Road traffic flow forecasting provides critical information for the operational management of road mobility challenges, and models are used to generate the forecast. This paper uses a random process to present a novel...Road traffic flow forecasting provides critical information for the operational management of road mobility challenges, and models are used to generate the forecast. This paper uses a random process to present a novel traffic modelling framework for aggregate traffic on urban roads. The main idea is that road traffic flow is random, even for the recurrent flow, such as rush hour traffic, which is predisposed to congestion. Therefore, the structure of the aggregate traffic flow model for urban roads should correlate well with the essential variables of the observed random dynamics of the traffic flow phenomena. The novelty of this paper is the developed framework, based on the Poisson process, the kinematics of urban road traffic flow, and the intermediate modelling approach, which were combined to formulate the model. Empirical data from an urban road in Ghana was used to explore the model’s fidelity. The results show that the distribution from the model correlates well with that of the empirical traffic, providing a strong validation of the new framework and instilling confidence in its potential for significantly improved forecasts and, hence, a more hopeful outlook for real-world traffic management.展开更多
Internet of Car, resulting from the Internet of Things, is a key point for the forthcoming smart city. In this article, GPS technology, 3G wireless technology and cloud-processing technology are employed to construct ...Internet of Car, resulting from the Internet of Things, is a key point for the forthcoming smart city. In this article, GPS technology, 3G wireless technology and cloud-processing technology are employed to construct a cloud-processing network platform based on the Internet of Car. By this platform, positions and velocity of the running cars, information of traffic flow from fixed monitoring points and transportation videos are combined to be a virtual traffic flow data platform, which is a parallel system with real traffic flow and is able to supply basic data for analysis and decision of intelligent transportation system.展开更多
In this paper, we present the conditions under which the traffic processes in a pure jump Markov process with a general state space are Poisson processes, and give a simple proof of PASTA type theorem in Melamed (1982...In this paper, we present the conditions under which the traffic processes in a pure jump Markov process with a general state space are Poisson processes, and give a simple proof of PASTA type theorem in Melamed (1982) and Walrand (1988). Furthermore, we consider a generalized network with phase type negative arrivals and show that the network has a product-form invariant distribution and its traffic processes which represent the customers exiting from the network are Poisson processes.展开更多
The frequent traffic jams at major intersections call for an effective management system. The paper suggests implementing a smart traffic controller using real-time image processing. The sequence of the camera is anal...The frequent traffic jams at major intersections call for an effective management system. The paper suggests implementing a smart traffic controller using real-time image processing. The sequence of the camera is analyzed using different edge detection algorithms and object counting methods. Previously they used matching method that means the camera will be installed along with traffic light. It will capture the image sequence. To set an image of an empty road as a reference image, the captured images are sequentially matched using image matching;but in my paper, we used filtering method, which filtered the image and released all waste objects and only showed the cars, and after it well showed the number of cars in image. My paper is software that takes a picture or video. It has been customized to be used in the future to control the traffic light sign by giving each sign sufficient time, depending on the number of cars on each direction.展开更多
