物候和波谱--位置分析在城镇绿化植物群分类中的应用  

Applications of Phenology Analysis and Spectrum-Location Analysis in the Classification of Urban Vegetation Population

作　　者：秦俊 冷寒冰[3] 赵广琦[3] 景军[3] 周坚华 

机构地区：[1]上海辰山植物园,上海201602 [2]上海植物园,上海200231 [3]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,上海200231 [4]华东师范大学地理科学学院,上海200241

出　　处：《遥感技术与应用》2017年第5期948-957,共10页Remote Sensing Technology and Application

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

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。Classification from remote sensing imagery has been proved to be a quick and effective method for automatic mapping of vegetation population.However,the interference from background noises and the weak spectral separability between vegetation populations have a negative impact on the identification of urban vegetation populations from remote sensing images.This make the classification accuracy unsatisfied as using conventional classification methods.In order to solve this problem,a new method of classification from remote sensing imagery,named as SLPA,is proposed.SLPA consists of two parts：Spectrum-Location joint analysis（abbreviated as S-L analysis or SL）and Phenology remote sensing Analysis（PA）.By adding density descriptors to a feature space and by combining the descriptors derived from winter and summer images of a scene into the same feature space,these two kinds of analyses can be embedded into the classification process.The embedding increases the number of available independent descriptors therefore making the feature space rich enough to adapt to such a complex classification;meanwhile,uncertainty in the classification can be reduced and the classification accuracy can be improved significantly.The data from accuracy assessments show that with the S-L analysis and the phenology analysis,the overall accuracies of the classification for urban vegetation population will increase by 15%and 29.3%respectively.In addition,the S-L analysis almost does not increase the computational complexity because a frequently used computation for the density of gray elements is replaced with a calculation of binary mean value,a much more time saving operation.Experiments indicate that SLPA is good at robustness and universality in the classification of urban vegetation population.

关 键 词：城镇植物群 图像分类 物候遥感 波谱—位置分析 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]