Improvement and Enhancement of YOLOv5 Small Target Recognition Based on Multi-module Optimization

TL;DR

This paper enhances YOLOv5s for small target detection by integrating multiple modules like GhostNet, RepGFPN, CA, Transformer attention, and NWD loss, significantly improving accuracy in complex scenarios.

Contribution

The study introduces a multi-module optimization framework that significantly improves YOLOv5s performance on small target detection tasks.

Findings

Improved model shows higher precision, recall, and mAP.

Enhanced robustness in complex backgrounds and tiny targets.

Provides a comprehensive optimization strategy for small target detection.

Abstract

In this paper, the limitations of YOLOv5s model on small target detection task are deeply studied and improved. The performance of the model is successfully enhanced by introducing GhostNet-based convolutional module, RepGFPN-based Neck module optimization, CA and Transformer's attention mechanism, and loss function improvement using NWD. The experimental results validate the positive impact of these improvement strategies on model precision, recall and mAP. In particular, the improved model shows significant superiority in dealing with complex backgrounds and tiny targets in real-world application tests. This study provides an effective optimization strategy for the YOLOv5s model on small target detection, and lays a solid foundation for future related research and applications.