防止人-车碰撞的交叉口过街行人位置预测

为减少交叉口人-车碰撞事故的发生,利用单目视觉技术和行人横道线特征建立图像像素坐标与实际路面坐标的映射关系,进行行人检测,在获得实时、可靠的过街行人交通参数的基础上采用卡尔曼滤波器预测过街行人的位置,用于判断行人-车辆潜在冲突点,为驾驶员提供行人信息,以便驾驶员采取相应措施保障过街行人的安全;最后进行了相关试验验证。     

基于单视觉的车辆防碰撞预警系统研究

应用单目视觉车辆防碰撞预警系统能够发识别周围车辆并估算出与其之间的距离,利用预警系统及时提醒驾驶员。解决了车道线和车辆的检测识别问题,提出了新的边缘检测算法识别道路线,然后利用特征识别算法识别车辆,并根据车辆之间的距离判定危险等级。实验结果表明,系统能够有效的识别车道线和车辆,并能很好的测量车间距,实现预警输出。     

用于行人头部特征提取的目标区域匹配方法

为了准确地定位与跟踪序列图像中的运动行人以获取精确的客流量信息,提出了一种基于目标区域匹配的行人头部特征提取新方法。与常用的基于致密视差图的头部区域视差获取方法不同,该方法基于“先分割后匹配”的思想,即首先借助单目图像处理方法对基准图进行分割,获取候选头部区域;然后直接将这些候选头部区域作为目标区域,在匹配图中搜索其匹配对应区域以获取候选头部区域的视差;再借助候选头部区域的视差提取出候选头部区域的深度与透视特征,用于去除虚假头部区域以获取最终的头部检测结果。性能测试与实验结果表明,该方法不仅视差提取精度高、实时性好,并且借助该方法获取的头部特征具有较高的区分度,可以有效去除候选头部区域中的虚假头部区域,使客流量检测的准确率达到90%以上。     

基于Kinect的疲劳驾驶综合监测系统研究

基于Kinect传感器研究设计一种驾驶员疲劳状态综合监测系统,通过对Kinect红外图像数据的预处理,减弱了夜晚光照不足的影响;进而利用Kinect提供的人脸识别功能获取驾驶员头部、嘴部、眼部等部位的特征信息,并利用RBF神经网络进行信息融合,分级判断驾驶员的疲劳状态;同时利用滑动平均法及数据库技术,使疲劳状态监测更加准确可靠。模拟实验结果表明,本系统在白天甚至夜晚都能较有效地监测驾驶员疲劳状态。     

基于车载视频的行人预警系统

行车安全一直是社会生活中的研究热点问题之一,本文设计实现一种基于聚合通道特征的实时行人预警系统。系统包括行人检测模块、区域划分模块、单目测距模块和预警模块,其中行人检测模块使用聚合通道特征和级联Adaboost分类器相结合的方法构造通道金字塔,对车载视频中行人进行快速检测并获取目标关键信息和运动属性;单目测距模块利用检测时获取的信息估测人车距离;预警模块利用得到的运动信息判断前方行人的危险程度并给出相应的响应类型。通过使用城市道路条件下的实录视频进行实验,验证了系统的实时性与准确性。      

基于单目视觉的车辆碰撞预警系统

基于单目视觉的车辆碰撞预警系统能够发现道路前方的车辆并估算出与前方车辆之间的距离,利用预警机制及时提醒驾驶员危险状况。车道检测和车辆识别是该系统需要解决的两个主要难题，提出了利用边缘分布函数EDF检测车道标线，利用车辆底部纹理和对称性特征识别车辆，并根据图像坐标系和世界坐标系之间的几何映射关系测距。实验结果表明，提出的方法能够有效检测出车道标线，并能很好地测定与前车的距离。     

基于DSP与ARM技术在汽车防追尾系统中的应用

本系统的研究是将DSP技术与嵌入式ARM技术应用到汽车防尾追系统中。它利用DSP技术在图像信号的处理中突出特点，首创目标运动的模型，进行目标状态的识别和判定，用ARM强大的运算功能完成数据处理识别与判断，准确迅速提供前方车辆信息与本车的距离等并适时预警，自动提醒驾驶员与目标保持安全行驶距离。     

基于ZYNQ的车载行人检测系统设计与实现

为了防止驾驶员因注意力分散而导致的行人交通事故,提出了一种基于ZYNQ的车载行人检测系统设计方案。方案在FPGA部分完成图像采集、缓存和显示,在ARM部分运行梯度方向直方图和支持向量机算法对图像进行行人检测,获取行人的位置,并根据行人位置给出相应的预警信息。该系统实现了行人检测的目的,具有准确度高、体积小、功耗低等特点。     

