Object Recognition and Location Based on Mask R-CNN and Structured Light Camera

DOI

10.2991/cnci-19.2019.55How to use a DOI?

Keywords

Mask RCNN, Instance segmentation, Structured light camera, Computer vision.

Abstract

In the indoor construction site scene, binocular cameras rely on natural light in the environment for collecting images. However, with the influence of environmental factors such as changes in illumination angle and changes in illumination intensity, the effect of the binocular vision algorithm is drastically reduced in the case of darker illumination. In order to make the robot recognize the object better and judge the position of the object, this paper proposes a method based on Mask R-CNN model and structured light camera for object recognition and localization. The Mask R-CNN model is used to segment the RGB image, and then extracting the depth information of the target object, and calculating the distance between the target object and the robot through the principle of triangulation. The experiment shows that the proposed object recognition and positioning system can still complete the recognition and location of the trained object in the case of dark indoor light.

Copyright

© 2019, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Weipeng Mao
AU  - Hongyang Yu
PY  - 2019/05
DA  - 2019/05
TI  - Object Recognition and Location Based on Mask R-CNN and Structured Light Camera
BT  - Proceedings of the 2019 International Conference on Computer, Network, Communication and Information Systems (CNCI 2019)
PB  - Atlantis Press
SP  - 402
EP  - 408
SN  - 2352-538X
UR  - https://doi.org/10.2991/cnci-19.2019.55
DO  - 10.2991/cnci-19.2019.55
ID  - Mao2019/05
ER  -