摘 要:遥感图像是获取土地覆被信息的重要数据源,基于遥感数据的及时性、高效性、精准性等优点进行土地覆被分类是当前的研究热点。文章采用决策树分类技术,将京津冀地区作为研究区,结合京津冀地区土地覆被的分类特征,建立一套符合京津冀地区的图像分类体系。以 Landsat 8 数据为原始数据,基于不同类型的光谱特征构建决策树模型,获取研究区地表覆被分类结果,所得结果直观地反映了京津冀地区土地覆被的分布特征。
关键词:决策树;分类;遥感影像
DOI:10.19850/j.cnki.2096-4706.2023.05.010
中图分类号:TP751 文献标识码:A 文章编号:2096-4706(2023)05-0041-04
Landsat Image Classification in Beijing-Tianjin-Hebei Region Based on Decision Tree
WEI Xiangyi 1, KONG Lingran1, XIAO Lei 2
(1.China Water Resources Beifang Investigation, Design and Research Co. Ltd., Tianjin 300222, China; 2.Hydrology and Water Resources Management Center of Tianjin City, Tianjin 300061, China)
Abstract: Remote sensing image is an important data source for obtaining land cover information. Land cover classification based on the advantages of timeliness, efficiency and accuracy of remote sensing data is a current research hotspot. In this paper, taking the BeijingTianjin-Hebei region as the study area, and combining the classification characteristics of the land cover in the Beijing-Tianjin-Hebei region, the decision tree classification technology is used to establish a set of image classification system that conforms to the Beijing-TianjinHebei region. Taking Landsat8 data as the original data, a decision tree model is built according to different types of spectral characteristics to obtain the classification results of land cover in the study area. The obtained results directly reflect the distribution characteristics of land cover in Beijing-Tianjin-Hebei region.
Keywords: decision tree; classification; remote sensing image
参考文献:
[1] 张蓬涛,周雁,刘晓庄,等 . 人工神经网络在农业自然资源研究中的应用 [J]. 安徽农业科学,2007(27):8711-8713.
[2] 汤国安 . 遥感数字图像处理 [M]. 北京:科学出版社,2004.
[3] 邸凯昌,李德仁,李德毅 . 基于空间数据发掘的遥感图象分类方法研究 [J]. 武汉测绘科技大学学报,2000,25(1):42-48.
[4] 李爽,张二勋 . 基于决策树的遥感影像分类方法研究 [J].地域研究与开发,2003,22(1):17-21.
[5] 魏强 . 基于 MODIS 和 TM 数据的京津冀地区土地覆被分类方法研究 [D]. 石家庄:河北师范大学,2010.
[6] 贾涛,韩萌,王少峰,等 . 基于 McDiarmid 不等式的决策树分类算法 [J]. 山西大学学报:自然科学版,2019,74(4):718-728.
作者简介:魏向祎(1990—),女,汉族,河北石家庄人,工程师,硕士研究生,研究方向:摄影测量与遥感。