3D Sensors for Sewer Inspection: A Quantitative Review and Analysis

Abstract

Automating inspection of critical infrastructure such as sewer systems will help utilities optimize maintenance and replacement schedules. The current inspection process consists of manual reviews of video as an operator controls a sewer inspection vehicle remotely. The process is slow, labor-intensive, and expensive and presents a huge potential for automation. With this work, we address a central component of the next generation of robotic inspection of sewers, namely the choice of 3D sensing technology. We investigate three prominent techniques for 3D vision: passive stereo, active stereo, and time-of-flight (ToF). The Realsense D435 camera is chosen as the representative of the first two techniques wheres the PMD CamBoard pico flexx represents ToF. The 3D reconstruction performance of the sensors is assessed in both a laboratory setup and in an outdoor above-ground setup. The acquired point clouds from the sensors are compared with reference 3D models using the cloud-to-mesh metric. The reconstruction performance of the sensors is tested with respect to different illuminance levels and different levels of water in the pipes. The results of the tests show that the ToF-based point cloud from the pico flexx is superior to the output of the active and passive stereo cameras.

System for 3D Acquisition and 3D Reconstruction using Structured Light for Sewer Line Inspection

“…However, as all the aforementioned algorithms rely on tracked keypoints, the generated 3D models are sparse, and the success is susceptible to the illumination conditions and the structures visible in the scene. In (Bahnsen et al, 2021) the authors assess different alternative 3D sensors to facilitate the inspection of sewer pipes. They compare passive stereo, active stereo and time-of-flight sensors and identify the last one as most suitable, as it works reliably under all simulated lightning conditions and also in the presence of water.…”

A Study on Long-Close Distance Coordination Control Strategy for Litchi Picking

“…In fact, while consumer-grade depth cameras have made it easier to build vision systems for fruit-picking robots, there are still many hardware and depth-measurement technical limitations [29,30]. For example, point clouds obtained by depth cameras based on structured-light methods frequently include hole regions caused by missing depth information, which, along with lighting variations, have a large impact on image quality [31].…”

Multi-scale hybrid vision transformer and Sinkhorn tokenizer for sewer defect classification

“…However, with more than 1.2 million kilometers of public sewerage infrastructure in just the U.S. [1], this becomes an unimaginable task to perform manually on a regular basis, as each inspection has to be performed by a professional sewer inspector. Therefore, the task of automating the sewer inspection process has been researched for more than three decades, through the development and application of sensors and computer vision algorithms [2][3][4][5].…”