| 结合多尺度纹理的高分辨率遥感影像决策树分类 |
| |
| 引用本文： | 陈亮,张友静,陈波.结合多尺度纹理的高分辨率遥感影像决策树分类[J].地理与地理信息科学,2007,23(4):18-21. |
| |
| 作者姓名： | 陈亮 张友静 陈波 |
| |
| 作者单位： | 1. 河海大学水文水资源及水利工程国家重点实验室,江苏,南京,210098
2. 河海大学水文水资源学院,江苏,南京,210098 |
| |
| 摘 要： | 地物具有多尺度特点,遥感影像包含的地物纹理信息很难用单一尺度来描述。通过选择最佳纹理尺度组合,利用光谱数据结合多尺度纹理对高分辨率影像进行决策树分类。研究结果表明:结合多尺度纹理的高分辨遥感影像决策树分类,能够更好地描述地物并有效解决光谱数据分类中存在的地物破碎问题,其分类精度为81.7%,kap-pa系数为0.78;与光谱数据分类和结合单尺度纹理数据分类结果比较,分类精度分别提高了11.2%和6%,该方法有助于提高高分辨率影像的分类精度。
|
| 关 键 词： | 多尺度纹理 高分辨率 决策树分类 |
| 文章编号： | 1672-0504（2007）04-0018-04 |
| 修稿时间： | 2007-01-242007-04-09 |
High Spatial Resolution Remote Sensing Image Classification Based on Decision Tree Classification Combined with Multiscale Texture |
| |
| CHEN Liang,ZHANG You-jing,CHEN Bo.High Spatial Resolution Remote Sensing Image Classification Based on Decision Tree Classification Combined with Multiscale Texture[J].Geography and Geo-Information Science,2007,23(4):18-21. |
| |
| Authors: | CHEN Liang ZHANG You-jing CHEN Bo |
| |
| Institution: | State Key Laboratory of Hydrology-Water Resource and Hydraulic Engineering, Hohai University ,Nanjing 210098 College of Hydrology-Water Resource, Hohai University, Nanjing 210098, China |
| |
| Abstract: | Scales of urban features are different.No unique scale is possible for characterizing the texture of objects in a remote sensing image.In this paper,a set of optimized textures was chosen,which was used to classify the high spatial resolution remote sensing image based on decision tree classification(DTC) combined with multiscale texture data.And comparing to the results based on single spectrum data and integration of spectrum data and single scale texture data,it shows that decision tree classification with multiscale texture data can character the textures precisely and solve the image classification fragmentation better.The accuracy of the method is 81.7% and the kappa is 0.78.Compared with those two methods,the accuracy has improved 11.2% and 6% respectively.It is proven that the method can be aid for improving the accuracy of high resolution image classification. |
| |
| Keywords: | multiscale texture high resolution decision tree classification |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
