合模型的纹理图像分割.测绘学报,2006,35(3):224-228)
[57] R. M. Haralick, Dinstein, K. Shanmugam. Textural feature for image classification.IEEE Transactions on Systems, Man, Cybernetics,1973, SMC-3(6):610-621
[58] G. Cross, A. Jain. Markov random fields texture models. IEEE Transactions onSystem Man Cybernet,1987,17:1087-1095
[59]罗晓燕,王成儒.基于自相关图像的纹理特征检索的研究.计算机应用与软件,2008,25(7):241-243
[60] F. Zhou, J. Feng, Q. Shi. lmage Segmentation Based on Local Fourier CoefficientsHistogram, International Conference on Multispectral Image Processing and PatternRecognition, Wuhan, China,2001
[61] B. S Manjunath, W. Y. Ma. Texture feature for browsing and retrieval of image data.IEEE Transaction on Patten Aanlysis and Machine Intelligence,1996,18(8):837-422
[62] E. Persoon, K. S. Fu. Shape diserimination using Foruier deseriptors. IEEETransaetionson Systems, Manand Cyberneties,1977,7(3):170-179
[63] D. G. Lowe. Distinctive image features from scale-invariant keypoints. InternationalJournal of Computer Vision,2004,60(2):91-110
[64] T. Kadir, M. Brady. Scale, saliency and image description. International Journal ofComputer Vision,2001,45(2):83-105
[65] K. Mikolajczyk, C. Schmid. Indexing based on scale invariant interest points. IEEEProceedings of the8th International Conference on Computer Vision, Vancouver,Canada,2001:525-531
[66] K. Mikolajczyk, C. Schmid. Scale and affine invariant interest point detectors.International Journal of Computer Vision,2004,1(60):63-86
[67] J. Matas, O. Chum., Urban M. Robust wide baseline stereo from maximally stableextremal regions. Image and Vision Computing,2004,22(10):761-767
[68] H. L. Deng, W. Zhang, E. Mortensen. Principal curvature-based region detector forobject recognition. IEEE Conference on Computer Vision and Pattern Recognition,2007,1-8
[69] R. Fergus, P. Perona, A. Zisserman. A sparse object category model for efficientlearning and exhaustive recognition. IEEE Conference on Computer Vision andPattern Recognition, San Diego,2005,1:380-387
[70] N. Dalal, B. Triggs. Histograms of oriented gradients for human detection. IEEEConference on Computer Vision and Pattern Recognition, San Diego, CA,2005:886-893
[71] Q. Zhu, S. Avidan, M. Ye. Fast human detection using a cascade of histograms oforiented gradients. IEEE Conference on Computer Vision and Pattern Recognition,New York, USA,2006:1491-1498
[72] Y. Ke, R. Sukthankar. PCA-SIFT: a more distinctive representation for local imagedescriptors. IEEE Conference Computer Vision and Pattern Recognition,2004:511-517
[73] S. Belongie, J. Malik, and J. Puzicha. Shape Matching and Object RecognitionUsing Shape Contexts. IEEE Transactions on Pattern Analysis and MachineIntelligence,2002,24(24):509-521
[74] W. Freeman, E. Adelson. The design and use of steerable filters. IEEE Transactionson Pattern Analysis and Machine Intelligence,1991,13(9):891-906
[75] K. Mikolajczyk, C. Schmid. A performance evaluation of local descriptors. IEEETransactions on Pattern Analysis and Machine Intelligence,2005,27,1615-1630
[76] F. Schaffalitzky, A. Zisserman. Multi-view Matching for Unordered Image Sets, or“How Do I Organize My Holiday Snaps?”. In Proceedings of the EuropeanConference on Computer Vision.2002,1:414-431
[77]陈星星,张荣.基于多尺度相位特征的图像检索方法.电子与信息学报,2009,31(5):1193-1196
[78]赵一凡,夏良正.基于轮廓波特征的纹理图像识别方法.东南大学学报(自然科学版),2008,38(11):128-131
[79] Baumberg. Reliable feature matching across widely separated views. IEEEConference on Computer Vision and Pattern Recognition.2000. Hilton Head Island,South Carolina, USA
