物候和波谱——位置分析在城镇绿化植物群分类中的应用

发布时间：2018-01-26 09:28

  本文关键词： 城镇植物群 图像分类 物候遥感 波谱—位置分析　出处：《遥感技术与应用》2017年05期 　论文类型：期刊论文

【摘要】：遥感图像植物群分类已被证明是植物群分布自动制图快速有效的方法。然而,场景噪声和植物群之间光谱可分性差等形成的负面影响,使传统的分类方法无法满足必要的精度要求。为了解决这个问题,提出了一种称为SLPA的遥感图像植物群分类方法。它由波谱—位置联合分析(S-L分析)和植物物候遥感分析(PA)两部分组成。通过向特征空间添加密度描述符以及在特征空间叠加冬、夏季图像特征数据,可以将这两类分析嵌入分类过程。这种改进增加了可用描述符的数量,使分类特征空间足够丰富,以适应复杂分类;同时又降低了分类不确定性,使分类精度获得显著改善。精度测试显示,增加S-L分析和物候分析,将使植物群分类的全局精度分别平均提高15.0%和29.3%。另外,由于采用二值邻域均值替代灰度邻域密度,使得加入S-L分析几乎不引起运算复杂性增大。Matlab测试结果显示,SLPA在城镇植物群遥感自动分类方面具有鲁棒和普适性。
[Abstract]:The classification of flora in remote sensing images has been proved to be a rapid and effective method for automatic mapping of flora distribution. However, the negative effects of scene noise and poor spectral separability among flora have been found. In order to solve this problem, the traditional classification method can not meet the necessary precision requirements. In this paper, a classification method of plant flora in remote sensing images called SLPA is proposed, which is based on spectral and position analysis (S-L analysis) and phytophenological remote sensing analysis (PAA). By adding density descriptor to feature space and overlaying winter in feature space. The summer image feature data can be embedded into the classification process. This improvement increases the number of available descriptors and makes the classification feature space abundant enough to adapt to complex classification. At the same time, the uncertainty of classification is reduced, and the accuracy of classification is improved significantly. The precision test shows that S-L analysis and phenological analysis are added. The global accuracy of plant taxonomy will be improved by 15.0% and 29.3, respectively. In addition, the binary neighborhood mean is used to replace the gray neighborhood density. The results of Matlab test show that SLPA is robust and universal in remote sensing automatic classification of urban flora.
【作者单位】： 上海辰山植物园;上海植物园;上海城市植物资源开发应用工程技术研究中心上海植物园;华东师范大学地理科学学院;
【基金】：十二五国家科技支撑计划项目“绿地低碳效益综合提升和评价技术研究”(2013BAJ02B01-4)资助
【分类号】：S731;TP751
【正文快照】： 1　引　言植物群类别是生态功能定量估算的一个重要变量,比如它曾用于植物生物量和净生产力的估算[1],也曾用于与植物生态功能相关的生物多样性[2]、动物和昆虫的栖息地质量[3]、植物碳储存量[4]等的评估。因此,开发植物群类别识别技术对于生态建模非常重要。野外光谱测试表明，

本文编号：1465271