Improved human image density detection with comparison of YOLOv8 depth level architecture and drop-out implementation
Article Sidebar
Main Article Content
Abstract
Energy inefficiency due to Air Conditioners (AC) running in empty rooms contribute to unnecessary energy consumption and increased CO₂ emissions. This study explores how different depth levels of the YOLOv8 architecture and dropout regularization can enhance human density detection for smarter AC control systems. By evaluating model accuracy through Mean Average Precision (mAP50-95), we provide quantitative insights into how these modifications improve detection performance. Our dataset consists of 1363 images taken in an office environment at ITERA under varying lighting conditions and different human presence densities. The results show that the YOLOv8m model performs best, achieving an mAP50-95 score of 0.814 in training and 0.813 in validation, outperforming other YOLOv8 variants. Furthermore, applying dropout regularization improves model generalization, increasing mAP50-95 from 0.552 to 0.6 and effectively reducing overfitting. This study highlights the balance between architectural depth and dropout regularization in YOLOv8, demonstrating its effectiveness in energy-efficient smart buildings. The findings support the potential of deep learning-based human density detection in improving energy conservation strategies, making it a valuable solution for intelligent automation systems.
Downloads
Article Details
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
References
Y. Peng et al., “Energy Consumption Optimization for Heating, Ventilation and Air Conditioning Systems Based on Deep Reinforcement Learning,” IEEE Access, vol. 11, no. July, hal. 88265–88277, 2023, doi: 10.1109/ACCESS.2023.3305683.
V. L. Erickson dan A. E. Cerpa, “Occupancy based demand response HVAC control strategy,” BuildSys’10 - Proc. 2nd ACM Work. Embed. Sens. Syst. Energy-Efficiency Build., no. May, hal. 7–12, 2010, doi: 10.1145/1878431.1878434.
A. M. Vicedo-Cabrera et al., “The burden of heat-related mortality attributable to recent human-induced climate change,” Nat Clim Chang, vol. 11, no. 6, hal. 492–500, 2021.
I. P. Sary, S. Andromeda, dan E. U. Armin, “Performance Comparison of YOLOv5 and YOLOv8 Architectures in Human Detection using Aerial Images,” Ultim. Comput. J. Sist. Komput., vol. 15, no. 1, hal. 8–13, 2023, doi: 10.31937/sk.v15i1.3204.
S. Rajakumar dan R. B. Azad, “A novel YOLOv8 architecture for human activity recognition of occluded pedestrians,” Int. J. Electr. Comput. Eng., vol. 14, no. 5, hal. 5244–5252, 2024, doi: 10.11591/ijece.v14i5.pp5244-5252.
D. Huang, Y. Li, J. Qu, S. Zhang, dan Q. Wang, “YOLOv8-MGH: Dense Crowd Object Detection,” Adv. Comput. Graph. , vol. 15338, hal. 133–144, 2025, doi: https://doi.org/10.1007/978-3-031-81806-6_10.
J. Terven, D. M. Córdova-Esparza, dan J. A. Romero-González, “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.
A. Lemay et al., “Improving the repeatability of deep learning models with Monte Carlo dropout,” npj Digit. Med., vol. 5, no. 1, hal. 1–11, 2022, doi: 10.1038/s41746-022-00709-3.
Z. Zhou, H. Li, S. Zhou, L. Yan, dan H. Yang, “A deep learning-based algorithm for fast identification of multiple defects in tunnels,” Eng. Appl. Artif. Intell., vol. 145, 2025, doi: https://doi.org/10.1016/j.engappai.2025.110035.
Z. Qi, D. Pan, T. Niu, Z. Ying, dan P. Shi, “Bridge the gap between practical application scenarios and cartoon character detection: A benchmark dataset and deep learning model,” Displays, vol. 84, 2024, doi: https://doi.org/10.1016/j.displa.2024.102793.
J. Zhao et al., “A deep learning-based study on visual quality assessment of commercial renovation of Chinese traditional building facades,” Environ. Impact Assess. Rev., vol. 113, 2025, doi: https://doi.org/10.1016/j.eiar.2025.107862.
N. P. Damayanti, M. N. D. Ananda, dan F. W. Nugraha, “Lung cancer classification using convolutional neural network and DenseNet,” J. Soft Comput. Explor., vol. 4, no. 3, hal. 133–141, 2023, doi: 10.52465/joscex.v4i3.177.
D. Magdaleno, M. Montes, B. Estrada, dan A. Ochoa-Zezzatti, “Utilizing a YOLOv8 Segmentation-Based Model for Automated Defect Detection in Bread Images,” Innov. Appl. Artif. Neural Networks to Data Anal. Signal Process., hal. 499–532, 2024, doi: https://doi.org/10.1007/978-3-031-69769-2_20.
P. Wspanialy, J. Brooks, dan M. Moussa, “An Image Labeling Tool and Agricultural Dataset for Deep Learning,” no. September, 2020, doi: 10.48550/arXiv.2004.03351.
A. Alkalbani, M. Saqib, A. S. Alrawahi, A. Anwar, C. Adak, dan S. Anwar, “RDD4D: 4D Attention-Guided Road Damage Detection And Classification,” hal. 1–13, 2025, [Daring]. Tersedia pada: http://arxiv.org/abs/2501.02822.
S. Kale, P. D., Mahajan, T., Kanawade, A., Shete, S., & Jadhav, “Comparative Analysis of Image Annotation Tools: LabelImg, VGG Annotator, Label Studio, and Roboflow,” vol. 11, no. 5, hal. n398–n403, 2024.
K. Kim, K. Kim, dan S. Jeong, “Application of YOLO v5 and v8 for Recognition of Safety Risk Factors at Construction Sites,” Sustainability, vol. 15, no. 20, hal. 15179, 2023, doi: 10.3390/su152015179.
U. Patel, R. Vaghela, Y. Popat, H. Patel, J. Sarda, dan A. Bhoi, “Multi-Class Event Classification using YOLOv8,” 15th Int. Conf. Inf. Commun. Technol. Converg., hal. 2078–2081, 2024.
M. Khan, “Understanding dropout in deep neural networks,” IEEE Comput. Intell. Mag, vol. 18, hal. 45–53, 2023.
S. R. Vinta, B. Dhanalaxmi, A. T. M. Lavanya, dan Y. Raju, “Banana Disease and Deficiency Detection Using YOLOv8,” Mach. Vis. Augment. Intell., vol. 1211, 2024, doi: https://doi.org/10.1007/978-981-97-4359-9_42.
J. Xie, Z. Ma, J. Lei, G. Zhang, J.-H. Xue, dan Z.-H. Tan, “Advanced Dropout: A Model-free Methodology for Bayesian Dropout Optimization,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 44, no. 9, hal. 4605–4625, 2022, doi: 10.1109/TPAMI.2021.3083089.
P. Hidayatullah, N. Syakrani, M. R. Sholahuddin, T. Gelar, dan R. Tubagus, “YOLOv8 to YOLO11: A Comprehensive Architecture In-depth Comparative Review,” 2025, [Daring]. Tersedia pada: http://arxiv.org/abs/2501.13400.
M. R. Ningsih et al., “Sign Language Detection System Using YOLOv5 Algorithm to Promote Communication Equality People with Disabilities,” Sci. J. Informatics, vol. 11, no. 2, hal. 549–558, 2024, doi: 10.15294/sji.v11i2.6007.
X. Li, Q. Wang, X. Yang, K. Wang, dan H. Zhang, “Track Fastener Defect Detection Model Based on Improved YOLOv5s,” Sensors, vol. 23, no. 14, 2023, doi: 10.3390/s23146457.