Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM68815
Volume:
10 |
11 |
Nov 2026
Pages:
1-7
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.

Artificial Intelligence And Machine Learning
VIvekananda College of Engineering and Technology Puttur

Artificial Intelligence And Machine Learning
VIvekananda College of Engineering and Technology Puttur

Artificial Intelligence And Machine Learning
VIvekananda College of Engineering and Technology Puttur

Artificial Intelligence And Machine Learning
VIvekananda College of Engineering and Technology Puttur
Vehicle counts alone describe road conditions poorly: a crowded carriageway may be flowing freely, while a handful of stationary vehicles can block it. This paper presents Cognitive Traffic Flow AI, a prototype decision-support framework that passes recorded traffic video through YOLOv8s detection and ByteTrack tracking, derives density, occupancy and image-space movement features, and labels each observation as Free Flow, Moderate or Congested with a multilayer perceptron whose output is then revised by movement rules. A Dynamic Congestion Index and a Traffic Risk Score, both scaled from 0 to 100 and defined within this project, add continuous severity information, and an LSTM forecasts seven traffic variables from 30-step windows. A Flask service connects these modules to a dashboard offering route planning (OpenRouteService, Nominatim, Leaflet and OpenStreetMap), locally generated explanations (Ollama) and e-mail alerts that fire only after sustained congestion. The classifier was assessed on 4,338 samples (3,138 video-derived from 22 videos and 1,200 synthetic) using five-fold grouped cross-validation. Mean accuracy was about 0.734 with a standard deviation of 0.167 across folds. Congested recall reached only 0.48, because 838 congested samples were labelled Moderate; we regard this as the principal weakness. The LSTM was trained on synthetic series, so its small errors show that the pipeline functions but do not demonstrate forecasting skill on real roads. Retraining with verified feature dimensions, speed calibration and validation on real data are identified as next steps.
KEYWORDS
Cognitive Traffic Flow AI combines YOLOv8s detection, ByteTrack tracking, movement-aware classification, two project-defined 0–100 indicators, LSTM forecasting, open-service route display, local explanations and sustained-congestion e-mail alerts within a Flask dashboard. On 4,338 samples the classifier reached a mean grouped cross-validation accuracy of about 0.734, but with wide variation between folds and a Congested recall of only 0.48. The forecasting results rest on synthetic data and demonstrate functionality only. The prototype therefore establishes feasibility of the integrated design, while real-world validation, verified model dimensions and improved severe-congestion detection remain the main tasks ahead.
[1] N. Kumar and M. Raubal, “Applications of deep learning in traffic congestion detection, prediction and alleviation: A survey,” Transportation Research Part C, 2021; arXiv:2102.09759. [2] J. Redmon, S. Divvala, R. Girshick and A. Farhadi, “You only look once: Unified, real-time object detection,” in Proc. IEEE CVPR, 2016, pp. 779–788. [3] G. Jocher, A. Chaurasia and J. Qiu, “Ultralytics YOLOv8,” software, 2023. [Online]. Available: https://github.com/ultralytics/ultralytics [4] Y. Zhang et al., “ByteTrack: Multi-object tracking by associating every detection box,” in Proc. ECCV, 2022; arXiv:2110.06864. [5] S. Hochreiter and J. Schmidhuber, “Long short-term memory,” Neural Computation, vol. 9, no. 8, pp. 1735–1780, 1997. [6] Hari K.C., M. Pokharel and S. Shrestha, “The prediction of road traffic congestion patterns: An optimized YOLOv8-based approach,” in Proc. 3rd Int. Electronic Conf. on Processes, MDPI, 29–31 May 2024. [7] M. M. Fahrezi and E. A. Laksana, “Traffic flow and congestion detection with YOLOv8 and ByteTrack,” JUTIF, vol. 5, no. 4, Aug. 2024, doi: 10.52436/1.jutif.2024.5.4.2063. [8] S. Pudaruth, I. M. Boodhun and C. W. Onn, “Reducing traffic congestion using real-time traffic monitoring with YOLOv8,” Int. J. Advanced Computer Science and Applications (IJACSA), vol. 15, no. 10, 2024. [9] A. Bewley, Z. Ge, L. Ott, F. Ramos and B. Upcroft, “Simple online and realtime tracking,” in Proc. IEEE ICIP, 2016, pp. 3464–3468. [10] N. Wojke, A. Bewley and D. Paulus, “Simple online and realtime tracking with a deep association metric,” in Proc. IEEE ICIP, 2017,pp. 3645–3649.
Publication Date
11/6/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
0
Total views since publication
PDF Downloads
4
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...