Lightweight Convolutional Neural Network based Resource-Aware Energy-Efficient Detector within Edge–Fog-enabled Industrial IoT systems

Ashwin M; Phani Kumar Solleti; Sarangam Kodati; T. Ravi; Gayathri Parasa; M. Vamsikrishna; D. Vetrithangam

doi:10.55938/ijgasr.v5i1.339

Authors

Ashwin M AMC Engineering College
Phani Kumar Solleti Koneru Lakshmaiah Education Foundation
Sarangam Kodati CVR College of Engineering
T. Ravi Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology
Gayathri Parasa Vignan's Foundation for Science, Technology & Research
M. Vamsikrishna Aditya University
D. Vetrithangam Chandigarh University

DOI:

https://doi.org/10.55938/ijgasr.v5i1.339

Keywords:

Object Detection, Industrial IoT, Edge Computing, Resource Awareness, Deep Learning, Fog Computing, Energy Efficiency

Abstract

The fast development of the applications of the Industrial Internet of Things (IIoT) requires the real-time object detection which has the capability to work effectively in energy-restrained edgefog conditions. Unlike the former YOLO-based and lightweight detectors and RL-based edgefog offloading mechanisms, the proposed RAEED framework combines lightweight detection and training-free adaptive inference offloading to optimize the accuracy, latency and energy consumption jointly in IIoT settings. The deep learning-based detectors are very accurate and capable of handling resource-intensive IIoT deployments, however, their systems are frequently computation- and powerintensive, which is not feasible in resource-constrained IIoT applications. In this effort, the present work suggests that a resource-conscious energy-efficient detector (RAEED) system is proposed which allocates inference resources between the edge and fog nodes dynamically projected by resource availability and network characteristics. The framework integrates a convolutional backbone which is lightweight together with adaptive offloading strategy to trade-off between detection accuracy, latency, and energy consumption. The applications to the COCO 2017 dataset under constrained deployment conditions indicate that RAEED can have 82.4% mAP, 83.1% precision and 81.7% recall, with a latency of 47.6 ms and with a consumption of 0.91 mJ/frame. These findings indicate that IIoT systems can greatly manage to find and identify objects with a much better energy and resource savings.

Downloads

Download data is not yet available.

References

Hanifi S, Alkali B, Lindsay G et al. Optimizing energy and air consumption in smart manufacturing: An industrial internet of things-based monitoring and efficiency enhancement solution. Applied Sciences ; 15(6): 3222. URL https://doi.org/10.3390/app15063222.

Gore S, Sharma H, Singh JP et al. Evolution of big data adoption in industries using topsis method. In 2024 International Conference on Sustainable Communication Networks and Application (ICSCNA). IEEE, p. 680–686. URL https://doi.org/10.1109/ICSCNA63714.2024. 10864346.

Fernando N, Shrestha S, Loke SW et al. On edge-fog-cloud collaboration and reaping its benefits: A heterogeneous multi-tier edge computing architecture. Future Internet ; 17(1): 22. URL https://doi.org/10.3390/fi17010022.

Kuchuk H and Malokhvii E. Integration of iot with cloud, fog, and edge computing: A review. Advances in Information Systems ; 8(2): 65–78. URL https://doi.org/10.20998/ 2522-9052.2024.2.08.

Mittal P. A comprehensive survey of deep learning-based lightweight object detection models for edge devices. Artificial Intelligence Review ; 57(9): 242. URL https://doi.org/10.1007/ s10462-024-10877-1.

Zhang J, Chen Z, Yan G et al. Faster and lightweight: An improved yolov5 object detector for remote sensing images. Remote Sensing ; 15(20): 4974. URL https://doi.org/10.3390/ rs15204974.

Liu G, Hu Y, Chen Z et al. Lightweight object detection algorithm for robots with improved yolov5. Engineering Applications of Artificial Intelligence ; 123: 106217. URL https://doi.org/10. 1016/j.engappai.2023.106217.

Liu J, Cai Q, Zou F et al. Biga-yolo: A lightweight object detection network based on yolov5 for autonomous driving. Electronics ; 12(12): 2745. URL https://doi.org/10.3390/ electronics12122745.

Luo H, Wei J, Wang Y et al. An improved lightweight object detection algorithm for yolov5. PeerJ Computer Science ; 10: e1830. URL https://doi.org/10.7717/peerj-cs.1830.

Di Fabrizio G, Calisti L, Contoli C et al. A study on the energy-efficiency of the object tracking algorithms in edge devices. In Proceedings of the IEEE/ACM 16th International Conference on Utility and Cloud Computing. New York, NY, USA: ACM, p. 1–6. URL https://doi.org/ 10.1145/3603166.3632541.

Suganya B, Gopi R, Kumar AR et al. Dynamic task offloading edge-aware optimization framework for enhanced uav operations on edge computing platform. Scientific Reports ; 14(1): 16383. URL https://doi.org/10.1038/s41598-024-67285-2.

Khaleel MI. Failure-aware resource provisioning for hybrid computation offloading in cloud- assisted edge computing using gravity reference approach. Swarm and Evolutionary Computation ; 91: 101704. URL https://doi.org/10.1016/j.swevo.2024.101704.

Pandeeswari ST, Padmavathi S, Kabilan D et al. Resource-aware fog service placement using deferred acceptance in edge computing. Journal of Engineering Research ; 13(2): 786–796. URL https://doi.org/10.1016/j.jer.2024.03.003.

Abdelghany HM. Dynamic resource management and task offloading framework for fog computing. Journal of Grid Computing ; 23(2): 19. URL https://doi.org/10.1007/ s10723-025-09804-7.

Jiang W, Han H, Feng D et al. Energy-efficient and accuracy-aware dnn inference with iot device-edge collaboration. IEEE Transactions on Services Computing ; 18(2): 784–797. URL https://doi.org/10.1109/TSC.2025.3536311.

Millar J, Haddadi H and Madhavapeddy A. Energy-aware deep learning on resource-constrained hardware. arXiv Preprint arXiv:250512523 ; URL https://arxiv.org/abs/2505.12523.

Mendula M, Bellavista P, Levorato M et al. A novel middleware for adaptive and efficient split computing for real-time object detection. Pervasive and Mobile Computing ; 108: 102028. URL https://doi.org/10.1016/j.pmcj.2025.102028.

Li J and Ye J. Edge-yolo: Lightweight infrared object detection method deployed on edge devices. Applied Sciences ; 13(7): 4402. URL https://doi.org/10.3390/app13074402.

Kim K, Jang SJ, Park J et al. Lightweight and energy-efficient deep learning accelerator for real- time object detection on edge devices. Sensors ; 23(3): 1185. URL https://doi.org/10. 3390/s23031185.

Huang YX, Liu HI, Shuai HH et al. Dq-detr: Detr with dynamic query for tiny object detection. In Computer Vision – ECCV 2024. Lecture Notes in Computer Science, volume 15134. Cham: Springer, p. 290–305.

Halder S, Rafiqul Islam M, Mamun Q et al. A comprehensive survey on ai-enabled secure social industrial internet of things in the agri-food supply chain. Smart Agricultural Technology ; 11: 100902. URL https://doi.org/10.1016/j.atech.2025.100902.