Improving Robot Mobility by Combining Downward-Looking and Frontal Cameras
This paper presents a novel attempt to combine a downward-looking camera and a forward-looking camera for terrain classification in the field of off-road mobile robots. The first camera is employed to identify the terrain beneath the robot. This information is then used to improve the classification...
Main Authors: | , , |
---|---|
Format: | info:eu-repo/semantics/article |
Language: | English |
Published: |
MDPI
2020
|
Subjects: | |
Online Access: | http://hdl.handle.net/10835/7403 |
_version_ | 1789406325882486784 |
---|---|
author | González Sánchez, Ramón Rituerto, Alejandro Guerrero, José J. |
author_facet | González Sánchez, Ramón Rituerto, Alejandro Guerrero, José J. |
author_sort | González Sánchez, Ramón |
collection | DSpace |
description | This paper presents a novel attempt to combine a downward-looking camera and a forward-looking camera for terrain classification in the field of off-road mobile robots. The first camera is employed to identify the terrain beneath the robot. This information is then used to improve the classification of the forthcoming terrain acquired from the frontal camera. This research also shows the usefulness of the Gist descriptor for terrain classification purposes. Physical experiments conducted in different terrains (quasi-planar terrains) and different lighting conditions, confirm the satisfactory performance of this approach in comparison with a simple color-based classifier based only on frontal images. Our proposal substantially reduces the misclassification rate of the color-based classifier (∼10% versus ∼20%). |
format | info:eu-repo/semantics/article |
id | oai:repositorio.ual.es:10835-7403 |
institution | Universidad de Cuenca |
language | English |
publishDate | 2020 |
publisher | MDPI |
record_format | dspace |
spelling | oai:repositorio.ual.es:10835-74032023-04-12T19:25:05Z Improving Robot Mobility by Combining Downward-Looking and Frontal Cameras González Sánchez, Ramón Rituerto, Alejandro Guerrero, José J. mobile robotics computer vision terrain classification color and gist descriptors This paper presents a novel attempt to combine a downward-looking camera and a forward-looking camera for terrain classification in the field of off-road mobile robots. The first camera is employed to identify the terrain beneath the robot. This information is then used to improve the classification of the forthcoming terrain acquired from the frontal camera. This research also shows the usefulness of the Gist descriptor for terrain classification purposes. Physical experiments conducted in different terrains (quasi-planar terrains) and different lighting conditions, confirm the satisfactory performance of this approach in comparison with a simple color-based classifier based only on frontal images. Our proposal substantially reduces the misclassification rate of the color-based classifier (∼10% versus ∼20%). 2020-01-16T12:11:39Z 2020-01-16T12:11:39Z 2016-11-28 info:eu-repo/semantics/article 2218-6581 http://hdl.handle.net/10835/7403 en https://www.mdpi.com/2218-6581/5/4/25 Attribution-NonCommercial-NoDerivatives 4.0 Internacional http://creativecommons.org/licenses/by-nc-nd/4.0/ info:eu-repo/semantics/openAccess MDPI |
spellingShingle | mobile robotics computer vision terrain classification color and gist descriptors González Sánchez, Ramón Rituerto, Alejandro Guerrero, José J. Improving Robot Mobility by Combining Downward-Looking and Frontal Cameras |
title | Improving Robot Mobility by Combining Downward-Looking and Frontal Cameras |
title_full | Improving Robot Mobility by Combining Downward-Looking and Frontal Cameras |
title_fullStr | Improving Robot Mobility by Combining Downward-Looking and Frontal Cameras |
title_full_unstemmed | Improving Robot Mobility by Combining Downward-Looking and Frontal Cameras |
title_short | Improving Robot Mobility by Combining Downward-Looking and Frontal Cameras |
title_sort | improving robot mobility by combining downward-looking and frontal cameras |
topic | mobile robotics computer vision terrain classification color and gist descriptors |
url | http://hdl.handle.net/10835/7403 |
work_keys_str_mv | AT gonzalezsanchezramon improvingrobotmobilitybycombiningdownwardlookingandfrontalcameras AT rituertoalejandro improvingrobotmobilitybycombiningdownwardlookingandfrontalcameras AT guerrerojosej improvingrobotmobilitybycombiningdownwardlookingandfrontalcameras |