Earticle

Robust object tracking is important in many real applications. This study proposes a novel robust object tracking algorithm that combines particle filter and geometric active contour. In contrast to conventional combinations, the proposed algorithm establishes mutualism, wherein the particle filter provides curve initialization for the active contour by using the alpha shape algorithm, and the active contour provides evolved boundary for particle filter classification. The refined particles are used to estimate the states which reduce the distance error. Discriminable object feature is proposed to enhance the tracking robustness, and prior information-based curve evolution is used to improve the accuracy of boundary segmentation. Comparable experiments are performed using a stand particle filter and continuously adaptive mean-shift methods. The proposed algorithm is found to be highly robust for long video sequences with non-linear motion, scale changes and clutter background.

This paper adumbrates a novel metamaterial named Closed Ring Pair (CRP). Ansoft HFSS has been used to design and analyze the MTM having Mylar (εr=2.89) as substrate material. Magnetic properties of CRP MTM have been used to mathematically demonstrate the negative permeability region. Nicolson Ross Weir (NRW) method has been used to retrieve the material parameters from transmission and reflection coefficient. Upon simulation the resonant frequency has been found to be 30.55 THz.

Based on the problem that data simple weight is unchanged in current feature selection algorithms, we proposed a new boosting feature selection algorithm framework using data sampling technique to describe the change of data sample weight in the process of feature selection which can solve the problem that dynamic mutual information is sensitive to noise data. This method evaluates mutual information value among features by dynamic adjust data sample weights in the process of feature selection so that every selected feature can describe characteristics of data classification more precisely. Efficiency of this algorithm has been testified by experience results on UCI common test data sets.

The traditional clustering algorithms need some prior knowledge to determine the initial parameters, and the selection of initial parameters usually will produce great influence on the result of clustering. In order to achieve this purpose, we put forward a new clustering algorithm based on "idea", through optimizing the given data and mathematical model between the adaptive data to model the best match. This algorithm combined with an "idea" algorithm to learn the hybrid model because of the gaussian mixture model can be thought of as a distribution of soft clustering method. This clustering algorithm can overcome the shortcomings of k-means algorithm. Experimental result shows that the algorithm has better clustering effect.

Islanding detection method of active current disturbances in three-phase photovoltaic-to-grid system makes the common connection point (PCC) voltage appear high frequency components on account of load variation, the surge current occurrence and other factors. This high frequency component cause islands misjudgment. Using Db10 wavelet to detect and analysis PCC’s voltage high-frequency components on real-time based on the active current disturbances principle. And selecting an effective wavelet domain values as the voltage harmonic detection amount islanding detection. Islands simulation test is conducted in the case of the inverter output power and load input power to match, the test results show that this method can quickly detect islanding, and effectively prevent pseudo-island phenomenon of false positives.

Multi-Criteria Evaluation (MCE)is one of applications of multiple criteria decision making（MCDM）. GIS is an information system that is designed to work with data referenced by spatial or geographic coordinates. GIS combined with MCE can achieve measurable evaluation of drought risk. Technologies of evaluating agriculture meteorological drought risk with GIS-MCE are introduced, Taking precipitation anomaly as main drought evaluation index, the paper calculated the spatial distribution of meteorological factors to the whole region by the interpolation method of IDW. In addition, use multiple regression analysis method to study the correlation of meteorological factors, geography factors and social economic factors. The study fully reflects the important role of population density and regional economic development for drought division, making the drought spatial distribution model more accurate and comprehensive. Ultimately, In considering each factor effects, calculate the multi-factor comprehensive division map by GIS-MCE method, which including establishing evaluation criteria and weight by Delphi method, obtaining the spatial distribution of factors of the meteorological drought risk by diffusing spatial attribute value and implementing evaluation by spatial overlay calculation. The results indicated that technology of GIS-MCE can combine multiple source information associating with agriculture meteorological drought risk and achieve measurable result.

This paper addresses the color modeling selection problem for moving cast shadows detection in video sequences. This paper discusses issues related to moving cast shadow detection in video, and also illustrated the performance of shadow detection can be improved significantly through appropriate selection of color space. A Selection of appropriate color space is a very important step in building and detecting cast shadow from video. This paper also reported comparison results of several well-known color models to build shadow model embedded into a globally optimal MRF framework. Experimental results show that for real-time videos CIE L*u*v* color space is the most reliable for shadow detection and suppression and in the second place it is HSV color space.

A corner event detection method based on HMM (Hidden Markov Model) in soccer videos. Through analysis of the semantic structure of corner events, define and extract six multi- modal semantic clues to describe shot sequences, which constitute observation sequences as the HMM model input. By the iterative training of the HMM model and the continuous optimization of model parameters, construct the HMM model of corner events. From two aspects of audio and video, dig the inherent pattern of corner events and realize corner events automatic detection based on HMM accurately. The experimental results show that the present method achieves 89.6% recall and 96.3% precision, which has the better performance.

This paper proposed a new method for surface texture synthesis using improved self-organizing maps as the synthesis logic. The method will first loop map the sample image to the target surface until the surface is full filled, then construct a improved self-organizing maps model, in which use sample image as the input level and the target three-dimensional surface as out put level, to adjust the pixel position in target surface, and the mapping area is controlled in the neighborhood of local extremum by reducing search area of variables. The new algorithm not only speed up the synthesis progress, and also enhances the image quality of target surface texture, and compare with the previous algorithm, it is not necessary for the user’s intervention. Compared with previous results for the proof of our concepts, we have successfully implemented the experimental results and the proposed algorithm.