检索结果-内蒙古大学图书馆

Mixed-attention-based regional soft partition network for vehicle reidentification

IET IMAGE PROCESSING 2022年第13期16卷 3648-3658页

作者： Li, Zhiyong Luo, Yunzhong Li, Qiaochu Song, Lulu Liu, Weiyi Sichuan Agr Univ Coll Informat Engn Yaan 625000 Sichuan Peoples R China Sichuan Key Lab Agr Informat Engn Yaan Sichuan Peoples R China JD Technol JD iCity Beijing Peoples R China JD Intelligent Cities Res Beijing Peoples R China

The purpose of vehicle reidentification is to distinguish the same vehicle from different vehicles using different cameras. The main challenge of this task is the significant intra-instance discrepancy of the same vehicle from different views and the subtle inter-instance differences of similar vehicles from the same views. To address this problem, researchers have attempted to align features from different views, such as using additional metadata (colour, type, key points, mask, etc.) to improve performance. Although these attempts improve the performance of vehicle reidentification, considerable efforts are required to create additional precise annotations to the data before using these methods. This results in expensive data preparation costs, rendering these methods less convenient to use. Therefore, we propose a novel deep learning network, called a mixed-attention-based regional soft partition network, to address this problem. This network does not require additional metadata annotations;it only trains the identity label as a supervision signal and uses soft partition attention to identify specific vehicle regions. Experiments showed that the performance of the proposed method was comparable to that of the state-of-the-art method with additional annotations on the VeRri-776 and VERI-Wild datasets.

关键词： specific vehicle regions deep learning (artificial intelligence) meta data object detection soft partition attention deep learning network traffic engineering computing additional precise annotations mixed-attention-based regional soft partition network similar vehicles vehicle reidentification learning (artificial intelligence) subtle inter-instance differences additional metadata annotations significant intra-instance discrepancy feature extraction

来源：评论

学校读者我要写书评

暂无评论

Reinforcement learning based mainline dynamic speed limit adjustment of expressway off-ramp upstream under connected and autonomous vehicles environment

引用

IET INTELLIGENT TRANSPORT SYSTEMS 2022年第12期16卷 1809-1819页

作者： Xiao, Daiquan Kang, Shengyang Xu, Xuecai Shen, Zhenwu Huazhong Univ Sci & Technol Sch Civil & Hydraul Engn Wuhan 430074 Peoples R China Shenzhen Urban Transport Planning Ctr Co Ltd Shenzhen Peoples R China Wuhan Huake Quanda Transport Planning & Design Co Wuhan Peoples R China

With the rapid progress of urbanization and continuous increasing of automobiles, expressway on- and off-ramp area becomes the bottleneck, and recurrent congestion occurs frequently. In order to solve the problem of traffic jam caused around off-ramps, various methods have been employed. Among all, mainline variable speed limit (VSL) control accounts for some proportion. In this study, mainline VSL adjustment of off-ramp upstream is investigated with the reinforcement learning algorithm under the connected vehicles environment to alleviate the traffic congestion. First, the assumptions are made to be suitable for the traffic conditions of mainline VSL control on off-ramp upstream, and then VSL algorithm based on reinforcement learning is presented, and Q-learning is chosen as the main algorithm. Next, the state space, action space, and reward function required by Q-learning are constructed orderly, and the related parameters are labelled. After that, according to the platform based on Python and VISSIM, three schemes, free control (Scheme 0), mainline VSL adjustment of off-ramp upstream based on rule (Scheme 1), and mainline VSL adjustment of off-ramp upstream based on Q-learning algorithm (Scheme 2), are designed, and the three schemes are simulated and compared quantitatively to reflect the off-ramp travel efficiency. The results indicate that mainline dynamic VSL adjustment of off-ramp upstream based on Q-learning algorithm performs the best in terms of general and specific indexes. The results provide potential insights for relieving the traffic congestion and traffic flow control under CAVs environment.

