Scalable framework of intelligent RFI flagging for large-scale HI survey data from FAST

Radio frequency interference (RFI) identification is a key step in radio data processing. In order to efficiently process huge volumes of data produced by modern large radio telescopes, such as the Five-hundred-meter Aperture Spherical radio Telescope (FAST), exceptional balance between accuracy and performance (throughput) is required for RFI flagging algorithms. RFI-Net is a single-process RFI identification package based on deep learning technique, and has achieved a higher flagging accuracy than the classical SumThreshold method. In this paper, we present a scalable RFI flagging toolkit, which can drive parallel workflows on multi-CPU and multi-GPU clusters, with RFI-Net as its core detector. It can automatically schedule the workload and aggregate itself after errors according to the running environment. Moreover, its main components are all pluggable, and can be easily customized according to requirements. The experiments with real data of FAST showed that using eight parallel workflows, the toolkit can process sky survey data at a speed of 66.79 GB/h, which means quasi-real-time RFI flagging can be achieved considering the data rate of FAST extragalactic spectral line observations.     

Quaternary chronology of Goulburn Valley sediments and their correlation in Southeastern Australia

Radiocarbon dates from the Goulburn Valley, Victoria, throw new light on the age of Quaternary tectonics, ancestral streams, fluviatile sediments and associated soils.

Sand dunes were formed during glacial times from ancestral rivers in which discharges were higher than at present. Tectonic movement occurred on the Cadell Fault between 13,000 and 20,000 B.P.

Stratigraphic correlation between the various units currently in use on the Riverine Plain is sometimes ambiguous, whereas wider correlation based on ground‐surface similarities in the valleys of the Goulburn, Maribyrnong and Shoalhaven rivers is invalid.

The sediments dated provide limiting ages for the red‐brown earth and minimal prairie soils developed on them in this area.     

An automatic HyLoggerTM mineral mapping method using a machine-learning-based computer vision technique

Abstract

HyLogger profile scanning is commonly utilised for drill-core logging but the limited scanning area may not detect all important geological features. The study presented in this paper aims to develop a mineral mapping solution for this core-logging process by leveraging the colour image captured during the scanning process. A machine-learning-based computer vision program was developed by implementing a k-means clustering and a global colour profiling algorithm. A suite of drill-core images was used to validate the developed program. Results indicate that there is a direct correlation between the mineral assemblage of a rock type and its colour specifications. The identified mineral type and relative abundance were comparable with HyLogger scan results.     

Opposition-based firefly algorithm for earth slope stability evaluation

This paper introduces a new approach of firefly algorithm based on opposition-based learning （OBFA） to enhance the global search ability of the original algorithm. The new algorithm employs opposition based learning concept to generate initial population and also updating agents’ positions. The proposed OBFA is applied for minimization of the factor of safety and search for critical failure surface in slope stability analysis. The numerical experiments demonstrate the effectiveness and robustness of the new algorithm.     

Does Social Learning Lead to Better Natural Resource Management? A Case Study of the Modern Farming Community of Practice in Sweden

This study investigates whether social learning among large-scale farmers in central Sweden leads to better natural resource management in the agricultural landscape. Three different frames of social learning are first identified: social learning as a fundamental social phenomenon, social learning as collaborative learning, and social learning as deeper learning. This article investigates the role of social learning and other factors through semistructured in-depth interviews. Results show that learning among farmers is inherently social, but that this learning does not necessarily improve natural resource management or lead to better environmental governance. The article discusses when social learning can be expected to influence natural resources management, and finds that without the presence of policy, individual leadership, or facilitation, it is not an important factor. Furthermore, the call for social learning based on results from successful instrumental application risks obscuring findings indicating that both social learning and better natural resource management are conditioned on the same external factors.     

改进的OIF和SVM结合的高光谱遥感影像分类

文章提出了一种结合改进的最佳指数法(OIF)和支持向量机(SVM)进行高光谱遥感影像分类新方法.利用本文提出的稳定系数进行波段初选择,根据相关系数选择波段组合生成新影像,并对新影像进行OIF计算,得到OIF值最大的波段组合为最佳波段组合;构建SVM分类器,对最佳波段组合分类;最后将分类结果与其他监督分类方法比较,并在相同核函数下与PCA和SVM结合的方法进行精度比较分析.实验结果表明,本文方法能够有效提取最佳波段组合,在SVM算法下获得较高分类精度.     

