Development of a Modular Web Platform for Detection, Segmentation, and Object Tracking with Dynamic Model Management

Authors

  • Sylvia✉ Politeknik Negeri Lampung, Bandar Lampung, Indonesia Author
  • Atika Arpan Politeknik Negeri Lampung, Bandar Lampung, Indonesia Author
  • Rizka Permata Politeknik Negeri Lampung, Bandar Lampung, Indonesia Author
  • Muhammad Reza Redo Islami Politeknik Negeri Lampung, Bandar Lampung, Indonesia Author
Diajukan 30 Sep 2025 Diterbitkan 30 Des 2025
DOI: https://doi.org/10.59025/48b50fbf

Keywords:

web platform, object detection, image segmentation, dynamic model management, YOLO

Abstract

Computer vision research within campus environments is frequently slowed by lengthy model-testing cycles: every model swap forces a server restart, code modification, and pipeline reconfiguration. This study develops Poli SEG, a modular Flask-based web platform that unifies four core computer vision tasks (object detection, instance segmentation, object tracking, and automatic annotation) within a single integrated interface, supported by a dynamic model management mechanism backed by MD5 hashing. Development follows an iterative SDLC model with seven stages, from literature study to functional evaluation. The central contribution is a model_manager module that loads and swaps models on the fly (hot-swap) without restarting the server, paired with a Flask Blueprint layer that isolates each feature as an independent module. Black-box testing across six main modules with 60 scenarios produced an average pass rate of 96.2%, an average inference time of 38.4 ms for 640x640 pixel images, and a 100% success rate for hot-swapping six different models. The Slicing Aided Hyper Inference mode successfully improved small-object detection on high-resolution images. Poli SEG can serve as a multi-task research sandbox and a cross-researcher model testing platform that removes the technical friction of model loading and reconfiguration.

