Papers by Author: Xue Zhang Zhao

Abstract: Fuzzy kernel clustering algorithm is a combination of unsupervised clustering and fuzzy set of the concept of image segmentation techniques, But the algorithm is sensitive to initial value, to a large extent dependent on the initial clustering center of choice, and easy to converge to local minimum values, when used in image segmentation, membership of the calculation only consider the current pixel values in the image, and did not consider the relationship between neighborhood pixels, and so on segmentation contains noise image is not ideal. This paper puts forward an improved fuzzy kernel clustering image segmentation algorithm, the multi-objective problem, change the single objective problem to increase the secondary goals concerning membership functions, Then add the constraint information space; Finally, using spatial neighborhood pixels corrected membership degree of the current pixel. The experimental results show that the algorithm effectively avoids the algorithm converges to local extremism and the stagnation of the iterative process will appear problem, significantly lower iterative times, and has good robustness and adaptability.

Abstract: In the practical need in order to make the most effective image compression in this paper, a new image compression used wavelet neural network model, and gives the corresponding calculation formula and algorithm procedures, By using wavelet transform good time-frequency local area on the characteristics and neural network self-learning function characteristics, overcome traditional BP neural network of hidden-layer points are difficult to be determined and the convergence speed is slow and easy to converge to a local minimum points shortcomings. The results of the simulation experiment prove wavelet neural network image compression characteristic and the convergence speed are much better than traditional BP neural network, and show that the algorithm is effective and feasible.

Abstract: In order to better solve asynchronous motor complex fault characteristics, improve the reliability of the diagnosis and accuracy, combined with wavelet transform technique, construct a wavelet neural network, wavelet transform technology feature extraction asynchronous motor as a signal wavelet neural network's input vector, and the wavelet neural network algorithm was used to optimize, realize the motor identify types of fault, through the simulation experiment data diagnosis results show that this method is effective and feasible. Based on the wavelet analysis and neural network fault diagnosis method of research.

Abstract: In order to improve large amount of computing and slowly convergence speed, an improved radial basis function (RBF) neural network is raised in this paper. According to feature that the more recent data should be the more important in time-series data, it converts width value from original constant value to step function and accelerates the iterative convergence by using nearest neighbor clustering algorithm only at center, training weight by using gradient descent algorithm to correct network parameters and deleting input neurons adaptively. Network size is streamlined through network optimization training. Simulation shows that the restored image is good in visual and quantitative with faster image restoration processing. The algorithm based on improved RBF neural network has significantly improved the image restoration compared to other methods, but also well keeps image detail.

Abstract: GPU technology release CPU from burdensome graphic computing task. The nVIDIA company, the main GPU producer, adds CUDA technology in new GPU models which enhances GPU function greatly and has much advantage in computing complex matrix. General algorithms of image rotation and the structure of CUDA are introduced in this paper. An example of rotating an image by using HALCON based on CPU instruction extensions and CUDA technology is to prove the advantage of CUDA by comparing two results.