Please wait a minute...
img

官方微信

遥感技术与应用  2014, Vol. 29 Issue (3): 489-497    DOI: 10.11873/j.issn.1004-0323.2014.3.0489
图像与数据处理     
综合优度法和不一致性法的最优分割参数选择方法
郭钇宏,王博,刘勇,杨亦宁
(兰州大学资源环境学院,甘肃 兰州730000)
Integrated Methods with Goodness Measures and Discrepancy Measures for Selecting Optimal Segmentation Parameter
Guo Yihong,Wang Bo,Liu Yong,Yang Yining
(College of Earth and Environmental Sciences,Lanzhou University,Lanzhou 730000,China)
 全文: PDF(14297 KB)  
摘要:

分割参数的选择直接决定着影像对象的大小、形状。因此如何选择最优的分割参数显得尤为重要。用局部标准差和局部Moran指数构建了新的优度度量函数,然后与通过不一致性法选择的分割参数进行比较,进而选取综合的最优分割参数。通过对高空间分辨率IKONOS影像上农田、草地、池塘和建筑物等4种不同类型地物进行实验,表明不同地物类型具有不同的最优分割参数区间,且分别使用本研究所提出的优度法和不一致性法所得度量参数具有基本一致的最优参数分布区间。进而,通过对最优分割参数和其他分割参数下的分割结果进行影像分类,分类结果的精度评价表明综合两种度量方法得到的最优分割参数可获得最佳的分类结果。

关键词: 影像分割最优分割参数优度法不一致性法分类精度评价    
Abstract:

Image segmentation is the first step for object\|based image analysis.The size and quality of segmented objects directly affect the accuracy of the subsequent classification.Once the algorithm for image segmentation is determined,the choice of image segmentation parameter will directly determine the size and shape of image objects.How to choose optimal segmentation parameter is becoming the key important.The paper proposes a new goodness measure based on an inner\|segment homogeneity measurement with local standard deviation and an inter\|segment heterogeneity measure with local Moran index.The optimal segmentation parameter is chosen by discrepancy measures,including Potential Segmentation Error (PSE),Number\|of\|Segments Ratio (NSR),and Euclidean Distance (ED),were compared with this method to obtain an integrated optimal segmentation parameter.Four different categories of land cover,including cropland,grassland,ponds and buildings in a high\|resolution IKONOS image are experimented.The experiment demonstrates that different categories of land cover have different optimal interval for segmentation parameter,and the intervals derived from goodness measures and discrepancy measures are consistent on the whole.Then the optimal image segments are classified using nearest distance to means.The accuracy assessment of the results using optimal segmentation parameter are the best by comparing with classification results when using commonly selected three parameters.

Key words: Image segmentation    Optimal segmentation parameter    Goodness measure    Discrepancy measure    Classification accuracy assessment
收稿日期: 2013-10-10 出版日期: 2014-06-23
ZTFLH:  TP 751  
基金资助:

国家自然科学基金项目“遥感影像多尺度分割质量评价与参数优选方法研究”(41271360)。

通讯作者: 刘勇(1964-),男,甘肃清水人,教授,博士生导师,主要从事遥感与地理信息系统研究。Email:liuy@lzu.edu.cn。    
作者简介: 郭钇宏(1988-),女,河南安阳人,硕士研究生,主要从事基于对象影像分析方法的研究。Email:guoyh114@163.com。
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  
郭钇宏
王博
刘勇
杨亦宁

引用本文:

郭钇宏,王博,刘勇,杨亦宁 . 综合优度法和不一致性法的最优分割参数选择方法[J]. 遥感技术与应用, 2014, 29(3): 489-497.

Guo Yihong,Wang Bo,Liu Yong,Yang Yining. Integrated Methods with Goodness Measures and Discrepancy Measures for Selecting Optimal Segmentation Parameter. Remote Sensing Technology and Application, 2014, 29(3): 489-497.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2014.3.0489        http://www.rsta.ac.cn/CN/Y2014/V29/I3/489

[1]Li Shihua,Wang Jinliang,Bi Yan,et al.A Review of Methods for Classification of Remote Sensing Images[J].Remote Sensing for Land & Resources,2005,5(2):1-6.[李石华,王金亮,毕艳,等.遥感图像分类方法研究综述[J].国土资源遥感,2005,5(2):1-6.]

[2]Zhou Chenghu,Luo Jiancheng.Geo Computation of High Resolution Satellite Image[M].Beijing:Science Press,2009.[周成虎,骆剑承.高分辨率影像地学计算[M].北京:科学出版社,2009.]

[3]Zhang Chunxiao,Hou Wei,Liu Xiang,et al.Remote Sensing Image Classification based on Object-oriented and Image Cognition——A Case Study in Xiang’e,Dujiangyan[J].Bulletin of Survey and Mapping,2010(4):11-14.[张春晓,侯伟,刘翔,等.基于面向对象和影像认知的遥感影像分类方法——以都江堰向峨乡区域为例[J].测绘通报,2010(04):11-14.]

[4]Navalur K.Multispectral Image Analysis Using the Object-oriented Paradigm[M].UK:Taylor & Francis Group,2007.

[5]He Jun,Ge Hong,Wang Yufeng.Survey on the Methods of Images Segmentation Research[J].Computer Engineering & Science,2009,31(12):58-61.[何俊,葛红,王玉峰.图像分割算法研究综述[J].计算机工程与科学,2009,31(12):58-61.]

