On-line vehicle and pedestrian detections based on sign pattern

Real time detection methods of moving vehicles and pedestrians for navigation of the mobile robot are proposed. The method is based on a locomotion strategy, viz. signature-based stereotype motion. Signature of the moving vehicle is the shadow underneath the vehicle which is darker than any other parts of the asphalt paved road. Signature of the pedestrian is rhythm of walking. Rhythm of walking is unique to the pedestrian, and not influenced by time, weather, sunlight, shadow, and distance. Moreover, it is independent from clothes the pedestrian puts on. The result of experiments verify the validity of the methods     

一种基于多源触发行人过街控制系统研究

文章结合实际项目,研究一种基于多源触发的行人过街控制系统,解决目前很多路段行人过街路口行人等待时间过长和对机动车干扰较大等问题。系统通过机动车检测器、行人过街按钮、热成像行人等待检测器等多种检测器,采集不同交通流数据,进行综合判断决策,在兼顾绿波情况下,有效减少行人过街等待时间,缓解由于行人过街造成的交通拥堵,同时,减少机动车无效等待时间,缓解驾驶员的焦虑情绪,提高道路通行能力。     

Robust step detection method for pedestrian navigation systems

A robust step detection method for pedestrian navigation systems is presented. The method utilises a two-axis accelerometer and is based on an acceleration pattern of a step. Experimental results show that the method gives robust performance for ground surface variation and walking behaviour.     

基于惯性传感器的动态步长算法研究

行人步态参数的精确估计是行人自主导航系统和行人健康监测的关键技术之一。针对当前行人自主导航系统中步长估算算法精度低和弱适应性的问题,提出了一种计算行人动态步长算法。首先对行人的步态特征进行分解,利用改进的零速检测确定行人运动状态,采用卡尔曼滤波技术降低惯性传感器中累积误差的影响,再对进行滤波和坐标转换后的加速度进行双重积分,最终得到行人脚尖的运动轨迹。通过采用MTI-700惯性模块设计实验并进行实验验证。结果表明,该文提出的步长算法计算的步长与行人实际步长的误差低于3.0%。与现有的行人动态步长算法相比,该算法首次计算出行人脚尖的运动轨迹,精度较高且适应强,在行人自主导航及行人健康监测领域具有较大的应用价值。     

基于步行者航位推算的井下人员辅助定位

针对目前井下人员定位系统存在的问题,提出一种基于步行者航位推算的辅助定位方法。利用低成本的惯性测量单元(IMU)和磁力计,设计稳定的姿态与航向参考系统(AHRS),利用惯性导航的相关理论,并通过分析人员步行姿态,进行零速修正(ZUPT),组成步行者航位推算系统,对井下人员进行实时航位推算。以实验室大楼走廊模拟井下矿道环境进行航位推算实验,实验结果表明,本方法对人员行走距离和方向做出良好的推算,能成为现有井下定位的有效补充,提高人员定位精确度。     

行人动态实时监测系统的设计与实现

针对视频场景中行人动态信息监测的需求,设计实现了一种行人动态实时监测系统。首先,通过YOLOv3检测算法对场景中行人进行目标检测,在此基础上结合改进的KCF实现多行人目标的跟踪并获取对应下底边中心点。之后结合场景标定结果完成行人图像与三维空间位置监测、场景行人计数和行人行走速度等动态信息监测。通过实验表明,该系统不仅能够较好地完成视频场景下行人目标检测与跟踪,也能够精准完成以上信息动态实时监测的任务,为实际理论研究与工程应用奠定重要基础。     

Deep Convolutional Neural Networks for pedestrian detection

Pedestrian detection is a popular research topic due to its paramount importance for a number of applications, especially in the fields of automotive, surveillance and robotics. Despite the significant improvements, pedestrian detection is still an open challenge that calls for more and more accurate algorithms. In the last few years, deep learning and in particular Convolutional Neural Networks emerged as the state of the art in terms of accuracy for a number of computer vision tasks such as image classification, object detection and segmentation, often outperforming the previous gold standards by a large margin. In this paper, we propose a pedestrian detection system based on deep learning, adapting a general-purpose convolutional network to the task at hand. By thoroughly analyzing and optimizing each step of the detection pipeline we propose an architecture that outperforms traditional methods, achieving a task accuracy close to that of state-of-the-art approaches, while requiring a low computational time. Finally, we tested the system on an NVIDIA Jetson TK1, a 192-core platform that is envisioned to be a forerunner computational brain of future self-driving cars.     

