基于颜色名和OpponentSIFT特征的鳞翅目昆虫图像识别

【目的】本研究旨在探索使用先进的计算机视觉技术实现对昆虫图像的自动分类方法。【方法】通过预处理对采集的昆虫标本图像去除背景,获得昆虫图像的前景蒙板,并由蒙板确定的轮廓计算出前景图像的最小包围盒,剪切出由最小包围盒确定的前景有效区域,然后对剪切得到的图像进行特征提取。首先提取颜色名特征,把原来的RGB（Red-Green-Blue）图像的像素值映射到11种颜色名空间,其值表示RGB值属于该颜色名的概率,每个颜色名平面划分成3×3像素大小的网格,用每格的概率均值作为网格中心点的描述子,最后用空阈金字塔直方图统计的方式形成颜色名视觉词袋特征;其次提取OpponentSIFT(Opponent Scale Invariant Feature Transform)特征,首先把RGB图像变换到对立色空间,对该空间每通道提取SIFT特征,最后用空域池化和直方图统计方法形成OpponentSIFT视觉词袋。将两种词袋特征串接后得到该昆虫图像的特征向量。使用昆虫图像样本训练集提取到的特征向量训练SVM（Support Vector Machine）分类器,使用这些训练得到的分类器即可实现对鳞翅目昆虫的分类识别。【结果】该方法在包含10种576个样本的昆虫图像数据库中进行了测试,取得了100%的识别正确率。【结论】试验结果证明基于颜色名和OpponentSIFT特征可以有效实现对鳞翅目昆虫图像的识别。     

Support vector machines for learning to identify the critical positions of a protein

A method for identifying the positions in the amino acid sequence, which are critical for the catalytic activity of a protein using support vector machines (SVMs) is introduced and analysed. SVMs are supported by an efficient learning algorithm and can utilize some prior knowledge about the structure of the problem. The amino acid sequences of the variants of a protein, created by inducing mutations, along with their fitness are required as input data by the method to predict its critical positions. To investigate the performance of this algorithm, variants of the beta-lactamase enzyme were created in silico using simulations of both mutagenesis and recombination protocols. Results from literature on beta-lactamase were used to test the accuracy of this method. It was also compared with the results from a simple search algorithm. The algorithm was also shown to be able to predict critical positions that can tolerate two different amino acids and retain function.     

Reduced amino acid alphabet is sufficient to accurately recognize intrinsically disordered protein

Intrinsically disordered proteins are an important class of proteins with unique functions and properties. Here, we have applied a support vector machine (SVM) trained on naturally occurring disordered and ordered proteins to examine the contribution of various parameters (vectors) to recognizing proteins that contain disordered regions. We find that a SVM that incorporates only amino acid composition has a recognition accuracy of 87+/-2%. This result suggests that composition alone is sufficient to accurately recognize disorder. Interestingly, SVMs using reduced sets of amino acids based on chemical similarity preserve high recognition accuracy. A set as small as four retains an accuracy of 84+/-2%; this suggests that general physicochemical properties rather than specific amino acids are important factors contributing to protein disorder.     

High‐Performance Protonic Ceramic Fuel Cells with 1 µm Thick Y:Ba(Ce,Zr)O3 Electrolytes

This study demonstrates the effectiveness of using thin‐film electrolytes to enhance protonic ceramic fuel cells (PCFCs). The material tested in this study is yttrium‐doped barium cerate‐zirconate (BCZY), which is a representative electrolyte material of PCFCs. The thickness of the electrolyte membrane is as small as 1 µm and designed to minimize ohmic loss in proton transport pathways. Integration of this thin BCZY electrolyte is attempted on a multilayered anode comprised of two‐step supports with bulk nickel‐yttria stabilized zirconia cermet as a base and thin nickel‐BCZY as an anode functional layer atop the base. The compatibility of this support with the deposited thin electrolyte is able to be confirmed from the results of iterated tests. The power of the fabricated cell is greater than 1.1 W cm^?2 at 600 °C, which is a record high for PCFCs and is reproducible. In this paper, the origin of this high power is discussed and improvements that could be made to cell performance are further suggested.     

A protein structural class prediction method based on novel features

In this study, a 12-dimensional feature vector is constructed to reflect the general contents and spatial arrangements of the secondary structural elements of a given protein sequence. Among the 12 features, 6 novel features are specially designed to improve the prediction accuracies for α/β and α + β classes based on the distributions of α-helices and β-strands and the characteristics of parallel β-sheets and anti-parallel β-sheets. To evaluate our method, the jackknife cross-validating test is employed on two widely-used datasets, 25PDB and 1189 datasets with sequence similarity lower than 40% and 25%, respectively. The performance of our method outperforms the recently reported methods in most cases, and the 6 newly-designed features have significant positive effect to the prediction accuracies, especially for α/β and α + β classes.     

