Reference spectra to classify Amazon water types

Reference spectra extracted from spectral libraries can distinguish different water types in images when associated with limnological information. In this study, we compiled available databases into a single spectral library, using field water reflectance spectra and limnological data collected by different researchers and campaigns in the Amazonian region. By using an iterative clustering procedure based on the combination of reflectance and optically active components (OACs), reference spectra representative of the major Amazonian water types were defined from this library. Differences between the resultant limnological classes were also evaluated by paired t-tests at significance level 0.05. Finally, reference spectra were tested for Spectral Angle Mapper (SAM) classification of waters in Hyperion/Earth Observing-One (EO-1) and Medium Resolution Imaging Spectrometer (MERIS)/Environment Satellite (Envisat) images acquired simultaneously as the field campaigns. Results showed highly variable concentrations of OACs due to the complexity of the Amazonian aquatic environments. Ten classes were defined to represent this complexity, broadly grouped into four limnological characteristics: clear waters with low concentrations of OACs (class 1); black waters rich in dissolved organic carbon (DOC) (class 2); waters with large concentrations of inorganic suspended solids (ISSs) (classes 3–7); and waters dominated by chlorophyll-a (chl-a) (classes 8–10). Using the ten reference spectra, SAM classification of the field water curves produced an overall accuracy of 86% with the highest values observed for classes 3, 4, 6 and 7 and the lowest accuracy for classes 1 and 2. The results of paired t-tests confirmed the class differences based on the concentrations of OACs. SAM classification of the Hyperion and MERIS images using ground truth information resulted in overall classification accuracies of 48% and 67%, respectively, with the highest errors associated with specific portions of the scenes that were not adequately represented in the spectral library.     

A novel multi-parameter support vector machine for image classification

The support vector machine (SVM) classification algorithm has received increasing attention in recent years in remote sensing for land-cover classification. However, it is well known that the performance of the SVM is sensitive to the choice of parameter settings. The traditional single optimized parameter SVM (SOP-SVM) attempts to identify globally optimized parameters for multi-class land-cover classification. In this article, a novel multi-parameter SVM (MP-SVM) algorithm is proposed for image classification. It divides the training set into several subsets, which are subsequently combined. Based on these combinations, sub-classifiers are constructed using their own optimum parameters, providing votes for each pixel with which to construct the final output. The SOP-SVM and MP-SVM were tested on three pilot study sites with very high, high, and low levels of landscape complexity within the Sanjiang Plain – a typical inland wetland and freshwater ecosystem in northeast China. A high overall accuracy of 82.19% with kappa coefficient (κ) of 0.80 was achieved by the MP-SVM in the very high-complexity landscape, statistically significantly different (z-value = 3.77) from the overall accuracy of 72.50% and κ of 0.69 produced by the traditional SOP-SVM. Besides, for the moderate-complexity landscape a significant increase in accuracy was achieved (z-value = 2.44), with overall accuracy of 84.03% and κ of 0.80 compared with an overall accuracy 76.05% and κ of 0.71 for the SOP-SVM. However, for the low-complexity landscape the MP-SVM was not significantly different from the SOP-SVM (z-value = 0.80). Thus, the results suggest that the MP-SVM method is promising for application to very high and high levels of landscape complexity, differentiating complex land-cover classes that are spectrally mixed, such as marsh, bare land, and meadow.     

Using k-nn and discriminant analyses to classify rain forest types in a Landsat TM image over northern Costa Rica

