Improving the Performance of the YOLOv8 Algorithm for Air Conditioner Detection in Urban Areas: Data Augmentation Techniques and Model Optimization

Year 2025, Volume: 5 Issue: 2, 49 - 66, 01.10.2025

Can Aydın

Abstract

The aim of this study is to evaluate the performance of the YOLOv8 algorithm in the task of climate detection and optimize it for urban environments. Currently, YOLOv8 is a state-of-the-art deep learning model designed by an architecture that gives superior performance results in object detection tasks simultaneously with speed and accuracy. YOLOv8s have been characterized to be a deep learning model that will do well in object detection in complex urban environments. Performance characteristics of a deep learning model will include the influence of background intensity and the effect of different camera illumination while detecting climate. It implements several data augmentation techniques, model architecture variations, and training strategies that enable the model to perform even better. Experimental results identified that data augmentation techniques make YOLOv8 perform much better, while model architecture variations present a trade-off between speed and accuracy. In particular, the complicated data augmentation methods, such as Mosaic and perspective switching, greatly improve the generalization capability of the model. It would mean that YOLOv8 performs quite well in recognizing air-conditioners and similar objects in an urban context; hence, until now, everything looks very encouraging to support the consideration of YOLOv8 in applications having to do with energy efficiency, heat island effect monitoring, and other environmental analyses.

Keywords

object detection , deep learning , data augmentation , yolo v8 , urban areas

References

• [1] Farhadi, A., & Redmon, J. (2018, June). Yolov3: An incremental improvement. In Computer vision and pattern recognition (Vol. 1804, pp. 1-6). Berlin/Heidelberg, Germany: Springer.Ref-2
• [2] Flores-Calero, M., Astudillo, C. A., Guevara, D., Maza, J., Lita, B. S., Defaz, B., ... & Armingol Moreno, J. M. (2024). Traffic sign detection and recognition using YOLO object detection algorithm: A systematic review. Mathematics, 12(2), 297.Ref-4
• [3] Vinh, T. Q., & Long, P. H. (2023, November). Pedestrian Detection Using YOLO with Improved Attention Module. In 2023 International Conference on Advanced Computing and Analytics (ACOMPA) (pp. 93-98). IEEE.
• [4] Cong, X., Li, S., Chen, F., Liu, C., & Meng, Y. (2023). A review of YOLO object detection algorithms based on deep learning. Frontiers in Computing and Intelligent Systems, 4(2), 17-20.
• [5] Koonce, B., & Koonce, B. (2021). SqueezeNet. Convolutional neural networks with swift for tensorflow: image recognition and dataset categorization, 73-85.
• [6] Geiger, A., Lenz, P., & Urtasun, R. (2012, June). Are we ready for autonomous driving? the kitti vision benchmark suite. In 2012 IEEE conference on computer vision and pattern recognition (pp. 3354-3361). IEEE.
• [7] Ruan, J., & Wang, Z. (2020). An improved algorithm for dense object detection based on YOLO. In The 8th international conference on computer engineering and networks (CENet2018) (pp. 592-599). Springer International Publishing.
• [8] Özcan, İ., Altun, Y., & Parlak, C. (2024). Improving YOLO Detection Performance of Autonomous Vehicles in Adverse Weather Conditions Using Metaheuristic Algorithms. Applied Sciences, 14(13), 5841.
• [9] Chen, L. C., Sheu, R. K., Peng, W. Y., Wu, J. H., & Tseng, C. H. (2020). Video-based parking occupancy detection for smart control system. Applied Sciences, 10(3), 1079.
• [10] Kong, L., Wang, J., & Zhao, P. (2022). YOLO-G: A lightweight network model for improving the performance of military targets detection. IEEE Access, 10, 55546-55564.
• [11] Cao, Y., Li, C., Peng, Y., & Ru, H. (2023). MCS-YOLO: A multiscale object detection method for autonomous driving road environment recognition. IEEE Access, 11, 22342-22354.
• [12] Fang, W., Wang, L., & Ren, P. (2019). Tinier-YOLO: A real-time object detection method for constrained environments. Ieee Access, 8, 1935-1944.
• [13] Ling, H., Zhao, T., Zhang, Y., & Lei, M. (2024). Engineering Vehicle Detection Based on Improved YOLOv6. Applied Sciences, 14(17), 8054.
• [14] Kim, M., Jeong, J., & Kim, S. (2021). ECAP-YOLO: Efficient channel attention pyramid YOLO for small object detection in aerial image. Remote Sensing, 13(23), 4851.
• [15] Doshi, Y. (2024). Comparison of YOLO Models for Object Detection from Parking Spot Images. Educational Administration: Theory and Practice, 30 (4), 10401-10411.
• [16] Haimer, Z., Mateur, K., Farhan, Y., & Madi, A. A. (2023, May). Yolo algorithms performance comparison for object detection in adverse weather conditions. In 2023 3rd International Conference on Innovative Research in Applied Science, Engineering and Technology (IRASET) (pp. 1-7). IEEE.
• [17] Ren, S., He, K., Girshick, R., & Sun, J. (2016). Faster R-CNN: Towards real-time object detection with region proposal networks. IEEE transactions on pattern analysis and machine intelligence, 39(6), 1137-1149.
• [18] Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C. Y., & Berg, A. C. (2016). Ssd: Single shot multibox detector. In Computer Vision–ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11–14, 2016, Proceedings, Part I 14 (pp. 21-37). Springer International Publishing.
• [19] Liu W, Anguelov D, Erhan D, Szegedy C, Reed S, Fu CY, Berg AC (2016). Ssd: Single shot multibox detector. In Computer Vision–ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11–14, 2016, Proceedings, Part I 14 (pp. 21-37). Springer International Publishing.