Yonggao Yue, Shang Zhang, Zhiyuan Wu, Jianpu Xi, Zonglin Shi, Lei Wang, Lijuan Deng
{"title":"Research on nighttime road visibility monitoring based on video images.","authors":"Yonggao Yue, Shang Zhang, Zhiyuan Wu, Jianpu Xi, Zonglin Shi, Lei Wang, Lijuan Deng","doi":"10.1080/15389588.2025.2495203","DOIUrl":null,"url":null,"abstract":"<p><strong>Objective: </strong>Road traffic accidents have become a serious social problem, with a significant proportion of accidents caused by insufficient visibility on roads at night. Therefore, nighttime road visibility detection based on video images has become one of the difficulties and a key issue in domestic and international research.</p><p><strong>Methods: </strong>This study analyzes the importance of nighttime road visibility monitoring, introduces the structure, working principle, and monitoring method of a video image nighttime visibility monitoring system, and proposes a nighttime road visibility monitoring method based on video images. Based on the characteristics of nighttime images, an improved dark channel prior method was adopted to calculate the nighttime road visibility. This method mainly includes eight steps: video image acquisition, image grayscale processing, calculation of image average variance, image average gradient, drawing grayscale histograms, image enhancement based on the calculated values, calculation of transmittance, and calculation of visibility.</p><p><strong>Results: </strong>The experimental results show that the proposed night road visibility monitoring method based on video images can effectively realize real-time monitoring of night road visibility, effectively overcome the inherent defects of traditional methods, and the constructed night visibility monitoring framework can realize high-precision visibility calculation, and has broad application prospects.</p><p><strong>Conclusions: </strong>Through adaptive threshold and adaptive filtering technology, the improved dark channel algorithm has shown competitive advantages in both image quality index and practical application effect, especially in noise suppression and edge preservation. However, under extreme illumination conditions, the algorithm still has room for improvement in the processing of the strong light source region, and the dark channel prior may lead to bias in the transmission estimation.</p>","PeriodicalId":54422,"journal":{"name":"Traffic Injury Prevention","volume":" ","pages":"1-9"},"PeriodicalIF":1.6000,"publicationDate":"2025-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Traffic Injury Prevention","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1080/15389588.2025.2495203","RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"PUBLIC, ENVIRONMENTAL & OCCUPATIONAL HEALTH","Score":null,"Total":0}
引用次数: 0
Abstract
Objective: Road traffic accidents have become a serious social problem, with a significant proportion of accidents caused by insufficient visibility on roads at night. Therefore, nighttime road visibility detection based on video images has become one of the difficulties and a key issue in domestic and international research.
Methods: This study analyzes the importance of nighttime road visibility monitoring, introduces the structure, working principle, and monitoring method of a video image nighttime visibility monitoring system, and proposes a nighttime road visibility monitoring method based on video images. Based on the characteristics of nighttime images, an improved dark channel prior method was adopted to calculate the nighttime road visibility. This method mainly includes eight steps: video image acquisition, image grayscale processing, calculation of image average variance, image average gradient, drawing grayscale histograms, image enhancement based on the calculated values, calculation of transmittance, and calculation of visibility.
Results: The experimental results show that the proposed night road visibility monitoring method based on video images can effectively realize real-time monitoring of night road visibility, effectively overcome the inherent defects of traditional methods, and the constructed night visibility monitoring framework can realize high-precision visibility calculation, and has broad application prospects.
Conclusions: Through adaptive threshold and adaptive filtering technology, the improved dark channel algorithm has shown competitive advantages in both image quality index and practical application effect, especially in noise suppression and edge preservation. However, under extreme illumination conditions, the algorithm still has room for improvement in the processing of the strong light source region, and the dark channel prior may lead to bias in the transmission estimation.
期刊介绍：
The purpose of Traffic Injury Prevention is to bridge the disciplines of medicine, engineering, public health and traffic safety in order to foster the science of traffic injury prevention. The archival journal focuses on research, interventions and evaluations within the areas of traffic safety, crash causation, injury prevention and treatment.
General topics within the journal''s scope are driver behavior, road infrastructure, emerging crash avoidance technologies, crash and injury epidemiology, alcohol and drugs, impact injury biomechanics, vehicle crashworthiness, occupant restraints, pedestrian safety, evaluation of interventions, economic consequences and emergency and clinical care with specific application to traffic injury prevention. The journal includes full length papers, review articles, case studies, brief technical notes and commentaries.
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号
