A robust instance segmentation framework for underground sewer defect detection. (28th February 2022)

• Record Type: Journal Article
• Authors: Li, Yanfen; Wang, Hanxiang; Dang, L.Minh; Jalil Piran, Md; Moon, Hyeonjoon
• Is Part Of: Measurement. Volume 190(2022)
• Publication Date: 2022-02-28
• Subjects: Deep learning -- Defect inspection -- Underground sewer -- Instance segmentation; Weights and measures -- Periodicals; Measurement -- Periodicals; Measurement; Weights and measures; Periodicals; 530.8
• DOI: 10.1016/j.measurement.2022.110727
• Languages: English
• ISSNs: 0263-2241

Abstract

Highlights: A dehazing algorithm is integrated into the proposed defect inspection framework in order to increase the detection rate. An efficient backbone structure (Res2Net-Mish-BN-101) is proposed to improve the learning ability of SOLOv2. An enhanced BiFPN (EBiFPN) is designed for better feature fusion. A deep learning-based segmentation model (Pipe-SOLO) and a manually annotated dataset are firstly proposed to segment sewer defects at the instance level. Abstract: The inspection of underground sewer defects plays a considerable role in estimating the structural integrity and avoiding various unforeseen functional failures. However, the conventional sewer defect inspection approaches suffer from the blurry and vaporous environment inside the sewer pipes, which significantly lowers the performance. Besides, it is challenging to achieve efficient and accurate condition assessment by the common manual inspection. Therefore, this manuscript introduces an automatic instance segmentation-based defect analysis framework. The main contributions include 1) a novel defect segmentation model called Pipe-SOLO is firstly presented to segment six common types of defects at the instance level by proposing an efficient backbone structure (Res2Net-Mish-BN-101) and designing an enhanced BiFPN (EBiFPN), 2) a GAN-based dehazing model is applied to effectively solve the image blurring problem, and 3) a publicly available sewer defect segmentation dataset. The experimental results show the proposed Pipe-SOLO achieved an improvement of 7.3% compared with the state-of-the-art method in terms of the mean Average Precision (mAP). Therefore, the proposed defect segmentation method is promising to be integrated with real-life applications that require defect localization and estimation.