Zhang Xuewu

An Object Recognition and Location Based on Binocular Stereo Vision

The image includes much useful information, so it’s very important in intelligence traffic system. In this paper, one method of object recognition and location based on binocular stereo vision is introduced. Through dealing with two images what are captured from difference points in the ca, the system can analyze the object in images and compute its location, with the real-time and accuracy requirements.

Algorithm of Classifying Classes of Social Networks

Social Networks have become embedded in our daily life so much that we no longer realize it. But researches on the structures of networks are correspondingly fewer than their statistical properties, however, the studies of structural features are necessary and significant. Here we present an approach to study the structures of social networks based on Role-to-role Connectivity Profiles (RCP). We conduct experiments in some classical social networks, we find that the social networks with different functions exhibit obvious different profiles. Different from the classification of complex networks, social networks are divided into two distinct classes with strict judgement basis rather than vague judgement. In addition, according to the RCP, we can find important nodes in social networks, which is more significant than those with high degree. Through the use of the algorithm in social networks, we can thoroughly understand a social network by the comprehensive network classifications rather than only relying on community detection and other global properties.

Research on Complicated Background Suppression of Infrared Image Based on Vision Perception Mechanism

Aimed at the characteristics of infrared images for moving target, such as complicated background, low signal-to-noise ratio and low contrast, a new method for complicated background suppression is proposed based on Large Field (LF) and Small Field (SF) system in the paper. Firstly, the CCD sensors are used to obtain infrared video data for moving target. Then, after analyzing the visual motion detection mechanism of insect compound eye, the paper establishes background suppression model based on LF&SF. In this model, LF is used to suppress noise of edge detection while SF is used to do nonlinear enhancement of target features and estimation of target movement direction. Finally, the paper uses the primary motion detector to integrate LF with SF. Real scene experiment shows that this proposed method has obvious advantages in suppressing the cloud cover, grass lands and other noise and enhancing target.

A Fast Target Tracking Algorithm Based on Wavelet Transformation and Kalman Filter

Directing against the huge amounts of calculation for traditional target tracking, a fast target tracking algorithm based on image correlation matching is proposed in this paper. Firstly, for the continuity of movement target in the real-time image sequences, the image matching algorithm based on wavelet transform is used to match the first frame image to determine the best match point, then the best match point is used as initial point and combined with prediction characteristics of Kalman filter to predict the possible position of the target on the next frame and optimal matching point of the next frame. Eventually, it realized a fast and accurate target tracking algorithm. The experimental results show that the proposed matching algorithm can provide favorable tracking performance and improve the speed and accuracy of target tracking, which meets the real-time request of system.