垦利县耕地利用动态监测研究

对土地利用变化进行监测是全球变化的一项基本内容,基于遥感技术的两个最主要的应用是通过影像分类制作土地覆被图和通过变化监测分析土地覆被变化,从而为土地利用分析提供了有力的工具与手段。论文以山东省垦利县为研究对象,探讨了利用TM数据进行耕地信息提取及监测,包括遥感图像的几何精校正、数据配准、研究区数据提取及图像增强等,分别采用屏幕目视解译法和自动分类与目视解译结合法两种分类方法提取耕地信息,在此基础上对耕地分类结果进行叠加监测,并对分类和监测结果进行了验证;最后对耕地资源的变化做了一个系统的分析。结果表明,在计算机自动分类结果的基础上,进行目视修正,可以有效地提高分类精度,适宜于农用土地区尤其是滨海盐渍土区的耕地信息提取。

A study on dynamic monitoring of cultivated land use in Kenli County

Remote sensing provides a powerful tool for study on Land Use /Cover Change.Two of the most common uses of satellite images in Land Use /Cover Change are land cover mapping via image classification and land cover change via change detection.Taking Kenli County in Shandong Province as a basic unit,using remote sensing image processing software ENVI 3.4and TM data,the whole process of cultivated land information extracting and dynamic change monitoring with TM data is described in this paper.The pretreatment of TM data includes images correction,data registration,data extraction and image enhancement.The image is classified by two methods,screen visual interpretation and supervised classification combined with screen visual interpretation.Both methods could acquire a good result,but the former method needs more manual work.Monitoring results are acquired based on the information extracting results by means of spatial analysis.It concludes that in agricultural regions,the methods of supervised classification combined with screen visual interpretation and post classification comparisons are effective in cultivated land information extracting and dynamic monitoring,especially in the coastal soil salinized regions.