基于卫星遥感数据的库尔勒香梨识别提取方法研究

针对目前国内利用卫星遥感数据的光谱特征及其植被指数对香梨识别提取方法鲜有研究的现状,探讨研究了Landsat8_OLI数据对不同树龄香梨的识别提取方法和实际应用。以库尔勒市中部的英下乡和沙依东园艺场为试验区域。采用多时相Landsat8_OLI数据。通过分析试验区内典型地物的光谱特征和植被指数的变化规律以及作物物候历的不同,选择不同树龄香梨树林遥感识别的最佳时相,利用决策树方法构建了不同树龄香梨树林的识别提取模型,实现了香梨种植面积的有效提取。结果表明:1香梨成林(种植5a以上的香梨树林)遥感提取效果较为满意,精度为93.51%,而香梨幼林由于受多种因素的影响,其提取精度只有78.79%,有待进一步研究;2 Landsat8_OLI数据具有较高的时空分辨率,是遥感识别提取库尔勒香梨的潜力数据源。

Study on Extracting Method of Korla Pear Information based on Satellite Remote Sensing data

Korla pear industry plays a significant role in the Xinjiang fruit planting and production and masters the korla pear information such as the planting area, the spatial distribution is of great significance to develop fruit industry. However, studies using the spectral characteristics and vegetation index of multi-temporal Remote Sensing data to extract the korla pear"s information were still few in our country. the research took Korla city as the study area,used Landsat8_OLI satellit data , and analyzed the difference of spectral characteristics and vegetation index between korla pear and other typical land cover types to model a decision tree in Korla city of Xinjiang Uygur Autonomous Region, then used the model to extract korla pear information precisely. The results suggested that :(1) the mature korla pear was extracted effectively with an overall accuracy of 93.51%, However the accuracy of young korla pear was only 78.79% due to the influence of various factors. it need further on young korla pear information .(2) Landsat8_OLI satellit data has high spatial resolution and convenient access,so it is a potential data source for korla pear information identification, monitoring, extracting in the future.