[80] F. Schaffalitzky, A. Zisserman. Multi-view matching for unordered image sets. In:the7thEuropean Conference on Computer Vision.2002. Copenhagen, Denmark
[81] D. Gabor. Theory of communication. Journal of Institute for Electrical Engineering,1946,93:429-457
[82] T. Leung, J. Malik. Representing and recognizing the visual appearance of materialsusing three-dimensional textons. International Journal of Computer Vision,2001,43(1):29-44
[83] M. Varma, A. Zisserman. Classifying images of materials: achieving viewpoint andillumination independence. In: Proceedings of the European Conference onCompute Vision,2002
[84] M. Varma, A. Zisserman. A Statistical Approach to Texture Classi_cation fromSingle Images. International Journal of Computer Vision,2005,62(1):61-81
[85] Witkin A. P. Scale-space filtering. In: Proc. International Joint Conf. on ArtificialIntelligence. Karlsruhe. Germany: Springer Berlin,1983:1019-1022
[86] Lindeberg T. Scale-space theory: A basic tool for analysing structures at differentscales. Journal of Applied Statistics,1994,21(2):224-270
[87]赵玲玲,翁苏明.模式分析的核方法.北京:机械工业出版社,2006
[88]张学工.关于统计学习理论与支持向量机.自动化学报,2000,26(1):32-42
[89]边肇祺,张学工.模式识别（第2版）.北京:清华大学出版社,2000
[90]李晋博,特征提取的核方法与非线性多核学习的研究:[硕士学位论文].上海:华东师范大学武汉,2009
[91] F. Bach, G. Lanckriet, M. Jordan. Multiple kernel learning, conic duality, and theSMO algorithm. In Proceedings of the21st International Conference on MachineLearning, New York, NY, USA,2004:41-48
[92] G. Lanckriet, N. Cristianini, L. E. Ghaoui, et al. Learning the kernel matrix withsemi-definite programming. Journal of Machine Learning Research,2004,5:27-72
[93] M. Varma, D. Ray. Learning the discriminative power-invariance trade-off. IEEEConference on Computer Vision and Pattern Recognition,2007:1150-1157
[94] A. Rakotomamonjy, F. Bach, S. Canu, et al. More efficiency in multiple kernellearning. In IEEE Conference on Machine Learning,2007
[95] A. Bosch, A. Zisserman, X. Munoz. Representing shape with a spatial pyramidkernel. Proceedings of the6th ACM international conference on Image and videoretrieval, Amsterdam, The Netherlands, July09-11,2007:401-408
[96] X. Huagng, L. P. Zhang, P. Li. Classification and extraction of spatial features inurban areas using high resolution multispectral imagery. IEEE Geoscience andRemote Sensing letters,2007,4(2):260-264
[97]王田苗,王君臣,杨艳等.基于Harris角点的内窥镜图像变形全自动校正算法.自动化学报,2011,37(11):1360-1367
[98]吴秀芸,李艳,周华.基于角点检测的建筑物轮廓矢量化方法.遥感信息,2010,5:95-99
[99]C. G. Harris, M. Stephens. A Combined Corner and Edge Detector. In Proceedings ofFourth Alley Vision Conference,1988:182-192
[100] J. Shi, C. Tomasi. Good Features to Track, in: Proceedings of IEEE Conference onComputer Vision and Pattern Recognition,1994:593-600
[101] S. M. Smith, M. Brady. SUSAN—a new approach to low level image processing.International Journal of Computer Vision,1997,23(1):45-78
[102] J. Canny. A computational approach to edge detection. IEEE Transaction on patternrecognition and machine intelligence,1986,8(6):679-698
[103] D. H. Douglas, T. Pecuker. Algorithm for the reduction of the number of pointsrequired to represent a digitized lie or it caricature,1973,10(2):112-122
[104] M. Sezgin, B. Sankur. Survey over image thresholding techniques and quantitativeperformance evaluation. Journal of Electronic Imaging,2003,13(1):146-165