Conservation and land use planning in humid tropical lowland forests urgently need accurate remote sensing techniques to distinguish among floristically different forest types. We investigated the degree to which floristically and structurally defined Costa Rican lowland rain forest types can be accurately discriminated by a non-parametric k nearest neighbors (k-nn) classifier or linear discriminant analysis. Pixel values of Landsat Thematic Mapper (TM) image and Shuttle Radar Topography Mission (SRTM) elevation model extracted from segments or from 5 × 5 pixel windows were employed in the classifications. 104 field plots were classified into three floristic and one structural type of forest (regrowth forest). Three floristically defined forest types were formed through clustering the old-growth forest plots (n = 52) by their species specific importance values. An error assessment of the image classification was conducted via cross-validation and error matrices, and overall percent accuracy and Kappa scores were used as measures of accuracy. Image classification of the four forest types did not adequately distinguish two old-growth forest classes, so they were merged into a single forest class. The resulting three forest classes were most accurately classified by the k-nn classifier using segmented image data (overall accuracy 91%). The second best method, with respect to accuracy, was the k-nn with 5 × 5 pixel windows data (89% accuracy), followed by the canonical discriminant analysis using the 5 × 5 pixel window data (86%) and the segment data (82%). We conclude the k-nn classifier can accurately distinguish floristically and structurally different rain forest types. The classification accuracies were higher for the k-nn classifier than for the canonical discriminant analysis, but the differences in Kappa scores were not statistically significant. The segmentation did not increase classification accuracy in this study.     

Disposable microfluidic chip for rapid pathogen identification with DNA microarrays

This work presents the combination and acceleration of PCR and fluorescent labelling within a disposable microfluidic chip. The utilised geometry consists of a spiral meander with 40 turns, representing a cyclic-flow PCR system. The used reaction chemistry includes Cy3-conjugated primers leading to a one-step process accelerated by cyclic-flow PCR. DNA of three different bacterial samples (Staphylococcus aureus, Escherichia coli and Pseudomonas aeruginosa) was processed and successfully amplified and labelled with detection limits down to 10² cells per reaction. The specificity of species identification was comparable to the approach of separate PCR and labelling. The overall processing time was decreased from 6 to 1.5 h. We showed that a disposable polycarbonate chip, fabricated by injection moulding is suitable for the significant acceleration of DNA microarray assays. The reaction output led to high-sensitivity bacterial identification in a short time, which is crucial for an early and targeted therapy against infectious diseases.     

Estimating boreal forest species type with airborne polarimetric synthetic aperture radar

We have applied a non-parametric classifier (k nearest neighbour) to two calibrated orthogonal passes of airborne polarimetric synthetic aperture radar (POLSAR) image data over boreal forest for the purpose of discriminating canopy tree species of predefined stands. We found that a single classifier based on a single feature space (i.e. one set of POLSAR variables for all species) was less accurate than a hierarchical two-stage classifier that used different POLSAR variables for each species. We designed a two-stage classifier that first grouped stands into broad classes: pine, spruce and deciduous, and then classified each sample within the broad classes into individual species. We found that the most effective feature spaces had two or three dimensions. The two-stage classifier attained overall accuracies of between 60% and 75%.

We provide a first use of an equivalency test applied to remote-sensing classification. We use Lloyd's test of equivalency to find equivalent classifiers and thus infer informative POLSAR variables. The POLSAR variables that were most informative varied between the two passes and between the various elements of the hierarchical classifier. For the initial three-class classifier the most informative POLSAR variables were the two circular polarization ratios, several of Touzi's Stokes vector variables, HHVV coherence, several texture measures such as the variance of several scattering coefficients and the order parameter of the K-distribution and characteristics of the polarization signature pedestal. These results demonstrate that C-band POLSAR has great potential for mapping boreal forest cover either on its own or in concert with other geospatial data.     

Construction of the Algorithm Graph by Autotracing Method

An autotracing approach to the problem of the algorithm graph construction based on the possibility of overloading operators in the C++ language is suggested. The basic idea of the approach is to replace the standard double type by a special class number, which supports basic operations on numbers (arithmetic, input/output) and constructs the graph in a background mode. A class graph is responsible for general control of the graph construction process. Classes vector and matrix are introduced to support the construction of the graph for vector and matrix operations. The library of classes developed is a powerful and flexible tool for analysis of the algorithm graphs.     

