The DOI matches its Crossref record; 2 of the 2 references it cites were found in a registry or an index.
Quantized vs. Full-Precision YOLO Models on Edge Devices: A Performance Benchmark for Real-Time License Plate Detection in Smart Parking Systems
Abstract
The deployment of deep learning-based vision systems on edge devices introduces a complex trade-off between computational efficiency and detection accuracy. In this work, we investigate this trade-off in the context of a multi-stage Automatic License Plate Recognition (ALPR) pipeline, evaluated in two heterogeneous edge execution environments: a general-purpose Raspberry Pi 5 single-board computer and the ARTPEC-8 system-on-chip integrated into an Axis smart camera, where neural network inference is accelerated by the on-chip DLPU. All experiments were performed using pre-recorded images loaded from the file system; neither the Axis camera sensor nor a live video stream was used. This study evaluates the impact of model architecture, numerical precision, and input resolution on both inference latency and detection performance. YOLOv5- and YOLOv8-based models were analyzed under multiple quantization schemes (FP32, FP16, dynamic, and INT8), while a cross-platform benchmark was conducted to assess the benefits and limitations of hardware acceleration. The results show that dedicated accelerators provide significant latency reduction at higher resolutions; however, this advantage is accompanied by reduced flexibility and increased sensitivity to quantization effects. In contrast, CPU-based execution enables the use of more recent and quantization-robust model architectures, which can partially compensate for the lack of hardware acceleration when combined with resolution scaling. Furthermore, the analysis hig ights the importance of hybrid-resolution processing in multi-stage pipelines, where different stages can operate at different input resolutions to balance accuracy and performance. The findings demonstrate that optimal system design requires a joint consideration of hardware characteristics, model architecture, and quantization strategy, rather than relying on a single optimization dimension. The presented results provide practical insights for the design of efficient and robust edge-based ALPR systems, with direct implications for real-world industrial deployments.
Overview
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
You Only Look Once: Unified, Real-Time Object Detection
Redmon, J., Divvala, S., Girshick, R., & Farhadi, A. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.91 · Cited by 53,803 · cited 2×
“In this work, we use models from the YOLO (You Only Look Once) family [22], which are widely adopted in real-time embedded vision systems due to their favorable speed–accuracy trade-off. https://doi.org/10.3390/s26165034 Alternative architectures exist.”
InfluentialMazzia, V., Khaliq, A., Salvetti, F., & Chiaberge, M. (2020) · IEEE Access · doi:10.1109/access.2020.2964608 · Cited by 212
Research works printed but not linked: 19
- Li, 2020, Edge AI: On-Demand Accelerating Deep Neural Network Inference via Edge Computing, IEEE Trans. Wirel. Commun., 19, 447, 10.1109/TWC.2019.2946140 · DOI · Paper page
- Han, S., Mao, H., and Dally, W.J. (2015). Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding. arXiv.
- Sze, 2017, Efficient Processing of Deep Neural Networks: A Tutorial and Survey, Proc. IEEE, 105, 2295, 10.1109/JPROC.2017.2761740 · DOI · Paper page
- Redmon, J., and Farhadi, A. (2018). YOLOv3: An Incremental Improvement. arXiv.
- Ferrari, 2018, License Plate Detection and Recognition in Unconstrained Scenarios, Computer Vision–ECCV 2018, Volume 11216, 593, 10.1007/978-3-030-01258-8_36 · DOI
- Tang, X., Han, S., Zhang, L.L., Cao, T., and Liu, Y. (2021, January 8–11). To Bridge Neural Network Design and Real-World Performance: A Behaviour Study for Neural Networks. Proceedings of the Machine Learning and Systems 3 (MLSys 2021), Xiamen, China.
