遥感图像中建筑物自动识别与标绘方法研究
|
摘要
自动目标识别(Automatic Target Recognition，ATR)技术，是遥感图像处理领域的核心技术之一。近年来，随着几颗商用遥感卫星的成功发射，如IKONOS、QuickBird等，高分辨率遥感图像资源越来越丰富，遥感图像中的自动目标识别技术再次成为研究的热点。遥感图像中自动目标识别的任务是将感兴趣的目标，如森林、湖泊、农作物、建筑物等，从图像背景中自动地识别出来。为了实现这一目标，需要综合应用图像处理，特征提取，图像分割等多项技术。针对实际的应用，在完成目标的自动识别后，还需要对其进行几何重建，构造出目标的规则几何外形，以实现地图的自动标绘。遥感图像中的自动目标识别与标绘技术现在已广泛应用于环境监测、军事侦察、农业估产、灾害控制等方面，也是未来地理信息系统(GIS)、数字地球、数字城市构想的关键技术。
本文研究了当前遥感图像中建筑物自动识别与标绘领域的相关技术和方法。在对已有方法进行归纳总结的基础上，提出了一种建筑物自动识别与标绘的应用框架。该框架包含三个部分：首先从城区遥感图像中找出建筑物的候选区域，然后从这些候选区域中确定真实的建筑物目标，最后对建筑物进行标绘。
在建筑物候选区域的寻找阶段，本文主要采用了一种基于灰度共生矩阵的方法对图像进行分割，从分割结果中得到建筑物的候选区域。
在建筑物目标的确认阶段，本文提出了一种基于主线段的建筑物判别方法。该方法通过区域生长、边缘提取、Hough变换求取建筑物候选区域的主线段，然后通过主线段长度来判别该候选区域是否为真实的建筑物目标。
在建筑物外形标绘阶段，本文提出了一种网格匹配与形态学处理相结合的方法，可获得建筑物的规则几何外形。
本文所进行的各项实验，均采用近年来QuickBird所采集的世界各大城市的高分辨率图像。实验结果表明，本文提出的应用框架对城区建筑物目标进行识别和标绘，有较好的准确性和鲁棒性。
Automatic Target Recognition (ATR) is a core technology in the field of remote sensing image processing. Recent commercial satellites, like IKONOS and QuickBird, provide a large quantity of high resolution remote sensing images. The ATR technology has become more useful and important. The goal of ATR in remote sensing images is to recognize the regions of interest, such as forests, lakes, crops, and buildings. In order to achieve this goal, multiple techniques like image processing, image feature extraction, image segmentation are used. After a target is recognized, a mapping procedure is needed to acquire the regular outline of this target. Automatic target recognition and mapping techniques are widely used in environment monitoring, military detective, agricultural estimation, and disaster control. They will be the key techniques of future GIS, digital globe, digital city system.
The researches of this paper focus on automatic building recognition and mapping techniques in urban remote sensing images. By concluding and summarizing former methods, this paper proposes an application framework for automatic building recognition and mapping. There are three steps in this framework: candidate building area searching, building target verification, and building mapping. In the candidate building area searching step, an image segmentation method based on grey level co-occurrence matrix (GLCM) is adopted to get candidate building areas in an urban image.
In the building target verification step, a method based on the dominant line segment is proposed. The dominant line segment (the longest line segment) of every candidate area is extracted by a procedure using region growing, edge detection, and Hough transform. Then, the real buildings are verified by judging the length of their dominant line segments.
本文研究了当前遥感图像中建筑物自动识别与标绘领域的相关技术和方法。在对已有方法进行归纳总结的基础上，提出了一种建筑物自动识别与标绘的应用框架。该框架包含三个部分：首先从城区遥感图像中找出建筑物的候选区域，然后从这些候选区域中确定真实的建筑物目标，最后对建筑物进行标绘。
在建筑物候选区域的寻找阶段，本文主要采用了一种基于灰度共生矩阵的方法对图像进行分割，从分割结果中得到建筑物的候选区域。
在建筑物目标的确认阶段，本文提出了一种基于主线段的建筑物判别方法。该方法通过区域生长、边缘提取、Hough变换求取建筑物候选区域的主线段，然后通过主线段长度来判别该候选区域是否为真实的建筑物目标。
在建筑物外形标绘阶段，本文提出了一种网格匹配与形态学处理相结合的方法，可获得建筑物的规则几何外形。
本文所进行的各项实验，均采用近年来QuickBird所采集的世界各大城市的高分辨率图像。实验结果表明，本文提出的应用框架对城区建筑物目标进行识别和标绘，有较好的准确性和鲁棒性。
Automatic Target Recognition (ATR) is a core technology in the field of remote sensing image processing. Recent commercial satellites, like IKONOS and QuickBird, provide a large quantity of high resolution remote sensing images. The ATR technology has become more useful and important. The goal of ATR in remote sensing images is to recognize the regions of interest, such as forests, lakes, crops, and buildings. In order to achieve this goal, multiple techniques like image processing, image feature extraction, image segmentation are used. After a target is recognized, a mapping procedure is needed to acquire the regular outline of this target. Automatic target recognition and mapping techniques are widely used in environment monitoring, military detective, agricultural estimation, and disaster control. They will be the key techniques of future GIS, digital globe, digital city system.