[105] T. Kanungo, D. M. Mount, N. S. Netanyahu. An Efficient k-Means ClusteringAlgorithm: Analysis and Implementation. IEEE Transaction on pattern recognitionand machine intelligence,2002,24(7):881-892
[106] A. Y. Ng, M. I. Jordan, Y. Weiss. On spectral clustering: analysis and an algorithm.Advances in Neural Information Processing Systems,2002,14:849-856
[107] U. Luxburg, M. Planck. A tutorial on spectral clustering. Journal of Statistics andComputing,2007,17(4):225-246
[108] T. Xiang, S. Gong. Spectral clustering with eigenvector selection. PatternRecognition,2008,41(3):1012-1029
[109] Y. Boykov, O. Versker, R. Zabih, Fast approximate energy minimization via graphcuts. IEEE transaction on Pattern Analysis and Machine Intelligence,2001,23(11):1222-1329
[110] T. J. Palmeri, I. Gauthier. Visual object understanding. Nature ReviewsNeuroscience,2004,5(1):291-303
[111] P. Blomgren, T. F. Chan. Color TV: Total Variation Methods for Restoration ofVector Valued Images. IEEE Transactions on Image Processing,1998,7(3):304-309
[112]李书晓,常红星.基于总变分和形态学的航空图像道路检测算法.计算机学报,2007,30(12):2173-2179
[113] F. Cheng, A. N. Venesanopoulous. Adaptive Morphological filter for imageprocessing. IEEE transactions on Image Processing,1992,1(4):533-539
[114] N. Vassis, A. Likas. The Kurtosis-Based Dynamic Approach to Gaussian MixtureModeling. IEEE Transaction on Systems, Man and Cybernetics,1999,29(4):393-399
[115] L. Xu. Bayesian Ying-Yang Machine, Clustering and Number of Clusters. PatternRecognition Letters,1997,18(11):1167-1178
[2]黄昕.高分辨率遥感影像多尺度纹理、形状、特征提取与面向对象分类方法研究:[博士学位论文].武汉:武汉大学,2009
[3]江万寿.航空影像多视匹配与规则建筑物自动提取方法研究:[博士学位论文].武汉:武汉大学,2001
[4] B. Sirmacek, C. Unsalan. Urban Area Detection Using Local Feature Points andSpatial Voting. IEEE Geoscience and Remote Sensing Letter.2010,7(1):146:150
[5] P. Zhon, R. Wang. A multiple conditional random fields ensemble model for urbanarea detection in remote sensing optical images. IEEE Transaction on Geoscienceand Remote Sensing,2007,45(12):3978-3988
[6] W. Sun, G. Xu, P. Gong. Textual and Local Spatial Statistics for the Object-orientedClassification of Urban Areas Using High Resolution Imagery. International.Journal of Remote Sensing,2006,27(22):4963-4990
[7] M. Pesaresi, A. Gerhardinger, F. Kayitakire. A robust built-up area presence indexby anisotropic rotation-invariant textural measure. IEEE Journal of Selected Topicsin Applied Earth Observations and Remote Sensing,2008,1(3):180-192
[8] J. A. Benediktsson, J. A. Palmason, J. R. Sveinsson, Classification of hyperspectraldata from urban areas based on extended morphological profiles. IEEE Transactionon Geoscience and Remote Sensing,2005,43(3):480-491
[9] M. Pesaresi, J. A. Benediktsson. A new approach for the morphologicalsegmentation of high-resolution satellite imagery. IEEE Transaction on Geoscienceand Remote Sensing,2001,39(2):309-320
[10] C. ünsalan, K. L. Boyer. Classifying land development in high-resolutionpanchromatic satellite images using straight-line statistics. IEEE Transaction onGeoscience and Remote Sensing,2004,42(4):907-919
[11] X. Huang, L. P. Zhang, P. Li. Classification of very high spatial resolution imagerybased on the fusion of edge and multispectral information. PhotogrammetriEngineering and Remote Sensing,2008,74(12):1585-1596
[12] L. Weizman, J. Goldberger. Detection of Urban Zones in Satellite Images UsingVisual Words. IEEE conference on Geoscience and Remote Sensing Symposium(IGARSS),2008:160-163