小波包信息熵特征矢量光谱角高光谱影像分类

目的 针对高光谱数据波段多、数据存在冗余的特点,将小波包信息熵特征引入到高光谱遥感分类中。方法 通过对光谱曲线进行小波包分解变换,定义了小波包信息熵特征矢量光谱角分类方法(WPE-SAM),基于USGS光谱库中4种矿物光谱数据的分析表明,WPE-SAM可增大类间地物的可区分性。在特征矢量空间对Salina高光谱影像进行分类计算,并讨论了小波包最佳分解层的确定,分析了WPE-SAM与光谱角制图(SAM)方法的分类精度。结果 Salina数据实例计算表明：小波包信息熵矢量能较好地描述原始光谱特征,WPE-SAM分类方法可行,总体分类精度(OA)由SAM的78.62%提高到WPE-SAM的78.66%,Kappa系数由0.769 0增加到0.769 5,平均分类精度(AA)由83.14%提高到84.18%。此外,通过Pavia数据验证了WPE-SAM分类方法具有较强的普适性。结论 小波包信息熵特征可较好地表示原始光谱波峰、波谷等特征信息,定义的小波包信息熵特征矢量光谱角分类方法(WPE-SAM)可增大类间地物可区分性,有利于分类。实验结果表明,WPE-SAM分类方法技术可行,总体精度及Kappa系数较SAM有一定的提高,且有较强的普适性。但WPE-SAM方法精度与效率有待进一步提高。     

基于节点向量表达的复杂网络社团划分算法

社团结构划分对复杂网络研究在理论和实践上都非常重要.借鉴分布式词向量理论,提出一种基于节点向量表达的复杂网络社团划分方法（CDNEV）.为了构建网络节点的分布式向量,提出启发式随机游走模型.利用节点启发式随机游走得到的节点序列作为上下文,采用SkipGram模型学习节点的分布式向量.选择局部度中心节点作为K-Means算法的聚类中心点,然后用K-Means算法进行聚类,最终得到社团结构.在真实和模拟两种网络上做了丰富的实验,与主流的全局社团划分算法和局部社团划分算法作了比较.在真实网络上CDNEV算法的F1指标比其他算法平均提高19%;在模拟网络上,F1指标则可以提高15%.实验结果表明,相对其他算法,CDNEV算法的精度和效率都较高.     

Linguistic and architectural requirements for personalized digital libraries

information tokens as supplied by information providers on the net; value is added to such tokens by individually constructing information artifacts over them with the goal of information consumer satisfaction; (2) the services required for artifact construction and use – on the information level as well as on the level of the software artifacts required for these processes – rely heavily on powerful binding environments for multi-medial, persistent and networked information; (3) the processes of artifact construction and use are in themselves valuable sources of information about artifacts; for the exploitation of such process information, digital libraries employ advanced tracing environments. We derive linguistic and architectural requirements for digital libraries from these above essentials. On the language level we concentrate on generalized requirements for the typing, binding and scoping of library entities and services. On the system level we discuss architectural requirements in terms of orthogonal persistence, open extensibility, platform independence, mobility and reflection. We present Tycoon, a polymorphic, higher-order language and its system, and demonstrate its potential for digital libraries. We evaluate Tycoon's rich conceptual basis (data, functions and threads), library-based extensibility, powerful binding mechanisms, its orthogonal persistence and its capability of network-wide data, code and thread migration. We conclude by referring to an interdisciplinary digital library project in Art History Research based on icons, texts and data. Here, Tycoon effectively supports the process of individually customizing and scaling library services thus generalizing the notion of a query language into that of a persistent personal reference library. Received: 1 August 1996 / Accepted: 15 November 1996     

FFTs in external or hierarchical memory

Conventional algorithms for computing large one-dimensional fast Fourier transforms (FFTs), even those algorithms recently developed for vector and parallel computers, are largely unsuitable for systems with external or hierarchical memory. The principal reason for this is the fact that most FFT algorithms require at least m complete passes through the data set to compute a 2 ^m -point FFT. This paper describes some advanced techniques for computing an ordered FFT on a computer with external or hierarchical memory. These algorithms (1) require as few as two passes through the external data set, (2) employ strictly unit stride, long vector transfers between main memory and external storage, (3) require only a modest amount of scratch space in main memory, and (4) are well suited for vector and parallel computation.Performance figures are included for implementations of some of these algorithms on Cray supercomputers. Of interest is the fact that a main memory version outperforms the current Cray library FFT routines on the CRAY-2, the CRAY X-MP, and the CRAY Y-MP systems. Using all eight processors on the CRAY Y-MP, this main memory routine runs at nearly two gigaflops.A condensed version of this paper previously appeared in the Proceedings of Supercomputing '89.     