关键词： mobile robots road vehicles mainline VSL control intelligent transportation systems reinforcement learning algorithm traffic jam connected vehicles environment off-ramp area mainline dynamic speed limit adjustment learning (artificial intelligence) mainline VSL adjustment VSL algorithm traffic flow control mainline variable speed limit control accounts automobiles off-ramp travel efficiency traffic congestion mainline dynamic VSL adjustment road traffic road traffic control reinforcement learning Python traffic engineering computing Q-learning algorithm autonomous vehicles environment

来源：评论

学校读者我要写书评

暂无评论

A dynamic object filtering approach based on object detection and geometric constraint between frames

引用

IET IMAGE PROCESSING 2022年第6期16卷 1636-1647页

作者： Wei, Jiansheng Pan, Shuguo Gao, Wang Zhao, Tao Southeast Univ Sch Instrument Sci & Engn Nanjing 210096 Peoples R China

In order to eliminate the influence of moving targets in visual positioning, a dynamic object filtering approach based on object detection and inter-frame geometric constraints is proposed to filter the dynamic objects in monocular images. The object detection algorithm is firstly used to identify and locate objects in the single-frame image, and the object matching between frames is performed. Then, the depth estimation network and pose recovery network are trained jointly to output estimated depth along with transformation matrix of object centroids between frames respectively. Finally, the mapping centroid is obtained from the estimated depth and inter-frame transformation matrix. The joint constraint function is performed to complete the detection and filtering of dynamic objects. The experimental results show that the proposed dynamic object filtering approach not only can filter moving objects in sequence images accurately but also allows to reserve the dynamic objects that are temporarily in the stop state. The generalization ability of this approach is also verified in a real urban road scene and meets the requirements of follow-up research in visual positioning.

关键词： image sequences object detection object detection algorithm image matching dynamic object filtering approach single-frame image traffic engineering computing image motion analysis inter-frame geometric constraints object centroids

来源：评论

学校读者我要写书评

暂无评论

Deep learning-based hybrid model for the behaviour prediction of surrounding vehicles over long-time periods

引用

IET INTELLIGENT TRANSPORT SYSTEMS 2022年第10期16卷 1404-1412页

作者： Gao, Xiang Chen, Long Tai, Kangsheng Cai, Yingfeng Wang, Hai Li, Yicheng Jiangsu Univ Automot Engn Res Inst Zhenjiang 212013 Jiangsu Peoples R China Jiangsu Univ Sch Automot & Traff Engn Zhenjiang Jiangsu Peoples R China

Autonomous vehicles need to have the ability to predict the future behaviour of surrounding vehicles, which helps with proper trajectory planning and tracking. Many behaviour prediction methods have limited application because they have a very limited prediction horizon. This paper proposes a deep learning-based hybrid model for behaviour prediction over long-time periods, including manoeuvre recognition and a behaviour prediction module. In manoeuvre recognition module, the CNN extracts the social characteristics of the target vehicle and LSTM outputs the manoeuvre probability vector to form a contextual feature vector with the social features. In the behaviour prediction module, LSTM and Attention are based on the contextual feature vector to capture multi-time step information in the behaviour time window to complete the prediction of the target vehicle behaviour. Real-car collection and open-source vehicle trajectory datasets were used for training and testing. The results show that the proposed algorithm could predict vehicle behaviour with an accuracy of 89.73% and an average prediction time of 2.032 s, which has a high engineering application value.

关键词： LSTM mobile robots autonomous vehicles target vehicle behaviour prediction trajectory control deep learning (artificial intelligence) contextual feature vector long-time periods control engineering computing automobiles real-car collection tracking path planning convolutional neural nets deep learning-based hybrid model recurrent neural nets open-source vehicle trajectory datasets road traffic control social features CNN manoeuvre recognition feature extraction trajectory planning manoeuvre probability vector behaviour time window probability multitime step information motion control attention mechanism surrounding vehicle behaviour average prediction time traffic engineering computing prediction horizon

来源：评论

学校读者我要写书评

暂无评论

Event-triggered optimisation of overtaking decision-making strategy for autonomous driving on highway