Effectiveness of SIMBLIGHT1 and SIMPHYT1 models for predicting Phytophthora infestans in north-eastern United States

Accurate prediction of Phytophthora infestans outbreak in a cropping season is crucial for the effective management of late blight on potato. The SIMBLIGHT1, SIMPHYT1 and modified SIMPHYT1 models were assessed for predicting late blight outbreaks relative to NOBLIGHT model by using climatic and crop data from field experiments at Presque Isle, ME, from 2004 to 2009. The dynamics of late blight infection pressures were computed by SIMPHYT3 model to assess the conduciveness of environmental conditions for late blight infection and potential for disease development. Infection pressure indicated conditions were moderately conducive for late blight development but varied across years. The SIMPHYT1 model recommended fungicide applications to commence on 11 July, 21 July, 8 July, 10 July, 7 July and 7 July for 2004, 2005, 2006, 2007, 2008 and 2009, respectively. The modified SIMPHYT1 model (US-version) recommended similar fungicide application dates for the same years with the exception of 2007. Model simulations of disease outbreak differed from actual recorded observations in untreated plots by 24–65 days. Simulation of SIMBLIGHT1 based on high and low soil moisture conditions in the field resulted in vast differences in dates of first fungicide application. Validation of the models (differences in the number of days between recommendation of fungicide treatment and late blight outbreak) indicated a better goodness of fit for the models (intervals of 6–20 days from the start of fungicide application to first disease outbreak). The NOBLIGHT model was accurate in forecasting the timing of first fungicide applications for late blight control. Significant improvements in late blight predictions could result if these models were modified to account for external sources of inoculum, by combining weather-based forecasts with spore traps, disease detection methods or complimentary systems.     

Ecological land suitability analysis through spatial indicators: An application of the Analytic Network Process technique and Ordered Weighted Average approach

Multicriteria-Spatial Decision Support Systems (MC-SDSS) are increasingly popular tools in decision-making processes and in policy making, thanks to their significant new capabilities in the use of spatial or geospatial information.Many spatial problems are complex and require the use of integrated analysis and models. The present paper illustrates the development of a MC-SDSS approach for studying the ecological connectivity of the Piedmont Region in Italy. The MC-SDSS model considers ecological and environmental spatial indicators which are combined by integrating the Multicriteria Decision Aiding (MCDA) technique named Analytic Network Process (ANP) and the Ordered Weighted Average (OWA) approach. The ANP is used for the elicitation of attribute weights while the OWA operator function is used to generate a wide range of decision alternatives for addressing uncertainty associated with interaction between multiple criteria. The usefulness of the approach is illustrated by different OWA scenarios that report the ecological connectivity index on a scale between 0 and 1. The OWA scenarios are intended to quantify the level of risk taking (i.e., optimistic, pessimistic, and neutral) and to facilitate a better understanding of patterns that emerge from decision alternatives involved in the decision-making process.The purpose of the research is to generate a final map representing the ecological connectivity index of each area in the region under analysis, to be used as a decision variable in spatial planning. In particular, by using the resulting index map as a means of analysis, it is possible to identify, for the sake of nature conservation, some critical areas needing mitigation measures. In addition, areas with high ecological connectivity values can be identified and monitoring procedures can therefore be planned. The study concludes highlighting that the applied methodology is an effective tool in providing decision support for spatial planning and sustainability assessments.     

Design considerations for ceramic resurfaced femoral head: effect of interface characteristics on failure mechanisms

Ceramic hip resurfacing may offer improved wear resistance compared to metallic components. The study is aimed at investigating the effects of stiffer ceramic components on the stress/strain-related failure mechanisms in the resurfaced femur, using three-dimensional finite element models of intact and resurfaced femurs with varying stem–bone interface conditions. Tensile stresses in the cement varied between 1 and 5 MPa. Postoperatively, 20–85% strain shielding was observed inside the resurfaced head. The variability in stem–bone interface condition strongly influenced the stresses and strains generated within the resurfaced femoral head. For full stem–bone contact, high tensile (151–158 MPa) stresses were generated at the cup–stem junction, indicating risk of fracture. Moreover, there was risk of femoral neck fracture due to elevated bone strains (0.60–0.80% strain) in the proximal femoral neck region. Stresses in the ceramic component are reduced if a frictionless gap condition exists at the stem–bone interface. High stresses, coupled with increased strain shielding in the ceramic resurfaced femur, appear to be major concerns regarding its use as an alternative material.     

Prediction of prostate cancer using hair trace element concentration and support vector machine method

A change in the normal concentration of essential trace elements in the human body might lead to major health disturbances. In this study, hair samples were collected from 115 human subject, including 55 healthy people and 60 patients with prostate cancer. The concentrations of 20 trace elements (TEs) in these samples were measured by inductively coupled plasma-mass spectrometry. A support vector machine was used to investigate the relationship between TEs and prostate cancer. It is found that, among the 20 TEs, 10 (Mg P, K, Ca, Cr, Mn, Fe. Cu, Zn, and Se) are related to the risk of prostate cancer. These 10 TEs were used to build the prediction model for prostate cancer. The model obtained can satisfactorily distinguish the healthy samples from the cancer samples. Furthermore, the cross-validation by leaving-one method proved that the prediction ability of this model reaches as high as 95.8%. It is practical to predict the risk of prostate cancer using this model in the clinics