New semi-automated mapping of asbestos cement roofs using rule-based object-based image analysis and Taguchi optimization technique from WorldView-2 images

The use of asbestos cement (AC) roofing materials is a significant concern because of their deleterious effects on human health and the environment. The main objective of this study was to map AC roofs from WorldView-2 (WV-2) images using object-based image analysis (OBIA). A robust Taguchi optimization technique was used to optimize segmentation parameters for WV-2 images in heterogeneous urban areas. In this research, two subsets of WV-2 satellite image sets were utilized to map AC roofs. Rule-based OBIA framework was developed on the first study area. Different supervised OBIA classifiers, such as Bayes, k-nearest neighbour (k-NN), support vector machine (SVM), and random forest (RF), were tested on the first image of the study areas to evaluate the performance of a rule-based classifier. Results of the supervised classifiers showed confusion between AC roof class and some urban features, with overall accuracies of 72.21%, 77%, 81.75%, and 82.02% for Bayes, k-NN, SVM, and RF, respectively. To assess the transferability of the proposed method, the adopted classification framework was applied to larger subsets of WV-2 of the second study area. The results of the proposed approach showed outstanding performance, with overall accuracies of 93.10% and 90.74% for the first and second classified images, respectively. The McNemar test emphasized the statistical reliability of rule-based result (in the first site) compared with supervised classification results. Therefore, the proposed framework of using rule-based classification and Taguchi optimization technique provide an efficient and expeditious approach to mapping and monitoring the presence of AC roofs and help local authorities in their decision-making strategies and policies.     

A tale of two greps

Text searching programs such as the UNIX system tools grep and egrep require more than just good algorithms; they need to make efficient use of system resources such as I/O. I describe improving the I/O management in grep and egrep by using a new fast I/O library fio to replace the normal I/O library stdio. I also describe incorporating the Boyer-Moore algorithm into egrep; egrep is now typically 8–10 (for some common patterns 30–40) times faster than grep.     

Solving Finite Mixture Models: Efficient Computation in Economics Under Serial and Parallel Execution

Many economic models are completed by finding a parameter vector θ that optimizes a function f(θ), a task that can only be accomplished by iterating from a starting vector θ₀. Use of a generic iterative optimizer to carry out this task can waste enormous amounts of computation when applied to a class of problems defined here as finite mixture models. The finite mixture class is large and important in economics and eliminating wasted computations requires only limited changes to standard code. Further, the approach described here greatly increases gains from parallel execution and opens possibilities for re-writing objective functions to make further efficiency gains.Documented code that implements the algorithm described is available from the author for objectives written in C and other languages. It runs in both serial and parallel mode using the MPI library.JEL Classification: C61; C63; D58     

New Least Squares Support Vector Machines Based on Matrix Patterns

Support vector machine (SVM), as an effective method in classification problems, tries to find the optimal hyperplane that maximizes the margin between two classes and can be obtained by solving a constrained optimization criterion using quadratic programming (QP). This QP leads to higher computational cost. Least squares support vector machine (LS-SVM), as a variant of SVM, tries to avoid the above shortcoming and obtain an analytical solution directly from solving a set of linear equations instead of QP. Both SVM and LS-SVM operate directly on patterns represented by vector, i.e., before applying SVM or LS-SVM to a pattern, any non-vector pattern such as an image has to be first vectorized into a vector pattern by some techniques like concatenation. However, some implicit structural or local contextual information may be lost in this transformation. Moreover, as the dimension d of the weight vector in SVM or LS-SVM with the linear kernel is equal to the dimension d ₁ × d ₂ of the original input pattern, as a result, the higher the dimension of a vector pattern is, the more space is needed for storing it. In this paper, inspired by the method of feature extraction directly based on matrix patterns and the advantages of LS-SVM, we propose a new classifier design method based on matrix patterns, called MatLSSVM, such that the new method can not only directly operate on original matrix patterns, but also efficiently reduce memory for the weight vector (d) from d ₁ × d ₂ to d ₁ + d ₂. However like LS-SVM, MatLSSVM inherits LS-SVM’s existence of unclassifiable regions when extended to multi-class problems. Thus with the fuzzy version of LS-SVM, a corresponding fuzzy version of MatLSSVM (MatFLSSVM) is further proposed to remove unclassifiable regions effectively for multi-class problems. Experimental results on some benchmark datasets show that the proposed method is competitive in classification performance compared to LS-SVM, fuzzy LS-SVM (FLS-SVM), more-recent MatPCA and MatFLDA. In addition, more importantly, the idea used here has a possibility of providing a novel way of constructing learning model.     

