Comparative analysis of YOLOv8 and YOLOv9 algorithms for object detection tasks
Rubrics: SCIENTIFIC ARTICLES
Authors:
1.
Kuban State University
student
student
Type:
Article
Pages:
from 234
to 237
Status:
Published
Received:
01.10.2025
Accepted:
11.10.2025
Published:
11.10.2025
Subject area:
UDC 004.93
Language:
Russian
Keywords:
artificial intelligence, computer vision, object detection, object classification, object detection algorithms, YOLO
Abstract (English):
This article presents a comparative analysis of the YOLOv8 and YOLOv9 models designed for automatic object detection in images. Both models belong to the YOLO family, which is widely used in computer vision tasks, but differ in a number of characteristics that affect their practical efficiency. The article considers key differences in the speed and accuracy of recognition, as well as the features of using each model depending on the requirements of specific tasks. YOLOv8 stands out for its high performance and is suitable for real-time systems, such as video surveillance and autonomous devices. YOLOv9 is focused on improving recognition accuracy, which makes it preferable in areas where reliability is critical, such as medicine or industrial diagnostics. The presented analysis can be useful when choosing a suitable model for various computer vision applications.
This article presents a comparative analysis of the YOLOv8 and YOLOv9 models designed for automatic object detection in images. Both models belong to the YOLO family, which is widely used in computer vision tasks, but differ in a number of characteristics that affect their practical efficiency. The article considers key differences in the speed and accuracy of recognition, as well as the features of using each model depending on the requirements of specific tasks. YOLOv8 stands out for its high performance and is suitable for real-time systems, such as video surveillance and autonomous devices. YOLOv9 is focused on improving recognition accuracy, which makes it preferable in areas where reliability is critical, such as medicine or industrial diagnostics. The presented analysis can be useful when choosing a suitable model for various computer vision applications.
Keywords:
artificial intelligence, computer vision, object detection, object classification, object detection algorithms, YOLO
artificial intelligence, computer vision, object detection, object classification, object detection algorithms, YOLO
1. Computer Vision Algorithms : [sayt]. – 2024. – URL: https://www.geeksforgeeks.org/computer-vision/computer-vision-algorithms/ (data obrascheniya: 25.03.2025).
2. Platonov, A. V. Mashinnoe obuchenie : uchebnoe posobie dlya vuzov / A. V. Platonov. – 2-e izd. – Moskva : Izdatel'stvo Yurayt, 2025. – 89 s. –ISBN 978-5-534-20732-3.
3. Shapiro, L. Komp'yuternoe zrenie : uchebnik / L. Shapiro, D. Stokman. – 5-e izd. – Moskva : Laboratoriya znaniy, 2024. – 762 s. – ISBN 978-5-93208-725-1.
4. Ultralytics YOLO Docs : [sayt]. – 2025. – URL: https://docs.ultralytics.com/ (data obrascheniya: 27.03.2025).