[13] L. Fei-Fei, P. Perona. A Bayesian hierarchical model for learning natural scenecategories. IEEE Conference on Computer Vision and Pattern Recognition,2005,2:524-531
[14] R. Fergus, P. Perona, A. Zisserman. Object class recognition by unsupervisedscale-invariant learning. IEEE Conference on Computer Vision and PatternRecognition,2003,2:264-271
[15] B. Sirmacek, C. Unsalan. Urban-Area and Building Detection Using SIFTKeypoints and Graph Theory. IEEE Transaction on Geoscience and. RemoteSensing,2009,47(4):1156-1167
[16] B. Sirmacek, C. Unsalan. Using Local Features to Measure Land Development inUrban Regions. Pattern Recognition Letters,2010,31(10):1155:1159
[17]朱江洪,李江风,叶菁.利用决策树工具的土地利用类型遥感识别方法研究.武汉大学学报,2011,36(3):301-305
[18] V. Vapnik. The Nature of Statistical Learning Theory. New York: Springer-Verlag,1995
[19] M. Pal, P. M. Mather. Support Vector Machines for Classification in Remote Sensing.International Journal of Remote Sensing,2005,26(5):1007-1011
[20]骆剑承,周成虎,杨艳.人工神经网络遥感影像分类模型及其与知识集成方法研究.遥感学报,2001,5(2):121-129
[21] Y. Freund, R. E. Schapire. A Decision-theoretic generalization of online learningand an application to boosting. Journal of Computer and System Sciences,1997,55(1):119-120
[22] M. J. Barnsley, S. L. Rarr. Inferring Urban Land Use From Satellite Sensor UsingKernel-based Spatial Reclassification. Photogrammetric Engineering and RemoteSensing,1996,62(7):949-958
[23] M. Fauvel, J. Chanussot, J. A. Benediktsson. Decision fusion for the classificationof urban remote sensing images. IEEE Transaction on Geoscience and RemoteSensing,2006,44(10):2828-2838
[24]黄昕,张良培,李平湘.高空间分辨率遥感图像分类的SSMC方法.中国图象图形学报,2006,11(4):529-534
[25]潘欣,张树清,李晓峰等.利用α-torrent粗集的遥感影像分类方法研究.武汉大学学报:信息科学版,2009,34(10):1240-1244
[26]陶超,谭毅华,彭碧发等.一种基于概率潜在语义模型的高分辨率遥感影像分类方法.测绘学报,2011,40(2):156-162
[27]唐亮,谢维信,黄建军等.从航空影像中自动提取高层建筑物.计算机学报.2005,28(7):1199-1204
[28] Z. W. Kim, R. Nevatia. Automatic description of complex buildings from multipleimages. Computer Vision and Image Understanding.2004,96:60-95
[29] E. Simonton, H. Oriot, R. Garello. Rectangular Building Extraction fromStereoscopic Airborne Radar Images. IEEE Transactions on Geoscience and RemoteSensing.2005,43(10):2386-2395
[30]孙显,王宏琦,张正.基于对象的Boosting方法自动提取高分辨率遥感图像中建筑物目标.电子与信息学报.2009,31(1):177-181
[31] X. Jin, C. Davis. Automated Building Extraction from High-Resolution SatelliteImagery in Urban Areas Using Structural, Contextual, and Spectral Information.EURASIP Journal on Applied Signal Processing,2005,14(5):2196-2206
[32]陶超,谭毅华,蔡华杰等.面向对象的高分辨率遥感影像城区建筑物分级提取方法.测绘学报.2010,39(1):39-45
[33]周亚男,沈占锋,骆剑承等.阴影辅助下的面向对象城市建筑物提取.地理与地理信息科学.2010,26(3):37-40
[34] K. Karantzalos, N. Paragios. Recognition-Driven Two-Dimensional CompetingPriors Toward Automatic and Accurate Building Detection. IEEE Transaction onGeoscience and Remote Sensing,2009,47(1):133-144
[35] G. Gao. An Improved Scheme for Target Discrimination in High-Resolution SARImages. IEEE Transaction on Geoscience and Remote Sensing,2011,49(1):277-294
[36] L. Cheng, J. Y. Gong. Building Boundary Extraction Using Very High ResolutionImages and LiDAR. Acta Geodaetica et Cartographica Sinica,2008,37(3):391-393
[37] A. Sampath, J. Shan. Building Boundary Tracing and Regularization from AirborneLidar Point Clouds. Photogrammetric Engineering&Remote Sensing,2007,73(7):805-812
[38] C. Lin. Perception of3-D objects from an intensity image using simple geometricmodels[[PhD thesis]. America: the University of Southern California,1996