The researches of this paper focus on automatic building recognition and mapping techniques in urban remote sensing images. By concluding and summarizing former methods, this paper proposes an application framework for automatic building recognition and mapping. There are three steps in this framework: candidate building area searching, building target verification, and building mapping. In the candidate building area searching step, an image segmentation method based on grey level co-occurrence matrix (GLCM) is adopted to get candidate building areas in an urban image.
In the building target verification step, a method based on the dominant line segment is proposed. The dominant line segment (the longest line segment) of every candidate area is extracted by a procedure using region growing, edge detection, and Hough transform. Then, the real buildings are verified by judging the length of their dominant line segments.
引文
[1] Martin Herman and Takeo Kanade, Incremental Reconstruction of 3D Scenes from Multiple, Complex Images, Artificial Intelligence, Vol.30, pp.289-341, 1986
[2] A. Huertas and R. Nevatia, Detecting Buildings in Aerial Images, CVGIP-41, 1988, pp.131-152.
[3] Tao Wenbing, Tian Jinwen, Liu Jian. A New Approach To Extract Rectangle Building From Aerial Urban Images, ICSP'02 Proceedings, 2002
[4] Roth.G and Levine M. Geometric primitive extraction using a genetic algorithm. IEEE Trans. PAMI, 1994, 16(9):906-910.
[5] Lin C, Huertas A, Nevatia R. Detection of buildings using perceptual grouping and shadows. Proc. of IEEE computer science conference on Computer Vision and Pattern Recognition, Seattle, Washington, USA, 1994, 62-69
[6] Huertas A, Nevatia R. Detection building in aerial images. Computer Vision Graphics Image Process, 1988, 41 (2):131-152
[7] Kim T, Muller J P. Development of a graph-based approach for building detection, Image and Vision Computing, 1999, 17(1):3-14
[8] 张煜，张祖勋，张剑清。几何约束与影像分割相结合的快速半自动房屋提取。武汉测绘科技大学学报，2000，25(3)：238-242
[9] P.Fua and A.J.Hanson, Resegmentation Using Generic Shape:Locating General Cultural Objects, Technical Report, Artificial Intelligence Center, SRI International, May 1986.
[10] T.Matsuyama and V.Hwang, SIGMA: A Framework for Image Understanding: Integration of Bottom-up and Top-down Analyses, Proc. IJCAI, Los Angeles, CA, Aug 1985.
[11] A.Hanson and R.E.Riseman, VISIONS: A Computer System for Interpreting Scenes, New York: Academic, 1978.
[12] D.M.McKeown, W.A.Harvey and J.Mcdermott, Rule-Based Interpretation of Aerial Imagery, IEEE Trans. PAMI, Vol.7, pp.570-585, 1985.
[13] Yanfeng Wei, Zhongming Zhao, and Jianghong Song. Urban building extraction from high-resolution satellite panchromatic image using clustering and edge detection, IGARSS '04 Sept. 2004 vol.3, pp.2008- 2010, 2004.
[14] Jie Tian, Jinfei Wang, Peijun Shi. Urban Building Boundary Extraction From IKONOS Imagery, the Canadian Remote Sensing Society'04 Oct. 2004, poster session 3.
[15] Rakesh Mohan and Ramakant Nevatia, Using Perceptual Organization to Extract 3-D Structure, IEEE Trans. PAMI, Vol. 11, No. 11, pp. 1121-1139,1989.
[16] Pascal Fua and Andrew J. Hanson, Extracting Generic Shapes Using Models-Driven Optimization, DARPA Image Understanding Workshop. Vol.11,pp.994-1004, 1988.
[17] Radu Horaud and Thomas Skordas, Stereo Correspondence Through Feature Grouping and Maximal Cliques, IEEE Trans. PAMI, Vol. 11, No.11, pp.1168-1180, 1989.
[18] Yuan C. Hsieh, Frederic Perlant and David M. McKeown, Recovering 3D Information from Complex Aerial Imagery, Proceedings 10th int. Conference on Pattern Recognition, Vol. 1, pp. 136-146,1990.
[19] Zheng Wang and Toni Schenk, Urban Area Surface Analysis, ISPRS ⅩⅦ, Commission Ⅲ, International Archives of Photogrammetry and Rmote Sensing, Vol.29-3, pp.720-726, 1992.
