Crack Detection as a Weakly-Supervised Problem: Towards Achieving Less Annotation-Intensive Crack Detectors

TL;DR

This paper introduces a weakly-supervised crack detection framework that reduces annotation costs by combining low-quality annotations with pixel brightness cues, maintaining high accuracy across diverse environments.

Contribution

It presents a novel two-branched framework for crack detection that minimizes reliance on costly annotations, enabling more practical deployment in real-world scenarios.

Findings

Retains high detection accuracy with low-quality annotations

Reduces data annotation costs significantly

Effective across various target domains

Abstract

Automatic crack detection is a critical task that has the potential to drastically reduce labor-intensive building and road inspections currently being done manually. Recent studies in this field have significantly improved the detection accuracy. However, the methods often heavily rely on costly annotation processes. In addition, to handle a wide variety of target domains, new batches of annotations are usually required for each new environment. This makes the data annotation cost a significant bottleneck when deploying crack detection systems in real life. To resolve this issue, we formulate the crack detection problem as a weakly-supervised problem and propose a two-branched framework. By combining predictions of a supervised model trained on low quality annotations with predictions based on pixel brightness, our framework is less affected by the annotation quality. Experimental results show that the proposed framework retains high detection accuracy even when provided with low quality annotations. Implementation of the proposed framework is publicly available at https://github.com/hitachi-rd-cv/weakly-sup-crackdet.