摘要
河流作为一种重要的地表覆盖类型,是地理国情监测的重要内容和综合统计分析的基础信息。根据普查数据时点统一的要求,不同时相的影像中提取得到的河流面积需要修正成统一时点下的结果。针对这一问题,本文提出了高、中分辨率遥感影像逐步抽样计算面积调整系数的方法,建立不同月份河流覆盖面积之间的关系。以四川省和黑龙江省为例,利用Landsat ETM和QuickBird两种分辨率的影像进行抽样,并逐步计算不同类别、不同时相影像之间的调整系数,最终得到目标时相下实验区河流的解译面积。结果表明:该方法可操作性强,通过调整系数得到的解译结果与实际普查数据具有较好的一致性,可为地理国情监测提供方法参考。
As an important land cover type,river is the basis of information analysis and important content of geographic conditions monitoring and comprehensive statistics.According to the unified timing requirements of national geographic conditions survey,river area extracted from different images of different time and types need to be adjusted.To solve this problem,this paper provides the solution of monitoring a wide range of river area with high-precision,which is sampling the high-resolution and medium-resolution remote sensing images step by step,and establishing the relationships between the river area coverage in different months.Take Sichuan and Heilongjiang province for example,QuickBird images and Landsat ETM images were used to sample,calculate the adjustment coefficient stepwise between different types of images at different time. Finally,it obtained the river area in the target time.The result indicates that the method is in good operation,and the extraction results obtained by adjustment coefficient have good consistency with the actual census data,which can provide helpful method guidelines for the national geographic conditions monitoring.
出处
《遥感信息》
CSCD
2014年第4期26-30,36,共6页
Remote Sensing Information