[39]强永刚,殷建平,陈涛等,利用阴影信息检测高分辨率遥感图像中的建筑物.计算机科学,2006,33(8):338-340
[40] K. Y. Ren, H. X. Sun, Q. X. Jia, et al. Building Recognition from Aerial ImagesCombining Segmentation and Shadow, IEEE conference on Intelligent Computingand Intelligent System,2009:578-582
[41] F. Rottensteiner, A. C. Briese. A New Method for Building Extraction in UrbanAreas from High-Resolution LIDAR data. ISPRS Commission III Symposium,Austria,2002
[42] S. Vinson, L. D. Cohen. Multiple Rectangle Model for Buildings Segmentation and3D Scene Reconstruction. International Conference on Pattern Recognition,Quebec-City,2002,3:623-626
[43] S. Noronha, R. Nevatia. Detection and Modeling of Building from Multiple AerialImages. IEEE Transaction on Pattern Analysis and Machine Intelligence,2001,23(5):501-518
[44] J. Hu, S. You, U. Neumann. Approaches to Large-scale Urban Modeling. IEEEComputer Graphics and Applications,2003,23(6):62-69
[45]王继阳,文贡坚,吕金建等.建筑物三维重建方法综述.遥感技术与应用,2009,24(6):832-840
[46]柳嫁航.利用遥感技术进行城市建筑物震害的自动识别与分类方法研究:[博士学位论文].北京:中国地震局地质研究所,2008
[47] A. Huertas, R. Nevatia. Detecting changes in aerial view form man-made structures.Image Vision and Computing,2000,18:583-596
[48]刘亚文,叶晓新.城区人工地物变化检测方法的研究.测绘通报,2001,7:9-11
[49]刘直芳,张继平,张剑清.基于DSM和影像持征的城市变化检测.遥感技术与应用,2002,17(5):240-244
[50] H. Wersing, E. Koerner. Learning Optimized Features for Hierarchical Models forInvariant Recognition. Neural Computation,2003,15(7):1559-1588
[51] T. J. Palmeri, I. Gauthier. Visual object understanding. Nature ReviewsNeuroscience,2004,5(1):291-303
[52]章毓晋.图像工程（上册）:图像处理和分析.北京:清华大学出版社,1993:218-240
[53] J. W. Rouse, R. H. Haas, J. A. Schell, et al. Monitoring the vernal advancement ofnatural vegetation, Final report, NASA/GCSFC, Greenbelt, MD,1974
[54]陈君颖,田庆久.高分辨率遥感植被分类研究.遥感学报,2007,11(2):221-2
[55] L. Bruzzone, L. Carlin. A multilevel context-based system for classification of veryhigh spatial resolution images. IEEE Transactions on Geoscience and RemoteSensing,2006,44(9):2587-2600
[56]余鹏,张震龙,侯至群.基于高斯马尔可夫随机场混合模型的纹理图像分割.测绘学报,2006,35(3):224-228)
[57] R. M. Haralick, Dinstein, K. Shanmugam. Textural feature for image classification.IEEE Transactions on Systems, Man, Cybernetics,1973, SMC-3(6):610-621
[58] G. Cross, A. Jain. Markov random fields texture models. IEEE Transactions onSystem Man Cybernet,1987,17:1087-1095
[59]罗晓燕,王成儒.基于自相关图像的纹理特征检索的研究.计算机应用与软件,2008,25(7):241-243
[60] F. Zhou, J. Feng, Q. Shi. lmage Segmentation Based on Local Fourier CoefficientsHistogram, International Conference on Multispectral Image Processing and PatternRecognition, Wuhan, China,2001
[61] B. S Manjunath, W. Y. Ma. Texture feature for browsing and retrieval of image data.IEEE Transaction on Patten Aanlysis and Machine Intelligence,1996,18(8):837-422
[62] E. Persoon, K. S. Fu. Shape diserimination using Foruier deseriptors. IEEETransaetionson Systems, Manand Cyberneties,1977,7(3):170-179
[63] D. G. Lowe. Distinctive image features from scale-invariant keypoints. InternationalJournal of Computer Vision,2004,60(2):91-110
[64] T. Kadir, M. Brady. Scale, saliency and image description. International Journal ofComputer Vision,2001,45(2):83-105
[65] K. Mikolajczyk, C. Schmid. Indexing based on scale invariant interest points. IEEEProceedings of the8th International Conference on Computer Vision, Vancouver,Canada,2001:525-531
[66] K. Mikolajczyk, C. Schmid. Scale and affine invariant interest point detectors.International Journal of Computer Vision,2004,1(60):63-86
