Estimating winter wheat plant water content using red edge parameters

Remote sensing of plant water content is difficult because the absorption band sensitive to foliar liquid water is also sensitive to the atmospheric vapour. A method using non-water-absorption spectral parameters to evaluate plant water content (PWC) would be valuable. In our experiment, canopy spectra of 48 winter wheat treatments with different varieties, different fertilization and irrigation levels were measured by an ASD FieldSpec FR spectrometer in six different growth stages from erecting stage to milking stage, and the PWCs of the related wheat plant samples were also measured. Significant positive coefficients of correlation were observed between PWC and spectral reflectance in 740–930?nm region in all of the six different growth stages, which indicates that the NIR spectral reflectance increases due to the effect of PWC on the leaf internal structure. This mechanism also affects the red edge spectrum in 680–740?nm region. The spectral reflectance increases more rapidly and the red edge becomes steeper if PWC is higher. The coefficients of correlation between PWC and red edge width, derived from the inverted-Gaussian model, are significant at the 0·999 confidence-level, which is more reliable than WI and NDWI, and the statistical models for PWC based on red edge width were set up in all the six different growth stages. In addition, LAI and canopy chlorophyll density (CCD) are also related to red edge parameter, such as red edge position and red edge width. It seems that PWC plays a more important role in red edge width than LAI and CCD due to the effect of PWC on the leaf internal structure, and that CCD plays a more important role in red edge position than LAI and PWC.     

基于新型存储器件的分布式文件系统性能优化

新型存储器件的I/O性能通常比传统固态驱动器（SSD）高一个数量级,然而使用新型存储器件的分布式文件系统相对于使用SSD的分布式文件系统性能并没有显著的提高,这说明目前的分布式文件系统并不能充分发挥新型存储器件的性能。针对这个问题,对Hadoop分布式文件系统（HDFS）的数据写入流程及传输过程进行了量化分析。通过量化分析HDFS数据写入过程各阶段的时间开销,发现在写入数据的各个阶段中,节点间数据传输的时间占比较大。因此提出了对应的优化方案,通过异步写入的方式并行化数据传输与处理过程,使得不同数据包的处理阶段叠加起来,减少了数据包整体的处理时间,从而提升了HDFS的写入性能。实验结果表明,所提方案将HDFS的写入吞吐量提升了15%~24%,总体的写入执行时间降低了28%~36%。     

基于新型存储器件的分布式文件系统性能优化

新型存储器件的I/O性能通常比传统固态驱动器（SSD）高一个数量级，然而使用新型存储器件的分布式文件系统相对于使用SSD的分布式文件系统性能并没有显著的提高，这说明目前的分布式文件系统并不能充分发挥新型存储器件的性能。针对这个问题，对Hadoop分布式文件系统（HDFS）的数据写入流程及传输过程进行了量化分析。通过量化分析HDFS数据写入过程各阶段的时间开销，发现在写入数据的各个阶段中，节点间数据传输的时间占比较大。因此提出了对应的优化方案，通过异步写入的方式并行化数据传输与处理过程，使得不同数据包的处理阶段叠加起来，减少了数据包整体的处理时间，从而提升了HDFS的写入性能。实验结果表明，所提方案将HDFS的写入吞吐量提升了15%~24%，总体的写入执行时间降低了28%~36%。     

Microfluidic electrodischarge devices with integrated dispersion optics for spectral analysis of water impurities

This paper reports a microfluidic device that integrates electrical and optical features required for field-portable water-chemistry testing by discharge spectroscopy. The device utilizes a dc-powered spark between a metal anode and a liquid cathode as the spectral source. Impurities are sputtered from the water sample into the microdischarge and characteristic atomic transitions due to them are detected optically. A blazed grating is used as the dispersion element. The device is fabricated from stacked glass layers, and is assembled and used with a charge-coupled device (CCD) sensing element to distinguish atomic spectra. Two structural variations and optical arrangements are reported. Detection of Cr and other chemicals in water samples has been successfully demonstrated with both devices. The angular resolution in terms of angular change per unit variation in wavelength (/spl part//spl theta///spl part//spl lambda/) is experimentally determined to be approximately 0.10 rad//spl mu/m, as opposed to the idealized theoretical estimate of 0.22 rad//spl mu/m. This is because the microdischarge is uncollimated and not a point source. However, this is sufficient angular resolution to allow critical spectra of metal impurities to be distinguished.     

