计算机与现代化

• 图像处理 • 上一篇    下一篇

基于Landsat8 OLI图像的土地利用分类方法

  

  1. 西藏民族大学信息工程学院,陕西咸阳712082
  • 收稿日期:2015-07-29 出版日期:2015-09-21 发布日期:2015-09-24
  • 作者简介:刘炜(1976-),男,陕西咸阳人,西藏民族大学信息工程学院讲师,博士,研究方向:土地遥感与GIS应用技术。
  • 基金资助:
     国家自然科学基金资助项目(41361044); 西藏民族大学青年学人项目(13myQP09)

Land-use Classification Method Based on Landsat8 OLI Images

  1. College of Information Engineering, Xizang Minzu University, Xianyang 712082, China
  • Received:2015-07-29 Online:2015-09-21 Published:2015-09-24

摘要:

提出一种更合适的土地利用分类方法,比较该方法与最大似然法识别定边县OLI图像上11种主要土地利用类型的精度。对OLI图像小波滤波后进行LBV变换,以LBV变换的衍生波段作为有效特
征进行SVM监督分类,分类后执行基于数学形态学的开、闭运算操作,计算所得分类图的总体分类精度和Kappa系数,并与最大似然法分类结果比较。分类结果表明,与最大似然法相比,本文提出的OLI图
像分类方法能够更准确地识别定边县11种土地利用类型,有效降低“椒盐效应”。所得分类图的总体分类精度和Kappa系数分别为83.62%和0.785,比最大似然法分类结果分别提高12.82%和14.26%。

关键词:  , 图像分类, 定边县, 支持向量机, 小波滤波

Abstract:

This research aims to seek out the most suitable land use classification method for Landsat8 OLI images, by comparing 〖JP2〗supervised classification based on
maximum likelihood and the method proposed in this paper. There are 11 kinds of land use type in OLI images of Dingbian county. Firstly, OLI images were fused with panchromatic
image and then processed by 3 levels wavelet filtering after routine image preprocessing. Secondly, LBV transform were applied to OLI images. Thirdly, training samples set for
each kind land use type were collected, and then supervised classification based on SVM, opening-closing operation in mathematical 〖JP2〗morphology were carried out to get
precise information of each kind land use type. Fourthly, assessing the classification results of method proposed in this paper and maximum likelihood, by overall classification
accuracy and Kappa coefficient as evaluation indexes. Results show that: the overall accuracy and Kappa coefficient of classification image using the method proposed in this
paper were 83.62% and 0.785, with growth of 12.82% and 14.26% compared with classification image using maximum likelihood. Meanwhile, removal of salt and pepper noise in
classification image was more effective using method proposed in this paper.

Key words: image classification, Dingbian county, support vector machine, wavelet filtering