南方丘陵地区水稻种植面积遥感信息提取的可行性分析

以浙江省为试验区，在地理信息系统支持下综合利用多种地理信息，探讨丘陵地区大面积提取水稻种植面积信息的可行性。开展了分类识别方法的比较试验及训练样点相对稳定性试验。针对丘陵地区的复杂地形，在数字化地形图的基础上，建立数字地形模型（ＤＴＭ），并衍生出地面坡度等地貌因子的数字化图像，结合ＮＯＡＡ／ＡＶＨＲＲ数据，进行分类。试验结果表明，传统的分类识别方法中，最大似然法的分类精度可满足业务化运行的要求；建立在混合像元分解基础上的模糊监督分类，有较高的分类精度和较好的稳定性，具有较强的适应性；地貌因子参与遥感影像的分类，不仅可以有效地提高丘陵地区水稻种植面积信息的提取精度，而且还可以使面积信息提取精度保持一定的稳定性，提高空间精度；为探讨丘陵地区水稻种植面积信息遥感提取的可靠性和客观性，在训练样点保持相对稳定的前提下，对１９９６年和１９９７年浙江省水稻种植面积进行测算，两年的数量精度均在９２％以上。

Analysis of Feasibility of Professional Operating on Estimating Rice Planting Area of Hilly Region in Southern China Using Remote Sensing Technique

The objective of the study (Zhejiang province as our test region) was to investigate how to utilize synthetically a wide variety of geographical information,included topographic map,vegetation map,land cover/use map and other ancillary information,for estimating rice planting area of hilly region in southern China by remote sensing technique with NOAA/AVHRR data.The research contents mostly concerned both the contrast tests on practical approaches and the relative stability test on training sample.Both DEM and digital slope imagery derived from the digital relief map were used for the purpose of improving the classification accuracies of AVHRR data in large hilly region.The results indicated that the accuracies of maximum likehood classification could satisfy the professional requirement of estimating rice planting area and fuzzy supervised classification based on unmixing AVHRR imagery has better classification and stability than MLK.In addition,the results through using DEM and slope imagery as categorization data suggests slope imagery may improve the results of extracting paddy field signatures from AVHRR,particularly may improve the spatial precisions,while DEM contribute nothing to improve the accuracies mentioned above.The results also indicated that the precisions of estimating rice planting area with relative stable training sample exceeded 92%,up to 98%.