Analysis of a lidar voxel-derived vertical profile at the plot and individual tree scales for the estimation of forest canopy layer characteristics

The goal of the current study was to develop methods of estimating the height of vertical components within plantation coniferous forest using airborne discrete multiple return lidar. In the summer of 2008, airborne lidar and field data were acquired for Loblolly pine forest locations in North Carolina and Virginia, USA, which comprised a variety of stand conditions (e.g. stand age, nutrient regime, and stem density). The methods here implement both field plot-scale analysis and an automated approach for the delineation of individual tree crown (ITC) locations and horizontal extents through a marker-based region growing process applied to a lidar derived canopy height model. The estimation of vertical features was accomplished through aggregating lidar return height measurements into vertical height bins, of a given horizontal extent (plot or ITC), creating a vertical ‘stack’ of bins describing the frequency of returns by height. Once height bins were created the resulting vertical distributions were smoothed with a regression curve-line function and canopy layers were identified through the detection of local maxima and minima. Estimates from Lorey’s mean canopy height was estimated from plot-level curve-fitting with an overall accuracy of 5.9% coefficient of variation (CV) and the coefficient of determination (R²) value of 0.93. Estimates of height to the living canopy produced an overall R² value of 0.91 (11.0% CV). The presence of vertical features within the sub-canopy component of the fitted vertical function also corresponded to areas of known understory presence and absence. Estimates from ITC data were averaged to the plot level. Estimates of field Lorey’s mean canopy top height from average ITC data produced an R² value of 0.96 (7.9% CV). Average ITC estimates of height to the living canopy produced the closest correspondence to the field data, producing an R² value of 0.97 (6.2% CV). These results were similar to estimates produced by a statistical regression method, where R² values were 0.99 (2.4% CV) and 0.98 (4.9% CV) for plot average top canopy height and height to the living canopy, respectively. These results indicate that the characteristics of the dominant canopy can be estimated accurately using airborne lidar without the development of regression models, in a variety of intensively managed coniferous stand conditions.     

Inland waterbody mapping: towards improving discrimination and extraction of inland surface water features

Surface waterbodies in arid and semi-arid environments are threatened by both natural and anthropogenic pressures. Mapping the distribution of surface waterbodies is crucial for managing their dwindling quantities and quality. In this study, a fast and reliable method of water extraction has been introduced. A remote-sensing index called the simple water index (SWI) was formulated to differentiate waterbodies from vegetation class automatically, and to differentiate waterbodies from shadows or built-up areas (water-like features). Its performance was compared with the automated water extraction index (AWEI) and the modified normalized difference water index (MNDWI) on Landsat 8 Operational Land Imager (OLI) image of South Africa. The robustness of the algorithm was tested on images in Madagascar and the Democratic Republic of Congo (DRC) with different biomes. The overall accuracies and kappa coefficient (κ) were used to compare the performance of each index. The McNemar test was performed to assess the significance of the output map and the validation data set. The SWI showed the highest overall accuracy of 91.9% (κ = 0.83), whereas the AWEI and MNDWI yielded overall accuracies of 83.8% (κ = 0.65) and 78.4% (κ = 0.53), respectively. The McNemar test showed that there was no significant difference between the SWI map (p = 0.248), whereas both AWEI and MNDWI maps were significantly different from the validation data set at p = 0.041 and p = 0.013, respectively. The SWI approach reduces the thresholding problem by 50% over the conventional MNDWI and AWEI. It is expected that the SWI will also be useful for the accurate quantification of waterbodies for large areas.     

