Advanced Materials Research Vols. 718-720

Abstract: Artificial intelligence (AI) is an interdiscipline that aims to create and enhance the intelligence of machines and robots. Neuroscience has a tight connection with AI, which is also one of the earliest research fields that neuroscience attempted to carry out. This paper focused on the development and research trends of AI in neuroscience with the help of a latest scientometric tool, CiteSpace II. It allowed us to grasp the research frontiers and trends of AI in neuroscience through the analysis of data concerning AI and neuroscience between 1990 and 2012. We found that cluster #5 heart rate variability was most likely to be the emerging trends and some technologies will be more frequently used in neuroscience research.

Abstract: Web usage mining is the information about the user data extraction, transformation, analysis and model processing, extracted from the auxiliary business decision of key data. Intelligent site refers to the use of the Web server log for user access patterns and provide personalized service for users. The paper proposes the development and design of intelligent web site based on web usage mining. This paper presents the access interest measure method and the traditional consider only clicks visit interest measure method, recommend less deviation, has better recommendation results.

Abstract: Embedded system in general is formed by the embedded computer system and implementing device. Embedded system is a control program stored in ROM embedded processor control board. The basic function of Radio Frequency Identification (RFID) is to realize the precise data acquisition. RFID reader via the antenna and the RFID tag for wireless communication can be achieved on the label identification code and a memory data read or write operation. The paper presents the development of RFID data management system based on embedded system. Experimental results show that the proposed algorithm has high efficiency.

Abstract: Discrete Wavelet Transform (DWT) has been brought into wide use in image processing, but it cant fit the demand of the hugeimage data because the time of computing is vast. The GPU is an attractive platform for a broad fieldof applications,which remains asignificanthigharithmetic processingcapability. Therefore itcan beusedasa powerful accelerator without extra cost.CUDA(computeunifieddevicearchitecture) providesahardwareandsoftwareenvironment touse the GPU to accelerate the DWT for images. In this paper, we use the NVIDIA GeForce GT 650M that complies with the CUDA to improvethe execution time of theDiscrete Wavelet Transformfor images. TheresultofexperimentsindicatesthattheCUDAtechnology hastheadvantagesof parallel processingandtheefficiencyofimagetransform isimprovedgreatly. Whats more, it performs better on the larger size image (the max speedup is 15.9).

Abstract: A new denoising method was proposed in the paper according to the characteristics of insulator infrared image with impulse noise. First, based on the pulse coupled neural network (PCNN) to detect the location of the impulse noise pixels, while maintaining the same non-noise pixels. and then according to the characteristics of the impulse noise, the window size of the filter was adaptively determined by calculating the noise intensity of the image. The pixels with maximum and minimum gray value in filtering window are excluded, using the left pixels similarity calculation out weights. A new weighted filtering algorithm is used to filter noise pixels. The experiments show that the method is better than the median filter in peak signal-to-noise ratio (PSNR), and has better image edge details protection ability.

Abstract: The hard X-ray in-line phase contrast imaging (HXILPCI) is a phase contrast technique that generates excellent contrast of biological soft tissues compared to conventional X-ray absorption radiography. We explore the application of HXILPCI in the diagnosis of gastric cancer and pancreatic cancer. These nude mice cancer samples were checked by HXILPCI to obtain projection contrast images of 9μm image resolution with CCD camera. The texture extraction was based on gray level co-occurrence matrix (GLCM). The corresponding morphological features of abnormal and normal tissues are analyzed. The produced phase contrast images of nude mice cancer samples show clearly biological tissues architectures and the size of cancer. The paper results show that HXILPCI can be a potential noninvasive technique to diagnose early cancer.

Abstract: Some results on strong stability for weighted sums of ~½-mixingrandom variables and new strong laws of large numbers are presented, whichgeneralize the corresponding results of independent sequences.

Abstract: Naïve Bayes classifier was generally considered as a simple and efficient classification method. However, its classification performance was affected to some extent because of the assuming that the conditions properties were independent of each other. By analyzing the classification principle and improvement of Bayesian and the Attribute Reduction of Rough Set, this paper proposed a Naïve Bayes algorithm that the attribute order reduction and weighting were improved simultaneously. Experiment results demonstrated that the proposed method performed well in classification accuracy.

Abstract: Since the 7 famous Hus invariants had been introduced in 1960s, the moment invariants play an important role in image analysis and pattern recognition. In this paper, we propose a new moment called Gaussian-Geometric moment, and derived their translation and rotation invariants. One significant conclusion drawn is that the rotation invariants of Gaussian-Geometric moments have the identical forms to those of geometric moments.The Gaussian-Geometric moments and the geometric moments both can represent the image information, difference is that the Gaussian-Geometric moment can represent the center information of an image and the geometric moments represent the edge information of an imageonly. This is particularly evident in the performance of high order ones. Another important property of Gaussian-Geometric moments is that it has a scale parameter which allows choosing the best scale to represent the interest region of an image. A detailed comparison has been made to test the feature matching capability between the proposed moments and the geometric moments. The results show that the proposed moments perform much better than the geometric ones.

Abstract: New threats to and requirements of security come forth endlessly while the network technology progresses continually. Security equipment has little adaptive capacity in the existing rigid network. It has bad influence on network security in Mainland China. A reconfigurable security component is designed in this paper. It can fit the coming requirements of network development in the future. A prototype of this security component is implemented and analyzed.