Zhenkang Shen

Learn More
A contour edge detector combing surround facilitation and suppression is proposed in this paper. Specifically, the detector follows the flow of the Canny edge detector while incorporating surround facilitation and suppression into it. Firstly, edge gradient magnitude and edge density information is utilized to classify potential contour points and texture(More)
This paper presents a real-time infrared pedestrian detection algorithm based on multi-block LBP features. Local pattern feature has been successfully used in many computer vision applications. However, to our best knowledge, it has not been utilized for pedestrian detection in infrared imagery. We exploit the discriminative power and computational(More)
The coherent point drift (CPD) algorithm is a powerful approach for point set registration. However, it suffers from a serious problem-there is a weight parameter w that reflects the assumption about the amount of noise and number of outliers in the Gaussian mixture model, and its value has an influence on the point set registration performance In the(More)
Training set is vary large in two-dimensional (2D) direction estimation which prevents neural network from being widely used in 2D DOA estimation. A dimension-degraded training (DDT) method is proposed in this paper to reduce the training set. In the DDT method, elevation and azimuth are estimated in two separate neural networks respectively. We construct(More)
A contour edge detector combing SUSAN principle and surround suppression is proposed in this paper. Specifically, the operator follows the flow of the Canny edge detector. Firstly, the edge gradient information and modified SUSAN principle are utilized to classify contour edge points and texture edge points approximately. Secondly, surround suppression is(More)
We propose an intermediate computational step, frequency domain filtering of gradient image, to improve contour detection performance of gradient-based edge detectors. This step is inspired by analyzing the spectrum distribution of object contours and texture edges in the frequency domain of gradient image. We illustrate the principle and effect of this(More)
A contour edge detector based on center-surround contrast is proposed in this paper.Specifically, the detector follows the flow of the Canny edge detector while using center-surround contrast instead of gradient magnitude to detect contour edges. Firstly, image gradient is computed in the same way as the traditional Canny detector. Secondly, center-surround(More)
An image registration technique based on the gray joint probability of two correlative images is developed and presented. In this paper, a new matching measurement is proposed, i.e. deviating measurement. This parameter quantifies aligning degree of two images, and an image registration scheme based on deviating measurement is developed. Comparing with(More)
This paper presents a novel framework for activities perception in video surveillance scenarios. Firstly, moving objects are detected by modeling the background using Gaussian Mixture Model (GMM). Secondly, a novel adaptive particle filter (APF) is introduced. The proposed APF has time-varying dimensions and can track multiple moving objects entering or(More)