Quantitative approach in the spectral reflectance-lithostratigraphy of the Wind River and southern Bighorn basins,Wyoming

A procedure developed for quantifying spectral variability was applied to visible and near-infrared spectra from a database of the Wind River and Bighorn basins. Principal Components Analysis (PCA) of 94 sedimentary rock spectra distinguished shales, sandstones, limestones, dolostones, bedded gypsum and bentonites from 25 stratigraphic units. However, only twelve spectrally distinct groups were delineated because of lithological homogeneity. Albedo and slope characteristics of the spectra were found to be related to the first and second PC, respectively, and absorption bands related to the remaining components. A second PCA phase, using vertically scaled spectra to minimize the albedo variability, indicated space distributions that enhanced major absorption features in the ncar-infrared region. Parameters describing absorption bands were also calculated from spectral continuum. The best parameters for spectra discrimination were band depths at 1900, 2200 and 2300 nm, in agreement with the PCA results using scaled spectra. Finally, the stratigraphic column was characterized by correlated log sections composed by PC scores of the original data and of scaled spectra, in lieu of band, band-ratios, area and depths of absorption bands. Reflectance-lithostratigraphic markers were identified in the log sections and indicated the potential of reflectance-lithostratigraphy in correlation studies. We concluded that PCA using a small number of narrow bands to represent the spectrum was an adequate mathematical approach for spectral analysis. PCA of scaled spectra was found to be a useful tool to enhance differentiations due to absorption features.     

A spectral matching quality indicator for material mapping using spectral library search methods

Spectral library search methods are being used increasingly as an efficient approach for exploiting hyperspectral remotely sensed data in material identification and mapping applications. The aim of this study was to develop a quantitative method, using an indicator called the Quality factor (Q-factor), for providing quantitative information on the reliability of spectral identifications in the interpretation (classification) of unknown spectra by library search methods. This was achieved by summing the two main requirements of a typical reflectance spectral library search for material mapping: (1) a reliable correlation between spectral matching scores and material similarity, and (2) a reliable separation ability between the relevant and non-relevant parts of the candidate reference spectra. These form a metric whose values reflect the closeness of the output reference spectra to the input unknown spectra for a chosen library search method. The Q-factor was tested as an indicator of the reliability of the material identifications by the library search for a range of unknown reflectance spectra of various types of vegetation, soils and minerals collected from the US Geological Survey (USGS) Spectral Library and from our in-house spectral database. The results indicate that this approach has the potential to separate correct and incorrect spectral identifications resulting from a particular spectral library search method using a reference similarity logic. The method may be applied to any combination of deterministic spectral matching alternatives using reflectance spectra. Spectrum-level quality information provided by the Q-factor is useful for optimizing a particular search method or for choosing the most appropriate method for distinct identification and classification problems.     

基于F范数群组效应和谱聚类的无监督特征选择

基于谱聚类的无监督特征选择主要涉及相关系数矩阵和聚类指示矩阵, 在以往的研究中, 学者们主要关注于相关系数矩阵, 并为此设计了一系列约束和改进, 但仅关注相关系数矩阵并不能充分学习到数据内在结构. 考虑群组效应, 本文向聚类指示矩阵施加$F$范数, 并结合谱聚类以使相关系数矩阵学习更为准确的聚类指示信息, 通过交替迭代法求解两个矩阵. 不同类型的真实数据集实验表明文中方法的有效性, 此外, 实验表明$F$范数还可以使方法更加鲁棒.     

近红外高光谱图像数据预测技术

目的 受到传感器光谱响应范围的影响,可见光区域和近红外区域（400~2 500 nm）的高光谱数据通常使用不同的感光芯片进行成像,现有这一光谱区域典型的高光谱成像系统,如AVIRIS （airborne visible infrared imaging spectrometer）成像光谱仪,通常由多组感光芯片组成,整个成像系统成本和体积通常比较大,严重限制了该谱段高光谱探测技术的发展。为了能够扩展单感光芯片成像系统获得的高光谱图像的光谱范围,本文探索基于卷积神经网络的近红外光谱数据预测技术。方法 结合AVIRIS成像光谱仪的光谱配置,设计了基于残差学习的红外谱段图像预测网络,利用计算成像的方式从可见光范围的高光谱图像预测出近红外波段的光谱图像,并在典型的卫星高光谱遥感数据上进行红外光谱预测重构和基于重构的数据分类实验,以验证论文提出的红外光谱数据预测技术的可行性以及有效性。结果 本文设计的预测网络在Cuprite数据集上得到的预测近红外图像峰值信噪比为40.145 dB,结构相似度为0.996,光谱角为0.777 rad;在Salinas数据集上得到的预测近红外图像峰值信噪比为39.55 dB,结构相似性为0.997,光谱角为1.78 rad。在分类实验中,相比于只使用可见光图像,利用预测的近红外图像使得支持向量机（support vector machine,SVM）的准确率提升了0.6%,LeNet的准确率提升了1.1%。结论 基于AVIRIS传感器获取的两组典型卫星高光谱数据实验表明,本文提出的红外光谱数据预测技术不仅可基于计算成像的方式扩展可见光光谱成像系统的光谱成像范围,对于减小成像系统体积和质量具有重要意义,而且可有效提高可见光区域光谱图像数据在典型应用中的处理性能,对于提高高光谱数据处理精度提供新的技术支撑。     

