3-D Reconstruction and Measurement System Based on Multimobile Robot Machine Vision

Abstract

Visualization and measurement for civil structure have been a constant field of research. Typically, 3-D measurement systems are static instruments that offer low degrees of flexible motion and require a significant amount of time for successful 3-D measurement. In this article, a machine vision system is developed to facilitate the 3-D measurements and reconstruction process. The proposed method depends on multimobile robots to achieve a portable 3-D measurement and reconstruction system. It consists of a 3-D reconstruction and measurement pipeline, which deploys multiple vision sensors mounted on teleoperated mobile to acquire the 3-D scan and 2-D image data. It can operate in both indoor and outdoor environments as for each structure a sequence of images and LiDAR scans are processed. The outcomes of the proposed method are a 3-D visualization for the structure along with its dimensional and volumetric measurements. The performance of the proposed system is evaluated using indoor and outdoor experiments to validate its application for measuring real structures. The results obtained showed the robustness and mobility of the proposed method.