- Hosseininoorbin, 2023, Exploring Edge TPU for Network Intrusion Detection in IoT, J. Parallel Distrib. Comput., 179, 104712, 10.1016/j.jpdc.2023.05.001 · DOI
- Mittal, 2019, A Survey on Optimized Implementation of Deep Learning Models on the NVIDIA Jetson Platform, J. Syst. Archit., 97, 428, 10.1016/j.sysarc.2019.01.011 · DOI
- Febvay, M., and Bounekkar, A. (2022, January 26–27). Deep Learning Frameworks Evaluation for Image Classification on Resource Constrained Device. Proceedings of the 11th International Conference on Embedded Systems and Applications (EMSA 2022), Sydney, Australia. · DOI
- Mohammadi, M., Smith, H., Khan, L., and Zand, R. (2023, January 5–7). Facial Expression Recognition at the Edge: CPU vs. GPU vs. VPU vs. TPU. Proceedings of the ACM Great Lakes Symposium on VLSI, Knoxville, TN, USA. · DOI
- Baller, S.P., Jindal, A., Chadha, M., and Gerndt, M. (2021). DeepEdgeBench: Benchmarking Deep Neural Networks on Edge Devices. arXiv. · DOI · Paper page
- Banbury, C., Reddi, V.J., Torelli, P., Holleman, J., Jeffries, N., Kiraly, C., Montino, P., Kanter, D., Ahmed, S., and Pau, D. (2021). MLPerf Tiny Benchmark. arXiv.
- Amanatidis, P., Iosifidis, G., and Karampatzakis, D. (2021, January 26–28). Comparative Evaluation of Machine Learning Inference Machines on Edge-Class Devices. Proceedings of the 25th Pan-Hellenic Conference on Informatics, Volos, Greece. · DOI
- Howard, A., Sandler, M., Chen, B., Wang, W., Chen, L., Tan, M., Chu, G., Vasudevan, V., Zhu, Y., and Pang, R. (2019). Searching for MobileNetV3. Proceedings of the IEEE/CVF International Conference on Computer Vision, IEEE. · DOI · Paper page
- Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C., and Berg, A.C. (2016). SSD: Single Shot MultiBox Detector. Proceedings of the European Conference on Computer Vision, Springer. · DOI · Paper page
- Howard, A.G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., Andreetto, M., and Adam, H. (2017). MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications. arXiv.
- Huang, J., Rathod, V., Sun, C., Zhu, M., Korattikara, A., Fathi, A., Fischer, I., Wojna, Z., Song, Y., and Guadarrama, S. (2017, January 21–26). Speed/Accuracy Trade-Offs for Modern Convolutional Object Detectors. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA. · DOI · Paper page
- Lin, T.Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Dollár, P., and Zitnick, C.L. (2014). Microsoft COCO: Common Objects in Context. Proceedings of the European Conference on Computer Vision, Springer. · DOI · Paper page
- Wen, 2020, UA-DETRAC: A New Benchmark and Protocol for Multi-Object Detection and Tracking, Comput. Vis. Image Underst., 93, 102907, 10.1016/j.cviu.2020.102907 · DOI · Paper page
Metadata
- Authors
- Burkus, E., Lestyán, B., Dénes-Fazakas, L., & Eigner, G.
- Venue
- Sensors
- Publisher
- MDPI AG
- Published
- 2026-08-08
- Volume
- 26
- Issue
- 16
- Pages
- 5034
- ISSN
- 1424-8220
- Cited by
- 0
- References
- 2
- Language
- en
- Funders
- Óbuda University
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.
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: Quantized vs. Full-Precision YOLO Models on Edge Devices: A Performance Benchmark for Real-Time License Plate Detection in Smart Parking Systems. https://app.cochat.ai/papers/10.3390/s26165034
- The paper
- Burkus, E., Lestyán, B., Dénes-Fazakas, L., & Eigner, G. (2026). Quantized vs. Full-Precision YOLO Models on Edge Devices: A Performance Benchmark for Real-Time License Plate Detection in Smart Parking Systems. Sensors. https://doi.org/10.3390/s26165034
