利用SAR影像时间序列的耕地提取研究

卫星遥感是耕地资源调查的一种重要技术手段,利用遥感时间序列数据进行耕地提取具有很强的实践意义。光学遥感成像过程易受光照和大气条件影响,在云雨多发地区所能获取的可用数据十分有限;合成孔径雷达(SAR)能够全天时、全天气进行对地观测,但受斑点噪声影响,少见利用其构建时间序列进行信息提取的研究。本文研究了SAR影像时间序列在耕地提取中的适用性,利用江苏省徐州市2009年12月-2010年12月共11景ENVISAT ASAR 影像构建时间序列,目视选取30个5像元×5像元大小的耕地样区,分别统计样区内(相邻位置)与样区间(不同位置)耕地时域后向散射特征的一致性(变异系数);然后利用欧氏距离法、相关系数法以及动态时间弯曲法(DTW)进行研究区的耕地提取。结果显示：相邻位置耕地像元后向散射特性较为一致,平均变异系数为9.96%;不同位置耕地像元后向散射特性一致性也较好,平均变异系数为15.27%。在所选的3种方法中,相关系数法耕地提取精度最高,正确率与完整率分别为86.25%与80.70%;欧氏距离法精度次之,正确率与完整率分别为76.40%与71.93%;DTW效果较差,正确率和完整率分别为62.15%和77.78%。SAR影像时间序列作为一种新的数据组织形式,可用于耕地的有效提取。     

Context-awareness in similarity measures and pattern discoveries of trajectories: a context-based dynamic time warping method

Analyzing the spatial behaviors of moving-point objects (MPOs) and discovering their movement patterns have been of great interest to the geographic information science community recently. These interests can be explored through analyzing similarities in the MPO trajectories. Because movements of objects take place in various contexts, their trajectories are also highly influenced by such contexts. Therefore, it is essential to fully understand the contexts and to realize how they can be incorporated into movement analysis. This article first proposes a taxonomy for contexts. Then, a modified version of dynamic time warping called context-based dynamic time warping (CDTW) is introduced, to contextually assess the multidimensional weighted similarities of trajectories. Ultimately, the results of similarity searches are utilized in discovering the relative movement patterns of the MPOs. To evaluate the performance and effectiveness of our proposed CDTW method, we run several experiments on real datasets that were obtained from commercial airplanes in a constrained Euclidean space, taking into account contextual information. Specifically, these experiments were conducted to explore the role of contexts and their interactions in similarity measures of trajectories. The results yielded the robustness of CDTW method in quantifying the commonalities of trajectories and discovering movement patterns with 80% accuracy. Moreover, the results revealed the importance of exploiting contextual information because it can enhance and restrict movements.