lmplementasi YOLOv11 untuk Penghitungan Kerumunan Real-Time dalam Arsitektur Microservices pada Video CCTV

M.Sultonun  Naim; Salamun Rohman Nudin

doi:10.32664/smatika.v16i02.2378

Penulis

M.Sultonun Naim Universitas Negeri Surabaya
Salamun Rohman Nudin Universitas Negeri Surabaya

DOI:

https://doi.org/10.32664/smatika.v16i02.2378

Kata Kunci:

YOLOv11; Computer Vision; Deep Learning; Microservices; Crowd Counting.

Abstrak

Penyediaan informasi publik yang akurat dan mudah diakses mengenai kondisi taman kota masih menjadi tantangan dalam pengelolaan ruang terbuka publik. Masyarakat masih belum memiliki sistem yang memungkinkan mereka memantau tingkat keramaian taman secara langsung, sehingga pengamatan kondisi taman secara real-time menjadi sulit. Penelitian ini mengembangkan sistem informasi penghitungan pengunjung taman berbasis computer vision dengan menggunakan algoritma YOLOv11 untuk mendeteksi dan menghitung keramaian dari aliran video CCTV. Sistem ini dirancang menggunakan arsitektur microservices dengan kerangka kerja FastAPI untuk mendukung deteksi real-time dan integrasi data ke dalam situs web pemantauan taman Surabaya. Proses penelitian melibatkan beberapa tahap, termasuk persiapan dataset, pelabelan data menggunakan Roboflow, pelatihan model YOLOv11, pemilihan optimizer yang paling optimal, dan implementasi sistem pada backend deteksi. Hasil menunjukkan bahwa model YOLOv11m dengan optimizer SGD mencapai kinerja terbaik, dengan skor mAP@50 sebesar 92,76%, nilai recall sebesar 89,75%, dan skor F1 sebesar 90,07%. Selain itu, sistem berhasil melakukan deteksi dan penghitungan kerumunan secara real-time di bawah berbagai tingkat kepadatan kerumunan, kondisi pencahayaan, dan sudut kamera CCTV.

Referensi

[1] M. K. Kadri, R. Armanda, G. Purba, dan Y. Fitriani, “Kesesuaian Pengadaan Ruang Terbuka Hijau Taman Kota Berdasarkan Standar Minimal Pelayanan Penduduk di Kota Surabaya Suitability of City Park Green Open Space Procurement Based on Minimum Standards of Resident Services in the City of Surabaya,” vol. 02, no. 01, hal. 95–99, 2023, doi: 10.35718/compact.v2i1.853.

[2] D. D. Saputri, “Penilaian Fungsi Taman Kota Sebagai Ruang Terbuka Publik di Kota Surabaya,” vol. 13, no. 2, hal. 40–47, 2018, doi: 10.12962/j2716179X.v13i2.7113.

[3] T. Abdul, R. Shyaa, dan A. A. Hashim, “Enhancing real human detection and people counting using YOLOv8,” vol. 00061, hal. 1–8, 2024, doi: https://doi.org/10.1051/bioconf/20249700061.

[4] S. Hassan, “Systematic scalability analysis for microservices granularity adaptation design decisions,” no. December 2021, hal. 1378–1401, 2022, doi: 10.1002/spe.3069.

[5] F. Aldi, I. Nozomi, dan S. Rahmawati, “Real-Time Human Detection and Counting System Based on YOLOv11 for Smart Campus Environment,” JET-UMY, vol. 9, no. 2, hal. 64–73, 2025, doi: 10.18196/jet.v9i2.29213.

[6] A. A. Lubis, A. Prasasta, dan D. A. Sari, “Towards Efficient Crowd Counting and Behavior Analysis Using,” vol. 4, no. 1, hal. 35–47, 2025, doi: https://doi.org/10.63876/ijtm.v4i1.128.

[7] M. Khanam, R , Hussain, “YOLOV11: AN OVERVIEW OF THE KEY ARCHITECTURAL ENHANCEMENTS,” vol. 2024, hal. 1–9, 2024, doi: 10.48550/arXiv.2410.17725.

[8] N. Dwivedi, D. Katiyar, dan G. Goel, “A Comparative Study of Various Software Development Life Cycle ( SDLC ) Models,” IJRESM, vol. 5, no. 3, hal. 141–144, 2022, doi: 10.51316/ijresm.2022.1881.

[9] J. R. Terven dan D. M. Cordova-esparaza, “A COMPREHENSIVE REVIEW OF YOLO: FROM YOLOV1 TO YOLOV8 AND BEYOND,” Mach. Learn. Knowl. Extr., hal. 1–27, 2023, doi: 10.3390/make5040083.

[10] G. Aurelia, N. Lubis, R. Purnamasari, dan K. Saleh, “Designing a Real-Time TOXMAP Backend Based on FastAPI and Firebase for B3 Waste,” vol. 8, no. 1, 2025, doi: 10.32877/bt.v8i1.2888.

[11] F. Ciaglia, F. S. Zuppichini, P. Guerrie, M. Mcquade, dan J. Solawetz, “Roboflow 100: A Rich, Multi-Domain Object Detection Benchmark,” 2022, doi: 10.48550/arXiv.2211.13523.