一种多尺度远红外夜间交通行人识别方案

为降低夜间行车导致交通事故的概率,在能见度不理想的行车环境中为车辆提供主动安全系统,依据汽车辅助驾驶系统的基本要求,基于远红外传感技术,设计出夜间行人辅助模型。该模型通过远红外传感器来捕捉原始的数据源,利用灰度统计技术获取ROIs,构造出多尺度概率模板,在此基础上对数据源进行匹配检测,通过多帧校验综合处理技术将模型的漏检率和检测率进一步改善。实验表明,该模型的概率模板在匹配精度上相对于业内常用方法有了较大程度的改善,此外该模型还能够在郊区和市区两种交通路况下使用,实用性较好。     

红外交通场景下遮挡行人目标检测算法研究

针对交通十字路口等视野盲区往来行人间存在遮挡情况,如何高效准确地检测复杂道路中目标行人具有实际意义。为了实现夜间交汇路口场景行人检测,提出一种基于改进YOLOv5的行人目标检测算法,采用Non local和PSA模块对YOLOv5原网络的Bottleneck CSP进行改进,能够有效弥补遮挡中行人特征的帧间信息交互过程,增强长程范围通道特征依赖关系。设计更深的160×160检测层和自适应anthor,提升夜间行人检测的边界回归精确度。实验结果表明,针对夜间下交通路口场景,压缩改进后模型对行人检测鲁棒性高,相较于原始算法mAP_0.5和mAP_0.5:0.95值分别提升了14.2和12.7,说明所提算法对夜间行人检测的有效性。     

改进YOLOv4算法的复杂视觉场景行人检测方法

复杂视觉场景下存在过暗或者过曝的光照、恶劣的天气、严重遮挡、行人尺寸差别大以及图像模糊等问题,大大增加了行人检测的难度。因此,针对复杂视觉场景下行人检测准确度低、漏检严重的问题,提出了改进的YOLOv4算法以增强复杂视觉场景下的行人检测效果。首先,构建复杂视觉场景下的行人数据集。然后,在主干网中加入混合空洞卷积,提高网络对行人特征的提取能力。最后,提出空间锯齿空洞卷积结构,代替空间金字塔池化结构,获取更多细节特征。实验表明,在本文构建的行人数据集上,改进后的 YOLOv4算法的平均精度（average precision,AP）达到了90.08%,相比原YOLOv4算法提高了7.2%,对数平均漏检率（log-average miss rate,LAMR）降低了13.69%。     

融合多尺度信息和跨维特征引导的轻量行人检测

针对复杂道路场景下行人检测精度与速度难以提升的问题,提出一种融合多尺度信息和跨维特征引导的轻量级行人检测算法。首先以高性能检测器YOLOX为基础框架,构建多尺度轻量卷积并嵌入主干网络中,以获取多尺度特征信息。然后设计了一种端到端的轻量特征引导注意力模块,采用跨维通道加权的方式将空间信息与通道信息融合,引导模型关注行人的可视区域。最后为减少模型在轻量化过程中特征信息的损失,使用增大感受野的深度可分离卷积构建特征融合网络。实验结果表明,相比于其他主流检测算法,所提算法在KITTI数据集上达到了71.03%的检测精度和80 FPS的检测速度,在背景复杂、密集遮挡、尺度不一等场景中都具有较好的鲁棒性和实时性。     

Vehicle detection and inter-vehicle distance estimation using single-lens video camera on urban/suburb roads