References

  1. J. Terven, D.-M. Cordova-Esparza, dan J.-A. Romero-Gonzalez, "A Comprehensive Review of YOLO Architectures in Computer Vision: From YOLOv1 to YOLOv8 and YOLO-NAS," Mach. Learn. Knowl. Extr., vol. 5, no. 4, hal. 1680-1716, 2023, doi: 10.3390/make5040083.
  2. R. Sapkota et al., "YOLO advances to its genesis: a decadal and comprehensive review of the You Only Look Once (YOLO) series," Artif. Intell. Rev., vol. 58, no. 7, 2025, doi: 10.1007/s10462-025-11253-3.
  3. L. Wang et al., "A systematic literature review of lightweight YOLO models for object detection," PeerJ Comput. Sci., vol. 11, 2025, doi: 10.7717/peerj-cs.3357.
  4. Y. Zhao et al., "DETRs Beat YOLOs on Real-time Object Detection," dalam Proc. IEEE/CVF Conf. Computer Vision and Pattern Recognition (CVPR), 2024, hal. 16965-16974, doi: 10.1109/CVPR52733.2024.01605.
  5. F. M. A. Mazen dan Y. Shaker, "Small Object Detection in Complex Images: Evaluation of Faster R-CNN and Slicing Aided Hyper Inference," Int. J. Adv. Comput. Sci. Appl., vol. 16, no. 3, hal. 951-960, 2025, doi: 10.14569/IJACSA.2025.0160393.
  6. M. Ishtiaq dan J.-U. Won, "YOLO-SIFD: YOLO with Sliced Inference and Fractal Dimension Analysis for Improved Fire and Smoke Detection," Comput. Mater. Continua, vol. 82, no. 3, hal. 5343-5361, 2025, doi: 10.32604/cmc.2025.061466.
  7. S. Abba, A. M. Bizi, J.-A. Lee, S. Bakouri, dan M. L. Crespo, "Real-time object detection, tracking, and monitoring framework for security surveillance systems," Heliyon, vol. 10, no. 15, 2024, doi: 10.1016/j.heliyon.2024.e34922.
  8. N. N. Zhorif, R. K. Anandyto, A. U. Rusyadi, dan E. Irwansyah, "Implementation of Slicing Aided Hyper Inference (SAHI) in YOLOv8 to Counting Oil Palm Trees Using High-Resolution Aerial Imagery Data," Int. J. Adv. Comput. Sci. Appl., vol. 15, no. 7, 2024, doi: 10.14569/IJACSA.2024.0150786.
  9. S. Kang, H. Hu, dan L. Wang, "Object Detection YOLO Algorithms and Their Industrial Applications: Overview and Comparative Analysis," Electronics, vol. 14, no. 6, 2025, doi: 10.3390/electronics14061104.
  10. J. E. Gallagher dan E. J. Oughton, "Surveying You Only Look Once (YOLO) Multispectral Object Detection: Advancements, Applications, and Challenges," IEEE Access, vol. 13, hal. 7366-7395, 2025, doi: 10.1109/ACCESS.2025.3526458.
  11. M. F. Rashad dan Q. I. Ali, "Deploying Android-Based Smart RSUs with YOLOv8 and SAHI for Enhanced Traffic Management," Diyala J. Eng. Sci., vol. 18, no. 1, hal. 70-90, 2025, doi: 10.24237/djes.2025.18104.
  12. J. Wei, X. Zhao, X. Yang, dan G. Liu, "A Review of YOLO Algorithm and Its Applications in Autonomous Driving Object Detection," IEEE Access, vol. 13, hal. 93688-93711, 2025, doi: 10.1109/ACCESS.2025.3573376.
  13. M. Mao, Y. Hong, dan H. Liu, "YOLO Object Detection for Real-Time Fabric Defect Inspection in the Textile Industry: A Review of YOLOv1 to YOLOv11," Sensors, vol. 25, no. 7, 2025, doi: 10.3390/s25072270.
  14. M. G. Ragab et al., "A Comprehensive Systematic Review of YOLO for Medical Object Detection (2018 to 2023)," IEEE Access, vol. 12, 2024, doi: 10.1109/ACCESS.2024.3386826.
  15. Z. Wang, C. Li, H. Xu, dan X. Zhu, "Mamba YOLO: A Simple Baseline for Object Detection with State Space Model," dalam Proc. AAAI Conf. Artif. Intell., vol. 39, no. 8, 2025, hal. 8205-8213, doi: 10.1609/aaai.v39i8.32885.
  16. T. Cheng et al., "YOLO-World: Real-Time Open-Vocabulary Object Detection," dalam Proc. IEEE/CVF Conf. Computer Vision and Pattern Recognition (CVPR), 2024, hal. 16901-16911, doi: 10.1109/CVPR52733.2024.01599.
  17. J. Doherty, B. Gardiner, E. Kerr, dan N. Siddique, "BiFPN-YOLO: One-stage object detection integrating Bi-Directional Feature Pyramid Networks," Pattern Recognit., vol. 160, art. 111209, 2025, doi: 10.1016/j.patcog.2024.111209.
  18. D. Dodic et al., "SAHI-Tuned YOLOv5 for UAV Detection of TM-62 Anti-Tank Landmines: Small-Object, Occlusion-Robust, Real-Time Pipeline," Computers, vol. 14, no. 10, 2025, doi: 10.3390/computers14100448.
  19. Z. Cai dan N. Vasconcelos, "Cascade R-CNN: High Quality Object Detection and Instance Segmentation," IEEE Trans. Pattern Anal. Mach. Intell., vol. 43, no. 5, hal. 1483-1498, 2021, doi: 10.1109/TPAMI.2019.2956516.
  20. M. Kang, C.-M. Ting, F. F. Ting, dan R. C.-W. Phan, "ASF-YOLO: A novel YOLO model with attentional scale sequence fusion for cell instance segmentation," Image Vis. Comput., vol. 147, art. 105057, 2024, doi: 10.1016/j.imavis.2024.105057.
  21. X. Yuan, H. Yu, T. Geng, R. Ma, dan P. Li, "Enhancing sustainable Chinese cabbage production: a comparative analysis of multispectral image instance segmentation techniques," Front. Sustain. Food Syst., vol. 8, art. 1433701, 2024, doi: 10.3389/fsufs.2024.1433701.
  22. M. Lalinia dan A. Sahafi, "Colorectal polyp detection in colonoscopy images using YOLO-V8 network," Signal Image Video Process., vol. 18, no. 3, hal. 2047-2058, 2024, doi: 10.1007/s11760-023-02835-1.
  23. M. Alaftekin, I. Pacal, dan K. Cicek, "Real-time sign language recognition based on YOLO algorithm," Neural Comput. Appl., vol. 36, hal. 7609-7624, 2024, doi: 10.1007/s00521-024-09503-6.
  24. Z. Fei et al., "Comprehensive Performance Evaluation of YOLOv11, YOLOv10, YOLOv9, YOLOv8 and YOLOv5 on Object Detection of Power Equipment," arXiv preprint arXiv:2411.18871, 2024, doi: 10.48550/arXiv.2411.18871.
  25. E. H. Alkhammash, "A Comparative Analysis of YOLOv9, YOLOv10, YOLOv11 for Smoke and Fire Detection," Fire, vol. 8, no. 1, art. 26, 2025, doi: 10.3390/fire8010026.
  26. Y. Alhwaiti et al., "Leveraging YOLO deep learning models to enhance plant disease identification," Sci. Rep., vol. 15, art. 92143, 2025, doi: 10.1038/s41598-025-92143-0.
  27. M. Al Husaini, A. R. Raharja, V. H. C. Putra, dan H. H. Lukmana, "Enhanced Plant Disease Detection Using Computer Vision YOLOv11: Pre-Trained Neural Network Model Application," J. Comput. Netw. Archit. High Perform., vol. 7, no. 1, hal. 82-95, 2025, doi: 10.47709/cnahpc.v7i1.5146.
Cover JTDS Vol 3 No 2 (2025)

Downloads

How to Cite

Development of a Modular Web Platform for Detection, Segmentation, and Object Tracking with Dynamic Model Management. (2025). Journal of Technology and Data Science, 3(2), 229-244. https://doi.org/10.59025/48b50fbf