[20] M.A.O'Neil and M.I.Denos, Practical Approach to the Stereo Matching of Urban Imagery, Image Vision and Computing, Vol. 10, No.2,pp.89-98,1992.
[21] Haralick R M. Statistical and structural approaches to texture. Proceedings of IEEE, 1979, 67(5): 786～804
[22] Baraldi A, Parminggian F. An Investigation on the Texture Characteristics Associated with Gray Level Co-occurrence Matrix Statistical Parameters [J]. IEEE Trans. on Geoscience and Remote Sensing, 1995, 32(2):293～303
[23] 边肇祺，张学工等。模式识别。北京：清华大学出版社，1999
[24] 章毓晋。图像分割。北京：科学出版社，2001
[25] 章毓晋。图象处理与分析[M]。北京：清华大学出版社，1999：39-41，231-232．
[26] 裴继红，谢维信。势函数聚类自适应多阈值图像分割。计算机学报，1999，22(7)
[37] 谢兴灿，裴继红，黄建军，唐亮。城市航空影像中建筑群的FCM聚类分割方法。计算机仿真，2004，4
[38] 裴继红，谢维信。势函数自适应加权模糊C-均值聚类方法。系统工程与电子技术，1999，21(12)
[26] 田艳琴，郭平，卢汉清。基于灰度共生矩阵的多波段遥感图像纹理特征的提取。计算机科学，2004，31(12)
[27] John Canny. A Computational Approach to Edge Detection, IEEE Trans. Pattern Analysis and Machine Intelligence, Vol. PAMI-8, No. 1, pp.679-697, November 1986
[28] Gonzalez R C，Woods R E．数字图像处理(第二版)。北京：电子工业出版社，2003．3
[29] Gonzalez R C，Woods R E，Eddins S L．数字图像处理(MATLAB版)。北京：电子工业出版社，2005．9
[30] 赵小杰。合成孔径雷达图像变化检测方法研究，硕士论文。中国科学院电子学研究所，2001，6
[31] 陈志鹏。基于纹理特征的差值变化检测方法研究，硕士论文。中国科学院电子学研究所，2002，6
[32] 蔡国雷。基于各向异性扩散的纹理图像分割方法研究，硕士论文。中国科学院电子学研究所，2005，5
[33] 赵波，孙即祥，张学庆，张翠平。使用Radon变换快速检测直线目标的多应用方法。无线电工程，2005，35(5)
[34] 肖温格 RA著。遥感中的图像处理和分类计数[M]。北京：科学出版社，1991
[35] 张妙兰，付新文。一种纹理图像分类方法的研究。中国图像图形学报，1998，4(8)
[36] 王勇，吕扬生。基于纹理特征的超声医学图像检索。天津大学学报，2005，38 (1)
[39] 周俊，晏非，孙曼。基于区域分割合并的建筑物半自动提取方法。海洋测
绘，2005，25(1)
[40] 匡锦瑜，王颖。多尺度边缘检测与图像分割的马尔可夫随机场模型。北京师范大学学报(自然科学版)，1996，32(3)
[2] A. Huertas and R. Nevatia, Detecting Buildings in Aerial Images, CVGIP-41, 1988, pp.131-152.
[3] Tao Wenbing, Tian Jinwen, Liu Jian. A New Approach To Extract Rectangle Building From Aerial Urban Images, ICSP'02 Proceedings, 2002
[4] Roth.G and Levine M. Geometric primitive extraction using a genetic algorithm. IEEE Trans. PAMI, 1994, 16(9):906-910.
[5] Lin C, Huertas A, Nevatia R. Detection of buildings using perceptual grouping and shadows. Proc. of IEEE computer science conference on Computer Vision and Pattern Recognition, Seattle, Washington, USA, 1994, 62-69
[6] Huertas A, Nevatia R. Detection building in aerial images. Computer Vision Graphics Image Process, 1988, 41 (2):131-152
[7] Kim T, Muller J P. Development of a graph-based approach for building detection, Image and Vision Computing, 1999, 17(1):3-14
[8] 张煜，张祖勋，张剑清。几何约束与影像分割相结合的快速半自动房屋提取。武汉测绘科技大学学报，2000，25(3)：238-242
[9] P.Fua and A.J.Hanson, Resegmentation Using Generic Shape:Locating General Cultural Objects, Technical Report, Artificial Intelligence Center, SRI International, May 1986.
[10] T.Matsuyama and V.Hwang, SIGMA: A Framework for Image Understanding: Integration of Bottom-up and Top-down Analyses, Proc. IJCAI, Los Angeles, CA, Aug 1985.
[11] A.Hanson and R.E.Riseman, VISIONS: A Computer System for Interpreting Scenes, New York: Academic, 1978.
