摘要
近年来,利用遥感技术对湖泊、河流、海岸等水体信息的监测和评估成为时代发展趋势。本研究选取连云港市1988年、1998年、2008年和2018年四个时期同一月份的TM、ETM和OLI影像,基于RS和GIS平台,研究连云港市近30年水体信息变化。对多期遥感影像进行预处理和增强处理后,区别于以往普遍使用的单波段阈值分析法、比值法、差值法、水体指数法等水体信息提取方法,本研究尝试使用“马氏距离”法分类提取研究区水体信息,并使用决策树法分类统计连云港市近30年水体变化。研究发现:近三十年来,研究区水域面积整体呈下降趋势。其中,以最近十年水域面积减少更为突出,这一特征与该地区近年来社会经济建设和自然因素变化密切相关。
In recent years, remote sensing technology in monitoring and assessing the information of lakes, rivers, coasts and other water changes has become a trend of this times. This study tried to use ENVI and Arcgis to analyse water information changes in Lianyungang in recent 30 years. Distinguished from water information extraction methods such as single-band threshold analysis, band ratio method, difference value method and water index method, this study attempts to extract water information from the study area by using mahalanobis distance method to classify them and using decision trees method to make a statistics of water changes in Lianyungang in recent 30 years. It is found that in the past 30 years, the water information change of the study area has shown a downward trend. Among them, the decrease of water area in the last decade is more prominent, which is closely related to the social and economic construction and changes of natural factors in this area in recent years.
作者
蒋科迪
殷勇
范开桂
张凌
李月
JIANG Kedi;YIN Yong;FAN Kaigui;ZHANG Ling;LI Yue(School of Geography and Ocean Science,Nanjing University,Nanjing Jiangsu 210023,China;College of Oceanography,Hohai University,Nanjing Jiangsu 210098,China)
出处
《北京测绘》
2019年第6期620-624,共5页
Beijing Surveying and Mapping
基金
国家自然科学基金(40776023)
关键词
遥感技术
监督分类
水体变化
决策树
图像拟合
remote sensing technology
supervised classification
water information change
decision tree
image fitting