基于卷积神经网络的表情识别

在传统静态表情识别研究基础上,提出一种简单的人脸裁剪方法,再用浅层卷积神经网络进一步提取特征并进行表情识别。以CK+和JAFFE为实验数据集,进行预处理效果对比实验、数据增强实验、单种表情识别实验和跨数据集六分类实验。结果表明,针对数据量较少的情况,提出的表情识别方法效果明显且鲁棒性更优。     

Genome-wide effect of tetracycline,doxycycline and 4-epidoxycycline on gene expression in Saccharomyces cerevisiae

Tetracycline (Tet) and derivative chemicals (e.g., doxycycline or Dox) have gained widespread recognition for their antibiotic properties since their introduction in the late 1970s, but recent work with these chemicals in the lab has shifted to include multiple techniques in all genetic model systems for the precise control of gene expression. The most widely used Tet-modulated methodology is the Tet-On/Tet-Off gene expression system. Tet is generally considered to have effects specific to bacteria; therefore, it should have few off-target effects when used in eukaryotic systems, and a previous study in the yeast Saccharomyces cerevisiae found that Dox had no effect on genome-wide gene expression as measured by microarray. In contrast, another study found that the use of Dox in common cell lines and several model organisms led to mitonuclear protein imbalance, suggesting an inhibitory role of Dox in eukaryotic mitochondria. Recently, a new Dox derivative, 4-epidoxycycline (4-ED) was developed that was shown to have less off-target consequences on mitochondrial health. To determine the best tetracycline family chemical to use for gene expression control in S. cerevisiae, we performed RNA sequencing (RNA-seq) on yeast grown on standard medium compared with growth on media supplemented with Tet, Dox or 4-ED. We found each caused dozens of genes to change expression, with Dox eliciting the greatest expression responses, suggesting that the specific tetracycline used in experiments should be tailored to the specific gene(s) of interest when using the Tet-On/Tet-Off system to reduce the consequences of confounding off-target responses.     

Bacteriocins as food preservatives: Challenges and emerging horizons

ABSTRACT

The increasing demand for fresh-like food products and the potential health hazards of chemically preserved and processed food products have led to the advent of alternative technologies for the preservation and maintenance of the freshness of the food products. One such preservation strategy is the usage of bacteriocins or bacteriocins producing starter cultures for the preservation of the intended food matrixes. Bacteriocins are ribosomally synthesized smaller polypeptide molecules that exert antagonistic activity against closely related and unrelated group of bacteria. This review is aimed at bringing to lime light the various class of bacteriocins mainly from gram positive bacteria. The desirable characteristics of the bacteriocins which earn them a place in food preservation technology, the success story of the same in various food systems, the various challenges and the strategies employed to put them to work efficiently in various food systems has been discussed in this review. From the industrial point of view various aspects like the improvement of the producer strains, downstream processing and purification of the bacteriocins and recent trends in engineered bacteriocins has also been briefly discussed in this review.     

一种广义不可分支持向量机算法

针对标准的C-SVM(C-support vector machine)算法在处理很多实际分类问题时，对识别错误代价损失差异很大的极端情况表现出的局限性，提出一种通用的 广义支持向量机算法。根据识别错误后所付出的代价，可以把最优分类面向代价损失低的一方进行推移，留给代价损失高的一方更大的空间，提高其识别率，从而减小识别错误后带来的代价损失。该方法进一步提高了标准C SVM的适用性以及样本的正确识别率，将新算法应用到高分辨雷达距离像的识别中，实验证明,广义C-SVM能取得比传统C-SVM更好的识别效果。     

Rapidly finding CAD features using database optimization

Automatic feature recognition aids downstream processes such as engineering analysis and manufacturing planning. Not all features can be defined in advance; a declarative approach allows engineers to specify new features without having to design algorithms to find them. Naive translation of declarations leads to executable algorithms with high time complexity. Database queries are also expressed declaratively; there is a large literature on optimizing query plans for efficient execution of database queries. Our earlier work investigated applying such technology to feature recognition, using a testbed interfacing a database system (SQLite) to a CAD modeler (CADfix). Feature declarations were translated into SQL queries which are then executed.The current paper extends this approach, using the PostgreSQL database, and provides several new insights: (i) query optimization works quite differently in these two databases, (ii) with care, an approach to query translation can be devised that works well for both databases, and (iii) when finding various simple common features, linear time performance can be achieved with respect to model size, with acceptable times for real industrial models. Further results also show how (i) lazy evaluation can be used to reduce the work performed by the CAD modeler, and (ii) estimating the time taken to compute various geometric operations can further improve the query plan. Experimental results are presented to validate our main conclusions.     

Induction of Oxidative Stress and Antioxidative Mechanisms in Arabidopsis thaliana after Uranium Exposure at pH 7.5

To evaluate the environmental impact of uranium (U) contamination, it is important to investigate the effects of U at ecologically relevant conditions. Since U speciation, and hence its toxicity, strongly depends on environmental pH, the present study aimed to investigate dose-dependent effects of U at pH 7.5. Arabidopsis thaliana plants (Mouse-ear Cress) were exposed for three days to different U concentrations at pH 7.5. In the roots, the increased capacities of ascorbate peroxidase and glutathione reductase indicate an important role for the ascorbate-glutathione cycle during U-induced stress. However, a significant decrease in the ascorbate redox state was observed after exposure to 75 and 100 µM U, indicating that those roots are severely stressed. In accordance with the roots, the ascorbate-glutathione cycle plays an important role in the antioxidative defence systems in A. thaliana leaves exposed to U at pH 7.5 as the ascorbate and glutathione biosynthesis were upregulated. In addition, small inductions of enzymes of the antioxidative defence system were observed at lower U concentrations to counteract the U-induced stress. However, at higher U concentrations it seems that the antioxidative defence system of the leaves collapses as reductions in enzyme activities and gene expression levels were observed.     

Human action recognition in H.264/AVC compressed domain using meta-cognitive radial basis function network

In this paper, we propose a H.264/AVC compressed domain human action recognition system with projection based metacognitive learning classifier (PBL-McRBFN). The features are extracted from the quantization parameters and the motion vectors of the compressed video stream for a time window and used as input to the classifier. Since compressed domain analysis is done with noisy, sparse compression parameters, it is a huge challenge to achieve performance comparable to pixel domain analysis. On the positive side, compressed domain allows rapid analysis of videos compared to pixel level analysis. The classification results are analyzed for different values of Group of Pictures (GOP) parameter, time window including full videos. The functional relationship between the features and action labels are established using PBL-McRBFN with a cognitive and meta-cognitive component. The cognitive component is a radial basis function, while the meta-cognitive component employs self-regulation to achieve better performance in subject independent action recognition task. The proposed approach is faster and shows comparable performance with respect to the state-of-the-art pixel domain counterparts. It employs partial decoding, which rules out the complexity of full decoding, and minimizes computational load and memory usage. This results in reduced hardware utilization and increased speed of classification. The results are compared with two benchmark datasets and show more than 90% accuracy using the PBL-McRBFN. The performance for various GOP parameters and group of frames are obtained with twenty random trials and compared with other well-known classifiers in machine learning literature.