As the common feature of image, texture analysis is an important research method of CBIR, but most texture image retrieval systems are still incapable of providing retrieval result with high retrieval accuracy and less computational complexity. To address this problem, we compared the retrieval results of DT-CWT and the traditional method for extracting texture-GLCM (Gray- Level Co-occurrence Matrix), then propose a novel approach for texture image retrieval by using dual-tree complex wavelet transform (DT-CWT). It gives texture information oriented in six different directions. Experimental results indicate that the improvement in the precision of retrieval. Six different distance metrics are tested for their performance of image retrieval. We present a new Euclidean distance for images, which we call Image Euclidean Distance (IMED).Unlike the traditional Euclidean distance, IMED takes into account the spatial relationships of pixels. It is found that Canberra and Euclidean distance measures perform well for image retrieval, and achieved the higher retrieval performance.

The combined optimization problem of distribution network design and inventory decisions for chain retail enterprise under demand uncertainty is studied.Under the restriction of service levels at distribution centers and retail stores,two-echelon inventory system analysis is detailedly analyzed and a comprehensive cost optimization model for the distribution system can be established by considering transportation cost,facility location cost.Taking minimal system cost as its objective,the optimal solution of distribution network and system inventory strategy such as safety coefficient and the order cycle of retailer can be derived simultaneously through genetic algorithm(GA).The simulation results show that the model and algorithm are effective and efficient.Finally,the effects of demand,transportation distance and location cost parameters on the total system cost are given by simulation,which provides an important decision support for distribution network design and inventory control.

An image auto-recognition algorithm for determining the fusibility of coal ash is proposed in this paper.Accord- ing to the theory of pattern recognition and computer vision,first images are gotten from the CCD photography.After be- ing preproeessed,image segmentation is applied by using threshold iteration algorithm and Otsu segmentation algorithm. Then it is possible to extract the height,width and area as the geometric features exactly after locating the object region, which is used to recognize the four feature temperature images.The experiment shows that this algorithm is both easy and effective,and it can also meet the requirements of real-time and accuracy.

This paper mainly presents two approaches for image retrieval. There is some faintness in color locating in quantification boundary when image color is quantized. The membership function in fuzzy set theory can describe the fuzzy transition preferably. In this paper, the membership function in HSI color space is adopted to solve the problem above. At the same time, spatial information is integrated with color feature to describe the image content more truly. Due to the importance of texture feature, gray concurrence matrix is extracted. In the second part of this paper, retrieval scheme combining four image feature descriptors is presented, and then follows fuzzy hamming distance (FHD) which is used as a fuzzy similarity measure. We have performed an experiment on a 1000 images database and our results show higher retrieval accuracy.

This paper analyzes the feature and taxonomy of Web mining. Technologies and tools related to Web mining are presented. Considering lack of uniform and effective model in current Web usage mining, a Web usage mining architecture is presented and the processing of Web usage mining is analyzed in detail. The experimental results on Weblog data from an university Web server verify the effectiveness of the proposed Web usage mining architecture.