Image-Based Volume Estimation Using Stereo Vision

The current paper proposes an image-based volume estimation method and its implementation. Firstly, the camera positions are estimated based on the feature points found on the images using bundle adjustment. Using the camera positions, a dense point cloud is generated, onto which a Poisson surface a...

Full description

Saved in:
Bibliographic Details
Published in2020 IEEE 18th International Symposium on Intelligent Systems and Informatics (SISY) pp. 000055 - 000060
Main Authors Bandi, Nandor, Tunyogi, Rudolf-Balint, Szabo, Zoltan, Farkas, Eszter, Sulyok, Csaba
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.09.2020
Subjects
3D reconstruction
Androids
bundle adjustment
Cameras
Humanoid robots
Image reconstruction
point cloud processing
Servers
stereo vision
Surface reconstruction
volume estimation
Volume measurement
Online AccessGet full text

Cover

More Information
Summary:The current paper proposes an image-based volume estimation method and its implementation. Firstly, the camera positions are estimated based on the feature points found on the images using bundle adjustment. Using the camera positions, a dense point cloud is generated, onto which a Poisson surface and a convex hull is fit. Afterwards, an intersection of the surfaces is determined and refined. Finally, the volume of the final surface is calculated, and converted to metric scale using a reference object.The implementation of the software system consists of two main parts: an Android client and a server. The server is responsible for image processing whereas the Android client application provides the user interface for taking pictures, and rendering the three-dimensional reconstruction of the scanned object as a result.
ISSN:1949-0488
DOI:10.1109/SISY50555.2020.9217089