基于单固定摄像头的多目标跟踪框架

针对单固定摄像头的视频监控系统对合并遮挡目标跟踪效果不好导致跟踪失败的问题,提出了一个稀疏多目标跟踪框架.该框架对系统的目标的合并遮挡和跟踪滤波这两个部分作了改进.系统由运动目标检测、关联矩阵建立、目标交互处理和滤波四部分组成.首先提取前景区域并建立关联矩阵;然后用关联矩阵判断各目标运动状态并进行相应处理,当目标发生交互时,用TLD算法跟踪,为了提高TLD的跟踪效率和减少TLD的初始化异常情况,用双三次插值对目标和跟踪窗口进行同比例缩放;最后用分数阶卡尔曼滤波对跟踪结果进行滤波.实验结果证明,该框架能有效提高单固定摄像头对目标交互遮挡情况的处理能力.     

An information theoretic approach to camera control for crowded scenes

Navigation and monitoring of large and crowded virtual environments is a challenging task and requires intuitive camera control techniques to assist users. In this paper, we present a novel automatic camera control technique providing a scene analysis framework based on information theory. The developed framework contains a probabilistic model of the scene to build entropy and expectancy maps. These maps are utilized to find interest points which represent either characteristic behaviors of the crowd or novel events occurring in the scene. After an interest point is chosen, the camera is updated accordingly to display this point. We tested our model in a crowd simulation environment and it performed successfully. Our method can be integrated into existent camera control modules in computer games, crowd simulations and movie pre-visualization applications.     

Bayesian modeling of dynamic scenes for object detection

Accurate detection of moving objects is an important precursor to stable tracking or recognition. In this paper, we present an object detection scheme that has three innovations over existing approaches. First, the model of the intensities of image pixels as independent random variables is challenged and it is asserted that useful correlation exists in intensities of spatially proximal pixels. This correlation is exploited to sustain high levels of detection accuracy in the presence of dynamic backgrounds. By using a nonparametric density estimation method over a joint domain-range representation of image pixels, multimodal spatial uncertainties and complex dependencies between the domain (location) and range (color) are directly modeled. We propose a model of the background as a single probability density. Second, temporal persistence is proposed as a detection criterion. Unlike previous approaches to object detection which detect objects by building adaptive models of the background, the foregrounds modeled to augment the detection of objects (without explicit tracking) since objects detected in the preceding frame contain substantial evidence for detection in the current frame. Finally, the background and foreground models are used competitively in a MAP-MRF decision framework, stressing spatial context as a condition of detecting interesting objects and the posterior function is maximized efficiently by finding the minimum cut of a capacitated graph. Experimental validation of the proposed method is performed and presented on a diverse set of dynamic scenes.     

Evaluation of multiple features for violent scenes detection

Multimedia Tools and Applications - Violent scenes detection (VSD) is a challenging problem because of the heterogeneous content, large variations in video quality, and complex semantic meanings of...     

智能车前方目标车辆检测方法研究

为了准确检测智能车前方目标车辆以及确定目标车辆与本车的距离信息,提出了一种应用模糊逻辑融合车辆多特征检测目标车辆的算法.该算法充分考虑到车辆多个特征的重要性,将模糊理论中的模糊隶属度函数和特征置信度值相结合实现目标车辆的验证.在生成目标假设以及提取有效特征点后,通过对不同特征进行模糊化,根据模糊推理融合得到的目标车辆的最终置信度验证车辆假设,实现前方车辆的检测.试验数据表明,该算法具有较好的适应能力和抗干扰能力,能够及时准确地检测前方目标车辆.     

Sparse representation for robust abnormality detection in crowded scenes

In crowded scenes, the extracted low-level features, such as optical flow or spatio-temporal interest point, are inevitably noisy and uncertainty. In this paper, we propose a fully unsupervised non-negative sparse coding based approach for abnormality event detection in crowded scenes, which is specifically tailored to cope with feature noisy and uncertainty. The abnormality of query sample is decided by the sparse reconstruction cost from an atomically learned event dictionary, which forms a sparse coding bases. In our algorithm, we formulate the task of dictionary learning as a non-negative matrix factorization (NMF) problem with a sparsity constraint. We take the robust Earth Mover's Distance (EMD), instead of traditional Euclidean distance, as distance metric reconstruction cost function. To reduce the computation complexity of EMD, an approximate EMD, namely wavelet EMD, is introduced and well combined into our approach, without losing performance. In addition, the combination of wavelet EMD with our approach guarantees the convexity of optimization in dictionary learning. To handle both local abnormality detection (LAD) and global abnormality detection, we adopt two different types of spatio-temporal basis. Experiments conducted on four public available datasets demonstrate the promising performance of our work against the state-of-the-art methods.     

