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

The DOI matches its Crossref record; 122 of the 123 references it cites were found in a registry or an index; 7 of the 31 sources its claims cite were checked for support.

Vehicle Detection and Classification via YOLOv8 and Deep Belief Network over Aerial Image Sequences

Open access

Abstract

Vehicle detection and classification are the most significant and challenging activities of an intelligent traffic monitoring system. Traditional methods are highly computationally expensive and also impose restrictions when the mode of data collection changes. This research proposes a new approach for vehicle detection and classification over aerial image sequences. The proposed model consists of five stages. All of the images are preprocessed in the first stage to reduce noise and raise the brightness level. The foreground items are then extracted from these images using segmentation. The segmented images are then passed onto the YOLOv8 algorithm to detect and locate vehicles in each image. The feature extraction phase is then applied to the detected vehicles. The extracted feature involves Scale Invariant Feature Transform (SIFT), Oriented FAST and Rotated BRIEF (ORB), and KAZE features. For classification, we used the Deep Belief Network (DBN) classifier. Based on classification, the experimental results across the three datasets produced better outcomes; the proposed model attained an accuracy of 95.6% over Vehicle Detection in Aerial Imagery (VEDAI) and 94.6% over Vehicle Aerial Imagery from a Drone (VAID) dataset, respectively. To compare our model with the other standard techniques, we have also drawn a comparative analysis with the latest techniques in the research.

Overview

CoChat Summary

TL;DR

The paper proposes a five-stage pipeline for vehicle detection and classification in aerial image sequences, combining defogging/gamma preprocessing, Fuzzy C-Means segmentation, YOLOv8 detection, SIFT/ORB/KAZE feature extraction, and a Deep Belief Network classifier. It reports 95.6% accuracy on VEDAI and 94.6% on VAID.

Key findings

  • The proposed model achieved classification accuracy of 95.6% on the VEDAI dataset and 94.6% on the VAID dataset.
  • Vehicle detection mean scores on VEDAI were precision 0.957, recall 0.953, and F1-score 0.955; on VAID they were precision 0.964, recall 0.953, and F1-score 0.958.
  • In comparison with prior methods, the proposed model reported 95.6% on VEDAI versus Wang et al. 93.96%, Mandal et al. 51.95%, Terrail et al. 83.50%, Wang et al. 91.27%, Rafique et al. 92.2%, and Hou et al. 75.54%; on VAID it reported 94.6% versus Lin et al. 89.3%.
  • The paper states that YOLOv8 is efficient for detecting objects of different sizes and appearance, and that classification accuracy can be improved further by extracting more useful texture- and shape-based features.

Methods

  • Design: aerial videos are converted to frames, preprocessed with defogging for noise reduction and gamma correction for brightness, segmented with Fuzzy C-Means to separate foreground/background, passed to YOLOv8 for vehicle detection, then SIFT, ORB, and KAZE features are extracted and used to train a Deep Belief Network classifier.
  • Data: VEDAI, a public aerial vehicle detection dataset with nine categories (“car”, “truck”, “pick-up”, “plane”, “tractor”, “boat”, “camping car”, “van”, “other”), average 5.5 cars occupying about 0.7% of pixels per image; VAID, with 6000 vehicle photos divided into categories including minibus, truck, cement truck, sedan, pickup, bus, trailer, and truck, taken by drone at 90–95 m altitude, 23.98 fps, 2720 × 1530 resolution, across ten locations in southern Taiwan.
  • Analysis: experiments used an Intel Core i5-7200U 2.30 GHz processor, 8 GB RAM, Windows 10, and Google Colab; k-fold cross-validation was applied on all datasets; results are reported as accuracy, precision, recall, F1-score, confusion matrices, and comparison with prior methods.

Limitations

  • The paper states the proposed system needs to be trained with more vehicle classes and that more features can be added to improve classification accuracy.
  • The text makes evident an inconsistency: the abstract says experimental results were across “three datasets,” but the experimental section describes and evaluates only VEDAI and VAID.
  • The text also makes evident that VAID is described as having “seven categories” while listing eight category names, and the comparison table provides only one prior VAID result (Lin et al.) versus several for VEDAI.

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

Integrity signals

Registry

Crossref · 10.3390/su151914597

Cites

122 of 123 references verified0 citation errors · 1 not checkable

Statistics

Nothing to recompute.

Verification

Claims the paper makes

5 Valid · 2 Not valid · 16 Not checked · 8 Own results and interpretations

2 more claims

Create a free account to see all 31 claims and every source verdict.

See all claims — free account
  • Probably not a substantial claim. 18
  • It may state two separate claims as one. 9
  • The quote only partly states this claim. 1

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?

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

Cited works that were retracted

Works this paper cites that were later retracted, from the retraction notices publishers deposit. Major when the notice came out before this paper was published.

1 of 1

Citations

Cites

1 retracted

Show all 122 references

Cited by

See all 68

Metadata

Authors
Al Mudawi, N., Qureshi, A. M., Abdelhaq, M., Alshahrani, A., Alazeb, A., Alonazi, M., & Algarni, A.
Affiliations
  • Department of Computer Science, College of Computer Science and Information System, Najran University, Najran 55461, Saudi Arabia
  • Air University (PK)
  • Princess Nourah bint Abdulrahman University (SA)
  • University of Jeddah (SA)
  • Najran University (SA)
  • Prince Sattam Bin Abdulaziz University (SA)
  • Northern Border University (SA)
Corresponding author
  • Abdelhaq, Maha
  • Alonazi, Mohammed
Venue
Sustainability
Publisher
MDPI AG
Published
2023-10-08
Volume
15
Issue
19
Pages
14597
ISSN
2071-1050 (ISSN-L)
OpenAlex
W4387435696
Cited by
68
References
122
Citations by year
2026: 24 · 2025: 24 · 2024: 19 · 2023: 1
FWCI
6.15
Type
journal article
Language
en
Open access
Locations
  • Sustainability (publishedVersion, cc-by)
  • RePEc: Research Papers in Economics (submittedVersion)
Funders
  • Princess Nourah bint Abdulrahman University (PNURSP2023R97, NU/RG/SERC/12/40, PSAU/2023/R/1444)
  • Deanship of Scientific Research at Najran University (PNURSP2023R97, NU/RG/SERC/12/40, PSAU/2023/R/1444)
  • Prince Sattam bin Abdulaziz University (PNURSP2023R97, NU/RG/SERC/12/40, PSAU/2023/R/1444)
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 › Computer Science › Computer Vision and Pattern Recognition › Vehicle License Plate Recognition
Keywords
vehicle detection, vehicle classification, aerial imagery, UAV imagery, YOLOv8, deep belief network, image segmentation, foreground segmentation, SIFT, ORB, KAZE, feature extraction, intelligent traffic monitoring

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 2026-09-28, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-09-28 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

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: Vehicle Detection and Classification via YOLOv8 and Deep Belief Network over Aerial Image Sequences. https://app.cochat.ai/papers/10.3390/su151914597
The paper
Al Mudawi, N., Qureshi, A. M., Abdelhaq, M., Alshahrani, A., Alazeb, A., Alonazi, M., & Algarni, A. (2023). Vehicle Detection and Classification via YOLOv8 and Deep Belief Network over Aerial Image Sequences. Sustainability. https://doi.org/10.3390/su151914597

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.