Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-05-20 Cooperative journals: 《计算机应用研究》
Abstract： Texture similarity calculation is one of the basic methods of big-data analysis and data mining. For solving the problem that the existing texture features are not strong for color image discrimination, a texture similarity calculation method with improved local binary pattern features is proposed. This method proposes three modes for feature fusion, including extreme mode, addition mode and encoding mode. The LBP features acquired on the three channels of H, S and V of color image are fused by these modes to obtain the texture description of color image. The fusion operation is carried out in three stages including LBP calculation of neighborhood pixels, LBP calculation of central pixels, and histogram feature extraction, to improve the ability of feature discrimination. The results of texture similarity experiments on VisTex texture database show that, the false acceptance rate, flase rejection rate and equal error rate of this method are obviously lower than those of methods described in references [7, 8, 9].
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-05-20 Cooperative journals: 《计算机应用研究》
Abstract： In order to solve the problem that the current salt and pepper noise filtering algorithms for X-ray images are ineffective and low efficiency, this paper proposed a salt and pepper filtering algorithm with multi-level classification and adaptive fuzzy filtering, which included two parts: pixels’ multi-level classification and adaptive fuzzy filtering. In the process of multi-level classification, it designed a rapid rough classification according to priori knowledge, to divide the pixels into three categories: salt and pepper noise, signal and suspicious noise. For the suspicious noise, it extracted the histogram distribution features in the region, and designed the BP neural network classifier to classify the pixels, and finally classified all the pixels in the image into two kinds of signal and salt and pepper noise. In the process of adaptive fuzzy filtering, it created fuzzy membership function for three fuzzy sets, and calculated fuzzy membership value, and restored pixel brightness by fuzzy weighted summing. The experimental results show that the new method has high accuracy of pixel classification, high peak signal to noise ratio of filtered image, and less average time-consuming of filtering.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-05-18 Cooperative journals: 《计算机应用研究》
Abstract： Aiming at the problem that the super-resolution effect on different low-resolution images of existing super-resolution methods has large difference, this paper proposed a new single image super-resolution method based on sparse Bayesian estimation. In this method, it regarded the single image super-resolution problem as a regression problem. It used the Kronecker pulse functions as the regression basis functions, and obtained the optimal sparse solution of the specific prediction by combining the local information and global information of the image. It used the Bayesian method to estimate the weights, thereby reconstructing the super-resolution image. The experimental results show that this method can obtain high average peak signal to noise ratio, small variance and less time-consuming, when carried out on 14 testing images for single image super-resolution. It is proved that this method has good super-resolution effect, strong adaptability, and high efficiency.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-05-18 Cooperative journals: 《计算机应用研究》
Abstract： In order to meet the needs of the network under the background of big data, and eliminate inferior data interference data knowledge high precision requirements of large data transmission, variable size adjustment scheme based on the algorithm to expand the network of large data knowledge opportunistic scheduling is proposed. Based on the analysis of large data network characteristics, the adaptive vector encoding, capture the heterogeneous characteristics of large data network, using multi order back-propagation network of heterogeneous data is normalized, and then through the real-time transmission of large data network to achieve opportunistic scheduling. At the same time, the knowledge engineering system composed of network data segmentation of fine-grained big data based on the multidimensional feature dimension, the granularity of knowledge transformation is known, then adjust the size of the dynamic characteristics, making big data set of knowledge engineering with linear characteristics and clear geometric characteristics, improve the accuracy of knowledge acquisition through knowledge expansion. The experimental results are compared with the algorithm based on fine grained knowledge acquisition, which proves the high reliability, real time and high efficiency of network data transmission.
Hits
Hits
Downloads
Comment