Applied Mechanics and Materials Vols. 462-463

Abstract: According to the shortages of the standard Artificial Fish-school Algorithm (AFSA), this paper improves the behaviors of the fish-school by considering the group behaviors of neighborhood sensing effects, leader mode and Glowworm Swarm Optimization Algorithm (GSOA). Combining the self-adaptive step and vision, Gaussian mutation, Genetic Algorithm (GA), a new AFSA based on self-adaptive mutation operation is raised. By using the validation functions to test, the results show that the rate of convergence, optimization precision and the ability to avoid precocious phenomenon of the new algorithm are much better than the standard one.

Abstract: Sensorless control system for Permanent magnet synchronous motor (PMSM) has been widely used in many areas for its simple installing and easy realizing advantages. This paper mainly analysis sensorless control system of PMSM, estimate the angel of rotor and speed based on flux observation initially, PLL and state observer were used to obtain the accurate angle. Use Simulation toolbox design of MATLAB to implement the simulation of control system and analyze the simulation results. The results verify the correctness and feasibility flux observation method that used in paper.

Abstract: The stationary contourlet transform is built upon nonsubsampled pyramids and nonsubsampled directional filter banks and provides a shift invariant directional multiresolution image representation. Firstly, several SAR images can be decomposed into low-frequency coefficients and high-frequency coefficients with multi-scales and multi-directions using the stationary contourlet transform. For the low-frequency coefficients, the average fusion method is used. For the each directional high frequency sub-band coefficients, the larger value of horizontal and vertical direction gradient information measurement is used to select the better coefficients for fusion. At last the fused image can be obtained by utilizing inverse transform for fused contourlet coefficients. Experimental results show that the proposed algorithm gives more satisfactory results than the traditional image fusion algorithms in preserving the edges and texture information.

Abstract: The method of least squares support vector machine has been improved based on base vector space theory, which solved the problem of weak handling ability of nonlinear, sparsity and variable multiple correlation in the fault diagnosis process for LSSVM. This article proposed a double sections fault diagnosis algorithm combined PLS with LSSVMBVS, it first builds regression analysis model, then puts the detected fault data in the trained LSSVMBVS classifier, diagnosing troubles. It verified the algorithm has better prediction accuracy and generalization performance by the TE platform.

Abstract: Normal image processing, combined with automatic Matlab program, gives a way to deal with embedded date-and-time OSD in video surveillance. It has eliminated the influence on multi-frame averaging from date-and-time words occlusion. Results are applied in real cases and superior to ordinary multi-frame average method.

Abstract: We proposed an approach for estimating the density of salt-pepper noise in images with correlation inspection. The correlation coefficients histogram was introduced in this paper to depict the correlation distributions of images. Based on the fact that the correlation distributions of natural images are nearly independent of individual images, we revealed how the correlation coefficients histogram of the noisy image deviates from that of natural image along with the noise density in qua- ntitative form, thus we took advantage of this relation to make estimation. Simulation results showed that the proposed approach outperforms those of existing methods.

Abstract: People pay more attention to the security of RFID, this paper establishes the evaluation index architecture of RFID security, Then according to these indexes for the evaluation of the RFID system security which based on fuzzy synthetic evaluation model, all of this is to evaluate the security of RFID system.

Abstract: With the rapid development of high-speed railway, the equipment life-cycle management data are generated in large scales which run through the period of production, operation, maintenance, falling into a notion of Big Data. There is broad recognition of value of data and information obtained through analyzing it. The exponential growth in the amount of railway-related data means that revolutionary measures are needed for data management, analysis and accessibility. At present, the promise of data-driven decision-making is now being recognized broadly. How to store the big data efficiently, reliably and cheaply are important research topics. This paper proposes a framework of data management of high-speed railway equipment, where cloud computing provides a feasible technical solution combined with MapReduce programming model based on Hadoop platform. These models are capable of considering the characteristics of data and processing demand in management of High-speed railway equipment. Finally, we summarize the challenges and opportunities with Big Data for application of China railway and point out there is more than enough that we can work on.

Abstract: Saliency estimation has become a valuable tool in image processing and raised much interest in theory and applications. Despite significant recent progress, the performance of the best available saliency estimation approaches still lags behind human visual systems. In this paper we used saliency filters and domain knowledge in photography to estimate saliency. Experiments show that our method can successfully detect the true salient content from images.

Abstract: Face detection, pose estimation and facial landmark localization are three fundamental problems in pattern recognition. These three tasks have high request of algorithm efficiency and accuracy. Zhu and Ramanan proposed a model based on mixture of tree structures to solve the three tasks simultaneously and it obtains state-of-the-art result. However, the efficiency of their algorithm is relatively low. Our improved algorithm combines Viola Jones detector and tree-structured model and achieves a speed-up of tens of times even hundreds of times of original algorithm on ordinary laptop according to images of different sizes.