不同株型夏玉米群体冠层反射光谱特征及其应用研究

比较不同株型夏玉米在不同时期的反射光谱差异性,研究分析了红边位置(λred)、红边振幅(Dλred)、最小振幅(Dλmin)及Dλred/Dλmin与叶片全氮含量(LTN),叶绿素含量(Chl)及叶面积指数(LAI)间的相关性,并建立预测模型。结果表明,光谱差异随生育进程呈不同程度的规律性变化。在全生育期,用Dλred/Dλmin能更好地推算LTN,尤其在吐丝期,在开花前用Dλred也佳,在拔节期和喇叭口期用λred也有较高的精度。估算Chl时,在开花前用Dλred较可靠,在喇叭口期和抽雄期用λred也可考虑。估算LAI时,抽雄期后用Dλred推算有较高的可信度,在抽雄期用λred较好,在开花期和吐丝期用Dλred/Dλmin推算更为可靠。     

Multi-objective and constrained optimization for DS-CDMA code design based on the clonal selection principle

This paper proposes two new algorithms based on the clonal selection principle for the design of spreading codes for DS-CDMA. The first algorithm follows a multi-objective approach, generating complex spreading codes with “good” auto as well as cross-correlation properties. It also enables spreading code design with no restrictions on the number of users or code length. The algorithm maintains a repertoire of codes that are subject to cloning and undergo a process of affinity maturation to obtain better codes. Results indicate that the produced code sets lie very close to the theoretical Pareto front. A second penalty function-based constrained optimization algorithm based on clonal selection is proposed. It is applied to the design of spreading codes with pre-defined power spectral density requirement. The results suggest that the algorithm is capable of lowering significantly, the power spectra at undesired frequencies. Therefore, with the proposed algorithm, a DS-CDMA transmitter can, for the first time, selectively transmit power across the transmission bandwidth and adjust to jammers and other interferers. This study illustrates that using two stages of multi-objective and constrained optimization, using the proposed clonal selection algorithms, is an effective code design strategy.     

Mapping two Eucalyptus subgenera using multiple endmember spectral mixture analysis and continuum-removed imaging spectrometry data

Successful discrimination of a variety of natural and urban landscape components has been achieved with remote sensing data using multiple endmember spectral mixture analysis (MESMA). MESMA is a spectral matching algorithm that addresses spectral variability by allowing multiple reference spectra (i.e., endmembers) to represent each material class. However, materials that have a high-degree of spectral similarity between classes, such as similar plant-types or closely related plant species, and large variations in albedo present an ongoing challenge for accurate class discrimination with imaging spectrometry. Continuum removal (CR) analysis may improve class separability by emphasizing individual absorption features across a normalized spectrum. The spectral and structural characteristics common to most Eucalyptus trees make them notoriously difficult to discriminate in closed-canopy forests with imaging spectrometry. We evaluated whether CR applied to hyperspectral remote sensing data improved the performance of MESMA in classifying and mapping nine eucalypt tree species according to the two major Eucalyptus subgenera, Eucalyptus (common name “monocalypt”) and Symphyomyrtus (common name “symphyomyrtle”). Mixed-canopies comprised of monocalypts and symphyomyrtles are common in Australia, although their spatial distribution is not random. The ability to map these functional types on a landscape-scale could provide important information about ecosystem processes, landscape disturbance history and wildlife habitat. We created a spectral library of 229 pixels from 37 symphyomyrtle tree canopies and 406 pixels from 62 monocalypt tree canopies selected from HyMap imagery and verified with field data. Based on these reference data, we achieved overall classification accuracies at the subgenera-level of 75% (Kappa 0.48) for non-CR spectra and 83% (Kappa 0.63) for the CR spectra. We found that continuum-removal improved the classification performance of most endmember-models, although a larger portion of pixels remained unmodeled with the CR spectra (2%) compared to the non-CR spectra (0%). We utilized a new method for model optimization and created maps of monocalypt and symphyomyrtle distribution in our study area based on our best performing endmember-models. Our vegetation maps were largely consistent with our expectations of subgenera distribution based on our knowledge of the region.     

