Visual Odometry in Challenging Environments: An Urban Underground Railway Scenario Case
Localization is one of the most critical tasks for an autonomous vehicle, as position information is required to understand its surroundings and move accordingly. Visual Odometry (VO) has shown promising results in the last years. However, VO algorithms are usually evaluated in outdoor street scenar...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2022-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9810254/ |
| _version_ | 1818470919155220480 |
|---|---|
| author | Mikel Etxeberria-Garcia Maider Zamalloa Nestor Arana-Arexolaleiba Mikel Labayen |
| author_facet | Mikel Etxeberria-Garcia Maider Zamalloa Nestor Arana-Arexolaleiba Mikel Labayen |
| author_sort | Mikel Etxeberria-Garcia |
| collection | DOAJ |
| description | Localization is one of the most critical tasks for an autonomous vehicle, as position information is required to understand its surroundings and move accordingly. Visual Odometry (VO) has shown promising results in the last years. However, VO algorithms are usually evaluated in outdoor street scenarios and do not consider underground railway scenarios, with low lighting conditions in tunnels and significant lighting changes between tunnels and railway platforms. Besides, there is a lack of GPS, and it is not easy to access such infrastructures. This research proposes a method to create a ground truth of images and poses in underground railway scenarios. Second, the EnlightenGAN algorithm is proposed to face challenging lighting conditions, which can be coupled with any state-of-the-art VO techniques. Finally, the obtained ground truth and the EnlightenGAN have been tested in a real scenario. Two different VO approaches have been used: ORB-SLAM2 and DF-VO. The results show that the EnlightenGAN enhancement improves the performance of both approaches. |
| first_indexed | 2024-04-13T21:43:59Z |
| format | Article |
| id | doaj.art-fc80fcbff64e41718c374b92929dc176 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-13T21:43:59Z |
| publishDate | 2022-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-fc80fcbff64e41718c374b92929dc1762022-12-22T02:28:40ZengIEEEIEEE Access2169-35362022-01-0110692006921510.1109/ACCESS.2022.31872099810254Visual Odometry in Challenging Environments: An Urban Underground Railway Scenario CaseMikel Etxeberria-Garcia0https://orcid.org/0000-0001-9420-4490Maider Zamalloa1https://orcid.org/0000-0003-3872-9908Nestor Arana-Arexolaleiba2https://orcid.org/0000-0002-3305-8108Mikel Labayen3https://orcid.org/0000-0001-8136-5324Ikerlan Technology Research Centre, Basque Research and Technology Alliance (BRTA), Arrasate/Mondragón, SpainIkerlan Technology Research Centre, Basque Research and Technology Alliance (BRTA), Arrasate/Mondragón, SpainMGEP, Mondragon Unibertsitatea, Loramendi Kalea, Arrasate/Mondragón, Gipuzkoa, SpainCAF Signaling, Donostia, Gipuzkoa, SpainLocalization is one of the most critical tasks for an autonomous vehicle, as position information is required to understand its surroundings and move accordingly. Visual Odometry (VO) has shown promising results in the last years. However, VO algorithms are usually evaluated in outdoor street scenarios and do not consider underground railway scenarios, with low lighting conditions in tunnels and significant lighting changes between tunnels and railway platforms. Besides, there is a lack of GPS, and it is not easy to access such infrastructures. This research proposes a method to create a ground truth of images and poses in underground railway scenarios. Second, the EnlightenGAN algorithm is proposed to face challenging lighting conditions, which can be coupled with any state-of-the-art VO techniques. Finally, the obtained ground truth and the EnlightenGAN have been tested in a real scenario. Two different VO approaches have been used: ORB-SLAM2 and DF-VO. The results show that the EnlightenGAN enhancement improves the performance of both approaches.https://ieeexplore.ieee.org/document/9810254/Visual Odometryautonomous vehiclescomputer visiondata enhancementsimultaneous localization and mappingimage processing |
| spellingShingle | Mikel Etxeberria-Garcia Maider Zamalloa Nestor Arana-Arexolaleiba Mikel Labayen Visual Odometry in Challenging Environments: An Urban Underground Railway Scenario Case IEEE Access Visual Odometry autonomous vehicles computer vision data enhancement simultaneous localization and mapping image processing |
| title | Visual Odometry in Challenging Environments: An Urban Underground Railway Scenario Case |
| title_full | Visual Odometry in Challenging Environments: An Urban Underground Railway Scenario Case |
| title_fullStr | Visual Odometry in Challenging Environments: An Urban Underground Railway Scenario Case |
| title_full_unstemmed | Visual Odometry in Challenging Environments: An Urban Underground Railway Scenario Case |
| title_short | Visual Odometry in Challenging Environments: An Urban Underground Railway Scenario Case |
| title_sort | visual odometry in challenging environments an urban underground railway scenario case |
| topic | Visual Odometry autonomous vehicles computer vision data enhancement simultaneous localization and mapping image processing |
| url | https://ieeexplore.ieee.org/document/9810254/ |
| work_keys_str_mv | AT mikeletxeberriagarcia visualodometryinchallengingenvironmentsanurbanundergroundrailwayscenariocase AT maiderzamalloa visualodometryinchallengingenvironmentsanurbanundergroundrailwayscenariocase AT nestoraranaarexolaleiba visualodometryinchallengingenvironmentsanurbanundergroundrailwayscenariocase AT mikellabayen visualodometryinchallengingenvironmentsanurbanundergroundrailwayscenariocase |