[6]Huang Huiping.Scale Issues in Object-oriented Image Analysis[D].Beijing:Institute of Remote Sensing Applications,Chinese Academy of Sciences,2003.[黄慧萍.面向对象影像分析中的尺度问题研究[D].北京:中国科学院遥感应用研究所,2003.]

[7]Zhang Y J.A Survey on Evaluation Methods for Image Segmentation [J].Pattern Recognition,1996,29(8):1335-1346.

[8]Hou Gexian,Bi Duyan.Researches on Evaluation Methods for Image Segmentation[J].Journal of Image and Graphics,2005,5(1):39-43.[侯格贤,毕笃彦.图像分割质量评价方法研究[J].中国图象图形学报,2000,5(1):39-43.]

[9]Zhang Jun,Wang Yunjia,Li Yan,et al.An Object-oriented Optimal Scale Choice Method for High Spatial Resolution Remote Sensing Image[J].Science & Technology Review,2009,27(21):91-94.[张俊,汪云甲,李妍,等.一种面向对象的高分辨率影像最优分割尺度选择算法[J].科技导报,2009,27(21):91-94.]

[10]He Min,Zhang Wenjun,Wang Weihong.Optimal Segmentiation Scale Model based on Object-oriented Analysis Method[J].Journal of Geodesy and Geodynamics,2009,29(1):106-109.[何敏,张文君,王卫红.面向对象的最优分割尺度计算模型[J].大地测量与地球动力学,2009,29(1):106-109.]

[11]Hu Wenliang,Zhao Ping,Dong Zhangyu.An Improved Calculation Model of Object-oriented for the Optimal Segmentation Scale of Remote Sensing Image[J].Geography and Geo-Information Science,2010,26(6):15-18.[胡文亮,赵萍,董张玉.一种改进的遥感影像面向对象最优分割尺度计算模型[J].地理与地理信息科学,2010,26(6):15-18.]

[12]Lin Xiancheng,Li Yongshu.A Study of the Ssegmentation Scale of High-resolution Remote Sensing Image in Chengdu Plain[J].Remote Sensing for Land & Resources,2010,22(2):7-11.[林先成,李永树.成都平原高分辨率遥感影像分割尺度研究[J].国土资源遥感,2010,22(2):7-11.]

[13]Anselin L.Local Indicators of Spatial Association—LISA[J].Geographical Analysis,1995,27(2):93-115.

[14]Liu Y,Bian L,Meng Y H,et al.Discrepancy Measures for Selecting Optimal Combination of Parameter Values in Object-based Image Analysis[J].ISPRS Journal of Photogrammetry and Remote Sensing,2012,68:144-156.

[15]Clinton N,Holt A,Scarborough J,et al.Accuracy Assessment Measures for Object-based Image Segmentation Goodness[J].Photogrammetric Engineering and Remote Sensing,2010,76(3):289-299.

[16]Monteiro F C,Campilho A C.Image Analysis and Recognition[M].Guangzhou:Springer Berlin Heidelberg,2006:248-259.

[17]Yu Huan,Zhang Shuqing,Kong Bo,et al.Optimal Segmentation Scale Selection for Object-oriented Remote Sensing Image Classification[J].Journal of Image and Graphics,2010,15(2):352-360.[于欢,张树清,孔博,等.面向对象遥感影像分类的最优分割尺度选择研究[J].中国图象图形学报,2010,15(2):352-360.]

[18]Jin X Y.Segmentation-based Image Processing System,Patent Application 11/984,222[P].2007-11-14.

[19]Zhao Yingshi.Theory and Methods of Remote Sensing Application Analysis[M].Beijing:Science Press,2003.[赵英时.遥感应用分析原理与方法[M].北京:科学出版社,2003.]

[1] 宋德娟,魏青迪,张承明,李峰,韩颖娟,范克琦. 利用RefineNet模型提取冬小麦种植信息的方法[J]. 遥感技术与应用, 2019, 34(4): 720-726.
[2] 王苏芸,孙中昶,郭华东,申维. 基于面向对象的东营市城乡建设用地信息提取[J]. 遥感技术与应用, 2017, 32(4): 780-786.
[3] 潘一凡,张显峰,于泓峰,饶俊峰. 联合快舟一号影像纹理信息的城市土地覆盖分类[J]. 遥感技术与应用, 2016, 31(1): 194-202.
[4] 余其鹏,张晓祥,梅丹丹,徐盼. 结合地籍数据的高密度城区面向对象遥感分类 [J]. 遥感技术与应用, 2014, 29(2): 344-351.
[5] 别强,何磊,赵传燕. 基于影像融合和面向对象技术的植被信息提取研究[J]. 遥感技术与应用, 2014, 29(1): 164-171.
[6] 王志波,高志海,王琫瑜,徐先英,白黎娜,王红岩,吴俊君,孙 斌. 基于面向对象方法的沙化土地遥感信息提取技术研究[J]. 遥感技术与应用, 2012, 27(5): 770-777.
[7] 陈燕丽,莫伟华,莫建飞,王君华,钟仕全. 基于面向对象分类的南方水稻种植面积提取方法[J]. 遥感技术与应用, 2011, 26(2): 163-168.
[8] 胡进刚, 张晓东, 沈 欣, 张 婵. 一种面向对象的高分辨率影像道路提取方法[J]. 遥感技术与应用, 2006, 21(3): 184-188.