卫星遥感影像在农田灌排系统识别中的应用研究

利用监督分类、NDVI指数、屏幕数字化等遥感影像处理方法在黄河下游的山东簸箕李灌区开展农田灌排系统识别的应用研究。在对采用卫星遥感影像(SPOT、LANDSAT、ERS)识别的灌区农田灌排沟渠可视化效果进行分析、对比与评价的基础上，完成对农田灌排系统的数字化处理，得到灌排沟渠在灌区内的分布状况及总长度估值。研究表明，对SPOT XS影像进行屏幕数字化处理似乎是用来识别农田灌排沟渠的理想途径，估算的灌区干、支、斗三级沟渠的总长度具有较好的精度，然而要对农田灌排系统进行精确地分级划类还有赖于遥感识别技术、GPS定位方法与实地调查工作的紧密结合。

Applied study of satellite image in identification of irrigation and drainage system

By applying remote sensing techniques to the Bojili Irrigation District in Shandong, the applied study of satellite image in identification of irrigation and drainage system was conducted. Various image processing methods, including the supervised classification, the NDVI index and the screen digitization, were used to identify the ditch and canal conditions. The various satellite sensors (SPOT, LANDSAT and ERS) were applied, compared and evaluated. For the Bojili Irrigation District, the irrigation and drainage system were digitized by hand from the screen and mapping and statistics of the tertiary ditches and canals were carried out. The results show that the digitization of SPOT XS image processing seems to be satisfactory to identify the irrigation and drainage system because the quality of the image is better. However, a detailed classification in those tertiary ditches and canals must be dependent upon an integrative combination of RS technique and GPS measurements with local staff surveys and investigations.