Applied Mechanics and Materials Vols. 303-306

Abstract: This paper, by using the short CURE clustering algorithm and image SIFT identification method, the establishment of a kind of image semantic clustering fusion model (image text clustering fusion model, referred to as ITCFM). The model is based on component method, the original image components, original text member, image clustering member, text clustering components, clustering fusion member five parts. In ITCM model for image semantic clustering characteristics on the basis of the description and extraction. The experimental results show that ITCM model can effectively to image to describe the high-level semantic, the image retrieval effect is good, and have stable retrieval performance.

Abstract: The method to recognize facial expression in images quickly has been widely applied in many areas, but there is no way to take into account the accuracy and speed of recognition. This paper presents a method based on the weight calculation for fast expression recognition. It first calculates facial features based on active shape model, and then matches the features to the standard expression vector got by large number of image training and recognizes the expression in the static image. The experiments show that the system can provide fast, universal expression recognizing function under the premise of guaranteeing an ideal accuracy.

Abstract: This paper introduces a traditional method of concept extraction. Considered some defects which this method will miss some concepts having synonyms and the relationship of “is-a”. It gives the improved algorithm to extract them. The result of the experiment shows the feasibility of this method.

Abstract: In this information society, the traditional method of ontology construction can’t meet the demand for researchers. Because it’s a difficult task to build ontology by hands. The research of automatical construction of ontology is still in the primary stage, especially in the law field. Considering the above condition, this paper gives a Research on Ontology Construction Automatically in the Law Field; we put forward the method of using LLR formula in the step of term abstraction to improve the precision and applying K-means clustering algorithm to obtain better relations of concepts hierarchy. Compared with the approach of constructing ontology by hands, it gets better effect on the law ontology.

Abstract: Traditional cotton and bast fiber detection using artificial methods. Image processing techniques have been applied to the fiber detection and improve the automation and recognition efficiency of the detection. Image segmentation is the basic one of the steps for fiber identification. This paper using gradient edge detection method to segment the cotton and bast fiber longitudinal morphological image, and using morphological reconstruction operation method to the cross sectional fiber image. Both two kinds of images can be segmented by mathematical morphology method.

Abstract: Index investing is an important issue for researchers and practitioners. This paper proposes an index portfolio optimization model for index investing via employing CSI 300 as underlying index. Firstly, a self-organizing neural network clustering model is constructed to complete the stock clustering based on stock trend which regards stock price as input. The index portfolio optimization model is proposed to determine the optimal investment proportion of each cluster sampling and achieve the minimum tracking error. The constraint BP algorithm is improved to benefit the optimization calculation of stock weights. Empirical results show that our approach achieves smaller tracking error and better index tracking effect than the random sampling.

Abstract: In practical problems, there are usually no clear counterparts as reference to evaluate restoration results. So no-reference blur assessment is very important and necessary. In this paper, we proposed an objective measure named as Edge Factor (EF) to appraise image blurring. The fundamental rationale was that blurring effect was much more perceptible in edge transition zones. The pixel number of edge transition zones would decrease when blurring occured. We defined the pixel number ratio of the edge transition zones to the whole image as EF. Experimental results show the monotonic consistency of EF and RMS. The proposed method is further compared with some common edge detection algorithms to demonstrate the effectiveness of combining point-based entropy with Pulse Coupled Neural Network.

Abstract: υυυLet G be a simple graph, k be a positive integer, f be a mapping from V(G)∪E(G) to {1,2,...,k} . If ∀uv∈E(G) , we have f(u)≠f(v) , f(u)≠f(uv),f(v)≠f(uv) , C(u)≠C(v), where C(u)={f(u)}∪{f(uv)|uv∈E(G)}. Then f is called the adjacent vertex distinguishing E-total coloring of G. The number is called the adjacent vertex –distinguishing E-total chromatic number of χSubscript text(G)=min{k|G has a k-AVDETC} . The adjacent vertex distinguishing E-total chromatic numbers of the multiple join graph of wheel and complete graph are obtained in this paper.

Abstract: Naïve Bayes classification algorithm based on validity (NBCABV) optimizes the training data by eliminating the noise samples of training data with validity to improve the effect of classification, while it ignores the associations of properties. In consideration of the associations of properties, an improved method that is classification algorithm for Naïve Bayes based on validity and correlation (CANBBVC) is proposed to delete more noise samples with validity and correlation, thus resulting in better classification performance. Experimental results show this model has higher classification accuracy comparing the one based on validity solely.

Abstract: Schrödinger operator; Weighted BMO spaces; Reverse Hölder inequality; Commutator Abstract. In this paper, the Schrödinger operator on n dimensions Euclid space with the non-zero, nonnegative potential function satisfying the reverse Hölder inequality is considered. The weighted boundedness of the commutators composed of several Riesz transforms associated with the Schrödinger operator and weighted BMO function on weighted Lebesgue integral spaces are obtained, for some weighted function.