OPTIMIZING YOLOV8 FOR AUTONOMOUS DRIVING: BATCH SIZE FOR BEST MEAN AVERAGE PRECISION (MAP)

Zaids Syarif Hidayat; Yudhistira Arie  Wijaya; Rudi Kurniawan

doi:10.52436/1.jutif.2024.5.4.1626

Zaids Syarif Hidayat Informatics Engineering, STMIK IKMI Cirebon, Indonesia
Yudhistira Arie Wijaya Informatics Engineering, STMIK IKMI Cirebon, Indonesia
Rudi Kurniawan Software Engineering, STMIK IKMI Cirebon, Indonesia

DOI: https://doi.org/10.52436/1.jutif.2024.5.4.1626

Keywords: Autonomous Driving, Batch Size, Object Detection, YOLOv8, mAP

Abstract

Artificial intelligence (AI), especially computer vision, has made rapid progress in recent years. One of the rapidly developing fields in computer vision is object detection. The ability to detect objects accurately and quickly is essential for the development of autonomous driving technology or vehicles that can operate automatically without human intervention. However, the development of autonomous driving technology is still facing various challenges, especially related to the accuracy and speed of object detection by the system. The purpose of this study is to analyze the performance based on the mean average precision (mAP) value of the results of adjusting the number of epochs, batch size, and image size on one of the emerging object detection methods, YOLOv8, in the context of autonomous driving. The analysis focuses on the batch size hyperparameter on the object detection performance of YOLOv8. The research was conducted with an experimental approach where the YOLOv8 hyperparameters were modified and their performance was evaluated using the driver simulation dataset from the CARLA simulator. Object detection performance was evaluated using the mean average precision (mAP) metric. The research results with the highest mAP value are found in scheme VIII with an mAP value of 98.2% and a training time of 59.45 minutes. For scheme III, it gets the fastest training time of 36.25 minutes. Based on the mAP results, modifications to the number of batch sizes and the use of high image sizes can affect the performance and performance of object detection for autonomous driving.

Downloads

Download data is not yet available.

References

K. R. Chowdhary, Fundamentals of artificial intelligence. Springer India, 2020. doi: 10.1007/978-81-322-3972-7.

N. Jannah, A. Wibowo, and S. Siadari, ‘Eksploitasi Fitur Untuk Peningkatan Kinerja Deteksi Objek Berbasis Pada Pesawat Tanpa Awak’, e-Proceeding of Engineering, vol. 8, p. 2943, 2022.

H. Nurhadiati, A. S. Wibowo, and A. Pratondo, ‘Analisis Performansi Deteksi Objek Pada Metode Complex YOLOv4 Untuk Autonomous Driving’, Bandung, Dec. 2022.

O. Natan and J. Miura, ‘End-to-End Autonomous Driving With Semantic Depth Cloud Mapping and Multi-Agent’, IEEE Transactions on Intelligent Vehicles, vol. 8, no. 1, pp. 557–571, 2023, doi: 10.1109/TIV.2022.3185303.

R. Wang et al., ‘A Real-Time Object Detector for Autonomous Vehicles Based on YOLOv4’, Comput Intell Neurosci, vol. 2021, 2021, doi: 10.1155/2021/9218137.

T. A. A. H. Kusuma, K. Usman, and S. Saidah, ‘PEOPLE COUNTING FOR PUBLIC TRANSPORTATIONS USING YOU ONLY LOOK ONCE METHOD’, Jurnal Teknik Informatika (Jutif), vol. 2, no. 1, pp. 57–66, Feb. 2021, doi: 10.20884/1.jutif.2021.2.2.77.

J. Choi, D. Chun, H. Kim, and H.-J. Lee, ‘Gaussian YOLOv3: An Accurate and Fast Object Detector Using Localization Uncertainty for Autonomous Driving’, 2019.

