基于遥感影像时间序列的冬小麦种植监测最佳时相选择研究  

Selection of Optimum Periods for Extracting Winter Wheat Based on Multi-temporal Remote Sensing Images

作　　者：齐腊[1] 刘良云[1] 赵春江[1] 王纪华 王锦地[2] 

机构地区：[1]国家农业信息化工程技术研究中心,北京100097 [2]北京师范大学地理学与遥感科学学院,北京100875

出　　处：《遥感技术与应用》2008年第2期154-160,共7页Remote Sensing Technology and Application

基　　金：国家863项目(2006AA12Z138);国家科技支撑项目(2006BAD10A01)资助

摘　　要：遥感影像植被分类的最佳时相对作物种植面积遥感监测非常重要。根据2005～2006年北京冬小麦不同物候期的Landsat TM影像和2006年Spot-2影像，计算了各时期影像中主要植被类型的光谱可分性距离，分析了北京郊区主要植被物候差异和光谱可分性；对各生育期的遥感影像及其主要组合进行了监督分类，采用总体精度和分类效率指标两个参数，结合地面GPS调查数据，对分类结果进行了精度评价。结果表明：北京地区小麦监测最佳时相是4月上旬，影像分类的总体精度为92．9％，明显优于其它单时相影像的分类结果；发现北京郊区冬小麦光谱分类的最佳时相组合为4月上旬（起身期）和5月下旬（灌浆期），分类总体精度为94％。Selection of optimum periods of vegetation classification is important for extracting crop planting area. Based on Landsat TM images and Spot-2 images in Beijing, spectral distance between winter wheat and other ground targets was calculated,and the phenological difference of main vegetation types in Beijing was analyzed. Then winter wheat was classified from each individual image or their combinations using maximum likelihood classification algorithm. Finally, based on ground surveyed samples and visual interpretation,the classification results were evaluated by overall accuracy and classification efficiency index for different cases. The results shows that： （1） the optimum periods for extracting winter wheat is early April （rising stage） with an overall accuracy of 92.9 %;（2） the optimum combination for winter wheat identification is early April and late May （grain-seeding stage） with an overall accuracy of 94%.

关 键 词：冬小麦 分类 J—M距离 多时相 精度验证 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置]