Student or educator? Sign up with your school email and get 6 months of Pro free.

The DOI matches its Crossref record; 43 of the 43 references it cites were found in a registry or an index; 1 of the 3 sources its claims cite were checked for support.

Vision-based vehicle detection and counting system using deep learning in highway scenes

Open access
Song, Huansheng, Liang, Haoxiang, Li, Huaiyu, Dai, Zhe, Yun, Xu · 2019 · European Transport Research Review

Abstract

Intelligent vehicle detection and counting are becoming increasingly important in the field of highway management. However, due to the different sizes of vehicles, their detection remains a challenge that directly affects the accuracy of vehicle counts. To address this issue, this paper proposes a vision-based vehicle detection and counting system. A new high definition highway vehicle dataset with a total of 57,290 annotated instances in 11,129 images is published in this study. Compared with the existing public datasets, the proposed dataset contains annotated tiny objects in the image, which provides the complete data foundation for vehicle detection based on deep learning. In the proposed vehicle detection and counting system, the highway road surface in the image is first extracted and divided into a remote area and a proximal area by a newly proposed segmentation method; the method is crucial for improving vehicle detection. Then, the above two areas are placed into the YOLOv3 network to detect the type and location of the vehicle. Finally, the vehicle trajectories are obtained by the ORB algorithm, which can be used to judge the driving direction of the vehicle and obtain the number of different vehicles. Several highway surveillance videos based on different scenes are used to verify the proposed methods. The experimental results verify that using the proposed segmentation method can provide higher detection accuracy, especially for the detection of small vehicle objects. Moreover, the novel strategy described in this article performs notably well in judging driving direction and counting vehicles. This paper has general practical significance for the management and control of highway scenes.

Overview

CoChat Summary

## TL;DR

This paper proposes a highway-surveillance vehicle detection and counting system that combines road-surface segmentation, YOLOv3 detection, and ORB-based multi-object tracking. On the authors’ dataset of 11,129 images with 57,290 annotated instances, the system reports 87.88% mAP and near-real-time processing on tested videos.

## Key findings

  • The paper publishes a new high-definition highway vehicle dataset with 11,129 images and 57,290 annotated instances; the images are 1920 × 1080 RGB and average 5.15 instances per image.
  • The proposed detection model achieved 87.88% mAP, with class APs of 86.46% (car), 88.57% (bus), and 88.61% (truck); the paper also reports 71.32% average IoU and 83.46% overall correct rate for object detection.
  • For tracking/counting, the paper reports average accuracies of 92.3% for vehicle driving direction and 93.2% for vehicle counting.
  • Compared with full-image detection, the proposed road-surface segmentation approach improved small-object detection; the table reports overall correct rates of 92.14% vs 14.96% in the remote area and 97.82% vs 82.80% in the proximal area, respectively.

## Methods

  • Design: a pipeline that first extracts the highway road surface, then divides it into remote and proximal regions, detects vehicles with YOLOv3, and tracks them with ORB feature matching plus trajectory analysis.
  • Data/sample: a newly built dataset from highway surveillance video in Hangzhou, China, collected from 23 surveillance cameras across different scenes, times, and lighting conditions; the model was trained on an 80%/20% train-test split (8,904 training images and 2,225 test images).
  • Main analysis: YOLOv3 training and evaluation using mAP/AP/precision/recall, comparison of detection counts against actual vehicle numbers, and trajectory-based counting using a detection line to infer direction and vehicle flow.

## Limitations

  • The paper states that there are few public datasets for specific traffic scenes and that traffic images are often not publicly released due to copyright, privacy, and security issues.
  • The authors note that the method is less effective when cars are small, occluded, or parallel in lanes, and that detection deviation in the remote area can still occur; this is stated in the paper.
  • Evident from the text: the evaluation is limited to the authors’ own highway videos and dataset, so generalization beyond those scenes is not directly demonstrated in the provided text.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1186/s12544-019-0390-4

Cites

43 of 43 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Image 1
CC-BY · From the PDF
Image 2
CC-BY · From the PDF
Fig. 3 Annotated instances per image (average numbers of annotated instances are shown in parentheses)
Fig. 3 Annotated instances per image (average numbers of annotated instances are shown in parentheses)CC-BY · From the PDF
Image 4
CC-BY · From the PDF
Image 5
CC-BY · From the PDF
Image 6
CC-BY · From the PDF
Image 7
CC-BY · From the PDF
Image 8
CC-BY · From the PDF
Image 9
CC-BY · From the PDF
Image 10
CC-BY · From the PDF
Image 11
CC-BY · From the PDF
Image 12
CC-BY · From the PDF
Image 13
CC-BY · From the PDF

Verification

Claims the paper makes

1 Valid · 2 Not valid · 1 Not checked · 56 Own results and interpretations

19 more claims

Create a free account to see all 60 claims and every source verdict.

See all claims — free account
  • It may state two separate claims as one. 24
  • Probably not a substantial claim. 18
  • The quote only partly states this claim. 5

Create a free account to see these claims.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Citations

Cites

Show all 40 references

Cited by

See all 413

Metadata

Authors
Song, H., Liang, H., Li, H., Dai, Z., & Yun, X.
Affiliations
Chang'an University (CN)
Corresponding author
Liang, Haoxiang
Venue
European Transport Research Review
Published
2019-12
Volume
11
Issue
1
ISSN
  • 1866-8887 (ISSN-L)
  • 1867-0717
OpenAlex
W2997192890
Cited by
413
References
40
Citations by year
2026: 23 · 2025: 65 · 2024: 79 · 2023: 93 · 2022: 81 · 2021: 46 · 2020: 26
FWCI
14.75
Type
journal article
Language
en
Open access
Licence
  • cc-by
  • creativecommons.org/licenses/by/4.0 (tdm)
Locations
  • European Transport Research Review (publishedVersion, cc-by)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
  • National Natural Science Foundation of China (300102248402, No.61572083, 61572083)
  • Fundamental Research Funds for the Central Universities
Topics
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Video Surveillance and Tracking Methods
  • Physical Sciences › Engineering › Automotive Engineering › Autonomous Vehicle Technology and Safety
Keywords
vehicle detection, vehicle counting, highway monitoring, small object detection, road surface segmentation, YOLOv3, ORB feature matching, vehicle trajectory tracking, freeway traffic control

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Vision-based vehicle detection and counting system using deep learning in highway scenes. https://app.cochat.ai/papers/10.1186/s12544-019-0390-4
The paper
Song, H., Liang, H., Li, H., Dai, Z., & Yun, X. (2019). Vision-based vehicle detection and counting system using deep learning in highway scenes. European Transport Research Review. https://doi.org/10.1186/s12544-019-0390-4

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.