基于OLI影像多参数设置的SVM分类研究

在遥感影像自动分类中仅使用光谱特征很难产生正确的分类,OLI影像是波段数较多的多光谱影像,如果增加纹理、几何等多种特征以提高分类精度,就会使得特征的维度很高.支持向量机善于解决小样本、非线性和高维的影像分类问题,但是核函数和参数的设置只能依靠实验来获得.文中在OLI影像中提取了23个特征,逐个测试核函数和参数值对分类结果的影响.研究的主要结论如下：RBF核的支持向量机分类精度最高,Sigmoid核支持向量机分类精度最低;核函数的选择对分类精度的影响最大;核函数和参数值的变化不会影响重要特征的使用,3种核的支持向量机分类所使用的重要特征基本一致.     

Machine learning approaches to coastal water quality monitoring using GOCI satellite data

Since coastal waters are one of the most vulnerable marine systems to environmental pollution, it is very important to operationally monitor coastal water quality. This study attempts to estimate two major water quality indicators, chlorophyll-a (chl-a) and suspended particulate matter (SPM) concentrations, in coastal environments on the west coast of South Korea using Geostationary Ocean Color Imager (GOCI) satellite data. Three machine learning approaches including random forest, Cubist, and support vector regression (SVR) were evaluated for coastal water quality estimation. In situ measurements (63 samples) collected during four days in 2011 and 2012 were used as reference data. Due to the limited number of samples, leave-one-out cross validation (CV) was used to assess the performance of the water quality estimation models. Results show that SVR outperformed the other two machine learning approaches, yielding calibration R² of 0.91 and CV root-mean-squared-error (RMSE) of 1.74 mg/m³ (40.7%) for chl-a, and calibration R² of 0.98 and CV RMSE of 11.42 g/m³ (63.1%) for SPM when using GOCI-derived radiance data. Relative importance of the predictor variables was examined. When GOCI-derived radiance data were used, the ratio of band 2 to band 4 and bands 6 and 5 were the most influential input variables in predicting chl-a and SPM concentrations, respectively. Hourly available GOCI images were useful to discuss spatiotemporal distributions of the water quality parameters with tidal phases in the west coast of Korea.     

迁移学习支持下的遥感影像对象级分类样本自动选择方法

面向遥感大范围应用的目标,自动化程度仍是遥感影像分类面临的重要问题,样本的人工选择难以适应当前土地覆盖信息自动化提取的实际应用需求。为了构建一套基于先验知识的遥感影像全自动分类流程,本文将空间信息挖掘技术引入到遥感信息提取过程中,提出了一种面向遥感影像对象级分类的样本自动选择方法。该方法通过变化检测将不变地物标示在新的目标影像上,并将过去解译的地物类别知识迁移至新的影像上,建立新的特征与地物关系,从而完成历史专题数据辅助下目标影像的自动化的对象级分类。实验结果表明,在已有历史专题层的图斑知识指导下,该方法能有效地自动选择适用于新影像分类的可靠样本,获得较好的信息提取效果,提高了对象级分类的效率。     

属类概率距离构图的半监督高光谱图像分类

提出一种利用属类概率距离构图的半监督学习算法,并应用于高光谱图像分类。首先,该算法利用基于分类的稀疏表达方法来预估未标记样本的属类概率向量,然后,利用这个概率向量对描述数据相似性的距离函数进行改造,改造后的距离函数能有效扩大异类样本点之间的距离,在新的距离函数的度量下,每个样本点的邻域中可包含更多同类的样本点。最后,将该距离函数应用于半监督学习线性邻域传播算法和标签传播算法中。在Hyperion 和AVIRIS高光谱遥感图像上的实验结果表明：相比于传统的基于图的半监督学习算法,该算法能有效提高高光谱遥感图像分类精度。