基于有限状态自动机的人眼开度PERCLOS实现算法

针对驾驶员驾驶过程中因疲劳引起的眼睛开度变化问题, 在原有PERCLOS(percentage of eyelid closure over the pupil over time)标准的基础上, 提出了一种基于有限状态自动机的人眼开度PERCLOS计算方法, 并将其应用到疲劳驾驶预警系统中。该系统首先采用红外摄像头实时获取驾驶员的脸部视频图像, 使用ASM(active shape models)算法进行人脸检测, 在定位到的人脸范围内搜索人眼区域并计算人眼开度, 为了避免人与摄像头距离变化影响计算结果, 对人眼开度进行归一化处理; 然后依据建立的有限状态自动机模型计算PERCLOS值; 最后根据制定的预警机制实现基于人眼开度的疲劳预警。实验结果表明本方法能够实时监测驾驶员疲劳状况, 具有对光照变化、脸部配饰不敏感的特点。     

基于HOG特征的行人视觉检测方法

行人检测是近年来计算机视觉领域中备受关注的前沿方向和研究热点.以单目视觉传感器作为外界环境信息获取的主要手段,建立了一个包含行人分割、识别的检测系统.根据行人特有的一些特征,提出了基于垂直边缘和边缘对称性的行人分割方法,并进行精确定位.在行人识别阶段利用HOG特征进行特征提取,然后利用线性支持向量机进行行人识别.对大量...     

An integrated driver warning system for driver and pedestrian safety

A semi-integrated system for driver assistance and pedestrian safety is presented. This system is composed of a single camera which focuses on the driver for picking up visual cues and a stereo rig that focus on the road ahead for the detection of road obstructions and pedestrians. While the car is in motion, the driver's viewing direction is obtained and analyzed along with information of road condition and any moving vehicle ahead in order to determine if the current driving condition is safe. In addition, when the vehicle is moving slowly, the system can also detect the existence of a pedestrian ahead and warns the driver if the pedestrian moves in front of the car. This system contains algorithm-based safety analysis as well as fuzzy rules-based analysis for interaction between variables. Our experimental results show that the condition for driver safety can be accurately classified in 94.5% of the tested driving conditions, and the pedestrians can be identified in 93.18% of the tested cases. These were compared to the results of similar systems and shown to be superior.     

结合对象分割的运动行人检测

目标检测大量应用于监控系统的行人检测以及人脸识别,是当前深度学习的研究热点.监督学习利用人工标注大量数据集训练出针对特定场景的行人检测器.但是人工标注方法费时费力,本文针对监督学习需要人工标注数据集的缺点,研究了一种半自动标注行人的方法.针对静止的单目摄像机拍摄的监控视频,利用光流信息提供的初始前景可能性,以及跨越时间的视觉相似性来迭代地更新初始的前景可能性,分割出运动的行人,根据分割的前景对象,提出了一种半自动标注行人的方法.实验结果显示,本文的方法可以为行人检测系统提供大量数据集,且效率上明显优于传统人工标注的方法.     

A genetic fuzzy system to model pedestrian walking path in a built environment

A study on the pedestrian’s steering behaviour through a built environment in normal circumstances is presented in this paper. The study focuses on the relationship between the environment and the pedestrian’s walking trajectory. Owing to the ambiguity and vagueness of the relationship between the pedestrians and the surrounding environment, a genetic fuzzy system is proposed for modelling and simulation of the pedestrian’s walking trajectory confronting the environmental stimuli. We apply the genetic algorithm to search for the optimum membership function parameters of the fuzzy model. The proposed system receives the pedestrian’s perceived stimuli from the environment as the inputs, and provides the angular change of direction in each step as the output. The environmental stimuli are quantified using the Helbing social force model. Attractive and repulsive forces within the environment represent various environmental stimuli that influence the pedestrian’s walking trajectory at each point of the space. To evaluate the effectiveness of the proposed model, three experiments are conducted. The first experimental results are validated against real walking trajectories of participants within a corridor. The second and third experimental results are validated against simulated walking trajectories collected from the AnyLogic® software. Analysis and statistical measurement of the results indicate that the genetic fuzzy system with optimised membership functions produces more accurate and stable prediction of heterogeneous pedestrians’ walking trajectories than those from the original fuzzy model.     

Design and research of automobile anti-collision warning system based on monocular vision sensor with license plate cooperative target

At present, detection method for the target vehicle based on monocular vision sensor uses the whole vehicle as targets. The automobile anti-collision technology proposed in this paper adopts monocular vision sensor for automobile measurement based on vehicle license plate cooperative target. Monocular vision sensor has advantages of good real-time performance and low cost. The technique can improve the detection capability of vehicle collision avoidance warning systems. In addition to the target vehicle positioning, it can also realize attitude determination. This technology eliminates the limits of road surface roughness and fluctuation. This paper designs the realization scheme of collision warning system based on monocular vision sensor from the automobile license plate cooperative target. Technology roadmap of automobile collision warning system is given. In this paper, license plate frame location is as the research background. The paper presents an analytic solution of positioning method for the license plate frame image. The method uses four vertex characteristics of license plate frame image to locate. Positioning speed of the method is fast. And it has a unique solution. This method can be used to positioning for the license plate frame. Simulation experiment is done for the collision warning location. The simulation results show that this method can locate the position for license plate frame image. License plate is regular shape, uniform, with identity recognition function markers on the automobile body. In the previous research on automotive collision warning and intelligent vehicle, we have not seen the research methods similar to the method introduced in this paper. The research enriches automobile anti-collision technology and theory of intelligent vehicle technology. It can also provide an auxiliary method for navigation and obstacle avoidance research for unmanned vehicle. It has certain scientific significance. Vehicle collision warning system can help the driver judgment, prompting warning, improving driving safety, and has broad application prospects.     

