Adaptive visual target detection and tracking using weakly supervised incremental appearance learning and RGM-PHD tracker

Multiple visual target tracking is a challenging problem due to various uncertainties including occlusion, miss-detection and noisy measurement. Most tracking approaches utilize an object-specific detector, pre-trained on many labeled images, to provide suitable measurements for their tracking system. In this paper, we use a simple background subtraction detector which only needs the background image to localize targets independent of their shape or type. In order to cope with the uncertainties resulted by the detector, we propose an adaptive appearance model and develop an incremental appearance learning algorithm to learn the target appearances in time. The proposed method employs the background information and our defined keypoints’ miss-matched history to adapt the target appearances within different frames. Furthermore, we combine Refined Gaussian Mixture Probability Hypothesis Density (RGM-PHD) tracker with the detectors to keep target trajectories and handle uncertainties. The experiments conducted on several video datasets show the effectiveness of our proposed method.     

A velocity program using the Kanade–Lucas–Tomasi feature-tracking algorithm with demonstration for pressure and electroosmosis conditions

Investigating microfluidic flow profiles is of interest in the microfluidics field for the determination of various characteristics of a lab-on-a-chip system. Microparticle tracking velocimetry uses computational methods upon recording video footage of microfluidic flow to ultimately visualize motion within a microfluidic system across all frames of a video. Current methods are computationally expensive or require extensive instrumentation. A computational method suited to microparticle tracking applications is the robust Kanade–Lucas–Tomasi (KLT) feature-tracking algorithm. This work explores a microparticle tracking velocimetry program using the KLT feature-tracking algorithm. The developed program is demonstrated using pressure-driven and EOF and compared with the respective mathematical fluid flow models. An electrostatics analysis of EOF conditions is performed in the development of the mathematical using a Poisson's Equation solver. This analysis is used to quantify the zeta potential of the electroosmotic system. Overall, the KLT feature-tracking algorithm presented in this work proved to be highly reliable and computationally efficient for investigations of pressure-driven and EOF in a microfluidic system.     

基于实际路网情境的配送车辆调度优化

在实际路网情境下结合车道数、车道宽度、路口信号灯设置等路网物理特性,构建了考虑综合交通阻抗的多车型车辆调度模型,提出了两阶段求解策略:第1阶段设计了改进A-star精确解算法用于计算客户时间距离矩阵;第2阶段针对实际路网的特征设计了混合模拟退火算法求解调度方案。以大连市某配送中心运营实例进行路网情境仿真试验,结果表明:改进A-star算法较改进Dijkstra算法具有更短的路径搜索时间;混合模拟退火算法求解结果较实际调度方案优化了13.1% 的综合成本;路网增流、区域拥堵和路段禁行三类路网情境均能对配送方案的车辆配置、路径选择、客户服务次序、作业时间和违约费用等5方面内容产生干扰,调度计划的制定需要详细考虑这些因素的变化。     

Exact 3D simulation of Scanning Electron Microscopy images of semiconductor devices in the presence of electric and magnetic fields

Simulations of Scanning Electron Microscopy images of semiconductor devices in the presence of electric fields are usually too simplistic, since they just rely on approximated solutions of the Poisson equation. In this paper, the 3D Poisson equation is solved in a TCAD environment, which accounts for realistic boundary conditions, as well as for complex physical effects like the formation of space charge regions in semiconductors and the polarization of dielectrics. The calculated solution is then passed to a Monte Carlo code that implements a new electron tracking engine optimized for speed, stability, and accuracy. After introducing the new tracking engine, three simulation examples are presented dealing with the presence of an extraction field, self-charging of the irradiated sample, and potential contrast in a biased silicon junction.     

Multinational vehicle license plate detection in complex backgrounds

Many methods for multinational License Plate Detection (LPD) have been proposed in recent times but most of them are not sophisticated enough to handle complex backgrounds. Moreover, their ability to handle various environmental and illumination conditions has been limited and still needs improvement. In this paper, we propose a novel technique to detect license plates of vehicles regardless of their color, size, and content. As the rear vehicle lights are an essential part of any vehicle, we reduce the image processing area to eliminate the complex background by detecting the rear-lights as the license plates are in a certain range of these lights. Heuristic Energy Map (HEM) of the vertical edge information in the Region of Interest (ROI) is calculated and area with the dense edges is selected using a unique histogram approach which is considered to be the license plate. The proposed algorithm is tested on 855 images from various countries including China, Pakistan, Serbia, Italy and various states of America. Experimental results show that the proposed method is able to detect license plates 90.4% of times despite of complex backgrounds in 0.25 s on average that can achieve real time performance.     

Salient video cube guided nighttime vehicle braking event detection

In nighttime driving brake lights are particularly important because they offer a warning signal to prevent potential collisions. In this work, we propose a novel visual-based approach for nighttime brake light detection using three-dimensional Nakagami imaging to analyze tail lights of vehicles in front. Rather than heuristic features, such as symmetry of taillights and appearance of the third brake light, the proposed approach extracts invariant features by modeling the scattering of brake lights, thus allowing detection to proceed in a part-based manner. Experiments from extensive datasets show that the proposed system can effectively detect vehicle braking under different lighting and traffic conditions, making it a realistic option for real-world applications.     

Robust object tracking based on adaptive templates matching via the fusion of multiple features

Moving object tracking under complex scenes remains to be a challenging problem because the appearance of a target object can be drastically changed due to several factors, such as occlusions, illumination, pose, scale change and deformation. This study proposes an adaptive multi–feature fusion strategy, in which the target appearance is modeled based on timed motion history image with HSV color histogram features and edge orientation histogram features. The variances based on the similarities between the candidate patches and the target templates are used for adaptively adjusting the weight of each feature. Double templates matching, including online and offline template matching, is adopted to locate the target object in the next frame. Experimental evaluations on challenging sequences demonstrate the accuracy and robustness of the proposed algorithm in comparison with several state-of-the-art algorithms.     

一种基于标准化人眼图像的视线估计算法

本文提出了一种基于标准化人眼图像的视线估计算法,能有效解决头部运动对视线估计的影响.本算法在一个红外摄像机和三个红外光源的条件下,利用三个光斑(光源在角膜表面的反射点)的图像信息,将人眼图像变换到标准化人眼图像.通过引入标准化人眼图像,使屏幕平面到标准化人眼图像平面的射影变换为定值,从而使标准化人眼图像的瞳孔中心和屏幕上的视点的变换关系为定值,由此解决了头部运动对视线估计的影响.实验结果验证了该算法在头部运动情况下的有效性.     

基于视频多目标跟踪的高度测量算法

本文提出了一种基于视频多目标跟踪的高度测量算法.首先,采用码本模型检测前景,利用图割理论实现对多目标的跟踪.然后,提取每一帧中目标的头部特征点和脚部特征点,根据投影几何的约束关系计算每个目标的高度.最后,融合多帧的测量结果进行数据优化.本文提出的算法不需要对相机进行完全标定,只需计算摄像机的灭点和地平面的灭线,降低了计算的复杂度.实验结果表明,本文提出的算法具有较高的测量精度,对遮挡和运动状态变化具有较强的鲁棒性,同时,能够满足实时性要求.