[12] D.M.McKeown, W.A.Harvey and J.Mcdermott, Rule-Based Interpretation of Aerial Imagery, IEEE Trans. PAMI, Vol.7, pp.570-585, 1985.
[13] Yanfeng Wei, Zhongming Zhao, and Jianghong Song. Urban building extraction from high-resolution satellite panchromatic image using clustering and edge detection, IGARSS '04 Sept. 2004 vol.3, pp.2008- 2010, 2004.
[14] Jie Tian, Jinfei Wang, Peijun Shi. Urban Building Boundary Extraction From IKONOS Imagery, the Canadian Remote Sensing Society'04 Oct. 2004, poster session 3.
[15] Rakesh Mohan and Ramakant Nevatia, Using Perceptual Organization to Extract 3-D Structure, IEEE Trans. PAMI, Vol. 11, No. 11, pp. 1121-1139,1989.
[16] Pascal Fua and Andrew J. Hanson, Extracting Generic Shapes Using Models-Driven Optimization, DARPA Image Understanding Workshop. Vol.11,pp.994-1004, 1988.
[17] Radu Horaud and Thomas Skordas, Stereo Correspondence Through Feature Grouping and Maximal Cliques, IEEE Trans. PAMI, Vol. 11, No.11, pp.1168-1180, 1989.
[18] Yuan C. Hsieh, Frederic Perlant and David M. McKeown, Recovering 3D Information from Complex Aerial Imagery, Proceedings 10th int. Conference on Pattern Recognition, Vol. 1, pp. 136-146,1990.
[19] Zheng Wang and Toni Schenk, Urban Area Surface Analysis, ISPRS ⅩⅦ, Commission Ⅲ, International Archives of Photogrammetry and Rmote Sensing, Vol.29-3, pp.720-726, 1992.
[20] M.A.O'Neil and M.I.Denos, Practical Approach to the Stereo Matching of Urban Imagery, Image Vision and Computing, Vol. 10, No.2,pp.89-98,1992.
[21] Haralick R M. Statistical and structural approaches to texture. Proceedings of IEEE, 1979, 67(5): 786～804
[22] Baraldi A, Parminggian F. An Investigation on the Texture Characteristics Associated with Gray Level Co-occurrence Matrix Statistical Parameters [J]. IEEE Trans. on Geoscience and Remote Sensing, 1995, 32(2):293～303
[23] 边肇祺，张学工等。模式识别。北京：清华大学出版社，1999
[24] 章毓晋。图像分割。北京：科学出版社，2001
[25] 章毓晋。图象处理与分析[M]。北京：清华大学出版社，1999：39-41，231-232．
[26] 裴继红，谢维信。势函数聚类自适应多阈值图像分割。计算机学报，1999，22(7)
[37] 谢兴灿，裴继红，黄建军，唐亮。城市航空影像中建筑群的FCM聚类分割方法。计算机仿真，2004，4
[38] 裴继红，谢维信。势函数自适应加权模糊C-均值聚类方法。系统工程与电子技术，1999，21(12)
[26] 田艳琴，郭平，卢汉清。基于灰度共生矩阵的多波段遥感图像纹理特征的提取。计算机科学，2004，31(12)
[27] John Canny. A Computational Approach to Edge Detection, IEEE Trans. Pattern Analysis and Machine Intelligence, Vol. PAMI-8, No. 1, pp.679-697, November 1986
[28] Gonzalez R C，Woods R E．数字图像处理(第二版)。北京：电子工业出版社，2003．3
[29] Gonzalez R C，Woods R E，Eddins S L．数字图像处理(MATLAB版)。北京：电子工业出版社，2005．9
[30] 赵小杰。合成孔径雷达图像变化检测方法研究，硕士论文。中国科学院电子学研究所，2001，6
[31] 陈志鹏。基于纹理特征的差值变化检测方法研究，硕士论文。中国科学院电子学研究所，2002，6
[32] 蔡国雷。基于各向异性扩散的纹理图像分割方法研究，硕士论文。中国科学院电子学研究所，2005，5
[33] 赵波，孙即祥，张学庆，张翠平。使用Radon变换快速检测直线目标的多应用方法。无线电工程，2005，35(5)
[34] 肖温格 RA著。遥感中的图像处理和分类计数[M]。北京：科学出版社，1991
[35] 张妙兰，付新文。一种纹理图像分类方法的研究。中国图像图形学报，1998，4(8)
[36] 王勇，吕扬生。基于纹理特征的超声医学图像检索。天津大学学报，2005，38 (1)
[39] 周俊，晏非，孙曼。基于区域分割合并的建筑物半自动提取方法。海洋测
绘，2005，25(1)
[40] 匡锦瑜，王颖。多尺度边缘检测与图像分割的马尔可夫随机场模型。北京师范大学学报(自然科学版)，1996，32(3)