摘要
土地覆被变化监测对区域生态系统保护、环境变化研究具有重要的作用,研究旨在提供一种基于归一化植被指数(NDVI)的假彩色合成法的土地覆被变化监测方法。该研究以黄河三角洲为研究区,以3期Landsat TM影像(成像时间分别为1987年5月7日,1998年5月5日,2009年5月3日)为数据源,在进行相对辐射校正的基础上,生成3期NDVI图像,然后分别以三期的NDVI图像作为红、绿和蓝波段生成假彩色合成图像。基于彩色合成原理,对黄河三角洲的1987-2009年间的土地覆被变化进行了分析。结果表明:(1)假彩色合成图像上的灰白色区域表示其土地覆被状态稳定,三个时期的NDVI值均较大,黑色区域的土地覆被状态也较稳定,但是三个时期的NDVI值均较小,而青色、绿色、红色则反映相应地区的NDVI处在不稳定状态;(2)不同的颜色反映了不同的土地覆被变化方式,较为直观地反映了土地覆被的变化特点,尤其是自然植被与农田之间的转换;(3)限于NDVI的瞬时性,该方法需要与基于遥感影像分类的方法相结合,才能更好地监测土地覆被变化。
Land cover change has great impacts on regional ecosystem conservation and environmental change. The aim of this research is to provide an approach to monitor land cover change based on false color synthesis of NDVI. The study area is the Yellow River Delta and the data sources are three scenes of Landsat TM images （TM images were acquired on May 7, 1987, May 5, 1998 and May 3, 2009 respectively）. Relative radiomctric normalization was done to TM images and then NDVI was calculated. At last, NDVI images were used as red, green and blue band to generate RGB false color synthesized image. Based on the false color synthesized image and principles of color synthesis, the land cover change from 1987 to 2009 in the Yellow River Delta was analyzed. The results arc as follows. The grey-white color on the false color synthesized image means the NDVI of this area is always very high, the black color means the NDVI of this area is always very low, and the blue, green and red colors mean the vegetation of this area is in the status of dynamic change. Different color means different switch between land cover types; this is very straightforward, and well describes the characteristics of land cover change in the study area, especially, the switch between nature vegetation and cropland. However, for the instantaneity of NDVI, this approach based on false color synthesis of NDVI should be combined with remote sensing image classification to monitor the land cover change.
出处
《生态科学》
CSCD
北大核心
2013年第3期271-275,共5页
Ecological Science
基金
国家科技支撑计划(2012BAC01B08)
