Applied Mechanics and Materials Vols. 602-605

Abstract: In order to identify new words in huge Chinese corpus efficiently, this paper comes up with an algorithm based on ensemble methods. At first we perform Chinese word segmenting with Trie and build segment-tree. Then we select words pattern drawing method, frequency filtering, independent word probability and naive Bayes model to be sub-models of ensemble methods and train them independently. At last we integrate results from different sub-models with a multi-layer model. In experiment, this algorithm is proved to be quite fast as well as product precise and high-coverage results.

Abstract: In certain tank gun technical conditions integrated testing line, we firstly realize on line test of tank gun under non disassembly situation by using data acquisition card ----4472B produced in American NI Company. It realizes multiple data concurrent acquisition and processing and ensures the integral correlation of each testing data. The system adopts PXI main control platform and actualizes the technical conditions integrated test system of breechblock and recoil system based on Virtual Instrument technology. Through practical use, the result proves that the system has prominent character of integrated testing, online testing, visual data and convenient operation.

Abstract: An intrusion detection method based on RS-LSSVM is studied in this paper. Firstly, attribute reduction algorithm based on the generalized decision table is proposed to remove the interference features and reduce the dimension of input feature space. Then the classification method based on least square support vector machine (LSSVM) is analyzed. The sample data after dimension reduction is used for LSSVM training, and the LSSVM classification model is obtained, which forms the ability of detecting unknown intrusion. Simulation results show that the proposed method can effectively remove the unnecessary features and improve the performance of network intrusion detection.

Abstract: In this thesis, a moving object detection algorithm under dynamic scene is proposed, which is based on ORB feature. Firstly, we extract feature points and match them by using ORB. We then obtain global motion compensation image by parameters of transformation matrix based on the RANSAC method. Finally, we use the inter-frame difference method to achieve the detection of moving targets. The high speed and accuracy of ORB feature point matching method, as well as the effectiveness of the RANSAC method for removing outliers ensure accurate calculation of parameters of affine transformation model. Combined with inter-frame difference method, foreground objects can be detected entirely. Experiment results show that the algorithm can accurately detect moving objects, and to some extent, it can solve the issue of real-time detection.

Abstract: Chronic viral hepatitis, especially viral hepatitis B (HBV), has become a widespread infectious disease in the world. China is a big power of country in HBV, and people infected in China are the largest repository of HBV, which provides extensive research resources. The Data Mining and Aided Diagnosis System of Hepatopathy (DMADSH) embarks from the clinical situations and actual needs, combines the medical knowledge with computer data comprehensive analysis and mining technology, and through the knowledge extraction of the vast amounts of patient clinical data, image characteristics, location and shape of lesion, it gets the classification of hepatitis lesions and diagnosis automatically, so as to diagnose early to liver cancer and liver cirrhosis, and support for long-term dynamic monitoring in patients of hepatopathy and other health purposes.

Abstract: This paper introduces a novel on-line real-time monitoring system for acidic emission produced by acid pickling, which has been one of the greatest contaminant and major harness content. The system can monitor the HCL concentration in exhaust fumes on-line, and realize the real-time analysis and transmission of the acquisition data. HCL wet chemical sampling system form the whole on-line monitoring system. HCL wet chemical sampling system consists of integrated air velocity transducer, chloride ion analyzer, wet chemical sampling device and liquid level measurement device. It aims to collect HCL fog sample and convert the sampling analogue signal to digital data that can be analyzed. The system overcomes the influence by high temperature, high humidity and high concentration of dust. It realizes the monitoring, storage, analysis, warning and transmission of HCL concentration data in the exhaust. It is beneficial in practical industrial field and environmental protection.

Abstract: SAT (Separating Axis Theorem) is an algorithm applied in two-dimensional games of collision detection in the convex polygon. This paper applied SAT algorithm into video monitoring at intersections to detect the vehicles collision incidents. Established separation axis to extracted vehicles and projected the target vehicles into separation axis. Through comparing the scalar value in separation axis to determine whether the target vehicles collision. It's proved that the method is effective by using Matlab software to the simulation test in this algorithm.

Abstract: In this paper, an improved PIV algorithm is proposed for the velocity field of oil-water two-phase dispersed flow in horizontal pipe. In the proposed PIV algorithm, interrogation windows are overlapped by 50% in all iterations other than just overlapped in the final iteration. And what’s still different is that the interrogation window can also be a rectangular window just as [64 64; 64 64; 32 32; 32 32; 32 16;]. What’s more, if any element of the final interrogation window is different from the penultimate iteration, there is going to be another interrogation with the last interrogation window size, which can reduce the false vectors. Experimental results show that velocity measurements of oil-water two-phase flow can be realized by this advanced PIV algorithm with high accuracy. At the same time, it provides the basis for further studying the application of PIV in velocity measurements of oil-water two-phase flow.

Abstract: Empirical evidence shows that introducing additional structured priors can reduce complexity of coding data, and achieve better performance. To improve the performance of sparse representation-based classification (SRC), the article based on the potential correlations between the elements of dictionary gets a mixed group sparsity which is composed of dynamic group sparsity and fixed-length group sparsity. To solve the structured sparsity efficiently, structured greedy algorithm (structOMP) is redesigned to fit the new structure. The modification includes search space and its neighbor. Finally, three sparse models are compared by experiments of face recognition, and the results show that the mixed group sparsity can improve the face recognition rate of other sparse models by 10% or more in dealing with corrupted data.

Abstract: In our previous work, we proposed a motion edge detection method to extract the contour of the pedestrian in an image sequence. In order to locate and recognize the pedestrian in an image after its contour was extracted, we propose the BLBP method to describe the binary texture of the contour of the pedestrian in the image, and use the BLBP histogram to get the recognition feature of the pedestrian. And then we use the scatter matrix and the sequential forward selection method to select useful features, and use the SOM neural network to perform the recognition work. At the last part of this paper, some results of our experiments are illustrated there, which shows that our method is satisfactory.