This paper presents a driver assistance system for vehicle detection and inter-vehicle distance estimation using a single-lens video camera on urban/suburb roads. The task of vehicle detection on urban/suburb roads is more challenging due to their high scene complexity. In this work, the still area of frame inside the host vehicle is first removed using temporal differencing, followed by detecting vanishing point. Segmentation of road regions is then conducted using vanishing point and road’s edge lines. Shadow regions at the bottoms of vehicles verified using the HOG feature and an SVM classifier are utilized to detect vehicle positions. The distances between the host and its front vehicles are estimated based on the locations of detected vehicles and vanishing point. Experimental results show varied performance of vehicle detection with different scenes of urban/suburb roads and the detection rate can achieve up to 94.08%, indicating the feasibility of the proposed method.     

Vehicle and pedestrian detection method based on improved YOLOv4-tiny

Aiming at the problem of low detection accuracy of vehicle and pedestrian detection models, this paper proposes an improved you only look once v4 (YOLOv4)-tiny vehicle and pedestrian target detection algorithm. Convolutional block attention module (CBAM) is introduced into cross stage partial Darknet-53 (CSPDarknet53)-tiny module to enhance feature extraction capabilities. In addition, the cross stage partial dense block layer (CSP-DBL) module is used to replace the original simple convolutional module superposition, which compensates for the high-resolution characteristic information and further improves the detection accuracy of the network. Finally, the test results on the BDD100K traffic dataset show that the mean average precision (mAP) value of the final network of the proposed method is 88.74%, and the detection speed reaches 63 frames per second (FPS), which improves the detection accuracy of the network and meets the real-time detection speed.     

Driver Behavior and Situation Aware Brake Assistance for Intelligent Vehicles

This paper deals with the development of Human-Centric Intelligent Driver Assistance Systems. Rear-end collisions account for a large portion of traffic accidents. To help mitigate this problem, predictive braking systems and adaptive cruise control systems have been developed. However, these types of systems usually rely solely on the vehicle and vehicle surround sensors, either ignoring the human component of driving or learning the driver's control behavior using only these sensors. As with all human-computer interfaces, this has the potential to work against the driver, distract the driver further, or even annoy the driver so that the driver ignores or disables the system. It is, therefore, important to directly take the driver's intended actions into account when designing a driver assistance system. By using a probabilistic model for the system, warnings and preventative measures can be constructed based on varying levels of situational severity and driver attentiveness and intent. The research is based upon carefully conducted experimental trials involving a human subjects driving in natural manner and on typical freeways in the USA. The experiments, designed by inputs from cognitive scientist, were conducted in a specially designed instrumented vehicle to record important cues associated with driver's behavior, vehicle state, and vehicle surround in a synchronized manner. Quantitative results and analysis of the experimental trials are presented to show the feasibility and promise of this framework to predict the driver's intent to brake, the need for braking given the current situation, and at what level the driver should be warned     

基于YOLO模型的红外图像行人检测方法

针对基于传统特征提取方法的远红外图像行人检测存在准确率和实时性不足的问题,本文研究了一种基于改进YOLO模型的远红外行人检测方法,通过改进其深度卷积神经网络的输入分辨率,然后在基于实际道路采集的红外数据集上进行训练,得到检测效果最佳的检测模型,并提出基于车速的自适应图像分辨率模型,以提高车载系统的行人检测性能。在基于实际道路的红外数据集上的对比实验表明,该方法与传统方法相比,准确率从76.5%提高到89.2%,每秒传输帧数从0.01259 f/s提高到40.5 f/s,满足车载情况下的实时性需求。     

深度学习与图像融合的行人检测算法研究

为解决智能辅助驾驶技术中可见光摄像机受光照和气候影响而导致行人目标识别困难的问题。通过研究图像融合技术,结合深度卷积神经网络,实现并改进了一种道路行人目标检测算法。方法是利用多源传感器图像融合技术,采用可见光相机与红外热成像相机融合的策略,以Faster RCNN算法为基础,从改进网络结构、特征融合、优化模型训练等方面展开研究,对复杂环境下的行人检测与定位跟踪展开研究,提出一种基于图像融合技术和改进的深度卷积神经网络的道路行人目标检测算法。实验结果表明,该算法对复杂气候环境下行人目标检测提高了检测效率和准确率,增加了智能辅助驾驶汽车的安全性。     