引用

IET INTELLIGENT TRANSPORT SYSTEMS 2022年第12期16卷 1794-1808页

作者： Huang, Ping Zhang, Lin Chen, Hong Ding, Haitao Cao, Jianyong Jilin Univ Sch Artificial Intelligence Changchun Peoples R China Jilin Univ State Key Lab Automot Simulat & Control Changchun 130025 Peoples R China Tongji Univ Sch Automot Studies Postdoctoral Stn Mech Engn Shanghai Peoples R China Tongji Univ Coll Elect & Informat Engn Shanghai Peoples R China Shanghai Motor Vehicle Inspect Certificat & Tech Shanghai Peoples R China

Appropriate overtaking can improve the safety and efficiency of driving. Practically, overtaking consists of two asymmetric lane-changes: firstly to the left lane and then return back. Existing literatures treated the two lane-changes as the same behaviour, which may result in frequent and unnecessary lane-changes. For the more complex first lane-change, discretionary lane-change strategies have been developed to gain speed advantage with least deceleration. Most of them, however, described vehicles' dynamics with rough simplifications different from actual execution (e.g. constant speed and constant acceleration), which may fail them in practical applications. This paper presents an event-triggered overtaking decision-making strategy for autonomous vehicles to make the overtaking meet the driver's expectations. Firstly, the difference between the left-lane change and right-lane change is analysed. Left-lane change is mainly concerned with the efficiency, whereas driving safety is the main target of the right-lane change. Secondly, an event-triggering condition covering the speed difference between the ego vehicle and its preceding vehicle is set. Afterwards, based on quantified speed advantage, considering surrounding vehicles' changes in both speed and acceleration, an optimisation of overtaking decision-making strategy is proposed. During right-lane change in an overtaking, safety is ensured by predicting whether the ego vehicle will collide with surrounding vehicles. Finally, several scenarios are defined in the combination of SCANeR studio and Simulink, which are used to test and demonstrate the effectiveness and applicability of the overtaking decision-making strategy. The results show that the proposed overtaking decision-making strategy, by mathematical quantification of speed advantage, performs well in measuring the motivation of lane-change and avoiding frequent lane changes, which correspond to driver habits.

关键词： mobile robots discretionary lane-change strategies unnecessary lane-changes right-lane change road vehicles intelligent transportation systems frequent lane-changes optimisation surrounding vehicles driver information systems avoiding frequent lane changes left lane road traffic appropriate overtaking road safety left-lane change decision making asymmetric lane-changes event-triggered optimisation ego vehicle speed advantage overtaking decision-making strategy traffic engineering computing

来源：评论

学校读者我要写书评

暂无评论

Stacking-based ensemble learning method for the recognition of the preceding vehicle lane-changing manoeuvre: A naturalistic driving study on the highway

引用

IET INTELLIGENT TRANSPORT SYSTEMS 2022年第4期16卷 489-503页

作者： Zhang, Hongjia Guo, Yingshi Wang, Chang Fu, Rui Changan Univ Sch Automobile Xian 710064 Peoples R China

The recognition of the preceding vehicle lane-changing manoeuvre (LCM) is essential for improving the rationality and safety of the decision-making of driverless vehicles. However, traditional recognition researches for preceding vehicle LCM are generally characterized by problems such as low recognition accuracy and poor research foundation. In response to these problems, this paper carried out naturalistic driving study (NDS) on the highway, from which a large amount of on-road data consisting of lane-keeping (LK) and lane-changing left and right (LCL, LCR) manoeuvres were collected. A stacking-based ensemble learning method for the recognition of the LCM of the preceding vehicle, which integrates the random forest, support vector machine, long and short-term memory network (LSTM), and bi-directional LSTM based on attention mechanism (AT-Bi-LSTM) algorithms, is proposed. Compared with traditional machine learning methods, the proposed method exhibits greater advantages in terms of its recognition accuracy. It is particularly of note that the recognition accuracy of the model at 0.4 s and 0.8 s after the LCM reached 90.77% and 95.54%, respectively. The study reported is of great significance for the construction of more intelligent vehicle-vehicle collaboration and the promotion of the industrial applications of intelligent vehicle technology.

