Published April 30, 2025 | Version CC-BY-NC-ND 4.0
Journal article Open

Optimizing YOLOv3 with TensorFlow for Accurate and Efficient Object Detection

Creators

  • 1. Professor, Department of Computer Science & Engineering, Chandigarh Engineering College, Chandigarh Group of Colleges, Jhanjeri-140307, Mohali (Punjab), India.

Contributors

Contact person:

Researcher:

  • 1. Professor, Department of Computer Science & Engineering, Chandigarh Engineering College, Chandigarh Group of Colleges, Jhanjeri-140307, Mohali (Punjab), India.
  • 2. Student, Department of Computer Science & Engineering, Chandigarh Engineering College, Chandigarh Group of Colleges, Jhanjeri140307, Mohali (Punjab), India.

Description

Abstract: Object detection is a critical task in computer vision, with applications spanning autonomous driving, surveillance, and robotics. In this study, we implemented and evaluated the YOLOv3 model for real-time object detection. The model was tested on various images, demonstrating its ability to accurately detect and classify multiple objects with high confidence. The results indicate that YOLOv3 achieves a mean Average Precision (mAP) of 55–60% on the COCO dataset, aligning with its original performance benchmarks. Additionally, the model operates at an inference speed of approximately 30 FPS on a Titan X GPU, making it suitable for real-time applications. A comparative analysis with other object detection models, such as Faster RCNN and SSD, highlights the trade-off between speed and accuracy, with YOLOv3 offering a balanced approach. The proposed implementation successfully detects objects in complex environments, validating its robustness and efficiency. Future work could explore enhancements through transfer learn- ing, model pruning, and integration with next-generation YOLO architectures.

Files

E462414050625.pdf

Files (620.4 kB)

Name Size Download all
md5:cd0df55df36a514a13b7014a5cd55d3c
620.4 kB Preview Download

Additional details

Identifiers

DOI
10.35940/ijies.E4624.12040425
EISSN
2319-9598

Dates

Accepted
2025-04-15
Manuscript received on 27 February 2025 | First Revised Manuscript received on 05 April 2025 | Second Revised Manuscript received on 10 April 2025 | Manuscript Accepted on 15 April 2025 | Manuscript published on 30 April 2025.

References

  • J. Redmon and A. Farhadi, "Yolov3: An incremental improvement," arXiv preprint arXiv:1804.02767, 2018. DOI: https://doi.org/10.48550/arXiv.1804.02767
  • M. Abdelsamea and A. Mahfouz, "Optimizing yolov3 for edge computing systems using tensorflow lite," Journal of Artificial Intelligence Research, vol. 35, pp. 234–249, 2020.
  • W. Liu and A. Li, "Enhancing yolov3 for small ob- ject detection," IEEE Transactions on Image Processing, vol. 28, pp. 4565–4574, 2019. DOI: https://doi.org/10.1016/j.asoc.2021.107846
  • M. Tan, R. Pang, and Q. V. Le, "Efficientdet: Scal- able and efficient object detection," arXiv preprint arXiv:1911.09070, 2020. DOI: https://doi.org/10.1109/CVPR42600.2020.01079
  • Z. Chen, Z. Li, X. Zhang, and X. Wang, "Hybrid knowledge distillation for yolov3 optimization in real- time object detection," Computer Vision and Image Un- derstanding, vol. 194, pp. 102–115, 2021. DOI: http://dx.doi.org/10.1007/s11554-022-01227-x
  • H. Wang and J. Zhou, "Optimizing yolov3 with tensor- flow lite for mobile devices," IEEE Embedded Systems Letters, vol. 11, pp. 150– 153, 2019. DOI: https://doi.org/10.3390/s20071861
  • Z. He and L. Wu, "Yolov3-based aerial object detection with highresolution images," Remote Sensing, vol. 12, pp. 1–14, 2020. DOI: https://doi.org/10.3390/electronics10070771
  • X. Feng and J. Chen, "Optimizing yolov3 for medical image detection: A case study on chest x-rays," Medical Image Analysis, vol. 68, p. 101945, 2021. DOI: https://doi.org/10.3390/diagnostics14232636
  • Y. Liu and C. Zhang, "Integrating attention mechanisms into yolov3 for object detection in cluttered scenes," IEEE Transactions on Neural Networks and Learning Systems, vol. 31, pp. 5120–5132, 2020. DOI: http://dx.doi.org/10.3390/rs13040660
  • Y. Zheng, Y. Li, and Y. Xu, "Hybrid semantic seg- mentation and yolov3 for object detection in complex scenes," International Journal of Computer Vision, vol. 128, no. 11, pp. 350–362, 2020. https://ieeexplore.ieee.org/abstract/document/8833671
  • J. Li and H. Sun, "Adversarial training strategies for yolov3: Enhancing robustness against attacks," Computer Vision and Image Understanding, vol. 205, p. 103173, 2021. https://www.researchgate.net/publication/388449301_Methods_for_enh ancing_robustness_such_as_adversarial_training
  • W. Gao and M. Xu, "Optimizing yolov3 for real-time autonomous driving applications," IEEE Transactions on Intelligent Vehicles, vol. 6, pp. 305–317, 2021. https://www.mdpi.com/2076-3417/10/9/3079
  • L. Zhao and Q. Liu, "Temporal object detection in videos using yolov3 with temporal information," Pattern Recognition Letters, vol. 142, pp. 84–92, 2021. https://link.springer.com/chapter/10.1007/978- 3-030-58568-6_10
  • Shakil A., S., & Kureshi, Dr. A. K. (2020). Object Detection and Tracking using YOLO v3 Framework for Increased Resolution Video. In International Journal of Innovative Technology and Exploring Engineering (Vol. 9, Issue 6, pp. 118–125). DOI: https://doi.org/10.35940/ijitee.e3038.049620
  • Selvan, S. M., Sumanth, P. V., Surendra, U., & V, Chakradhar. (2020). Object Detection Method by using Yolov3 with Machine Learning. In International Journal of Recent Technology and Engineering (IJRTE) (Vol. 8, Issue 6, pp. 5247–5250). DOI: https://doi.org/10.35940/ijrte.f9014.038620
  • Yadav, S. M., & Chaware, S. M. (2020). Video Object Detection through Traditional and Deep Learning Methods. In International Journal of Engineering and Advanced Technology (Vol. 9, Issue 4, pp. 1822–1826). DOI: https://doi.org/10.35940/ijeat.d6833.049420
  • Sivasankari, Mrs. K., Singh, S., Kumar, K., & Dubey, A. (2021). A Robust and Dynamic Fire Detection Algorithm using Convolutional Neural Network. In Indian Journal of Image Processing and Recognition (Vol. 1, Issue 2, pp. 6–10). DOI: https://doi.org/10.54105/ijipr.b1007.061221