Accurate classification and prediction of future traffic conditions are essential for developing effective strategies for congestion mitigation on the highway systems. Speed distribution is one of the traffic stream p...Accurate classification and prediction of future traffic conditions are essential for developing effective strategies for congestion mitigation on the highway systems. Speed distribution is one of the traffic stream parameters, which has been used to quantify the traffic conditions. Previous studies have shown that multi-modal probability distribution of speeds gives excellent results when simultaneously evaluating congested and free-flow traffic conditions. However, most of these previous analytical studies do not incorporate the influencing factors in characterizing these conditions. This study evaluates the impact of traffic occupancy on the multi-state speed distribution using the Bayesian Dirichlet Process Mixtures of Generalized Linear Models (DPM-GLM). Further, the study estimates the speed cut-point values of traffic states, which separate them into homogeneous groups using Bayesian change-point detection (BCD) technique. The study used 2015 archived one-year traffic data collected on Florida’s Interstate 295 freeway corridor. Information criteria results revealed three traffic states, which were identified as free-flow, transitional flow condition (congestion onset/offset), and the congested condition. The findings of the DPM-GLM indicated that in all estimated states, the traffic speed decreases when traffic occupancy increases. Comparison of the influence of traffic occupancy between traffic states showed that traffic occupancy has more impact on the free-flow and the congested state than on the transitional flow condition. With respect to estimating the threshold speed value, the results of the BCD model revealed promising findings in characterizing levels of traffic congestion.展开更多
With the intelligent development of road traffic control and management,higher requirements for the accuracy and effectiveness of traffic data have been put forward.The issue of how to collect and integrate data for t...With the intelligent development of road traffic control and management,higher requirements for the accuracy and effectiveness of traffic data have been put forward.The issue of how to collect and integrate data for traffic scenes has sought importance in this field as various treatment technologies have emerged.A lot of research work have been carried out from the theoretical aspect to engineering application.展开更多
To process the traffic monitoring image, a local Histogram Equalization method based on fuzzy mathematics was proposed in this paper. In this paper, firstly, we define a function to measure the similarity degree of tw...To process the traffic monitoring image, a local Histogram Equalization method based on fuzzy mathematics was proposed in this paper. In this paper, firstly, we define a function to measure the similarity degree of two images. Then, a suitable Gaussian fuzzy distribution function was chose to generate a 3 × 3 matrix of influential factors. In order to reduce the artificial boundaries, we combined the 3 × 3 influential matrix with a 3 × 3 smooth filter matrix to get the final smooth-influ- ence matrix. Finally, the smooth-influence matrix was used to process the center block image. The simulation results demonstrated that the proposed method can reduce time consumption while improving the image contrast and can get satisfactory results.展开更多
The alpha stable self-similar stochastic process has been proved an effective model for high variable data traffic. A deep insight into some special issues and considerations on use of the process to model aggregated ...The alpha stable self-similar stochastic process has been proved an effective model for high variable data traffic. A deep insight into some special issues and considerations on use of the process to model aggregated VBR video traffic is made. Different methods to estimate stability parameter a and self-similar parameter H are compared. Processes to generate the linear fractional stable noise （LFSN） and the alpha stable random variables are provided. Model construction and the quantitative comparisons with fractional Brown motion （FBM） and real traffic are also examined. Open problems and future directions are also given with thoughtful discussions.展开更多
在智慧城市发展进程中,交通系统的精细化管理和智能化服务面临海量异构数据处理的挑战。传统交通信息查询系统存在数据源异构性强、自然语言交互能力不足、长尾查询场景覆盖有限等问题。文章基于ChatGLM3大语言模型,创新性地构建了融合N...在智慧城市发展进程中,交通系统的精细化管理和智能化服务面临海量异构数据处理的挑战。传统交通信息查询系统存在数据源异构性强、自然语言交互能力不足、长尾查询场景覆盖有限等问题。文章基于ChatGLM3大语言模型,创新性地构建了融合NL2SQL(Natural Language to Structured Query Language)技术的智能问数系统,通过动态Schema对齐、LoRA微调优化及多维度提示工程技术,实现了交通领域复杂自然语言查询到精准SQL指令的智能转换。实验结果表明,经过微调的模型在交通信息查询任务中准确率达到78.9%,较基线模型提升15.8个百分点。本研究为交通管理智能化转型提供了创新技术路径,并对大模型在垂直领域的深度适配进行了系统性探索。展开更多
