Feature Pyramid and Hierarchical Boosting Network for Pavement Crack Detection

TL;DR

This paper introduces a novel deep learning architecture called FPHBN that effectively detects pavement cracks by integrating semantic features and hierarchical sample reweighting, outperforming existing methods across multiple datasets.

Contribution

The paper presents a new network architecture combining feature pyramid and hierarchical boosting for improved pavement crack detection, addressing challenges of intensity inhomogeneity and complex backgrounds.

Findings

Outperforms state-of-the-art methods in accuracy.

Proven effective across five diverse crack datasets.

Demonstrates superior generality and robustness.

Abstract

Pavement crack detection is a critical task for insuring road safety. Manual crack detection is extremely time-consuming. Therefore, an automatic road crack detection method is required to boost this progress. However, it remains a challenging task due to the intensity inhomogeneity of cracks and complexity of the background, e.g., the low contrast with surrounding pavements and possible shadows with similar intensity. Inspired by recent advances of deep learning in computer vision, we propose a novel network architecture, named Feature Pyramid and Hierarchical Boosting Network (FPHBN), for pavement crack detection. The proposed network integrates semantic information to low-level features for crack detection in a feature pyramid way. And, it balances the contribution of both easy and hard samples to loss by nested sample reweighting in a hierarchical way. To demonstrate the superiority and generality of the proposed method, we evaluate the proposed method on five crack datasets and compare it with state-of-the-art crack detection, edge detection, semantic segmentation methods. Extensive experiments show that the proposed method outperforms these state-of-the-art methods in terms of accuracy and generality.