Applied Mechanics and Materials Vols. 135-136

Abstract: Traffic guide is a very important service for society. Current guide system is poor efficiency and complexly in operation. This made traffic guide information can’t be published regularly and in time. It also reduced down expressway net’s ability of providing traffic guide information for society. Regard CMS on expressway net as objective, after sorting and concluding guide information by scene analysis, the paper suggests a systematic method of processing traffic guide information based on dynamic traffic information automatic and prove it by an application. It can send detour information to each piece of CMS on expressway net quickly without redundant information. Traffic guide level on province expressway net would be improved extremely.

Abstract: support vector machine (SVM) has been shown to exhibit superior predictive power compared to traditional approaches in many studies, such as mechanical equipment monitoring and diagnosis. However, SVM training is very costly in terms of time and memory consumption due to the enormous amounts of training data and the quadratic programming problem. In order to improve SVM training speed and accuracy, we propose a modified incremental support vector machine (MISVM) for regression problems in this paper. The main concepts are that using the distance from the margin vectors which violate the Karush-Kuhn-Tucker (KKT) condition to the final decision hyperplane to evaluate the importance of each margin vectors, and the margin vectors whose distance is below the specified value are preserved, the others are eliminated. Then the original SVs and the remaining margin vectors are used to train a new SVM. The proposed MISVM can not only eliminate the unimportant samples such as noise samples, but also preserved the important samples. The effectiveness of the proposed MISVMs is demonstrated with two UCI data sets. These experiments also show that the proposed MISVM is competitive with previously published methods.

Abstract: As a basic aspect of computer vision, reliable tracking of multiple objects is still an open and challenging issue for both theory studies and real applications. A novel multi-object tracking algorithm based on multiple cameras is proposed in this paper. We obtain the foreground likelihood maps in each view by modeling the background using the codebook algorithm. The view-to-view homographies are computed using several landmarks on the chosen plane. Then, we achieve the location information of multi-target at chest layer and realize the tracking task. The proposed algorithm does not require detecting the vanishing points of cameras, which reduces the complexity and improves the accuracy of the algorithm. The experimental results show that our method is robust to the occlusion and could satisfy the real-time tracking requirement.

Abstract: A new algorithm to solve the fused complex-valued cost function for approximate joint diagonalization (AJD), named CVAJD (Complex-Valued Approximate Joint Diagonalization), is presented. The CVAJD algorithm adopts an iterative scheme to update the demixing matrix through the strictly diagonally-dominant residual mixing matrix obtained in each of iterations. Due to the relaxation of several constraints on the target matrices, it has more general utilizations. Besides, it is also easy to implement. A numerical simulation illustrates fast convergence and good performance of the CVAJD.

Abstract: Considering the inaccuracy of the traditional similarity method in the k-nearest neighbor algorithm (KNN), we put forward a hybrid algorithm crossing the principal components analysis (PCA) and KNN via a fresh hybrid way. The algorithm makes use of differences in the contribution rates of features mined by PCA to build the similarity model of users. The experiment results argue that the fresh hybrid algorithm makes personalized recommendations very effective.

Abstract: In order to improve the security of database system and resist threat from all aspects, especially the threat from database administrators, the paper designed a new database encryption system. In this system sensitive information is encrypted and establishes ciphertext address index table for it. Encryptions for character fields and numeric fields have different processing methods. Decryption key should be synthesized by both client and server. System can execute SQL query like equality queries, range queries and so on which are difficult to deal with after encryption. The analysis shows that the system has better security.

Abstract: Residual liquid inspection is an important process before filling up in the beer filling line. It aims to inspect that whether there is residual alkali liquid in the inspecting bottles and improve the quality of the beer. At present, manual inspection can’t meet the requirements of high filling speed and high-accuracy inspection. While the automatic residual liquid inspection system at present has some defect such as poor security, low accuracy and reliability. Therefore, this paper proposes a new method for residual liquid inspection which is based on the principle of capacitive coupling and has realized non-contact measurement for residual liquid remaining in empty bottle. The inspection prototype is also designed for detection. In order to further improve detection reliability, the statistical process control method is also introduced in the residual liquid inspection system. According to experiments, it is indicated that this method can effectively improve accuracy, reliability and security of the detection and has high practical value.

Abstract: In order to track the ballistic re-entry target, a new kind of ballistic target tracking algorithm, square-root quadrature Kalman filter (SRQKF) algorithm, was proposed. The proposed algorithm is the square-root implementation of the quadrature Kalman filter (QKF). The quadrature Kalman filter is a recursive, nonlinear filtering algorithm developed in the Kalman filtering framework and computes the mean and covariance of all conditional densities using the Gauss-Hermite quadrature rule. The square-root quadrature Kalman filter propagates the mean and the square root of the covariance. It guarantees the symmetry and positive semi-definiteness of the covariance matrix, improved numerical stability and the numerical accuracy, but at the expense of increased computational complexity slightly.

Abstract: A novel classification algorithm based on class association rules is proposed in this paper. Firstly, the algorithm mines frequent items and rules only in one phase. Then, the algorithm ranks rules that pass the support and confidence thresholds using a global sorting method according to a series of parameters, including confidence, support, antecedent cardinality, class distribution frequency, item row order and rule antecedent length. Classifier building is based on rule items that do not overlap in the training phase and rule items that each training instance is covered by only a single rule. Experimental results on the 8 datasets from UCI ML Repository show that the proposed algorithm is highly competitive when compared with the C4.5,CBA,CMAR and CPAR algorithms in terms of classification accuracy and efficiency. This algorithm can offer an available associative classification technique for data mining.

Abstract: At present, the web-based collaborative learning is the main form of long-distance learning and it has been widely used in the distance learning domain. However, the main body-intelligence degree of the existing Web learning system is low; the existing system lacks to the support of the personal study, which has hindered the individual skill of the users. According to the question that exists in the collaborative Learning, I constructed the intelligent agent model of collaborative Learning which is composed of main control program, student proxy, teaching proxy and the information proxy based on the intelligent agent technology which belongs to the domain of Artificial Intelligence and I discussed emphatically how to effectively realize the personal study by using the machine learning and the data mining method. It can cause the web-based cooperative learning developing along the direction of intelligent and personal