09. APPLICATION DEEP LEARNING YOLOv8 MODEL FOR OBJECT DETECTION

Tóm tắt

Detection of objects is a crucial aspect within the realm of computer vision, encompassing the identification and precise localization of objects within images or videos. This task holds significant importance in various applications, including but not limited to self-driving cars, robotics, and video surveillance systems. Throughout the years, numerous techniques and algorithms have been devised to detect objects within images and determine their spatial positions. The optimal performance in executing these tasks is achieved through the utilization of convolutional neural networks. Among the prominent neural networks designed for this purpose, YOLO stands out. Introduced in 2015 by Joseph Redmon, Santosh Divvala, Ross Girshick, and Ali Farhadi in their renowned research paper "You Only Look Once: Unified, Real-Time Object Detection," YOLO has become widely adopted. Since its inception, YOLO has undergone several iterations, with newer versions extending their capabilities beyond object detection. The latest release in this series is YOLOv8. In this article, the authors introduce a Deep Learning approach for object detection employing the YOLOv8 model. The testing outcomes reveal that the model attains a peak accuracy of 95 % for large and well-defined objects, whereas objects that are obscured and small in size exhibit an accuracy of 27.5 %.