Applied Mechanics and Materials Vols. 543-547

Abstract: This paper proposed of using stack filter technique to clears image noise. The key points of stack filter algorithm are threshold Decomposition and cascading features, and it is one of nonlinear filter based on the order statistics theory. Mat lab experiments show that stack filter can greatly reduce the time of clears image noise, and it performs better than median filters on noise reduction while retaining edges of an image.

Abstract: The prior distribution of the incidence of crime is based on a large-scale of criminals' investigation of historical data on their psychology problems and the new sample data is based on the measured people's sampled data of investigation on their psychological problems. The incidence of crime of the measured people is the posterior distribution of the measured that need to be predicted. With the application of Bayesian statistical methods we could compute the incidence of the crime of the measured and provide a basis for us to judge whether the suspect is a criminal. The paper has verified the feasibility of the method and pointed out its limitations and applied condition.

Abstract: A Mahalanobis distance based semi-supervised fuzzy clustering model is presented in this paper, whose objective function has a good explanation on how the labeled and unlabeled data are used in finding the underlying structure of matrix data. The iterative algorithm to solve this model is given. This algorithm can directly deal with matrix data like face images. We use 2DPCA on both row and column directions to reduce the dimension of image faces. The experimental result shows that using 2DPCA and semi-supervised algorithms can have a fairly good recognition rate if enough labeled data are given.

Abstract: For the local recovery of routing techniques in mobile Ad Hoc network (MANET), the fix problem with link failure is studied as a result of nodes moving relatively or wireless environment changing in mobile Ad Hoc network. It discusses the traditional local repairing methods which utilize the effect on service transmission delay and the successful delivery rate of packets by huge control overheads generated by the flooding throughout the network. In this paper on route-recovery algorithm in mobile Ad Hoc network based on on-the-fly strategy, it creates the repair model by limiting repair request zone within two hops and also builds the repair function so as to fulfill the objective of reducing flooding and controlling expenditures under the premise of meeting repair probability.

Abstract: VLF communication is one of the most effective means to command underwater platform. The spectrum resource of vlf communication is limited, so it is very important for enhance the reliability of vlf communication to strengthen the vlf communication spectrum management. It makes use of the well prediction of the vlf frequency band, to establish the signal field predicting model and atmospheric noise statistics predicting model respectively, predicts the field of signal and noise level, and then calculated the signal-to-noise ratio (SNR). According to the SNR distribution we can determine the relationship of the optimal communication frequency, time and location of receiver. So the efficiency of spectrum management for vlf communication can be enhanced.

Abstract: Combined with the characteristics of the image data, this study contrasted four kinds of data model. Then it analyzed the three kinds of realization methods of image database, comparative analysis of management modes of the distributed image database finally.

Abstract: In today's social life, communication between people is essential to do anything important way. The facial expressions are rich in human behavior information, it is very important means of communication, as a carrier of information, expression can convey a lot of voice can not convey the information. Expression recognition field of pattern recognition is a new task, is an essential part of intelligent machines. This paper studies the discrete wavelet transform feature extraction and expression using MATLAB software image feature extraction and treatment with an elastic template matching algorithm to do the appropriate test expression recognition.

Abstract: As the most important medium of communication in human beings life, speech carries abundant emotional information. In recent years, how to recognize the speakers emotional state automatically from the speech is attracting extensive attention of researchers in various fields. In this paper, we studied the method of speech emotion recognition. We collected a total of 360 sentences from four speakers with the emotional statement about happiness, anger, surprise, sadness, and extracted eight emotional characteristics from these voice data. Contribution analysis method is proposed to determine the value of emotion characteristic parameters. We also have used the weighted Euclidean distance template matching to identify the speech emotion, got more than 80% of the average emotional recognition rate.

Abstract: In order to make an intensive study of issues related on aircraft assembly operators, such as the behavior of humans habit, the typical posture of manufacturing, the limb internal coordination mechanism and the working environment cognitive mechanism, aircraft assembly posture data acquisition experiment is designed and implemented in this paper. Firstly, the purpose and object of the experiment are determined. Secondly, the design and implementation of the experiment scheme are completed, including the definition of tasks and parameters, layout of measuring signalized points and experiment environment, and implementation of data acquisition experiment. Finally, the data acquired is processed to establish the relationship between human postures and operating points, and the results are analyzed. By using the method proposed in this paper, formation rules base of aircraft assembly process posture can be founded, which is the foundation of human global approximate posture generation algorithm.

Abstract: Highly correlated signals exist in many applications of different fields of physics and engineering. But existing techniques for signal estimation and detection are either too complex or do not perform adequately when the observations consist of highly correlated signals in noise. Thus, this paper presents a review of three important super-resolution methods to discuss their characteristics and limitation for extracting the highly correlated signals in noise.