Optimized Deep Encoder-Decoder Methods for Crack Segmentation

TL;DR

This paper introduces optimized deep encoder-decoder architectures with novel components and data augmentation strategies, achieving state-of-the-art crack segmentation results across multiple datasets.

Contribution

It proposes a new decoder design, explores different encoder strategies, and introduces test-time augmentation and statistical analysis for improved crack segmentation.

Findings

Achieved new state-of-the-art results on four datasets.

Test-time augmentation improves segmentation performance.

Statistical analysis enhances reproducibility of results.

Abstract

Surface crack segmentation poses a challenging computer vision task as background, shape, colour and size of cracks vary. In this work we propose optimized deep encoder-decoder methods consisting of a combination of techniques which yield an increase in crack segmentation performance. Specifically we propose a decoder-part for an encoder-decoder based deep learning architecture for semantic segmentation and study its components to achieve increased performance. We also examine the use of different encoder strategies and introduce a data augmentation policy to increase the amount of available training data. The performance evaluation of our method is carried out on four publicly available crack segmentation datasets. Additionally, we introduce two techniques into the field of surface crack segmentation, previously not used there: Generating results using test-time-augmentation and performing a statistical result analysis over multiple training runs. The former approach generally yields increased performance results, whereas the latter allows for more reproducible and better representability of a methods results. Using those aforementioned strategies with our proposed encoder-decoder architecture we are able to achieve new state of the art results in all datasets.