Exploring Cost-effective Computer Vision Solutions for Smart Transportation Systems: Literature Review
Computer vision is reshaping the transportation industry and bringing its unique capabilities to the table to enable next generation smart transportation systems in many different ways. The state-of-the-art of Internet of Things (IoT) strategies and computer vision techniques is well-studied in the literature, some has already been tested and used for certain use cases, but this technology has not widely applied in the day-to-day operation to existing transportation infrastructures yet. A comprehensive review of both state-of-art and state-of-practices as well as gaps in terms of use cases and applications is needed. In addition, several major challenges that hinder further advances in computer vision-based smart transportation application development remain. This includes how to develop the transportation-specific computer vision techniques through advanced artificial intelligence (AI) and machine learning (ML) techniques; how to make use the outputs of the computer vision-based systems to enhance traffic safety and situational awareness; how to customize the solutions based on different objectives from the agencies and road users; how to improve the accuracy of these systems under conditions such as adverse weather; and finally, how to maintain the cost-effectiveness of these computer vision-based transportation solutions. The authors conducted a multi-facet literature review that first examined the current use cases and transportation related applications that utilize the computer vision methodologies with a focus in urban areas, especially work zone and safety applications, and evaluated their applicability to various tasks of urban analytics, state of adoption, and limitations. The literature review then assessed if and how transportation equity is considered in the current state of adoption of computer vision/AI technology, for example, whether state-of-the-art object detection systems have equitable predictive performance on pedestrians with different skin tones.
- Record URL:
- Record URL:
-
Supplemental Notes:
- This document was sponsored by the U.S. Department of Transportation, University Transportation Centers Program.
-
Corporate Authors:
Connected Cities with Smart Transportation (C2SMART)
New York University
Tandon School of Engineering
Brooklyn, NY United StatesOffice of the Assistant Secretary for Research and Technology
University Transportation Centers Program
Department of Transportation
Washington, DC United States 20590 -
Authors:
- Gao, Jingqin
-
0000-0002-1718-2432
- Ozbay, Kaan
-
0000-0001-7909-6532
- Xu, Chuan
-
0000-0001-5727-4561
- Zhang, Daniel
-
0009-0007-3372-4565
- Zuo, Fan
-
0000-0002-6761-2808
- Yang, Liu
-
0009-0009-7929-3582
- Hammami, Omar
-
0000-0002-4727-408X
- Publication Date: 2023
Language
- English
Media Info
- Media Type: Digital/other
- Features: Figures; Photos; References; Tables;
- Pagination: 13p
Subject/Index Terms
- TRT Terms: Artificial intelligence; Computer vision; Detection and identification systems; Equity; Internet of things; Literature reviews; Traffic safety; Urban areas; Work zones
- Subject Areas: Data and Information Technology; Highways; Pedestrians and Bicyclists; Safety and Human Factors; Society;
Filing Info
- Accession Number: 01898588
- Record Type: Publication
- Files: UTC, NTL, TRIS, ATRI, USDOT
- Created Date: Nov 9 2023 5:10PM