Pattern recognition methods in human-assisted reproduction

The aim of this paper is to present a method which allows to select, more accurately than it is currently done, embryos, at the desired stage of development, which will lead to births, although perhaps with some slight differences in precision, that are deemed to be important. Thus we show that embryos, pronuclei and oocytes consist at least of two types: those suitable for procreation and those not suitable and both types can be recognised easily by an automatic procedure at whatever stage is desired. From their digitilised images before transfer, specific characteristics are formulated automatically and through a particular pattern recognition algorithm the specimen is recognised as belonging to one of two groups with a high precision. The algorithm works in two stages: in the first, the images and their outcomes are used to train the algorithm and in a second stage, on the basis of the rule learnt in training, embryos, pronuclei or oocytes are classified. As multiple transfers and pregnancies occur the training must be carried out with an imprecise ‘teacher’ through a suitable algorithm which can handle this.     

On the application of model-driven engineering in data reengineering

Model-Driven Engineering (MDE) emphasizes the systematic use of models to improve software productivity and some aspects of the software quality such as maintainability or interoperability. Model-driven techniques have proven useful not only as regards developing new software applications but also the reengineering of legacy systems. Models and metamodels provide a high-level formalism with which to represent artefacts commonly manipulated in the different stages of a software evolution process (e.g., a software migration) while model transformation allows the automation of the evolution tasks to be performed. Some approaches and experiences of model-driven software reengineering have recently been presented but they have been focused on the code while data reengineering aspects have been overlooked. The objective of this work is to assess to what extent data reengineering processes could also take advantage of MDE techniques.The article starts by characterising data-reengineering in terms of the tasks involved. It then goes on to state that MDE is particularly amenable as regards addressing the tasks previously identified. We present an MDE-based approach for the reengineering of data whose purpose is to improve the quality of the logical schema in a relational data migration scenario. As a proof of concept, the approach is illustrated for two common problems in data re-engineering: undeclared foreign keys and disabled constraints. This approach is organised following the three stages of a software reengineering process: reverse engineering, restructuring and forward engineering. We show how each stage is implemented by means of model transformation chains. A running example is used to illustrate each stage of the process throughout the article. The approach is validated with a real widely-used database. An assessment of the application of MDE in each stage is then presented, and we conclude by identifying the main benefits and drawbacks of using MDE in data reengineering.     

Spectral discrimination of vegetation types in a coastal wetland

Remote sensing is an important tool for mapping and monitoring vegetation. Advances in sensor technology continually improve the information content of imagery for airborne, as well as space-borne, systems. This paper investigates whether vegetation associations can be differentiated using hyperspectral reflectance in the visible to shortwave infrared spectral range, and how well species can be separated based on their spectra. For this purpose, the field reflectance spectra of 27 saltmarsh vegetation types of the Dutch Waddenzee wetland were analysed in three steps. Prior to analysis, the spectra were smoothed with an innovative wavelet approach.In the first stage of the analysis, the reflectance spectra of the vegetation types were tested for differences between type classes. It was found that the reflectance spectra of saltmarsh vegetation types are statistically significantly different for various spectral regions.Secondly, it was tested whether this statistical difference could be enhanced by using continuum removal as a normalisation technique. For vegetation spectra, continuum removal improves the statistical difference between vegetation types in the visible spectrum, but weakens the statistical difference of the spectra in the near-infrared and shortwave infrared part of the spectrum.Thirdly, after statistical differences were found, it was determined how distant in spectral space the vegetation type classes were from each other, using the Bhattacharyya (BH) and the Jeffries-Matusita (JM) distance measures. We selected six wavelengths for this, based on the statistical analysis of the first step. The potential of correct classification of the saltmarsh vegetation types using hyperspectral remote sensing is predicted by these distance measures.It is concluded that the reflectance of vegetation types is statistically different. With high quality radiometric calibration of hyperspectral imagery, it is anticipated that vegetation species may be identified from imagery using spectral libraries that were measured in the field during the time of image acquisition.     