[67] J. Matas, O. Chum., Urban M. Robust wide baseline stereo from maximally stableextremal regions. Image and Vision Computing,2004,22(10):761-767
[68] H. L. Deng, W. Zhang, E. Mortensen. Principal curvature-based region detector forobject recognition. IEEE Conference on Computer Vision and Pattern Recognition,2007,1-8
[69] R. Fergus, P. Perona, A. Zisserman. A sparse object category model for efficientlearning and exhaustive recognition. IEEE Conference on Computer Vision andPattern Recognition, San Diego,2005,1:380-387
[70] N. Dalal, B. Triggs. Histograms of oriented gradients for human detection. IEEEConference on Computer Vision and Pattern Recognition, San Diego, CA,2005:886-893
[71] Q. Zhu, S. Avidan, M. Ye. Fast human detection using a cascade of histograms oforiented gradients. IEEE Conference on Computer Vision and Pattern Recognition,New York, USA,2006:1491-1498
[72] Y. Ke, R. Sukthankar. PCA-SIFT: a more distinctive representation for local imagedescriptors. IEEE Conference Computer Vision and Pattern Recognition,2004:511-517
[73] S. Belongie, J. Malik, and J. Puzicha. Shape Matching and Object RecognitionUsing Shape Contexts. IEEE Transactions on Pattern Analysis and MachineIntelligence,2002,24(24):509-521
[74] W. Freeman, E. Adelson. The design and use of steerable filters. IEEE Transactionson Pattern Analysis and Machine Intelligence,1991,13(9):891-906
[75] K. Mikolajczyk, C. Schmid. A performance evaluation of local descriptors. IEEETransactions on Pattern Analysis and Machine Intelligence,2005,27,1615-1630
[76] F. Schaffalitzky, A. Zisserman. Multi-view Matching for Unordered Image Sets, or“How Do I Organize My Holiday Snaps?”. In Proceedings of the EuropeanConference on Computer Vision.2002,1:414-431
[77]陈星星,张荣.基于多尺度相位特征的图像检索方法.电子与信息学报,2009,31(5):1193-1196
[78]赵一凡,夏良正.基于轮廓波特征的纹理图像识别方法.东南大学学报(自然科学版),2008,38(11):128-131
[79] Baumberg. Reliable feature matching across widely separated views. IEEEConference on Computer Vision and Pattern Recognition.2000. Hilton Head Island,South Carolina, USA
[80] F. Schaffalitzky, A. Zisserman. Multi-view matching for unordered image sets. In:the7thEuropean Conference on Computer Vision.2002. Copenhagen, Denmark
[81] D. Gabor. Theory of communication. Journal of Institute for Electrical Engineering,1946,93:429-457
[82] T. Leung, J. Malik. Representing and recognizing the visual appearance of materialsusing three-dimensional textons. International Journal of Computer Vision,2001,43(1):29-44
[83] M. Varma, A. Zisserman. Classifying images of materials: achieving viewpoint andillumination independence. In: Proceedings of the European Conference onCompute Vision,2002
[84] M. Varma, A. Zisserman. A Statistical Approach to Texture Classi_cation fromSingle Images. International Journal of Computer Vision,2005,62(1):61-81
[85] Witkin A. P. Scale-space filtering. In: Proc. International Joint Conf. on ArtificialIntelligence. Karlsruhe. Germany: Springer Berlin,1983:1019-1022
[86] Lindeberg T. Scale-space theory: A basic tool for analysing structures at differentscales. Journal of Applied Statistics,1994,21(2):224-270
[87]赵玲玲,翁苏明.模式分析的核方法.北京:机械工业出版社,2006
[88]张学工.关于统计学习理论与支持向量机.自动化学报,2000,26(1):32-42
[89]边肇祺,张学工.模式识别（第2版）.北京:清华大学出版社,2000
[90]李晋博,特征提取的核方法与非线性多核学习的研究:[硕士学位论文].上海:华东师范大学武汉,2009
[91] F. Bach, G. Lanckriet, M. Jordan. Multiple kernel learning, conic duality, and theSMO algorithm. In Proceedings of the21st International Conference on MachineLearning, New York, NY, USA,2004:41-48
