期刊文献+

一种高分辨率遥感影像的最优分割尺度自动选取方法 被引量:23

Automatic Selection of Optimal Segmentation Scale of High-resolution Remote Sensing Images
原文传递
导出
摘要 随着卫星遥感影像空间分辨率的不断提高,面向对象的地物信息提取技术发展迅速。图像分割作为面向对象分类的关键步骤之一,其分割尺度的参数设置目前仍以分类者的多次尝试和主观判断为依据,效率较低且分割结果因人而异。本文以WorldView2影像数据为例,结合当前现有的理论和方法,实现了一种计算机可自动进行主成分变换的高分辨率遥感图像全局最优分割尺度选取算法。改进后的算法以主成分变换所得的主成分影像作为图像分割的编辑层,主成分的特征值百分比作为计算异质性参数和分割质量评价值的权重,自动计算当分割尺度从20增至200时分割图像的分割质量评价值(GS),解决了人为确定图像分割编辑层的片面性问题,并利用三次样条插值选取出GS最高值所对应的尺度即为最优分割尺度。结果表明,该最优分割尺度选取方法可有效避免人为确定分割尺度的主观性、片面性和低效性,提升了高分辨率影像分割质量。 With the increasing of spatial resolution of imaging sensors, object-oriented feature information ex- traction technology is developing rapidly. The advantages of object-based classification over the traditional pix- el-based approach are well documented. Image segmentation is a key step to realize the object-oriented classifica- tion. The choice of scale parameter is very important and has a great influence on the segmentation effectiveness, but the choice of scale parameter is still decided by the repeated attempts and subjective judgments of operator, which are lacking in stability and reliability. Thus, an objective and unsupervised method is proposed for select- ing optimal parameter for image segmentation to ensure best quality results. In this paper, WorldView 2 as data source, a new method based on principal component transform is introduced to choose an optimal parameter for image segmentation. We choose principal component images as the editor of image segmentation and eigenval- ues as the weights of heterogeneity f and segmentation global score. Segmentation images, ranging from 20 to 200 scale, step by 10, are created in Definiens Professional 8.7. Then, the global intra-segment and inter-segment heterogeneity indexes are taken into account to identify the optimal segmentation scale (i. e. the highest GS val- ue) by using the cubic spline interpolation function method. After comparison with the results of image segmen- tation based on traditional three bands, image segmentation effect obtained by principal component transform has obvious advantages. As a result, the method in this paper can effectively avoid the subjectivity of the artifi- cial segmentation, one-sidedness and inefficiency, improve the quality of high-resolution image segmentation. The method also makes a good preprocessing work for later image classification and information extraction.
出处 《地球信息科学学报》 CSCD 北大核心 2013年第6期902-910,共9页 Journal of Geo-information Science
基金 上海市科委世博专项(13231203804) 国家自然科学基金项目(41201358)
关键词 面向对象 图像分割 最优分割尺度选择 主成分变换 Moran’s I指数 WorldView2 object-oriented image segmentation optimal segmentation scale selection principal componenttransform Moran' s Index WorldView2
  • 相关文献

参考文献5

二级参考文献27

  • 1冯德俊,李永树,兰燕.基于主成分变换的动态监测变化信息自动发现[J].计算机工程与应用,2004,40(36):199-202. 被引量:23
  • 2明冬萍,骆剑承,沈占锋,汪闽,盛昊.高分辨率遥感影像信息提取与目标识别技术研究[J].测绘科学,2005,30(3):18-20. 被引量:108
  • 3丁晓英.eCognition在土地利用项目中的应用[J].测绘与空间地理信息,2005,28(6):116-117. 被引量:23
  • 4Ioannis Z Gitas, George H Mitri and Gemma Ventura. Object -based image classification for burned area mapping of Creus Cape, Spain, using NOAA - AVHRR imagery [ J ]. Remote Sensing of Environment, 2004,92 (3) :409 - 413.
  • 5Renaud Mathieu, Claire Freeman and Jagannath Aryal. Map- ping private gardens in urban areas using object -oriented techniques and very high - resolution satellite imagery [ J ]. Landscape and Urban Planning, 2007,81 (3) :179 -192.
  • 6黄惠萍.面向对象影像分析中的尺度问题研究[D].中国科学院遥感应用研究所,2003.
  • 7Wulder M and Boots B. Local spatial autocorrelation characteristics of remotely sensed magery assessed with the Getis statistic [ J ]. International Journal of Remote Sensing, 1998, 19:223-2231.
  • 8Ashton Shortridge. Practical limits of Moran′s autocorrelation index for raster class maps[ J ]. Computers, Environment and Urban Systems, 2007,31 (3) :362 - 371.
  • 9[1]张孝灿,黄智才,赵元洪.遥感数字图像处理[M].杭州:浙江大学出版社,2003.
  • 10谭衢霖,刘正军,沈伟.一种面向对象的遥感影像多尺度分割方法[J].北京交通大学学报,2007,31(4):111-114. 被引量:72

共引文献99

同被引文献223

引证文献23

二级引证文献214

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部