Email this article 
(Universitas Ilsam Negeri Maulana Malik Ibrahim Malang, Indonesia)(2) Fachrul Kurniawan
(Universitas Ilsam Negeri Maulana Malik Ibrahim Malang, Indonesia)(3) Mochamad Imamudin
(Universitas Ilsam Negeri Maulana Malik Ibrahim Malang, Indonesia)*corresponding author
AbstractFace detection and tracking in crowded environments remain challenging due to occlusion, object overlap, and high visual similarity between individuals. In tracking-by-detection systems, detection quality plays a crucial role in tracking stability, yet the relationship between detection performance and identity consistency is not fully explored. This study proposes an integrated framework combining YOLOv9, an attention mechanism, and DeepSORT to enhance feature representation and improve identity tracking in dense environments, where the attention mechanism is embedded in the detection stage to strengthen feature discriminability and enable more stable identity association across frames. The system is evaluated using three dataset partitioning scenarios (90:10, 50:50, and 10:90) to analyze the impact of training data distribution. Experimental results show that the 90:10 configuration achieves the best performance, with precision 0.9398, recall 0.8869, F1-score 0.9126, MOTA 71.0%, and IDF1 73.0%. These findings confirm that improved feature representation significantly enhances detection quality and tracking stability, and demonstrate that feature stability is more critical than adopting newer detection architectures for achieving robust tracking performance in crowded environments.
KeywordsFace Tracking;YOLOv9;Attention Mechanism;DeepSORT;Crowded Environment
|
DOIhttps://doi.org/10.29099/ijair.v10i1.1705 |
Article metrics10.29099/ijair.v10i1.1705 Abstract views : 76 | PDF views : 12 |
Cite |
Full Text Download
|
References
D. Djarah, A. Benmakhlouf, G. Zidani, and L. Khettache, “Online Multi-object Tracking with YOLOv9 and DeepSORT Optimized by Optical Flow,” Engineering, Technology and Applied Science Research, vol. 14, no. 6, pp. 17922–17930, Dec. 2024, doi: 10.48084/etasr.8770.
X. Zhou, S. Chan, C. Qiu, X. Jiang, and T. Tang, “Multi-Target Tracking Based on a Combined Attention Mechanism and Occlusion Sensing in a Behavior-Analysis System,” Sensors, vol. 23, no. 6, p. 2956, Mar. 2023, doi: 10.3390/s23062956.
L. Almuqren, M. A. Hamza, A. Mohamed, and A. A. Abdelmageed, “Automated Video-Based Face Detection Using Harris Hawks Optimization with Deep Learning,” Computers, Materials and Continua, vol. 75, no. 3, pp. 4917–4933, 2023, doi: 10.32604/cmc.2023.037738.
A. A. Alsabei, T. M. Alsubait, and H. H. Alhakami, “Enhancing Crowd Safety at Hajj: Real-Time Detection of Abnormal Behavior Using YOLOv9,” IEEE Access, vol. 13, pp. 37748–37761, 2025, doi: 10.1109/ACCESS.2025.3545256.
R. Nasir, Z. Jalil, M. Nasir, T. Alsubait, M. Ashraf, and S. Saleem, “An enhanced framework for real-time dense crowd abnormal behavior detection using YOLOv8,” Artif. Intell. Rev., vol. 58, no. 7, p. 202, Apr. 2025, doi: 10.1007/s10462-025-11206-w.
F. Ngeni, J. Mwakalonge, and S. Siuhi, “Solving traffic data occlusion problems in computer vision algorithms using DeepSORT and quantum computing,” Journal of Traffic and Transportation Engineering (English Edition), vol. 11, no. 1, pp. 1–15, Feb. 2024, doi: 10.1016/j.jtte.2023.05.006.
Tommy, R. Siregar, and E. Rahman Syahputra, “Low-Resolution Face Image Reconstruction Using Multi-Stage FSRCNN to Improve Face Detection and Tracking Accuracy in CCTV Surveillance,” JOIV : International Journal on Informatics Visualization, 2025, doi: https://dx.doi.org/10.62527/joiv.9.3.3160.
K. Alkandary, A. S. Yildiz, and H. Meng, “A Comparative Study of YOLO Series (v3–v10) with DeepSORT and StrongSORT: A Real-Time Tracking Performance Study,” Electronics (Basel)., vol. 14, no. 5, p. 876, Feb. 2025, doi: 10.3390/electronics14050876.
G. Ciaparrone, F. Luque Sánchez, S. Tabik, L. Troiano, R. Tagliaferri, and F. Herrera, “Deep learning in video multi-object tracking: A survey,” Neurocomputing, vol. 381, pp. 61–88, Mar. 2020, doi: 10.1016/j.neucom.2019.11.023.
R. Sundararaman, C. De Almeida Braga, E. Marchand, and J. Pettre, “Tracking Pedestrian Heads in Dense Crowd,” in 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, Jun. 2021, pp. 3864–3874. doi: 10.1109/CVPR46437.2021.00386.
