Research on Target Detection and Recognition Algorithm Based on Improved YOLOv8

Object detection is a pivotal task in computer vision, with extensive applications across various fields. Despite the remarkable achievements of YOLOv8 in object detection, it still faces challenges in complex scenarios and multi-scale object detection. This study proposes an enhanced YOLOv8 algorithm, which integrates a multi-scale feature fusion module and an attention mechanism to address these limitations. The multi-scale feature fusion module leverages a Feature Pyramid Network (FPN) to integrate features from different scales, thereby improving detection accuracy for images of varying sizes. The attention mechanism based on Squeeze-and-Excitation (SE) blocks adaptively adjusts the channel weights of feature maps, enabling the model to focus on key regions of the detection targets and enhance detection performance in complex backgrounds. Experiments on the COCO128 dataset demonstrate that the improved YOLOv8 achieves an accuracy of 94.6% and a mean Average Precision (mAP) of 90.7%, outperforming the traditional YOLOv8 and other excellent improved models. The enhanced model exhibits faster convergence, better generalization ability, and maintains high computational efficiency for real-time detection.