Control strategy for an intelligent mobile vehicle

In this paper recent research into operating an intelligent mobile vehicle (denoted MV) is presented. A complex control procedure, having a two-input and twooutput fuzzy controller as kernel, is used. The inputs to the fuzzy controller are provided by a charge-coupled device (CCD) camera which transforms information on special objects by using an image processor. The MV is capable of tracking objects, searching for objects in space and recognizing a traffic signal. Results of laboratory testing are presented.     

Multipoint infrared laser-based detection and tracking for people counting

Laser devices have received increasing attention in numerous computer-aided applications such as automatic control, 3D modeling and virtual reality. In this paper, aiming at people counting, we propose a novel people detection and tracking method based on the multipoint infrared laser, which can further facilitate intelligent scene modeling and analysis. In our method, a camera with the infrared lens filter is utilized to capture the monitored scene where an array of infrared spots is produced by the multipoint infrared laser. We build a spatial background model based on locations of spots. Pedestrians are detected by clustering of foreground spots. Then, our method tracks and counts the detected pedestrians via inferring the forward–backward motion consistency. Both quantitative and qualitative evaluation and comparison are conducted, and the experimental results demonstrate that the proposed method achieves excellent performance in challenging scenarios.

     

Parametric ego-motion estimation for vehicle surround analysis using an omnidirectional camera

Omnidirectional cameras that give a 360° panoramic view of the surroundings have recently been used in many applications such as robotics, navigation, and surveillance. This paper describes the application of parametric ego-motion estimation for vehicle detection to perform surround analysis using an automobile-mounted camera. For this purpose, the parametric planar motion model is integrated with the transformations to compensate distortion in omnidirectional images. The framework is used to detect objects with independent motion or height above the road. Camera calibration as well as the approximate vehicle speed obtained from a CAN bus are integrated with the motion information from spatial and temporal gradients using a Bayesian approach. The approach is tested for various configurations of an automobile-mounted omni camera as well as a rectilinear camera. Successful detection and tracking of moving vehicles and generation of a surround map are demonstrated for application to intelligent driver support.Received: 1 August 2003, Accepted: 8 July 2004, Published online: 3 February 2005     

基于云台摄像机的自适应无人机跟踪系统

针对当前光电跟踪系统作用距离短、目标跟踪易丢失、视频图像抖动等问题,设计了一种适用于远近距离目标稳定跟踪的系统.首先利用云台摄像机连续采集图像;其次图像处理模块预测目标图像区域进行目标检测;然后利用目标大小进行镜头的变倍变焦自适应控制;最后将视场划分三个区域,根据连续时间目标偏移中心趋势及所在区域位置逐步控制云台的速度...     

基于对称性的快速车辆检测方法

提出了一种基于车辆边缘对称性和水平阴影的快速车辆检测方法.通过利用被检测图像减去道路背景图像的方法对被检测图像进行预处理,以最大限度去除背景图像中车道线等大量干扰,突出车辆图像,对图像分别做Sobel垂直和水平边缘检测.把垂直边缘图像分割放缩成远景、中景、近景3幅大小相同的图像,利用车辆对称性对分割放缩后3幅图像对称性进行检测,确定图像中所有车辆在图像中的水平位置.同时在水平边缘检测的图像中,利用水平阴影和汽车底部水平边缘确定汽车的垂直位置.实验结果表明,该方法简单、有效,能快速、精准地检测出汽车图像中所有的不同位置和不同大小的车辆.     

