Pedestrian detection and tracking at crossroads

This paper presents a system that can perform pedestrian detection and tracking using vision-based techniques. A very important issue in the field of intelligent transportation system is to prevent pedestrians from being hit by vehicles. Recently, a great number of vision-based techniques have been proposed for this purpose. In this paper, we propose a vision-based method, which combines the use of a pedestrian model as well as the walking rhythm of pedestrians to detect and track walking pedestrians. Through integrating some spatial and temporal information grabbed by a vision system, we are able to develop a reliable system that can be used to prevent traffic accidents happened at crossroads. In addition, the proposed system can deal with the occlusion problem. Experimental results obtained by executing some real world cases have demonstrated that the proposed system is indeed superb.     

Finite asymmetric generalized Gaussian mixture models learning for infrared object detection

The interest in automatic surveillance and monitoring systems has been growing over the last years due to increasing demands for security and law enforcement applications. Although, automatic surveillance systems have reached a significant level of maturity with some practical success, it still remains a challenging problem due to large variation in illumination conditions. Recognition based only on the visual spectrum remains limited in uncontrolled operating environments such as outdoor situations and low illumination conditions. In the last years, as a result of the development of low-cost infrared cameras, night vision systems have gained more and more interest, making infrared (IR) imagery as a viable alternative to visible imaging in the search for a robust and practical identification system. Recently, some researchers have proposed the fusion of data recorded by an IR sensor and a visible camera in order to produce information otherwise not obtainable by viewing the sensor outputs separately. In this article, we propose the application of finite mixtures of multidimensional asymmetric generalized Gaussian distributions for different challenging tasks involving IR images. The advantage of the considered model is that it has the required flexibility to fit different shapes of observed non-Gaussian and asymmetric data. In particular, we present a highly efficient expectation–maximization (EM) algorithm, based on minimum message length (MML) formulation, for the unsupervised learning of the proposed model’s parameters. In addition, we study its performance in two interesting applications namely pedestrian detection and multiple target tracking. Furthermore, we examine whether fusion of visual and thermal images can increase the overall performance of surveillance systems.     

基于MMPHDF的多机动目标联合检测、跟踪与分类算法

针对场景中存在新目标出现、旧目标消失(即目标数目变化)和密集杂波的复杂情形,利用多模型概率假设密度滤波器(MMPHDF)在多机动目标联合检测与跟踪上的优势,加入类别辅助信息,提出了一种多机动目标联合检测、跟踪与分类算法.该算法的基本思想是在MMPHDF中用属性向量扩展单目标状态向量,用位置和属性的组合测量似然函数代替单目标位置及杂波位置测量似然函数,提高了不同类目标与杂波测量间的鉴别能力,从而改善了目标数目及状态的估计精度;在更新目标状态后,对目标属性信息进行更新,更为精确的目标数目及状态估计又保证了目标分类性能.本文给出了该算法的粒子实现方法.仿真结果验证了上述结论.     

一种快速人眼定位与跟踪算法的实现

提出了一种在复杂背景、光照、姿势变化条件下的人脸眼睛定位与跟踪算法。首先采用基于OpenCV的级联式AdaBoost对象检测算法进行人脸检测,并提出了解决平面和深度旋转的方法;然后采用基于瞳孔定位后的分类器法精确定位眼睛;最后采用卡尔曼粒子滤波算法进行人眼跟踪。实验结果表明,该算法在复杂背景下极大地提高了眼睛定位与跟踪的速度和精确度,并对光照、姿势不敏感。     

红外小目标检测跟踪系统设计

从图像处理方法的角度对红外图像中小目标的检测、跟踪进行了研究,给出了系统总体设计框图,整个检测算法的前端图像滤波预处理采用FPGA实现,识别跟踪采用DSP实现。详细介绍了系统各功能模块的设计方法。实验表明该系统实现了小目标检测跟踪的功能,工作稳定可靠。     

A fast seed detection using local geometrical feature for automatic tracking of coronary arteries in CTA

We propose a fast seed detection for automatic tracking of coronary arteries in coronary computed tomographic angiography (CCTA). To detect vessel regions, Hessian-based filtering is combined with a new local geometric feature that is based on the similarity of the consecutive cross-sections perpendicular to the vessel direction. It is in turn founded on the prior knowledge that a vessel segment is shaped like a cylinder in axial slices. To improve computational efficiency, an axial slice, which contains part of three main coronary arteries, is selected and regions of interest (ROIs) are extracted in the slice. Only for the voxels belonging to the ROIs, the proposed geometric feature is calculated. With the seed points, which are the centroids of the detected vessel regions, and their vessel directions, vessel tracking method can be used for artery extraction. Here a particle filtering-based tracking algorithm is tested. Using 19 clinical CCTA datasets, it is demonstrated that the proposed method detects seed points and can be used for full automatic coronary artery extraction. ROC (receiver operating characteristic) curve analysis shows the advantages of the proposed method.     

Moving object classification using local shape and HOG features in wavelet-transformed space with hierarchical SVM classifiers

This paper proposes an integrated system for the segmentation and classification of four moving objects, including pedestrians, cars, motorcycles, and bicycles, from their side-views in a video sequence. Based on the use of an adaptive background in the red–green–blue (RGB) color model, each moving object is segmented with its minimum enclosing rectangle (MER) window by using a histogram-based projection approach or a tracking-based approach. Additionally, a shadow removal technique is applied to the segmented objects to improve the classification performance. For the MER windows with different sizes, a window scaling operation followed by an adaptive block-shifting operation is applied to obtain a fixed feature dimension. A weight mask, which is constructed according to the frequency of occurrence of an object in each position within a square window, is proposed to enhance the distinguishing pixels in the rescaled MER window. To extract classification features, a two-level Haar wavelet transform is applied to the rescaled MER window. The local shape features and the modified histogram of oriented gradients (HOG) are extracted from the level-two and level-one sub-bands, respectively, of the wavelet-transformed space. A hierarchical linear support vector machine classification configuration is proposed to classify the four classes of objects. Six video sequences are used to test the classification performance of the proposed method. The computer processing times of the object segmentation, object tracking, and feature extraction and classification approaches are 79 ms, 211 ms, and 0.01 ms, respectively. Comparisons with different well-known classification approaches verify the superiority of the proposed classification method.     

A novel feature-based tracking approach to the detection, localization, and 3-D reconstruction of internal defects in hardwood logs using computer tomography

A novel feature-based tracking approach based on the Kalman filter is proposed for the detection, localization, and 3-D reconstruction of internal defects in hardwood logs from cross-sectional computer tomography (CT) images. The defects are simultaneously detected, classified, localized, and reconstructed in 3-D space, making the proposed scheme computationally much more efficient than existing methods where the defects are detected and localized independently in individual CT image slices and the 3-D reconstruction of the defects accomplished via correspondence analysis across the various CT image slices. Robust techniques for defect detection and classification are proposed. Defect class-specific tracking schemes based on the Kalman filter, B-spline contour approximation, and Snakes contour fitting are designed which use the geometric parameters of the defect contours as the tracking variables. Experimental results on cross-sectional CT images of hardwood logs from select species such as white ash, hard maple, and red oak are presented.