基于YOLOv5算法的红外图像行人检测研究北大核心CSCD

目标检测是自动驾驶的重要前提,是与外界信息交互的重要环节。针对夜间远处行人检测识别精度低、漏检的问题,提出一种针对检测小尺寸行人的YOLOv5-p4的夜间行人识别模型。首先,通过增加更小目标的检测层,引入BiFPN特征融合机制,防止小目标被噪声淹没,使网络模型可以更聚焦于物体的细小特征;同时使用K-means先验框聚类出更小目标的锚框,并且使用了多尺度的数据增强方法,增加模型的鲁棒性。使用了MetaAcon-C激活函数与EIoU回归损失函数使模型收敛效果更好,提升了算法远距离行人的检测的准确率。最后在红外行人数据集FLIR上验证改进后的YOLOv5-p4模型对于行人的检测能力,实验结果表明该方法与传统方法相比,准确率从86.9%提升到90.3%,适合用于红外图像中的行人检测。     

Monocular precrash vehicle detection: features and classifiers.

Robust and reliable vehicle detection from images acquired by a moving vehicle (i.e., on-road vehicle detection) is an important problem with applications to driver assistance systems and autonomous, self-guided vehicles. The focus of this work is on the issues of feature extraction and classification for rear-view vehicle detection. Specifically, by treating the problem of vehicle detection as a two-class classification problem, we have investigated several different feature extraction methods such as principal component analysis, wavelets, and Gabor filters. To evaluate the extracted features, we have experimented with two popular classifiers, neural networks and support vector machines (SVMs). Based on our evaluation results, we have developed an on-board real-time monocular vehicle detection system that is capable of acquiring grey-scale images, using Ford's proprietary low-light camera, achieving an average detection rate of 10 Hz. Our vehicle detection algorithm consists of two main steps: a multiscale driven hypothesis generation step and an appearance-based hypothesis verification step. During the hypothesis generation step, image locations where vehicles might be present are extracted. This step uses multiscale techniques not only to speed up detection, but also to improve system robustness. The appearance-based hypothesis verification step verifies the hypotheses using Gabor features and SVMs. The system has been tested in Ford's concept vehicle under different traffic conditions (e.g., structured highway, complex urban streets, and varying weather conditions), illustrating good performance.     

Pedestrian trajectory prediction in crossing scenario using fuzzy logic and switching Kalman filter

Pedestrian trajectory prediction plays an important role in bothadvanced driving assistance system (ADAS) and autonomous vehicles. An algorithm for pedestrian trajectory prediction in crossing scenario is proposed. To obtain features of pedestrian motion, we develop a method for data labelling and pedestrian body orientation regression. Using the hierarchical features as domain of discourse, fuzzy logic rules are built to describe the transition between different pedestrian states and motion models. With derived probability of each type of motion model we further predict the pedestrian trajectory in the next 1.5 s using switching Kalman filter (KF). The proposed algorithm is further verified in our dataset, and the result indicates that the proposed algorithm successfully predicts pedestrian' s crossing behavior 0.4 s earlier before pedestrian moves. Meanwhile, the precision of predicted trajectory surpasses other methods including interacting multi-model KF and dynamic Bayesian network (DBN).     

基于上下文信息的监控场景行人检测方法

为了提高监控场景中行人检测的准确度,提出了一种基于上下文信息的行人检测方法.该方法将监控场景的上下文信息融入到卷积神经网络中,选择性地学习对行人检测有帮助的上下文信息.首先,利用一个截断的卷积神经网络提取输入图像的多张特征图.然后,将多张特征图通过两个包含上下文信息的卷积层,形成一张掩码图.最后,通过在掩码图上估计行人的边界框,获得行人检测的结果.实验表明,该方法能实现监控场景中准确且快速的行人检测.