[12] M. E. Khan, “A Comparative Study of White Box , Black Box and Grey Box Testing Techniques,” IJACSA, vol. 3, no. 6, hal. 12–15, 2012, doi: 10.14569/IJACSA.2012.030603.

[13] F. Zufari dan Y. Irawan, “Enhancing Real Time Crowd Counting Using YOLOv8 Integrated with Microservices Architecture for Dynamic Object Detection in High Density Environments,” KESATRIA, vol. 6, no. 1, 2025, doi: 10.30645/kesatria.v6i1.575.

[14] S. A. Rumapea dan M. F. Fawwaz, “Implementation of YOLOv8 in Object Recognition Systems for Public Area Security in Kebun Raya Bogor,” Ultim. J. Tek. Inform., vol. 17, no. 1, hal. 1–10, 2025, doi: 10.31937/ti.v17i1.4133.

[15] N. Sujana, M. M. Mutoffar, dan A. A. Haryanto, “ANALISIS KINERJA YOLOV8 OPTIMALISASI ROBOFLOW UNTUK DETEKSI EKSPRESI WAJAH EMOSIONAL DENGAN MACHINE LEARNING,” NARATIF, vol. 06, no. 02, hal. 115–124, 2024, doi: 10.53580/naratif.v6i2.292.

[16] C. Shorten dan T. M. Khoshgoftaar, “A survey on Image Data Augmentation for Deep Learning,” J. Big Data, 2019, doi: 10.1186/s40537-019-0197-0.

[17] J. Meng, Y. Li, Z. Li, dan W. Xie, “TARGET DETECTION FOR WEEDING ROBOTS BASED ON IMPROVED YOLOv11 MODELS,” engenhariaagricola, vol. 4430, 2025, doi: 10.1590/1809-4430-eng.agric.v45e20250032/2025.

[18] T. H. Pinem dan Z. P. Putra, “Evaluasi Kinerja Algoritma Klasifikasi Deep Learning dalam Prediksi Diabetes,” vol. 17, no. 1, hal. 17–28, 2025, doi: 10.22441/fifo.2025.v17i1.003.

[19] C. S. B. Sakti dan I. Hermawan, “IMPLEMENTASI ARSITEKTUR MICROSERVICE PADA BACK END SISTEM INFORMASI I ATLANTAS BERBASIS WEBSITE,” J. Teknol. terpadu, vol. 6, no. 2, hal. 96–104, 2020, doi: 10.54914/jtt.v6i2.281.

[20] S. Atmojo, R. Utami, S. Dewi, dan N. Widhiyanta, “Implementasi Sistem-informasi DesaBerbasisArsitektur Microservices,” SMATIKA STIKI Inform. J., vol. 12, no. 1, hal. 55–66, 2022, doi: https://doi.org/10.32664/smatika.v12i01.658.

[21] J. Chen, “Model Algorithm Research based on Python Fast API,” vol. 3, no. 9, hal. 7–10, 2023, doi: 10.54691/fse.v3i9.5591.

[22] K. Boudjit dan N. Ramzan, “Human detection based on deep learning YOLO-v2 for real-time UAV applications Human detection based on deep learning YOLO-v2 for real-time,” J. Exp. Theor. Artif. Intell., vol. 00, no. 00, hal. 1–18, 2021, doi: 10.1080/0952813X.2021.1907793.

[23] R. Shi, T. Li, dan Y. Yamaguchi, “An attribution-based pruning method for real-time mango detection with YOLO network,” Comput. Electron. Agric., vol. 169, no. July 2019, hal. 105214, 2020, doi: 10.1016/j.compag.2020.105214.

[24] C. Wang dan H. M. Liao, “YOLOv4: Optimal Speed and Accuracy of Object Detection,” 2020, doi: 10.48550/arXiv.2004.10934.

[25] H. T. Ngoc, K. H. Nguyen, H. K. Hua, H. Vu, N. Nguyen, dan L. Quach, “Optimizing YOLO Performance for Traffic Light Detection and End-to-End Steering Control for Autonomous Vehicles in Gazebo-ROS2,” vol. 14, no. 7, hal. 475–484, 2023, doi: 10.14569/IJACSA.2023.0140752.

[26] A. Syahputra, M. F. Azhary, A. Binti, A. Rahman, dan A. Saad, “Occupancy Measurement in Under-Actuated Zones : YOLO-based Deep Learning Approach,” vol. 15, no. 2, hal. 757–769, 2024, doi: 10.14569/IJACSA.2024.0150277.

[27] S. Woo, J. Park, J. Lee, dan S. Kweon, “CBAM: Convolutional Block Attention Module,” Eur. Conf. Comput. Vis., hal. 3–19, 2018, doi: 10.1007/978-3-030-01234-2_1.

[28] C. Guo et al., “Zero-Reference Deep Curve Estimation for Low-Light Image Enhancement,” Conf. Comput. Vis. Pattern Recognit., hal. 1780–1789, 2020, doi: 10.1109/CVPR42600.2020.00343.

[29] W. Liu, M. Salzmann, P. Fua, dan D. L. Epfl, “Context-Aware Crowd Counting,” Conf. Comput. Vis. Pattern Recognit., hal. 5094–5103, 2019, doi: 10.1109/CVPR.2019.00524.