A. Bochkovskiy, C.-Y. Wang, and H.-Y. M. Liao, ‘YOLOv4: Optimal Speed and Accuracy of Object Detection’, Apr. 2020, [Online]. Available: http://arxiv.org/abs/2004.10934

J. Terven and D. Cordova-Esparza, ‘A Comprehensive Review of YOLO: From YOLOv1 and Beyond’, ArXiv, Apr. 2023, [Online]. Available: http://arxiv.org/abs/2304.00501

Y. Cai et al., ‘YOLOv4-5D: An Effective and Efficient Object Detector for Autonomous Driving’, IEEE Trans Instrum Meas, vol. 70, 2021, doi: 10.1109/TIM.2021.3065438.

A. Sarda, S. Dixit, and A. Bhan, ‘Object detection for autonomous driving using YOLO [You only Look Once] algorithm’, in Proceedings of the 3rd International Conference on Intelligent Communication Technologies and Virtual Mobile Networks, ICICV 2021, Institute of Electrical and Electronics Engineers Inc., Feb. 2021, pp. 1370–1374. doi: 10.1109/ICICV50876.2021.9388577.

H. Lou et al., ‘DC-YOLOv8: Small-Size Object Detection Algorithm Based on Camera Sensor’, Electronics (Switzerland), vol. 12, no. 10, May 2023, doi: 10.3390/electronics12102323.

B. Mahaur and K. K. Mishra, ‘Small-object detection based on YOLOv5 in autonomous driving systems’, Pattern Recognit Lett, vol. 168, pp. 115–122, Apr. 2023, doi: 10.1016/J.PATREC.2023.03.009.

L. Suroiyah, Y. Rahmawati, and R. Dijaya, ‘FACEMASK DETECTION USING YOLO V5’, Jurnal Teknik Informatika (JUTIF), vol. 4, no. 6, pp. 1277–1286, 2023, doi: 10.52436/1.jutif.2023.4.6.1043.

I. Purwita Sary, E. Ucok Armin, S. Andromeda, E. Engineering, and U. Singaperbangsa Karawang, ‘Performance Comparison of YOLOv5 and YOLOv8 Architectures in Human Detection Using Aerial Images’, Ultima Computing : Jurnal Sistem Komputer, vol. 15, no. 1, 2023.

A. Dosovitskiy, G. Ros, F. Codevilla, A. López, and V. Koltun, ‘CARLA: An Open Urban Driving Simulator’, 2017.

FYP, ‘FYP2022 Dataset’. Accessed: Jul. 08, 2023. [Online]. Available: https://universe.roboflow.com/fyp-5uyrm/fyp2022

Shah Deval, ‘Mean Average Precision (mAP) Explained: Everything You Need to Know’, https://www.v7labs.com/blog/mean-average-precision.

M. Hussain, ‘YOLO-v1 to YOLO-v8, the Rise of YOLO and Its Complementary Nature toward Digital Manufacturing and Industrial Defect Detection’, Machines, vol. 11, no. 7. Multidisciplinary Digital Publishing Institute (MDPI), Jul. 01, 2023. doi: 10.3390/machines11070677.

R.-C. Chen, C. Dewi, Y.-C. Zhuang, and J.-K. Chen, ‘Contrast Limited Adaptive Histogram Equalization for Recognizing Road Marking at Night Based on Yolo Models’, IEEE Access, vol. 11, pp. 92926–92942, 2023, doi: 10.1109/ACCESS.2023.3309410.

J. Fan, T. Huo, and X. Li, ‘A review of one-stage detection algorithms in autonomous driving’, in 2020 4th CAA International Conference on Vehicular Control and Intelligence, CVCI 2020, Institute of Electrical and Electronics Engineers Inc., Dec. 2020, pp. 210–214. doi: 10.1109/CVCI51460.2020.9338663.

M. A. Yahya, S. Abdul-Rahman, and S. Mutalib, ‘Object detection for autonomous vehicle with Lidar using deep learning’, in 2020 IEEE 10th International Conference on System Engineering and Technology, ICSET 2020 - Proceedings, Institute of Electrical and Electronics Engineers Inc., Nov. 2020, pp. 207–212. doi: 10.1109/ICSET51301.2020.9265358.