X. Chen, Y. Jia, X. Tong, and Z. Li, “Research on Pedestrian Detection and DeepSort Tracking in Front of Intelligent Vehicle Based on Deep Learning,” Sustainability, vol. 14, no. 15, p. 9281, Jul. 2022, doi: 10.3390/su14159281.
M.-H. Guo et al., “Attention mechanisms in computer vision: A survey,” Comput. Vis. Media (Beijing)., vol. 8, no. 3, pp. 331–368, Sep. 2022, doi: 10.1007/s41095-022-0271-y.
Y. Wang, Y. Li, and H. Zou, “Masked Face Recognition System Based on Attention Mechanism,” Information, vol. 14, no. 2, p. 87, Feb. 2023, doi: 10.3390/info14020087.
Y. Wei, H. Li, Y. He, L. Li, Q. Lyu, and Y. Yang, “Robust face mask detection in complex scenarios using YOLOv8 and context-aware convolutions,” Sci. Rep., vol. 15, no. 1, p. 21350, Jul. 2025, doi: 10.1038/s41598-025-04768-w.
W. Liu, J. Yao, F. Jiang, and M. Wang, “An Improved Multi-Object Tracking Algorithm Designed for Complex Environments,” Sensors, vol. 25, no. 17, p. 5325, Aug. 2025, doi: 10.3390/s25175325.
S. Perikamana Narayanan, M. Sabarimalai Manikandan, and L. R. Cenkeramaddi, “YOLOv9-Based Human Face Detection and Counting Under Human-Animal Faces, Complex Imaging Environments, and Image Qualities,” IEEE Access, vol. 13, pp. 129600–129637, 2025, doi: 10.1109/ACCESS.2025.3591247.
D. T. Susetianingtias, E. Patriya, and R. Arianty, “Combination of YOLOv3 Algorithm and Blob Detection Technique in Calculating Nile Tilapia Seeds,” ILKOM Jurnal Ilmiah, vol. 15, no. 2, pp. 317–325, Aug. 2023, doi: 10.33096/ilkom.v15i2.1634.317-325.
N. L. Chusna and A. Khumaidi, “Comparison of Convolutional Neural Network Models for Feasibility of Selling Orchids,” ILKOM Jurnal Ilmiah, vol. 16, no. 3, pp. 296–304, Dec. 2024, doi: 10.33096/ilkom.v16i3.2006.296-304.
L. Fabrianto, T. Wahyuli Prihandayani, and N. Madhona Faizah, “Attention-based convolutional neural networks for interpretable classification of maritime equipment,” Jurnal Mandiri IT, vol. 14, no. 1, pp. 157–168, 2025, doi: https://doi.org/10.35335/mandiri.v14i1.426.
H. A. Salman and A. Kalakech, “Image Enhancement using Convolution Neural Networks,” Babylonian Journal of Machine Learning, vol. 2024, pp. 30–47, Jan. 2024, doi: 10.58496/BJML/2024/003.
P. A. Cahyani, M. Mardiana, P. B. Wintoro, and M. A. Muhammad, “Sistem Perhitungan Kendaraan Menggunakan Algortima YOLOv5 dan DeepSORT,” Jurnal Teknik Informatika dan Sistem Informasi, vol. 10, no. 1, May 2024, doi: 10.28932/jutisi.v10i1.7519.
F. Kurniawan, I. N. G. A. Astawa, I. M. A. D. S. Atmaja, and A. P. Wibawa, “Facemask Detection using the YOLO-v5 Algorithm: Assessing Dataset Variation and R esolutions,” Register, vol. 9, no. 2, pp. 95–102, Jul. 2023, doi: 10.26594/register.v9i2.3249.
S. Xue, “Facial recognition for surveillance videos based on RetinaFace and MobileFaceNet,” SPIE-Intl Soc Optical Eng, Jan. 2025, p. 155. doi: 10.1117/12.3048760.
M. G. Somoal and A. R. Dzikrillah, “Komparasi MobileNETV2 dengan Kustomisasi Transfer Learning dan Hyperparameter untuk Identifikasi Tumor Otak,” Jurnal Teknologi Informasi dan Ilmu Komputer, vol. 12, no. 1, pp. 229–240, Feb. 2025, doi: 10.25126/jtiik.2025129582.
M. Alansari et al., “EfficientFaceV2S: A lightweight model and a benchmarking approach for drone-captured face recognition,” Expert Syst. Appl., vol. 273, p. 126786, May 2025, doi: 10.1016/j.eswa.2025.126786.
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
________________________________________________________
The International Journal of Artificial Intelligence Research
Organized by: Prodi Teknik Informatika Fakultas Teknologi Bisnis dan Sains
Published by: Universitas Dharma Wacana
Jl. Kenanga No. 03 Mulyojati 16C Metro Barat Kota Metro Lampung
Email: jurnal.ijair@gmail.com
This work is licensed under Creative Commons Attribution-ShareAlike 4.0 International License.