Xiaowei Shi , Dongfang Zhao , Handong Yao , Xiaopeng Li , David K. Hale , Amir Ghiasi
{"title":"Video-based trajectory extraction with deep learning for High-Granularity Highway Simulation (HIGH-SIM)","authors":"Xiaowei Shi , Dongfang Zhao , Handong Yao , Xiaopeng Li , David K. Hale , Amir Ghiasi","doi":"10.1016/j.commtr.2021.100014","DOIUrl":null,"url":null,"abstract":"<div><p>High-granularity vehicle trajectory data can help researchers develop traffic simulation models, understand traffic flow characteristics, and thus propose insightful strategies for road traffic management. This paper proposes a novel vehicle trajectory extraction method that can extract high-granularity vehicle trajectories from aerial videos. The proposed method includes video calibration, vehicle detection and tracking, lane marking identification, and vehicle motion characteristics calculation. In particular, the authors propose a Monte-Carlo-based lane marking identification approach to identify each vehicle's lane. This is a challenging problem for vehicle trajectory extraction, especially when the aerial videos are taken from a high altitude. The authors applied the proposed method to extract vehicle trajectories from several high-resolution aerial videos recorded from helicopters. The extracted dataset is named by the High-Granularity Highway Simulation (HIGH-SIM) vehicle trajectory dataset. To demonstrate the effectiveness of the proposed method and understand the quality of the HIGH-SIM dataset, we compared the HIGH-SIM dataset with a well-known dataset, the NGSIM US-101 dataset, regarding the accuracy and consistency aspects. The comparison results showed that the HIGH-SIM dataset has more reasonable speed and acceleration distributions than the NGSIM US-101 dataset. Also, the internal and platoon consistencies of the HIGH-SIM dataset give lower errors compared to the NGSIM US-101 dataset. To benefit future research, the authors have published the HIGH-SIM dataset online for public use.</p></div>","PeriodicalId":100292,"journal":{"name":"Communications in Transportation Research","volume":null,"pages":null},"PeriodicalIF":12.5000,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S2772424721000147/pdfft?md5=9ef534362f55ca64e1d805fc202b1e16&pid=1-s2.0-S2772424721000147-main.pdf","citationCount":"30","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Communications in Transportation Research","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2772424721000147","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"TRANSPORTATION","Score":null,"Total":0}
引用次数: 30
Abstract
High-granularity vehicle trajectory data can help researchers develop traffic simulation models, understand traffic flow characteristics, and thus propose insightful strategies for road traffic management. This paper proposes a novel vehicle trajectory extraction method that can extract high-granularity vehicle trajectories from aerial videos. The proposed method includes video calibration, vehicle detection and tracking, lane marking identification, and vehicle motion characteristics calculation. In particular, the authors propose a Monte-Carlo-based lane marking identification approach to identify each vehicle's lane. This is a challenging problem for vehicle trajectory extraction, especially when the aerial videos are taken from a high altitude. The authors applied the proposed method to extract vehicle trajectories from several high-resolution aerial videos recorded from helicopters. The extracted dataset is named by the High-Granularity Highway Simulation (HIGH-SIM) vehicle trajectory dataset. To demonstrate the effectiveness of the proposed method and understand the quality of the HIGH-SIM dataset, we compared the HIGH-SIM dataset with a well-known dataset, the NGSIM US-101 dataset, regarding the accuracy and consistency aspects. The comparison results showed that the HIGH-SIM dataset has more reasonable speed and acceleration distributions than the NGSIM US-101 dataset. Also, the internal and platoon consistencies of the HIGH-SIM dataset give lower errors compared to the NGSIM US-101 dataset. To benefit future research, the authors have published the HIGH-SIM dataset online for public use.