Advanced Materials Research Vols. 605-607

Abstract: The shadow detecting algorithm based on the coherence and the Sigma filter is used to pick up the shadow of interferometric synthetic aperture sonar (InSAS), which can eliminate small separated shadow areas. To solve the problems such as great computer complexity of traditional Shepard interpolation method and large fluctuant of linear interpolation method for the large shadow area, an improved Shepard interpolation method is proposed. Interpolation boundary is picked up by using diffuse search, and interpolation source is adaptively chosen according to the size of shadow area. The method carries out a perfect tradeoff between performance and computer speed. Lake trial dataset is used to validate the performance of proposed method. The results indicate that the proposed method can eliminate the fluctuant from the linear interpolation method and can process in real time in the InSAS system.

Abstract: In order to achieve robust and rapid image preprocessing for finger vein, this paper presents a robust and rapid image preprocessing method. With this method, locating & cropping, normalization, equalization and filtering are executed step by step. Then, quality assessment is done to exclude the images of low quality. The results of simulate experiments show that this method is effective and it is able to ensure reliable feature extraction.

Abstract: Many factors influence vision neural network information processing process, for example: Signal initial value, weight, time and number of learning. This paper discussed the importance of weight in vision neural network information processing process. Different weight values can cause different results in neural networks learning. We structure a vision neural network model with three layers based on synapse dynamics at first. Then we change the weights of the vision neural network model’s to make the three layers a neural network of learning Chinese characters. At last we change the initial weight distribution to simulate the neural network of process of the learning Chinese words. Two results are produced. One is that weight plays a very important role in vision neural networks learning, the other is that different initial weight distributions have different results in vision neural networks learning.

Abstract: In wireless sensor network for nuclear power plant’s peripheral environmental radiation monitoring the gamma dose rate data may be missed affected by various factors, which will influence the validity of environmental radiation monitoring. To solve the problem, a missing data imputation algorithm is proposed based on particle swarm optimized least squares support vector machine. This algorithm imputes missing data utilizing node’s previous monitoring data and neighbor node’s current monitoring data jointly. Experimental results using the real radiation monitoring data around a nuclear power plant show that the proposed algorithm can impute the missing gamma dose rate data accurately.

Abstract: In order to realize the effective segmentation of the rice leaf lesion, how to remove background interference and isolating single rice leaf from collected images is the basis of disease classification and recognition. The main body information of the images collected under the simple background of lighting box and field contains only single leaf, so we can use the normal way to detect the blade edge. However, because of the outside interference under the background of field, parameters can't be quantified and the treatment effect of using ordinary edge detection methods is bad. This paper takes advantage of the characteristics that phase consistency test methods that protect them from the image contrast and brightness change and have a strong ability to resist noise to realize complete extraction of the main blade edge in the rice leaf image under the complex background.

Abstract: The objective of the well-known travelling salesman problem (TSP) is to search the optimal Hamiltonian circuit (OHC) in a tourist map. Finding the OHC becomes hard once the number of the cities and routes in the tourist map are large. The four vertices and three lines inequality was introduced as the constraints of the local optimal Hamiltonian paths (LOHPs) included in the OHC. The chaotic depth-priority algorithm was designed by adding the computation process with the chaotic operator to verify the rationality of the LOHPs generated with the depth-priority algorithm under the inequality constraints. A lot of non-LOHPs are abandoned in the search process and the search space of the OHC is reduced greatly. The method was verified with an example and it can be applied to the network optimization, path planning, task scheduling, assembly sequence planning etc.

Abstract: Objective To segment lung fields on digital chest radiographs automatically. Methods A morphological reconstruction filter was first applied to the original image to eliminate the local grey level extremes. Then, the Otsu-threshold method was used to segment the whole image into several connected regions, contours of which were then extracted by using a connected components labelling technique. Finally, a morphological closing operator was employed to smooth any small gaps or burrs of the lung field contours. Results Lung fields were segmented for 40 digital chest radiographs, resulting in an accurate segmentation and extraction of their lung fields for most images. Conclusion Integrating a series of algorithms, including morphological reconstruction filtering, Otsu-threshold segmentation, connected components labelling, and morphological operations for smoothing the outlines can effectively segment the lung fields on digital radiographs.

Abstract: This paper makes the case for explicit coordination of network transmission activities among virtual machines (VMs) in the data center Ethernet to proactively prevent network congestion. We think that virtualization has opened up new opportunities for explicit coordination that are simple, effective, currently feasible, and independent of switch-level hardware support. We show that explicit coordination can be implemented transparently without modifying any applications, standard protocols, network switches, or VMs.

Abstract: The principle and implementation steps of the method of feature extraction of weak multi-frequency signal based on the array of modulated stochastic resonance under the background of strong noise are described in the paper. By the modulation of the known multi-frequency weak signals under the strong noise background and the carrieres with different frequency respectively; multiple signals with the same frequency of 0.01Hz were generated. Then these generated signals were as the input signals of multiple parallel non-coupled resonant units. The Runge-Kutta algorithm was used to obtain the unit outputs and to analysis the frequency spectrum. According to the SNR of the 0.01Hz to determine whether the 0.01Hz frequency components were contained in the frequency spectrum. Finally the frequency characteristic vectors of the weak signals were generated by the systemization of the detection results of the stochastic resonance units.Results show that this method has obvious effect in the extraction of the feature of the weak multi-frequencies signals, and has a very good application prospect.

Abstract: A wavelet network was constructed through multi-spectral, high-pixel wavelet decomposition of remote sensing image, which replaced the traditional neuron activation function with a nonlinear wavelet system. So we found a data fusion model. Multidimensional information fusion was to integrate the multi-source information characteristics of a high dimensional feature space. With the genetic search based on the natural selection, we combine effective evaluation of fusion accuracy.