[92] G. Lanckriet, N. Cristianini, L. E. Ghaoui, et al. Learning the kernel matrix withsemi-definite programming. Journal of Machine Learning Research,2004,5:27-72
[93] M. Varma, D. Ray. Learning the discriminative power-invariance trade-off. IEEEConference on Computer Vision and Pattern Recognition,2007:1150-1157
[94] A. Rakotomamonjy, F. Bach, S. Canu, et al. More efficiency in multiple kernellearning. In IEEE Conference on Machine Learning,2007
[95] A. Bosch, A. Zisserman, X. Munoz. Representing shape with a spatial pyramidkernel. Proceedings of the6th ACM international conference on Image and videoretrieval, Amsterdam, The Netherlands, July09-11,2007:401-408
[96] X. Huagng, L. P. Zhang, P. Li. Classification and extraction of spatial features inurban areas using high resolution multispectral imagery. IEEE Geoscience andRemote Sensing letters,2007,4(2):260-264
[97]王田苗,王君臣,杨艳等.基于Harris角点的内窥镜图像变形全自动校正算法.自动化学报,2011,37(11):1360-1367
[98]吴秀芸,李艳,周华.基于角点检测的建筑物轮廓矢量化方法.遥感信息,2010,5:95-99
[99]C. G. Harris, M. Stephens. A Combined Corner and Edge Detector. In Proceedings ofFourth Alley Vision Conference,1988:182-192
[100] J. Shi, C. Tomasi. Good Features to Track, in: Proceedings of IEEE Conference onComputer Vision and Pattern Recognition,1994:593-600
[101] S. M. Smith, M. Brady. SUSAN—a new approach to low level image processing.International Journal of Computer Vision,1997,23(1):45-78
[102] J. Canny. A computational approach to edge detection. IEEE Transaction on patternrecognition and machine intelligence,1986,8(6):679-698
[103] D. H. Douglas, T. Pecuker. Algorithm for the reduction of the number of pointsrequired to represent a digitized lie or it caricature,1973,10(2):112-122
[104] M. Sezgin, B. Sankur. Survey over image thresholding techniques and quantitativeperformance evaluation. Journal of Electronic Imaging,2003,13(1):146-165
[105] T. Kanungo, D. M. Mount, N. S. Netanyahu. An Efficient k-Means ClusteringAlgorithm: Analysis and Implementation. IEEE Transaction on pattern recognitionand machine intelligence,2002,24(7):881-892
[106] A. Y. Ng, M. I. Jordan, Y. Weiss. On spectral clustering: analysis and an algorithm.Advances in Neural Information Processing Systems,2002,14:849-856
[107] U. Luxburg, M. Planck. A tutorial on spectral clustering. Journal of Statistics andComputing,2007,17(4):225-246
[108] T. Xiang, S. Gong. Spectral clustering with eigenvector selection. PatternRecognition,2008,41(3):1012-1029
[109] Y. Boykov, O. Versker, R. Zabih, Fast approximate energy minimization via graphcuts. IEEE transaction on Pattern Analysis and Machine Intelligence,2001,23(11):1222-1329
[110] T. J. Palmeri, I. Gauthier. Visual object understanding. Nature ReviewsNeuroscience,2004,5(1):291-303
[111] P. Blomgren, T. F. Chan. Color TV: Total Variation Methods for Restoration ofVector Valued Images. IEEE Transactions on Image Processing,1998,7(3):304-309
[112]李书晓,常红星.基于总变分和形态学的航空图像道路检测算法.计算机学报,2007,30(12):2173-2179
[113] F. Cheng, A. N. Venesanopoulous. Adaptive Morphological filter for imageprocessing. IEEE transactions on Image Processing,1992,1(4):533-539
[114] N. Vassis, A. Likas. The Kurtosis-Based Dynamic Approach to Gaussian MixtureModeling. IEEE Transaction on Systems, Man and Cybernetics,1999,29(4):393-399
[115] L. Xu. Bayesian Ying-Yang Machine, Clustering and Number of Clusters. PatternRecognition Letters,1997,18(11):1167-1178