EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Integrated Synthesis and Optimization of Low-Latency, Multi-Modal Visual Perception Architectures for Adaptive Metropolitan-Scale Traffic Supervision and Anomaly Detection

Mr. Santosh Fupate, Prof. Anita Yadav, Prof. Komal Naxine

International Journal of Scientific Research in Science and Technology · 25.04.2026 · 13 (2) · 1003-1012

DOI: 10.32628/ijsrst2613398 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Scientific Research in Science and Technology, online erschienen am 25.04.2026. Verfasst von Mr. Santosh Fupate und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This paper presents MM-ViPAD (Multi-Modal Visual Perception Architecture for Anomaly Detection), an integrated synthesis and optimization framework for low-latency, adaptive traffic supervision at metropolitan scale. The proposed architecture fuses RGB video, thermal-infrared (IR), and LiDAR point-cloud streams through a novel cross-modal attention transformer, enabling robust vehicle detection, pedestrian tracking, and anomaly identification under diverse environmental conditions including rain, fog, low-light, and high-density traffic scenarios. A hierarchical feature pyramid network (FPN) decoder with depthwise-separable convolutions reduces inference latency to 29.7 ms per frame (33.7 FPS) on an edge-embedded NVIDIA Jetson Orin NX platform while achieving a mean Average Precision (mAP) of 88.6% at IoU threshold 0.5 on the MetroVision-2k benchmark. Five anomaly categories—wrong-way driving, sudden braking, pedestrian intrusion, abandoned object, and traffic congestion—are detected with a mean F1-score of 90.4% and mean AUC-ROC of 0.960. A distributed edge-cloud co-processing protocol sustains near-linear throughput scaling to 128 camera nodes. Experimental validation on three benchmark datasets, including CityFlow-AD and KITTI-360, confirms statistically significant gains over existing single-stream and two-stream baselines.

Zitieren

APA

Fupate, M. S., Yadav, P. A., & Naxine, P. K. (2026). Integrated Synthesis and Optimization of Low-Latency, Multi-Modal Visual Perception Architectures for Adaptive Metropolitan-Scale Traffic Supervision and Anomaly Detection. International Journal of Scientific Research in Science and Technology, 13(2), 1003-1012. https://doi.org/10.32628/ijsrst2613398

Vancouver

Fupate MS, Yadav PA, Naxine PK. Integrated Synthesis and Optimization of Low-Latency, Multi-Modal Visual Perception Architectures for Adaptive Metropolitan-Scale Traffic Supervision and Anomaly Detection. International Journal of Scientific Research in Science and Technology. 2026;13(2):1003-1012. doi:10.32628/ijsrst2613398

BibTeX
@article{Fupate2026,
  author = {Fupate, Mr. Santosh and Yadav, Prof. Anita and Naxine, Prof. Komal},
  title = {Integrated Synthesis and Optimization of Low-Latency, Multi-Modal Visual Perception Architectures for Adaptive Metropolitan-Scale Traffic Supervision and Anomaly Detection},
  journal = {International Journal of Scientific Research in Science and Technology},
  year = {2026},
  volume = {13},
  number = {2},
  pages = {1003--1012},
  doi = {10.32628/ijsrst2613398}
}
Kostenlos starten →