关键词： road vehicles driverless vehicles intelligent transportation systems preceding vehicle lane-changing manoeuvre recognition long and short-term memory network lane-changing right learning (artificial intelligence) random forests lane-changing left recognition accuracy AT-Bi-LSTM intelligent vehicle-vehicle collaboration road traffic recurrent neural nets stacking-based ensemble learning method road safety naturalistic driving study random forest attention mechanism bi-directional LSTM support vector machine support vector machines vehicle LCM traffic engineering computing

来源：评论

学校读者我要写书评

暂无评论

Robust object detection under harsh autonomous-driving environments

引用

IET IMAGE PROCESSING 2022年第4期16卷 958-971页

作者： Kim, Youngjun Hwang, Hyekyoung Shin, Jitae Sungkyunkwan Univ Coll Informat & Commun Engn Suwon 16419 South Korea

In the autonomous driving environment, object instances in an image can be affected by various factors such as camera, driving state, weather, and system component. However, the deep learning-based vision systems are vulnerable to perturbation, which contains noise. Thus, robust object detection under harsh autonomous-driving environments is a more difficult than the generic situation. In this paper, it is found that not only the accuracy, but also the speed of the non-maximum suppression-based detector can be degraded under harsh environments. Therefore, object detection is handled under a harsh situation with adversarial mechanisms such as adversarial training and adversarial defence. Adversarial defence modules are designed to improve robustness in feature extraction level and define perturbations under a harsh environment for training object detectors to improve the robustness of the model's decision boundary. The proposed adversarial defence and training mechanisms improve the object detector in both accuracy and speed. The proposed method shows a 43.7% mean average precision for the COCO2015 dataset in generic object detection and 39.0% mean average precision for the BDD100K dataset in a driving environment. Furthermore, it achieves a real-time capability of 23 frames per second.

关键词： deep learning (artificial intelligence) training mechanisms object detection deep learning-based vision systems object detector nonmaximum suppression-based detector adversarial defence modules robust object detection traffic engineering computing harsh situation autonomous-driving environments autonomous driving environment computer vision harsh environment adversarial training generic object detection feature extraction

来源：评论

学校读者我要写书评

暂无评论

Real-time traffic cone detection for autonomous driving based on YOLOv4

引用

IET INTELLIGENT TRANSPORT SYSTEMS 2022年第10期16卷 1380-1390页

作者： Su, Qinghua Wang, Haodong Xie, Min Song, Yue Ma, Shaobo Li, Boxiong Yang, Ying Wang, Liyong Beijing Informat Sci & Technol Univ Minist Educ Key Lab Modern Measurement & Control Technol Beijing 100192 Peoples R China

A temporary road composed of traffic cones is an indispensable practical scene for the realization of automatic driving technology. However, the detection of traffic cones is a challenging issue because of their small volume and unfixed position. This work proposes a novel method that fuses colour and depth image information for traffic cone detection. traffic cones are captured by a special machine vision system consisting of two monochrome cameras for cone distance perception and two colour cameras for cone colour acquisition. Via the YOLOv4 algorithm based on the Darknet platform and a detection result matching algorithm, the position of the traffic cone can be obtained and path planning can be performed. The results of experiments show that the proposed method can recognize red, blue, and yellow traffic cones in colour images with an average detection time of 35.46 ms and respective accuracies of 97.51%, 98.63%, and 97.29%. Compared with the previous traffic cone detection research, the proposed algorithm was found to exhibit advantages in small target sensitivity and overall detection accuracy in both static and dynamic experiments.

