CoChat

Visual Detection of Traffic Incident through Automatic Monitoring of Vehicle Activities

Open access

The DOI matches its Crossref record; 85 of the 85 references it cites were found in a registry or an index; 4 of the 14 sources its claims cite were checked for support.

Abstract

Intelligent transportation systems (ITSs) derive significant advantages from advanced models like YOLOv8, which excel in predicting traffic incidents in dynamic urban environments. Roboflow plays a crucial role in organizing and preparing image data essential for computer vision models. Initially, a dataset of 1000 images is utilized for training, with an additional 500 images reserved for validation purposes. Subsequently, the Deep Simple Online and Real-time Tracking (Deep-SORT) algorithm enhances scene analyses over time, offering continuous monitoring of vehicle behavior. Following this, the YOLOv8 model is deployed to detect specific traffic incidents effectively. By combining YOLOv8 with Deep SORT, urban traffic patterns are accurately detected and analyzed with high precision. The findings demonstrate that YOLOv8 achieves an accuracy of 98.4%, significantly surpassing alternative methodologies. Moreover, the proposed approach exhibits outstanding performance in the recall (97.2%), precision (98.5%), and F1 score (95.7%), underscoring its superior capability in accurate prediction and analyses of traffic incidents with high precision and efficiency.

Overview

CoChat Summary

TL;DR

This paper integrates Roboflow, Deep-SORT, and YOLOv8 to detect traffic incidents from surveillance video, training on a traffic-incident image dataset. It reports YOLOv8 accuracy of 98.4%, recall of 97.2%, precision of 98.5%, and F1 score of 95.7%, outperforming several alternative methods.

Key findings

  • YOLOv8 achieved accuracy 98.4%, recall 97.2%, precision 98.5%, and F1 score 95.7%, which the paper says are greater than traditional methods.
  • The reported YOLOv8 accuracy of 98.4% is higher than Faster R-CNN (87.97%), FA-YOLO (92.95%), and FESSD (83%).
  • The precision-recall curve reports “all classes 0.933 mAP@0.5,” while the F1-confidence curve reports an F1 score of 0.89 or 89%.
  • Training losses dropped steadily over 40 epochs; mAP@0.5 stabilized around epoch 20, and mAP for IoU 0.50–0.95 showed an overall increasing trend.

Methods

  • Roboflow was used for data compilation and annotation; the dataset came from a publicly accessible Kaggle library of 523 traffic incident photos, with Pascal VOC annotations converted to COCO format and split 70:20:10 (335 training, 144 validation, 44 testing). Data augmentation using zooming, contrast correction, and image rotation added 756 photos.
  • Deep-SORT was used to scrutinize scenes over time and track vehicle behavior, while YOLOv8 was used for object detection and traffic incident detection; the model was trained through OpenCV and Google Colab and recognizes four categories: “car,” “truck,” “bus,” and “accident.”
  • Evaluation used metrics including mAP, accuracy, recall, precision, classification loss, and differential loss, with comparisons to Faster R-CNN, FA-YOLO, FESSD, and YOLO variants.

Limitations

  • The paper does not state a dedicated limitations section; it notes that no external funding was received and that data will be made available upon request.
  • Evident from the text: dataset size and split are inconsistently reported—the abstract says 1000 images for training and 500 for validation, while the methods section says 523 images split into 335 training, 144 validation, and 44 testing, with 756 photos added through augmentation.
  • Evident from the text: the F1 score is reported as 95.7% in the abstract, results, and conclusion but as 0.89 or 89% in Section 3.5, and no external validation or real-world deployment details are provided.

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

Integrity signals

Registry

Crossref · 10.3390/wevj15090382

Cites

85 of 85 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Verification

Claims the paper makes

4 Valid · 0 Not valid · 10 Not checked · 24 Own results and interpretations

6 more claims

Create a free account to see all 38 claims and to verify the 2 references still open.

See all claims — free account
  • Probably not a substantial claim. 16
  • It may state two separate claims as one. 21

Create a free account to see these claims.

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

3 of 3

Citations

Cites

Research works printed but not linked: 1

  1. Handiso, 2022, Trends and determinants of road traffic accident human death in Kembata Tembaro zone, SNNPR, Ethiopia, Sci. J. Appl. Math. Stat., 10, 85

Cited by

See all 23

Metadata

Authors
Karim, A., Raza, M. A., Alharthi, Y. Z., Abbas, G., Othmen, S., Hossain, M. S., Nahar, A., & Mercorelli, P.
Affiliations
  • Mehran University of Engineering and Technology (PK)
  • University of Hafr Al-Batin (SA)
  • Southeast University (CN)
  • Northern Border University (SA)
  • International University of Business Agriculture and Technology (BD)
  • American International University-Bangladesh (BD)
  • Leuphana University of Lüneburg (DE)
Corresponding author
  • Othmen, Salwa
  • Mercorelli, Paolo
Venue
World Electric Vehicle Journal
Publisher
MDPI AG
Published
2024-08-23
Volume
15
Issue
9
Pages
382
ISSN
2032-6653 (ISSN-L)
OpenAlex
W4401805438
Cited by
23
References
79
Citations by year
2026: 9 · 2025: 11 · 2024: 3
FWCI
3.20
Type
journal article
Language
en
Open access
Locations
  • World Electric Vehicle Journal (publishedVersion, cc-by)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
  • Multilingual Matters (Channel View Publications) (submittedVersion, cc-by)
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
traffic incident detection, YOLOv8, DeepSORT, object tracking, intelligent transportation systems, urban traffic, Roboflow dataset, real-time video analysis

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.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

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

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.