针对道路缺陷检测中伪缺陷干扰及现有方法对全局特征捕获能力不足而导致的误检和漏检问题,提出了基于YOLOv5的融合全局感知与坐标注意力的道路缺陷检测方法。首先,设计了全局感知增强模块(enhanced global perception block,EGPB),通过...针对道路缺陷检测中伪缺陷干扰及现有方法对全局特征捕获能力不足而导致的误检和漏检问题,提出了基于YOLOv5的融合全局感知与坐标注意力的道路缺陷检测方法。首先,设计了全局感知增强模块(enhanced global perception block,EGPB),通过在卷积神经网络(convolutional neural networks,CNN)中加入自注意力机制处理计算机视觉任务,增强对道路图像的局部细微缺陷的识别能力和全局上下文信息感知能力。其次,通过嵌入坐标注意力(coordinate attention,CA)模块,通过捕获跨通道信息增强定向感知和位置感知能力,提升缺陷区域空间位置的敏感度。此外,根据缺陷目标的实际物理特征对路径聚合网络进行策略性裁剪,促使模型集中分析更为关键的特征层次,实现降低模型冗余并保持高精度检测的效果。实验表明:相比原始YOLOv5模型,文中方法的关键指标均值平均精度(mean average precision,mAP)提升3.1%,F1分数提高0.031,准确率提升2.7%,并在多样化道路条件下表现出更优的稳定性和准确性。展开更多
文摘Road traffic flow forecasting provides critical information for the operational management of road mobility challenges, and models are used to generate the forecast. This paper uses a random process to present a novel traffic modelling framework for aggregate traffic on urban roads. The main idea is that road traffic flow is random, even for the recurrent flow, such as rush hour traffic, which is predisposed to congestion. Therefore, the structure of the aggregate traffic flow model for urban roads should correlate well with the essential variables of the observed random dynamics of the traffic flow phenomena. The novelty of this paper is the developed framework, based on the Poisson process, the kinematics of urban road traffic flow, and the intermediate modelling approach, which were combined to formulate the model. Empirical data from an urban road in Ghana was used to explore the model’s fidelity. The results show that the distribution from the model correlates well with that of the empirical traffic, providing a strong validation of the new framework and instilling confidence in its potential for significantly improved forecasts and, hence, a more hopeful outlook for real-world traffic management.
基金supported by National Basic Research Program of China (973 Program) 2012CB821200 (2012CB821206)National Natural Science Foundation under Grant No. 61170113, No.91024001, No.61070142+1 种基金Beijing Natural Science Foundation(No.4111002)KM201010011006, PHR201008242
文摘Internet of Car, resulting from the Internet of Things, is a key point for the forthcoming smart city. In this article, GPS technology, 3G wireless technology and cloud-processing technology are employed to construct a cloud-processing network platform based on the Internet of Car. By this platform, positions and velocity of the running cars, information of traffic flow from fixed monitoring points and transportation videos are combined to be a virtual traffic flow data platform, which is a parallel system with real traffic flow and is able to supply basic data for analysis and decision of intelligent transportation system.
基金This research is supported by the National Natural Science Foundation of China.
文摘In this paper, we present the conditions under which the traffic processes in a pure jump Markov process with a general state space are Poisson processes, and give a simple proof of PASTA type theorem in Melamed (1982) and Walrand (1988). Furthermore, we consider a generalized network with phase type negative arrivals and show that the network has a product-form invariant distribution and its traffic processes which represent the customers exiting from the network are Poisson processes.
文摘The frequent traffic jams at major intersections call for an effective management system. The paper suggests implementing a smart traffic controller using real-time image processing. The sequence of the camera is analyzed using different edge detection algorithms and object counting methods. Previously they used matching method that means the camera will be installed along with traffic light. It will capture the image sequence. To set an image of an empty road as a reference image, the captured images are sequentially matched using image matching;but in my paper, we used filtering method, which filtered the image and released all waste objects and only showed the cars, and after it well showed the number of cars in image. My paper is software that takes a picture or video. It has been customized to be used in the future to control the traffic light sign by giving each sign sufficient time, depending on the number of cars on each direction.