A hybrid data fusion approach to the hierarchical mapping of land-cover types in the urban core

It is difficult to map land covers in the urban core due to the close proximity of high-rise buildings. This difficulty is overcome with a proposed hybrid, the hierarchical method via fusing PAN-sharpened WorldView-2 imagery with light detection and ranging (lidar) data for central Auckland, New Zealand, in two stages. After all features were categorized into ‘ground’ and ‘above-ground’ using lidar data, ground features were classified from the satellite data using the object-oriented method. Above-ground covers were grouped into four types from lidar-derived digital surface model (nDSM) based on rules. Ground and above-ground features were classified at an accuracy of 94.1% (kappa coefficient or κ = 0.913) and 93.7% (κ = 0.873), respectively. After the two results were merged, the nine covers achieved an overall accuracy of 93.7% (κ = 0.902). This accuracy is highly comparable to those reported in the literature, but was achieved at much less computational expense and complexity owing to the hybrid workflow that optimizes the efficiency of the respective classifiers. This hybrid method of classification is robust and applicable to other scenes without modification as the required parameters are derived automatically from the data to be classified. It is also flexible in incorporating user-defined rules targeting hard-to-discriminate covers. Mapping accuracy from the fused complementary data sets was adversely affected by shadows in the satellite image and the differential acquisition time of imagery and lidar data.     

Mapping wetlands using ASTER data: a comparison between classification trees and logistic regression

This study compared a non‐parametric and a parametric model for discriminating among uplands (non‐wetlands), woody wetlands, emergent wetlands and open water. Satellite images obtained on 6 March 2005 and 16 October 2005 from the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) and geographic information system (GIS) data layers formed the input for analysis using classification and regression tree (CART^®) and multinomial logistic regression analysis. The overall accuracy of the CART model was 73.3%. The overall accuracy of the logit model was 76.7%. The accuracies were not statistically different from each other (McNemar χ ² = 1.65, p = 0.19). The CART producer's accuracy of the emergent wetlands was higher than the accuracy from the multinomial logit (57.1% vs. 40.7%), whereas woody wetlands identified by the multinomial logit model presented a producer's accuracy higher than that from the CART model (68.7% vs. 52.6%). A McNemar test between the two models and National Wetland Inventory (NWI) maps showed that their accuracies were not statistically different. Overall, these two models provided promising results, although they are not sufficiently accurate to replace current methods of wetland mapping based on feature extraction in high‐resolution orthoimagery.     

Proteomics integrated with Escherichia coli vector-based vaccines and antigen microarrays reveals the immunogenicity of a surface sialidase-like protein of Propionibacterium acnes

Proteomics is a powerful tool for the identification of proteins, which provides a basis for rational vaccine design. However, it is still a highly technical and time‐consuming task to examine a protein's immunogenicity utilizing traditional approaches. Here, we present a platform for effectively evaluating protein immunogenicity and antibody detection. A tetanus toxin C fragment (Tet‐c) was used as a representative antigen to establish this platform. A cell wall‐anchoring sialidase‐like protein (SLP) of Propionibacterium acnes was utilized to assess the efficacy of this platform. We constructed an Escherichia coli vector‐based vaccine by overexpressing Tet‐c or SLP in E. coli and utilized an intact particle of E. coli itself as a vaccine (E. coli Tet‐c or SLP vector). After ultraviolet (UV) irradiation, the E. coli vector‐based vaccines were administered intranasally into imprinting control region mice without adding exogenous adjuvants. For antibody detection, we fabricated antigen microarrays by printing with purified recombinant proteins including Tet‐c and SLP. Our results demonstrated that detectable antibodies were elicited in mice 6 weeks after intranasal administration of UV‐irradiated E. coli vector‐based vaccines. The antibody production of Tet‐c and SLP was significantly elevated after boosting. Notably, the platform with main benefits of using E. coli itself as a vaccine carrier provides a critical template for applied proteomics aimed at screening novel vaccine targets. In addition, the novel immunogenic SLP potentially serves as an antigen candidate for the development of vaccines targeting P. acnes‐associated diseases.