Fault detection for lateral and vertical vehicle dynamics

Mechatronic systems for vehicles have received increased importance for improving automotive safety and comfort. These systems are designed to aid the driver by preventing unstable or unpredictable vehicle behavior and to stabilize the horizontal and vertical motion of the vehicle. This is achieved by the integration of actuators, sensors and data processing. However, the attained benefits are paralleled by an increase in the complexity of the system requiring enhanced methods for fault detection and diagnosis. Therefore, concepts for model-based fault detection and diagnosis along with sensor fault tolerance are presented and realized for both a vehicle lateral dynamics system and an active suspension system.     

A low-cost near-infrared digital camera for fire detection and monitoring

The human visualization system is not optimally suited for fire detection. Smoke occlusion heavily limits flame visibility and low flames can be difficult to see. Thermal infrared (TIR) sensors mitigate these effects but come at high costs (>$3000) that limit use. Digital cameras modified to record near-infrared (NIR) are potential alternatives and are much more affordable (<$500). We examined the effectiveness of a converted NIR camera for fire detection. Eleven burning slash piles were simultaneously imaged with both a NIR and a camera-sensitive visible or red, green, and blue (RGB) light to compare performance. Quantitative differences in image fire-to-background contrast and flame size were compared between the paired NIR and RGB images. Quantitative analysis was facilitated by Maximum Likelihood Classifier in ENVI. Differences between contrast ratios and flame sizes were assessed for statistical significance by randomization test. Results showed statistically significant (p < 0.01) increases in both contrast ratio and flame size in all NIR images. The low-cost alternative offered by this technology relative to proven thermal sensors is attractive and immediately accessible. NIR sensors will never be as effective as TIR, but at a fraction of the cost, these cameras can fill a significant void in fire line situational awareness.     

Efficient sampling strategy and refinement strategy for randomized circle detection

Circle detection is fundamental in pattern recognition and computer vision. The randomized approach has received much attention for its computational benefit when compared with the Hough transform. In this paper, a multiple-evidence-based sampling strategy is proposed to speed up the randomized approach. Next, an efficient refinement strategy is proposed to improve the accuracy. Based on different kinds of ten test images, experimental results demonstrate the computation-saving and accuracy effects when plugging the proposed strategies into three existing circle detection methods.     

基于改进Faster RCNN的小尺度铁路侵限算法

针对目前传统铁路异物侵线检测算法识别精度不高、对于小尺度目标异物存在漏检等问题,提出一种基于改进Faster RCNN的小尺度铁路侵限算法。在特征提取网络中利用特征金字塔模型将高层特征与低层特征相融合;通过修改锚点框尺寸和增加锚点个数来提高对目标建议区域的精确性;提出一种基于衰减得分的NMS算法;在引入迁移学习思想同时利用在线难例挖掘训练网络以解决数据缺乏、训练难收敛的问题。实验结果表明,改进的Faster RCNN与传统的Faster RCNN网络相比,mAP (mean average precision)提高了2.1%,对小目标的识别有较好准确度。     

Learning spatial-temporal representation for smoke vehicle detection

Multimedia Tools and Applications - Vehicle exhaust emissions are notorious for being unhealthy both for humans and the environment. Smoke vehicle, emitting excess levels of visible black smoke, is...     

In-vehicle camera traffic sign detection and recognition

In this paper, we discuss theoretical foundations and a practical realization of a real-time traffic sign detection, tracking and recognition system operating on board of a vehicle. In the proposed framework, a generic detector refinement procedure based on mean shift clustering is introduced. This technique is shown to improve the detection accuracy and reduce the number of false positives for a broad class of object detectors for which a soft response’s confidence can be sensibly estimated. The track of an already established candidate is maintained over time using an instance-specific tracking function that encodes the relationship between a unique feature representation of the target object and the affine distortions it is subject to. We show that this function can be learned on-the-fly via regression from random transformations applied to the image of the object in known pose. Secondly, we demonstrate its capability of reconstructing the full-face view of a sign from substantial view angles. In the recognition stage, a concept of class similarity measure learned from image pairs is discussed and its realization using SimBoost, a novel version of AdaBoost algorithm, is analyzed. Suitability of the proposed method for solving multi-class traffic sign classification problems is shown experimentally for different feature representations of an image. Overall performance of our system is evaluated based on a prototype C++ implementation. Illustrative output generated by this demo application is provided as a supplementary material attached to this paper.