Enhancing Road Crack Detection Accuracy with BsS-YOLO: Optimizing Feature Fusion and Attention Mechanisms

TL;DR

This paper introduces BsS-YOLO, a novel model that enhances road crack detection accuracy by optimizing feature fusion and incorporating advanced attention mechanisms, demonstrating significant improvements over existing methods.

Contribution

The paper proposes BsS-YOLO, integrating improved feature fusion and attention modules to boost detection accuracy and robustness in complex environments.

Findings

Achieves a 2.8% increase in mAP for crack detection

Improves detection robustness across varied scenarios

Enhances feature representation with attention mechanisms

Abstract

Effective road crack detection is crucial for road safety, infrastructure preservation, and extending road lifespan, offering significant economic benefits. However, existing methods struggle with varied target scales, complex backgrounds, and low adaptability to different environments. This paper presents the BsS-YOLO model, which optimizes multi-scale feature fusion through an enhanced Path Aggregation Network (PAN) and Bidirectional Feature Pyramid Network (BiFPN). The incorporation of weighted feature fusion improves feature representation, boosting detection accuracy and robustness. Furthermore, a Simple and Effective Attention Mechanism (SimAM) within the backbone enhances precision via spatial and channel-wise attention. The detection layer integrates a Shuffle Attention mechanism, which rearranges and mixes features across channels, refining key representations and further improving accuracy. Experimental results show that BsS-YOLO achieves a 2.8% increase in mean average precision (mAP) for road crack detection, supporting its applicability in diverse scenarios, including urban road maintenance and highway inspections.