Deep Neural Network based Visual Inspection with 3D Metric Measurement of Concrete Defects using Wall-climbing Robot
Liang Yang, Bing Li, Guoyong Yang, Yong Chang, Zhaoming Liu, Biao Jiang, Jizhong Xiaol
Abstract
This paper presents a novel metric inspection robot system using a deep neural network to detect and measure surface flaws (i.e., crack and spalling) on concrete structures performed by a wall-climbing robot. The system consists of four modules: robotics data collection module to obtain RGB-D images and IMU measurement, visual-inertial SLAM module to generate pose coupled key-frames with depth information, InspectionNet module to classify each pixel into three classes (back-ground, crack and spalling), and 3D registration and map fusion module to register the flaw patch into registered 3D model overlaid and highlighted with detected flaws for spatial-contextual visualization. The system enables the metric model of each surface flaw patch with pixel-level accuracy and determines its location in 3D space that is significant for structural health assessment and monitoring. The InspectionNet achieves an average accuracy of 87.64% for crack and spalling inspection. We also demonstrate our InspectionNet is robust to view angle, scale and illumination variation. Finally, we design a metric voxel volume map to highlight the flaw in 3D model and provide location and metric information.
BibTeX
@inproceedings{iros2019_deepneuralnetwor,
title = {Deep Neural Network based Visual Inspection with 3D Metric Measurement of Concrete Defects using Wall-climbing Robot},
author = {Liang Yang and Bing Li and Guoyong Yang and Yong Chang and Zhaoming Liu and Biao Jiang and Jizhong Xiaol},
booktitle = {IROS 2019},
year = {2019}
}