An evaluation of Guided Regularized Random Forest for classification and regression tasks in remote sensing

New Earth observation missions and technologies are delivering large amounts of data. Processing this data requires developing and evaluating novel dimensionality reduction approaches to identify the most informative features for classification and regression tasks. Here we present an exhaustive evaluation of Guided Regularized Random Forest (GRRF), a feature selection method based on Random Forest. GRRF does not require fixing a priori the number of features to be selected or setting a threshold of the feature importance. Moreover, the use of regularization ensures that features selected by GRRF are non-redundant and representative. Our experiments based on various kinds of remote sensing images, show that GRRF selected features provides similar results to those obtained when using all the available features. However, the comparison between GRRF and standard random forest features shows substantial differences: in classification, the mean overall accuracy increases by almost 6% and, in regression, the decrease in RMSE almost reaches 2%. These results demonstrate the potential of GRRF for remote sensing image classification and regression. Especially in the context of increasingly large geodatabases that challenge the application of traditional methods.     

非线性洋面温度反演

本文从辐射方程出发，简述洋面温度与亮度温度之间的非线性关系，进行非线性洋面温度反演的方案设计，用正演算法 计算出非线性洋面温度反演公式的回归系数，对样进行检验。     

基于分块分形编码的多姿态、表情人脸图像检索

利用图像分形编码中定义域块和最小均方误差一一对应的特点，提出了1种基于分形编码的多姿态、表情的人脸图像检索方法。该方法将待检索图像分割为相同大小的值域块，然后将每一值域块按给定的定义域块进行分形编码，得到最小均方误差，计算该最小均方误差与图像库中最小均方误差的欧氏距离。将待检索图像所有值域块的欧氏距离求平均，此平均欧氏距离较小的几幅图像即为检索出的图像。实验证明该方法能够准确地检索出图像库中存储的同一人的不同姿态、表情的图像。     

Metridia pacifica in Dabob Bay, Washington: The diatom effect and the discrepancy between high abundance and low egg production rates

Information on life cycle strategies and reproductive parameters of Metridia pacifica is scarce, despite its importance in the zooplankton of the subarctic Pacific. In many regions it occurs in high abundance, but reproductive rates, when reported, are usually low. This discrepancy was studied in Dabob Bay, Washington, USA, in the context of an investigation of the effect of diatom blooms on the reproductive success of copepod grazers.In situ egg production rates of M. pacifica were measured in spring and mid-summer with standard methods (multi-wells) and a new incubation chamber (spawning towers) that separates the spawning female from its eggs and allows the eggs to develop undisturbed. Many females did not produce eggs, possibly due to a high fraction of immature individuals. Egg production rates were variable, but clutch sizes were higher in spawning towers, and estimates of female egg cannibalism revealed that females consume many eggs shortly after their release. Thus, a separation of females and eggs is mandatory for accurate measurements of M. pacifica egg production rates. The maximum clutch sizes recorded in our study were comparable to measurements for other calanoids. However, unviable eggs were a large fraction of those spawned, independent of incubation method, especially in late winter and early spring. In order to assess whether the diatom effect may be responsible for low viability of embryos and nauplii, we also measured in situ grazing. Adult females were omnivorous, but they ingested some diatoms that rank among the strongest anti-mitotic toxin producers known so far. Although M. pacifica’s vertical migration behavior suggests opportunistic feeding on abundant food during their short stay in the phytoplankton-rich surface, they often ignored the food items that contributed most to microplankton carbon concentrations. Thus, their feeding strategy remains ambiguous. Due to severe reproductive failure early in the season, recruitment was impaired in spring, while the population increased, reaching high abundance in mid-summer when reproductive output was low. While advection and interannual variability in bloom conditions might compensate for the losses described here, the paradox of high Metridia abundance versus low reproductive success still requires further investigation, and methodological constraints need to be ruled out in future studies.     

Ghost fishing in European waters: Impacts and management responses

In this paper we review levels of net loss, what happens to the gear once it has been lost, and the resulting levels of ‘ghost catches’ made in passive net fisheries in the EU. We also consider ghost catches resulting from lost gear in other types of fisheries, and the extent to which the value of ghost catches has been quantified. We consider why fishing gear is lost, and profile common management responses. We present a cost benefit model to assess the relative cost effectiveness of different management measures, and suggest that gear retrieval programmes may provide less value for money than other management responses.     

基于BP人工神经网络的水体遥感测深方法研究

利用Landsat7ETM＋遥感图像反射率和实测水深值之间的相关性,建立了动量BP人工神经网络水深反演模型,并对长江口南港河段水深进行了反演.结果表明：具有较强非线性映射能力的动量BP神经网络模型能较好地反演出长江口南港河段的水深分布情况;由于受长江口水体高含沙量的影响,模型对小于5 m的水深值反演精度较高,而对大于10 m的水深值反演精度较低.     

改进的边缘角度直方图在遥感图像检索中的应用

针对基于形状的遥感图像检索中区域分割困难这个问题,本文提出利用边缘角度直方图作为形状特征对遥感图像进行检索;针对遥感影像分辨率和颜色反差对检索性能的影响,本文提出采用多尺度分析的思想,提取图像在多个尺度上的边缘角度直方图特征进行检索。通过单一尺度边缘角度直方图特征和多尺度边缘角度直方图特征检索的对比实验,揭示了本文所提出的改进方案使遥感图像的检索精度提高了约31个百分点。     

Developing a SPASE Query Language

The advent of the Virtual Observatory has begun an evolution in the space physics data environment. A number of nascent and discipline specific Virtual Observatories have started to emerge with an emphasis on data search and retrieval. As this new data environment takes shape an emphasis will be placed on interdisciplinary communication in attempts to address large scale and global problems. To this end we formulate the development of a query language to facilitate Virtual Observatory to Virtual Observatory communication. Furthermore, we outline the goals of such a language, how it would work and how existing community efforts can be leveraged to speed the development of this query language.

长江口水域多光谱遥感水深反演模型研究

利用Landsat-7 ETM 遥感影像反射率和实测水深值之间的相关性可以探测水深。该文介绍单波段、双波段比值和多波段3种线性回归模型以及动量BP人工神经网络水深反演模型。选择长江口北港河道上段作为研究区,利用上述模型,分两种情况进行水深反演:一是以河道全部历史样本建模;二是将河道按自然水深划分为浅水区和深水区分别建模。结果表明:神经网络模型预测精度高于线性回归模型;水深分区后线性回归和神经网络模型预测误差均有所减小。