文摘Accurate classification and prediction of future traffic conditions are essential for developing effective strategies for congestion mitigation on the highway systems. Speed distribution is one of the traffic stream parameters, which has been used to quantify the traffic conditions. Previous studies have shown that multi-modal probability distribution of speeds gives excellent results when simultaneously evaluating congested and free-flow traffic conditions. However, most of these previous analytical studies do not incorporate the influencing factors in characterizing these conditions. This study evaluates the impact of traffic occupancy on the multi-state speed distribution using the Bayesian Dirichlet Process Mixtures of Generalized Linear Models (DPM-GLM). Further, the study estimates the speed cut-point values of traffic states, which separate them into homogeneous groups using Bayesian change-point detection (BCD) technique. The study used 2015 archived one-year traffic data collected on Florida’s Interstate 295 freeway corridor. Information criteria results revealed three traffic states, which were identified as free-flow, transitional flow condition (congestion onset/offset), and the congested condition. The findings of the DPM-GLM indicated that in all estimated states, the traffic speed decreases when traffic occupancy increases. Comparison of the influence of traffic occupancy between traffic states showed that traffic occupancy has more impact on the free-flow and the congested state than on the transitional flow condition. With respect to estimating the threshold speed value, the results of the BCD model revealed promising findings in characterizing levels of traffic congestion.
文摘With the intelligent development of road traffic control and management,higher requirements for the accuracy and effectiveness of traffic data have been put forward.The issue of how to collect and integrate data for traffic scenes has sought importance in this field as various treatment technologies have emerged.A lot of research work have been carried out from the theoretical aspect to engineering application.
文摘To process the traffic monitoring image, a local Histogram Equalization method based on fuzzy mathematics was proposed in this paper. In this paper, firstly, we define a function to measure the similarity degree of two images. Then, a suitable Gaussian fuzzy distribution function was chose to generate a 3 × 3 matrix of influential factors. In order to reduce the artificial boundaries, we combined the 3 × 3 influential matrix with a 3 × 3 smooth filter matrix to get the final smooth-influ- ence matrix. Finally, the smooth-influence matrix was used to process the center block image. The simulation results demonstrated that the proposed method can reduce time consumption while improving the image contrast and can get satisfactory results.
文摘The alpha stable self-similar stochastic process has been proved an effective model for high variable data traffic. A deep insight into some special issues and considerations on use of the process to model aggregated VBR video traffic is made. Different methods to estimate stability parameter a and self-similar parameter H are compared. Processes to generate the linear fractional stable noise （LFSN） and the alpha stable random variables are provided. Model construction and the quantitative comparisons with fractional Brown motion （FBM） and real traffic are also examined. Open problems and future directions are also given with thoughtful discussions.
文摘在智慧城市发展进程中,交通系统的精细化管理和智能化服务面临海量异构数据处理的挑战。传统交通信息查询系统存在数据源异构性强、自然语言交互能力不足、长尾查询场景覆盖有限等问题。文章基于ChatGLM3大语言模型,创新性地构建了融合NL2SQL(Natural Language to Structured Query Language)技术的智能问数系统,通过动态Schema对齐、LoRA微调优化及多维度提示工程技术,实现了交通领域复杂自然语言查询到精准SQL指令的智能转换。实验结果表明,经过微调的模型在交通信息查询任务中准确率达到78.9%,较基线模型提升15.8个百分点。本研究为交通管理智能化转型提供了创新技术路径,并对大模型在垂直领域的深度适配进行了系统性探索。
文摘针对道路缺陷检测中伪缺陷干扰及现有方法对全局特征捕获能力不足而导致的误检和漏检问题,提出了基于YOLOv5的融合全局感知与坐标注意力的道路缺陷检测方法。首先,设计了全局感知增强模块(enhanced global perception block,EGPB),通过在卷积神经网络(convolutional neural networks,CNN)中加入自注意力机制处理计算机视觉任务,增强对道路图像的局部细微缺陷的识别能力和全局上下文信息感知能力。其次,通过嵌入坐标注意力(coordinate attention,CA)模块,通过捕获跨通道信息增强定向感知和位置感知能力,提升缺陷区域空间位置的敏感度。此外,根据缺陷目标的实际物理特征对路径聚合网络进行策略性裁剪,促使模型集中分析更为关键的特征层次,实现降低模型冗余并保持高精度检测的效果。实验表明:相比原始YOLOv5模型,文中方法的关键指标均值平均精度(mean average precision,mAP)提升3.1%,F1分数提高0.031,准确率提升2.7%,并在多样化道路条件下表现出更优的稳定性和准确性。
