高光谱遥感反演土壤重金属含量研究进展

土壤重金属污染是当今重要的世界性环境问题之一。如何快速高效地获取土壤重金属含量及分布信息是进行污染风险评价及预测研究的前提。高光谱遥感具有无损、实时获取信息的优势,利用高光谱技术反演土壤重金属含量仍处于探讨阶段,对其反演机理和建模方法的研究成果进行了介绍和分析。认为充分研究不同类型土壤对重金属的吸附特性,是深入理解重金属含量反演机理的基础。加强不同土壤类型、不同污染程度的案例对比分析,发展光谱信号处理方法,引入非线性建模算法,构建重金属形态量反演模型,是土壤重金属含量遥感估算研究的重要研究方向。     

Estimation of Sea Surface Velocities from SAR Images Using the Doppler Shift

The sea surface velocities field plays an important role in seawater exchange and substance transportation.In this paper,the Wide Swath Mode(WSM) data derived from the ENVISAT Advanced Synthetic Aperture Radar(ASAR) are used to retrieve the high-resolution sea surface velocities field.based on the theoretical model of SAR Doppler shift,the errors caused by the relative motion between Earth and satellite are removed.Then we use the C Band Doppler Frequency Model(CDOP) and the Bragg scattering model to remove the errors caused by the sea surface wind and Bragg scattering,respectively.The data used to verify the accuracy of the retrieval results are AVISO(Archiving Validation and Interpretation of Satellite Oceanographic) velocities and the GLD(Global Drifting Buoy) velocities.Our method is applied both in the coastal area with land cover(Agulhas) and the open sea(Kuroshio) without any land.Results show that in Agulhas,the velocity ranges from -1.8 m/s to 1.8 m/s,and their directions agree very well.In the Kuroshio,the ASAR current can clearly reveal the flow path and direction of the Kuroshio,and it matches well with the AVISO current.The comprehensive results shows,the Root Mean Square Error(RMSE) of ASAR and AVISO is 0.17 m/s,and the RMSE of ASAR and GLD is 0.11 m/s.This implies that the methods used here not only simplified the processes but also has high accuracy to retrieve sea surface velocities both in the coastal area and in the open sea.     

一种基于形状特征的图像检索方法

基于内容的图像检索技术中,形状特征是最重要的图像特征部分。傅立叶描述子(FD),因为它高效性,稳定性,低计算复杂度和少量参数运行的优点而被广泛应用。目前已经存在的大多数傅立叶描述子往往直接使用它的傅立叶系数,本文我们应用主成分分析Principal Component Analysis(PCA)方法在傅立叶描述子中,从而提出基于PCA的傅立叶描述子的形状描述和检索方法。本论文的实验采用MPEG-7标准图像数据库来比较检索性能。实验结果表明PCA-FD的方法实现了更短时间内的更好检索精确率。     

Learning to rank academic experts in the DBLP dataset

Expert finding is an information retrieval task that is concerned with the search for the most knowledgeable people with respect to a specific topic, and the search is based on documents that describe people's activities. The task involves taking a user query as input and returning a list of people who are sorted by their level of expertise with respect to the user query. Despite recent interest in the area, the current state‐of‐the‐art techniques lack in principled approaches for optimally combining different sources of evidence. This article proposes two frameworks for combining multiple estimators of expertise. These estimators are derived from textual contents, from graph‐structure of the citation patterns for the community of experts and from profile information about the experts. More specifically, this article explores the use of supervised learning to rank methods, as well as rank aggregation approaches, for combining all of the estimators of expertise. Several supervised learning algorithms, which are representative of the pointwise, pairwise and listwise approaches, were tested, and various state‐of‐the‐art data fusion techniques were also explored for the rank aggregation framework. Experiments that were performed on a dataset of academic publications from the Computer Science domain attest the adequacy of the proposed approaches.     

西南地区不同山地环境梯度叶面积指数遥感反演

叶面积指数(LAI)遥感估算是植被定量遥感研究的热点之一,监测植被LAI时空变化对于研究陆地生态系统碳循环及全球变化等具有非常重要的意义。在我国西南山区设置10个50km×50km的观测样区作为研究区,其中包括5个森林生态系统样区、3个农田生态系统样区和2个草地生态系统样区。分别获取不同优势植被类型LAI地面实测数据,结合同期获取的遥感数据,考虑地形因素影响,基于偏最小二乘原理分别构建各样区LAI遥感估算模型,并采用交叉验证的方式对模型精度进行评价。结果表明:考虑了海拔、坡度和坡向等地形因子的森林LAI遥感反演模型与未考虑地形变量的模型相比,其验证精度有所提高,R2由0.30~0.75提高至0.50~0.80,RMSE由0.52~0.93m2/m2降低至0.48~0.89m2/m2;所有样区优势植被类型LAI反演模型验证R2在0.40~0.80之间,RMSE在0.22~0.89m2/m2之间。发展的LAI遥感估算方法有助于认知山地植被LAI反演的地形效应问题,可为进一步的山地植被长势监测提供科学依据。     

整合开源工具的小型搜索引擎构建

在介绍了搜索引擎基本原理和构架的基础上,提出了一种整合Larbin、Lucene、ICATLAS等多种开源工具快速构建小型搜索引擎的实验方案。实验表明,该引擎能有效完成网页的采集、处理、索引和检索等工作。     

视频分析技术的研究和应用

通过对视频分析关键技术的探讨及国内外研究小组科研成果的分析.揭示了视频编码技术、视频对象分割技术、基于内容检索技术、流媒体传输技术之间的内在联系以及各自的技术要点、难点.通过介绍视频分析技术在视频监控中的应用,展望了该技术在多个领域的广阔的应用前景.     

基于模糊颜色空间聚类的图像检索方法

鉴于图像信息固有的复杂性和不精确性,引入了模糊化的机理对HSI颜色模型的颜色信息进行模糊量化,得到关于颜色特征的模糊直方图,并应用Robust的视觉采样聚类方法对其进行聚类和匹配,取得了较为满意的结果。     

基于SVM的图像低层特征与高层语义的关联*

在基于内容的图像检索中,针对图像的低层可视特征与高层语义特征之间的鸿沟,提出了一种基于支持向量机(SVM)的语义关联方法。通过对图像低层特征的分析,提取了颜色和形状特征向量(221维),将它们作为支持向量机的输入向量,对图像类进行学习,建立图像低层特征与高层语义的关联,并应用于鸟类、花卉、海洋以及建筑物等几个典型的语义类别检索。实验结果表明,该方法可适应于不同用户的图像检索,并提高了检索性能。