Usefulness of spectral reflectance indices as durum wheat yield predictors under contrasting Mediterranean conditions

Early prediction of crop yield can be an important tool for identifying promising genotypes in breeding programmes. To assess whether measurements of canopy reflectance at given stages of development could be used for yield forecasting and to identify the most appropriate indices, locations and growth stages for durum wheat yield assessment, nine field experiments, each including 20 or 25 durum wheat (Triticum turgidum L. var durum) genotypes, were carried out under a wide range of Mediterranean conditions. Canopy reflectance was recorded with a portable field spectroradiometer at several times from booting to physiological maturity, and nine indices were further derived. Grain yield was measured at harvesting. The results indicated that milk-grain stage was the most appropriate developmental stage for yield assessment. However, some indices were also sensitive to yield variations when determined at anthesis or even heading or booting. The capacity of spectral reflectance indices to forecast grain yield increased on locations that allowed genotypes to express their yield potentiality. Reflectance at 550?nm (R550), water index (WI), photochemical reflectance index (PRI), structural independent pigment index (SIPI), normalized difference vegetation index (NDVI) and simple ratio (SR) explained jointly a 95.7% of yield variability when all the experiments were analysed together, 92% being explained by R550. When regression analyses were carried out separately for each experiment, spectral reflectance indices explained from 17.3% to 65.2% of total variation in yield, and the indices that best explained differences in yield were experiment-dependent. Our data suggest that reflectance at 680?nm (R680), WI and SR may be suitable estimators of durum wheat grain yield under Mediterranean conditions, when determined at milk-grain stage.     

PIPE (Pipelined Image-Processing Engine)

The Sensory-Interactive Robotics Group of the National Bureau of Standards' Industrial Systems Division is designing and constructing an experimental multistage pipelined image-processing device for research in machine vision. The device can acquire images from a variety of sources, such as analog or digital television cameras, ranging devices, and conformal mapping arrays. It can process sequences of images in real time, through a serial pipeline of local operations, under the control of an external device. Its output can be presented to such devices as monitors, robot vision systems, iconic-to-symbolic mapping devices, and image-processing computers. In addition to a forward flow of images through successive stages of operations in the pipeline, other paths between the stages of the device permit recursive operations within a single stage, and feedback of the results of operations from a stage to the preceding stage. This architecture facilitates a variety of relaxation operations, interactions of images over time, and other interesting functions. Numerous operations are supported, including arithmetic and Boolean neighborhood operations on images within each stage, and between-stage operations on each pixel such as thresholding, Boolean and arithmetic operations, functional mappings, and a variety of functions for combining pixel data converging via the multiple image paths. The device can also be used to implement several alternative processing modes. Some operate within each stage, for example, to control edge effects or to implement “MIMD” operations specific to regions of interest defined by the host device. Others operate between stages, for example, to support variable-resolution pyramids.     

The effects of bark beetle stress on the foliar spectral reflectance of lodgepole pine

Abstract

Abstract. This paper reports the results of a study of the foliar reflectance properties of lodgepole pine trees under mountain pine beetle attack which aimed to determine the sequence of changes which occur. The motivation for this study was to identify spectral regions showing the earliest signs of attack in order to choose spectral bands for airborne linear array sensors. Normal illumination, diffuse hemispherical reflectance spectra were obtained of needles from trees showing varying degrees of stress from beetle attack and for needles from unattacked trees for comparison. The more pronounced changes in the spectra were interpreted visually and compared to changes reported by other authors. The more subtle changes were studied by analysing variance methods. Three spectral bands (the green peak, red edge and near-infrared (NIR) shoulder regions) have been identified as most promising for detecting early effects of bark beetle attack. Three additional bands (the blue, red and NIR plateau regions) are identified as reference bands for calibration and comparison purposes. The difference between foliar reflectance of attacked and unattacked trees is found to be most significant in the visible and red edge regions for current foliage and in the NIR for previous foliage. A red edge red shift was observed in the spectra of current foliage from attacked trees, in contrast to red edge blue shifts associated with stress in other studies. The observed sequence of subtle to more pronounced changes cannot be explained qualitatively using current knowledge of the plant pigment and anatomical changes which occur at the cellular and needle levels in stressed conifers. Further study of the detailed changes in pigments anel cellular and foliar anatomy is recommended, both to elucidate on the cause-and-effect relationships which occur, and to indicate the extent to which this paper's findings can be generalized. The findings suggest that multispectral linear array airborne scanners may be able to detect stressed conifers long before the red attack stage, but further investigation is required to determine whether the differences at foliar level between attacked and unattacked trees can also be detected at the whole-tree level, and whether confusion with other ground cover types can be avoided.     