关键词： result matching algorithm detection robot vision mobile robots machine vision colour fusion red traffic cone recognition depth image information fusion neural nets path planning object recognition object detection advanced driver assistance systems YOLOv4 Darknet platform blue traffic cone recognition yellow traffic cone recognition cone distance perception image fusion real-time traffic cone detection cone colour acquisition traffic engineering computing image colour analysis

来源：评论

学校读者我要写书评

暂无评论

Speed behaviour of right-turn drivers at signalized intersections in China

引用

IET INTELLIGENT TRANSPORT SYSTEMS 2022年第9期16卷 1209-1221页

作者： Yuan, Tian Zhao, Xuan Liu, Rui Zhu, Xichan Wang, Shu Yu, Qiang Changan Univ Sch Automobile Xian Peoples R China Tongji Univ Sch Automot Studies Shanghai Peoples R China

Ensuring human-like driving performance is fundamental to the widespread adoption and acceptance of intelligent driving systems. The speed behaviour of right-turn drivers at signalized intersections was studied to improve the human-like driving capability. The study was conducted using natural driving data and 545 data for drivers turning right at signalized intersections were extracted. YOLOv4 was used to identify the different types of road users. A one-way ANOVA was used to study the influence of the scene complexity, number of lanes, and intersection shape on the speed selection behaviour. Linear regression and binary logistic regression analyses were used to study the influence of road users, number of lanes, intersection shape, and vehicle motion state on the speed control behaviour. The main results indicate the following: (1) Drivers drove slower when there were more traffic participants or fewer lanes. (2) The intersection shape did not have any effect on the speed behaviour of right-turn drivers, either speed selection behaviour or speed control behaviour. (3) The reaction distance was affected only by the approach speed in this study;the greater the speed, the longer the reaction distance. (4) In addition to the vehicle motion state, external environmental factors also play a critical role in the braking characteristics, that is, pedestrians, cyclists, lateral traffic, and number of lanes after the turn. (5) The driver's intention to brake after entering the intersection stemmed from road users who were in collision conflict with the host vehicle and fewer lanes after the turn, rather than the speed at the stop line. This research has a guiding significance for the design of human-like driving of right-turn driving assistance systems. The target classification method and the measurement method of the driving environment complexity also help improve the human-like perception of the right-turn driving assistance system.

关键词： driving environment complexity road vehicles right-turn drivers reaction distance driver information systems right-turn driving assistance system speed behaviour fewer lanes driving performance road users driving capability road traffic object detection natural driving data road safety road traffic control binary logistic regression analyses logistic regression intelligent driving systems driver vehicle motion state signalized intersections regression analysis intersection shape speed selection behaviour environmental factors approach speed pedestrians speed control behaviour traffic engineering computing

来源：评论

学校读者我要写书评

暂无评论

A novel method for calibration and verification of roadside millimetre-wave radar

引用

IET INTELLIGENT TRANSPORT SYSTEMS 2022年第3期16卷 408-419页

作者： Zhang, Changlong Wei, Jimin Hu, Albert Sibo Fu, Peipei Changsha Intelligent Driving Inst Co Ltd Intelligent Connected Transportat Dept Changsha Peoples R China

Roadside perception systems commonly rely on millimetre-wave radars that are stationary. Obtaining global coordinates of radar detection targets requires calibration of radars installed on roadways, which is dangerous if performed via conventional methods. Hazards such as live traffic, particularly on expressways, make safe calibration difficult without road closures. To avoid these issues, this paper proposes a novel calibration method for roadside millimetre-wave radar by using high-definition maps to provide global positioning for radar detection targets, combined with vehicle trajectory data from live traffic. After clustering collected traffic data, a polynomial curve fit provides the shape of lane centrelines. Calibration parameters for the relative-to-global position mapping are then obtained by matching fitted curve sample points to points in high-definition maps using non-linear optimization. The calibration results are verified experimentally using a connected vehicle and roadside unit on an urban street, demonstrating that the proposed method meets lane-level precision requirements for V2X applications. This data-driven method borrows live traffic for calibration and hence is safer and more convenient compared to existing calibration methods.

关键词： calibration methods calibration parameters curve fitting relative-to-global position mapping road vehicles intelligent transportation systems data-driven method borrows live traffic roadside millimetre-wave radar safe calibration road traffic radar detection targets connected vehicle global positioning high-definition maps novel calibration method road safety global coordinates road traffic control vehicular ad hoc networks calibration collected traffic data radar detection millimetre-wave radars traffic engineering computing fitted curve sample points roadside unit

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：