Deep infrared pedestrian classification based on automatic image matting

Infrared pedestrian classification plays an important role in advanced driver assistance systems. However, it encounters great difficulties when the pedestrian images are superimposed on a cluttered background. Many researchers design very deep neural networks to classify pedestrian from cluttered background. However, a very deep neural network associated with a high computational cost. The suppression of cluttered background can boost the performance of deep neural networks without increasing their depth, while it has received little attention in the past. This study presents an automatic image matting approach for infrared pedestrians that suppresses the cluttered background and provides consistent input to deep learning. The domain expertise in pedestrian classification is applied to automatically and softly extract foreground objects from images with cluttered backgrounds. This study generates trimaps, which must be generated manually in conventional approaches, according to the estimated positions of pedestrian’s head and upper body without the need for any user interaction. We implement image matting by adopting the global matting approach and taking the generated trimap as an input. The representation of pedestrian is discovered by a deep learning approach from the resulting alpha mattes in which cluttered background is suppressed, and foreground is enhanced. The experimental results show that the proposed approach improves the infrared pedestrian classification performance of the state-of-the-art deep learning approaches at a negligible computational cost.     

Tracking Pedestrians Under Occlusion in Parking Space

Many traffic accidents occur in parking lots. One of the serious safety risks is vehicle-pedestrian conflict. Moreover, with the increasing development of automatic driving and parking technology, parking safety has received significant attention from vehicle safety analysts. However, pedestrian protection in parking lots still faces many challenges. For example, the physical structure of a parking lot may be complex, and dead corners would occur when the vehicle density is high. These lead to pedestrians’ sudden appearance in the vehicle’s path from an unexpected position, resulting in collision accidents in the parking lot. We advocate that besides vehicular sensing data, high-precision digital map of the parking lot, pedestrians’ smart device’s sensing data, and attribute information of pedestrians can be used to detect the position of pedestrians in the parking lot. However, this subject has not been studied and explored in existing studies. To fill this void, this paper proposes a pedestrian tracking framework integrating multiple information sources to provide pedestrian position and status information for vehicles and protect pedestrians in parking spaces. We also evaluate the proposed method through real-world experiments. The experimental results show that the proposed framework has its advantage in pedestrian attribute information extraction and positioning accuracy. It can also be used for pedestrian tracking in parking spaces.     

Locality guided cross-modal feature aggregation and pixel-level fusion for multispectral pedestrian detection

Multispectral pedestrian detection has received much attention in recent years due to its superiority in detecting targets under adverse lighting/weather conditions. In this paper, we aim to generate highly discriminative multi-modal features by aggregating the human-related clues based on all available samples presented in multispectral images. To this end, we present a novel multispectral pedestrian detector performing locality guided cross-modal feature aggregation and pixel-level detection fusion. Given a number of single bounding boxes covering pedestrians in both modalities, we deploy two segmentation sub-branches to predict the existence of pedestrians on visible and thermal channels. By referring to the important locality information in the reference modality, we perform locality guided cross-modal feature aggregation to learn highly discriminative human-related features in the complementary modality by exploring the clues of all available pedestrians. Moreover, we utilize the obtained spatial locality maps to provide prediction confidence scores in visible and thermal channels and conduct pixel-wise adaptive fusion of detection results in complementary modalities. Extensive experiments demonstrate the effectiveness of our proposed method, outperforming the current state-of-the-art detectors on both KAIST and CVC-14 multispectral pedestrian detection datasets.     

基于视野域机制的行人轨迹预测

为提高行人在复杂交通场景中交互的安全性,提出一种基于social-GAN（social-generative adversarial network）的行人轨迹预测算法SAN-GAN（social angle norm-GAN）。该算法首先以行人历史位置信息与头部信息为输入,通过轨迹生成器LSTM网络（long short term memory networks）获取行人隐藏特征信息,并基于行人视野域模块捕捉行人视野域动态变化,对所有行人建立扇形视野域并筛选有效信息,从而驱动神经网络模型预测行人未来轨迹变化。将SAN-GAN与LSTM、social-LSTM（social-long short term memory networks）、social-GAN等轨迹预测算法进行对比实验,结果表明SAN-GAN算法相较于其他算法,在预测3.2 s的行人轨迹时,ADE分别平均降低65.8%、51.2%、10.7%,FDE分别平均降低73.6%、60.9%、10.4%。SAN-GAN能够有效地预测行人在复杂交通环境中进行交互的未来轨迹。