Reflectance assessment of seasonal and annual changes in biomass and CO2 uptake of a Mediterranean shrubland submitted to experimental warming and drought

We aimed to evaluate how the remote sensing vegetation indices NDVI and PRI responded to seasonal and annual changes in an early successional stage Mediterranean coastal shrubland canopy that was submitted to experimental warming and drought simulating predicted climate change for the next decades. These conditions were obtained by using a new non-intrusive methodological approach that increases the temperature and prolongs the drought period by using roofs that automatically cover the vegetation after the sunset or when it rains. On average, warming increased air temperature by 0.7 °C and soil temperature by 1.6 °C, and the drought treatment reduced soil moisture by 22%. We measured spectral reflectance at the canopy level and at the individual plant level seasonally during 4 years. Shrubland NDVI tracked the community development and activity. In control and warming treatments, NDVI increased with the years while it did not change in the drought treatment. There was a good relationship between NDVI and both community and individual plant biomass. NDVI also decreased in summer seasons when some species dry or decolour. The NDVI of E. multiflora plant individuals was lower in autumn and winter than in the other seasons, likely because of flowering. Shrubland PRI decreased only in winter, similarly to the PRI of the most dominant species, G. alypum. At this community scale, NDVI was better related than PRI to photosynthetic activity, probably because photosynthetic fluxes followed canopy seasonal greening in this complex canopy, which includes brevideciduous, annual and evergreen species and variable morphologies and canopy coverage. PRI followed the seasonal variations in photosynthetic rates in E. multiflora and detected the decreased photosynthetic rates of drought treatment. However, PRI did not track the photosynthetic rates of G. alypum plants which have lower LAIs than E. multiflora. In this community, which is in its early successional stages, NDVI was able to track biomass, and indirectly, CO₂ uptake changes, likely because LAI values did not saturate NDVI. Thus, NDVI appears as a valid tool for remote tracking of this community development. PRI was less adequate for photosynthetic assessment of this community especially for its lower LAI canopies. PRI usefulness was also species-dependent and could also be affected by flowering. These results will help to improve the interpretation of remote sensing information on the structure and physiological status of these Mediterranean shrublands, and to gain better insight on ecological and environmental controls on their ecosystem carbon dioxide exchange. They also show the possibility of assessing the impacts of climate change on shrubland communities.     

Enhancing the Tracking of Partials for the Sinusoidal Modeling of Polyphonic Sounds

This paper addresses the problem of tracking partials, i.e., determining the evolution over time of the parameters of a given number of sinusoids with respect to the analyzed audio stream. We first show that the minimal frequency difference heuristic generally used to identify continuities between local maxima of successive short-time spectra can be successfully generalized using the linear prediction formalism to handle modulated sounds such as musical tones with vibrato. The spectral properties of the evolutions in time of the parameters of the partials are next studied to ensure that the parameters of the partials effectively satisfy the slow time-varying constraint of the sinusoidal model. These two improvements are combined in a new algorithm designed for the sinusoidal modeling of polyphonic sounds. The comparative tests show that onsets/offsets of sinusoids as well as closely spaced sinusoids are better identified and stochastic components are better avoided.     

Random selection-based adaptive saliency-weighted RXD anomaly detection for hyperspectral imagery

With recent advances in hyperspectral imaging sensors, subtle and concealed targets that cannot be detected by multispectral imagery can be identified. The most widely used anomaly detection method is based on the Reed–Xiaoli (RX) algorithm. This unsupervised technique is preferable to supervised methods because it requires no a priori information for target detection. However, two major problems limit the performance of the RX detector (RXD). First, the background covariance matrix cannot be properly modelled because the complex background contains anomalous pixels and the images contain noise. Second, most RX-like methods use spectral information provided by data samples but ignore the spatial information of local pixels. Based on this observation, this article extends the concept of the weighted RX to develop a new approach called an adaptive saliency-weighted RXD (ASW-RXD) approach that integrates spectral and spatial image information into an RXD to improve anomaly detection performance at the pixel level. We recast the background covariance matrix and the mean vector of the RX function by multiplying them by a joint weight that in fuses spectral and local spatial information into each pixel. To better estimate the purity of the background, pixels are randomly selected from the image to represent background statistics. Experiments on two hyperspectral images showed that the proposed random selection-based ASW RXD (RSASW-RXD) approach can detect anomalies of various sizes, ranging from a few pixels to the sub-pixel level. It also yielded good performance compared with other benchmark methods.