Automatic Analysis of Sewer Pipes Based on Unrolled Monocular Fisheye Images

TL;DR

This paper presents a computer vision system that detects and classifies sewer pipe damages using low-quality fisheye images, employing image unwrapping, lighting modeling, and deep learning for semantic analysis.

Contribution

It introduces a novel pipeline combining fisheye image unwrapping, lighting modeling, and deep CNNs for damage detection in sewer pipes from challenging images.

Findings

Effective damage detection in low-quality fisheye images

Accurate pipe surface unwrapping and semantic labeling

Robust system for sewer pipe inspection

Abstract

The task of detecting and classifying damages in sewer pipes offers an important application area for computer vision algorithms. This paper describes a system, which is capable of accomplishing this task solely based on low quality and severely compressed fisheye images from a pipe inspection robot. Relying on robust image features, we estimate camera poses, model the image lighting, and exploit this information to generate high quality cylindrical unwraps of the pipes' surfaces.Based on the generated images, we apply semantic labeling based on deep convolutional neural networks to detect and classify defects as well as structural elements.