Applied Mechanics and Materials Vols. 278-280

Abstract: In this paper, a novel 3D face recognition method is proposed from the sparse representation point of view. Under the framework of sparse representation, the recognition problem is transformed to solve the problem of minimization L0-norm. Three types of facial geometrical features are extracted to describe 3D faces. According to the extracted features, 3D face recognition is conducted by applying to the ranking strategy of Fisher linear discriminant analysis. The experiments employed BJUT-3D datasets demonstrate the effectiveness of the proposed method.

Abstract: In the process of erotic image detection, a kind of beach images is often error detected. To solve this problem, a detection model of beach image is designed, which divides the image into two parts to detect: the blue sky in the upper portion and the beach in the lower part. At the same time, a beach naked body detection model is designed to deal with the beach images with naked body, in which OpenCV is used to detect face and naked body. The experimental results show that the proposed method can detect beach image accurately, effectively reduce the error rate of normal image.

Abstract: A fuzzy algorithm of customers evaluation based on attributes reduction is presented. The evaluation from the data objects based on key attributes can reduce the data size and algorithm complexity. After Clustering analysis of customers, then the evaluation analysis will process to the clustering data. There are a lot of uncertain data of customer cluster, so the traditional method of classification and evaluation to the incomplete data is very difficult. Superposition evaluation algorithm based on fuzzy set can improve the reliability and accuracy of e-commerce customer evaluation. Evaluation of the e-commerce customer also can improve efficiency, service quality and profitability of e-commerce businesses.

Abstract: Aiming at the defects and shortages of traditional moving vehicles detection algorithms, by the analysis and comparison of the existing detection algorithms, we propose an algorithm that combined with frames with symmetric difference and background difference to detect moving vehicle in this paper. First, two different difference images by using frames with symmetric difference and background difference are gained respectively and two binary images can be gained by the appropriate threshold, then the contour of moving vehicles can be extracted by applying OR operation in the two binary images. Finally, the precise moving vehicles will be gained by mathematic morphological methods. Experimental results show the algorithm used in this paper can detect the moving vehicles better than traditional moving vehicles detection algorithms.

Abstract: The license plate recognition(LPR) is the key technology in intelligent transportation system. This paper discusses the whole process of license plate recognition technology, include the license plate image preprocessing, license plate location, character segmentation and character recognition, and simulated it by MATLAB. The experimental result show this method can obtain good recognition effect.

Abstract: An improved algorithm using B-splines as weight functions for training neural networks is proposed. There is no need for training neural networks or solving linear equations. The most important advantage is that we can get the forms of weight functions by the given patterns directly. Each of weight function is a one-variable function and takes one associated input point (input neuron) as its argument. The form of each weight function is a linear combination of some B-splines defined on the sets of given input variables (input knots or input patterns), whose coefficients are associated with the given output patterns. Some examples are presented to illustrate good performance of the new algorithm.

Abstract: This paper proposed an improved Web text classification algorithm. It combined SVM algorithm and KNN algorithm and used the KNN algorithm to compensate for the deficiencies of traditional SVM algorithm and took simple ideas and smaller cost to improve traditional SVM algorithm. The experiments show that the proposed algorithm in this paper has gained good effect.

Abstract: This paper proposed the principal component analysis (PCA) and support vector machine-genetic algorithm (SVM-GA) to the real-time face recognition. The integrated scheme aims to apply the SVM-GA method to improve the validity of PCA based real-time recognition systems. Experimental results show that the proposed method simplifies features effectively and obtains a higher classification accuracy.

Abstract: Proposed one kind of improvement artificial immunization algorithm calculates the electrical power system most superior tidal current, this algorithm maintained the basic immunity algorithm comprehensive search ability, also the concept which is apart from through the introduction vector causes the immunity algorithm theoretically to guarantee the understanding the multiplicity. Through the IEEE-30pitch point system computed result indicated this algorithm is feasible. And so on compares with the heredity algorithm, this algorithm overall situation search ability strong, the convergence rate is quick.

Abstract: When ship navigates at sea, collision avoidance of ship’s speed alteration is frequently adopted by officer on watch in order to prevent from forming collision situation with target ship(s). Bacterial foraging algorithm (BFA) that imitates the social foraging behavior of Escherichia coli is an optimal search method suitable for complex problems. This research adopts the bacterial foraging algorithm to find the speed alteration collision avoidance strategy from an economical viewpoint, combining the international regulations for preventing collisions at sea (COLREGS) and the safety domain of ship. An optimal time of changing speed, amplitude of speed alteration and navigation restoration time will also be provided. The effectiveness of the algorithm has been verified by simulation. The study offers new thinking and a practical method for collision avoidance decision.