Crack Detection in Infrastructure Using Transfer Learning, Spatial Attention, and Genetic Algorithm Optimization

TL;DR

This paper presents a novel deep learning approach combining transfer learning, spatial attention, and genetic algorithm optimization for accurate infrastructure crack detection, especially effective with limited data.

Contribution

It introduces an Attention-ResNet50 model optimized with a genetic algorithm, improving crack detection accuracy over traditional methods.

Findings

Achieved precision of 0.9967 and F1 score of 0.9983

Outperformed conventional crack detection methods

Proved effective with limited annotated data

Abstract

Crack detection plays a pivotal role in the maintenance and safety of infrastructure, including roads, bridges, and buildings, as timely identification of structural damage can prevent accidents and reduce costly repairs. Traditionally, manual inspection has been the norm, but it is labor-intensive, subjective, and hazardous. This paper introduces an advanced approach for crack detection in infrastructure using deep learning, leveraging transfer learning, spatial attention mechanisms, and genetic algorithm(GA) optimization. To address the challenge of the inaccessability of large amount of data, we employ ResNet50 as a pre-trained model, utilizing its strong feature extraction capabilities while reducing the need for extensive training datasets. We enhance the model with a spatial attention layer as well as a customized neural network which architecture was fine-tuned using GA. A comprehensive case study demonstrates the effectiveness of the proposed Attention-ResNet50-GA model, achieving a precision of 0.9967 and an F1 score of 0.9983, outperforming conventional methods. The results highlight the model's ability to accurately detect cracks in various conditions, making it highly suitable for real-world applications where large annotated datasets are scarce.