基于多因子加权的遥感制图最佳波段组合分析  

Multi-factor Weighting Optimal Band Combination Analysis for Remote Sensing Mapping

作　　者：毛玉君 李志伟 沈焕锋[1] 苗静 MAO Yujun;LI Zhiwei;SHEN Huanfeng;MIAO Jing(School of Resource and Environmental Sciences,Wuhan University,Wuhan 430079,China;Wuhan Geomatics Institute,Wuhan 430022,China)

机构地区：[1]武汉大学资源与环境科学学院,湖北武汉430079 [2]武汉市测绘研究院,湖北武汉430022

出　　处：《测绘地理信息》2022年第6期70-75,共6页Journal of Geomatics

基　　金：中央高校基本科研业务费专项资金项目(2042021KF0078)。

摘　　要：遥感影像最佳波段组合的研究对遥感信息提取与专题制图具有重要意义。以Landsat-8影像为例,提出了多因子加权的遥感制图最佳波段组合评价方法,分别从全局加权与场景加权的角度,对长江经济带全域及城市、水体、植被典型场景进行最佳波段组合分析,并通过评估不同波段组合影像的土地覆盖分类精度进行结果验证。结果表明B6-B5-B4进行RGB彩色合成是适宜长江经济带区域Landsat-8影像遥感制图的最佳波段组合,在城市、水体、植被分场景下保证了较高的土地覆盖分类精度,同NASA(national aeronautics and space administration)的Landsat ARD(analysis ready data)产品展示采用的波段组合方案一致,验证了实验结论的可靠性,为大区域遥感制图中影像波段组合的优选提供决策依据。The study on the optimal band combination plays a critical role in remote sensing information extraction and thematic mapping.Taking Landsat-8 images as the study data,an evaluation method for the optimal band combination of remote sensing mapping was proposed based on multi-factor weighting.Specifically,it started with the global weighting and the scene weighting,aiming to analyze the optimal band combination for the entire Yangtze River economic zone and typical scenes of cities,water bodies,and vegetation.Then the results were verified by evaluating the accuracy of land cover classification for images with different band combination.The results showed that RGB composited image with 6-5-4 bands is the optimal band combination for Landsat-8 imagery mapping in the Yangtze River economic zone.Besides,it guaranteed high accuracy of land cover classification in different scenes and was consistent with the one adopted in the Landsat ARD(analysis ready data) products by NASA,which thus verified the reliability of the conclusions.This study will provide a decision basis for the selection of optimal band combinations in large-area remote sensing mapping.

关 键 词：遥感制图 Landsat-8影像 波段组合 多因子加权 长江经济带 

分 类 号：P237[天文地球—摄影测量与遥感]