MTGL40-5: A Multi-Temporal Dataset for Remote Sensing Image Geo-Localization
Image-based geo-localization focuses on predicting the geographic information of query images by matching them with annotated images in a database. To facilitate relevant studies, researchers collect numerous images to build the datasets, which explore many challenges faced in real-world geo-localiz...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-08-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/15/17/4229 |
| _version_ | 1797581928789966848 |
|---|---|
| author | Jingjing Ma Shiji Pei Yuqun Yang Xu Tang Xiangrong Zhang |
| author_facet | Jingjing Ma Shiji Pei Yuqun Yang Xu Tang Xiangrong Zhang |
| author_sort | Jingjing Ma |
| collection | DOAJ |
| description | Image-based geo-localization focuses on predicting the geographic information of query images by matching them with annotated images in a database. To facilitate relevant studies, researchers collect numerous images to build the datasets, which explore many challenges faced in real-world geo-localization applications, significantly improving their practicability. However, a crucial challenge that often arises is overlooked, named the cross-time challenge in this paper, i.e., if query and database images are taken from the same landmark but at different time periods, the significant difference in their image content caused by the time gap will notably increase the difficulty of image matching, consequently reducing geo-localization accuracy. The cross-time challenge has a greater negative influence on non-real-time geo-localization applications, particularly those involving a long time span between query and database images, such as satellite-view geo-localization. Furthermore, the rough geographic information (e.g., names) instead of precise coordinates provided by most existing datasets limits the geo-localization accuracy. Therefore, to solve these problems, we propose a dataset, MTGL40-5, which contains remote sensing (RS) satellite images captured from 40 large-scale geographic locations spanning five different years. These large-scale images are split to create query images and a database with landmark labels for geo-localization. By observing images from the same landmark but at different time periods, the cross-time challenge becomes more evident. Thus, MTGL40-5 supports researchers in tackling this challenge and further improving the practicability of geo-localization. Moreover, it provides additional geographic coordinate information, enabling the study of high-accuracy geo-localization. Based on the proposed MTGL40-5 dataset, many existing geo-localization methods, including state-of-the-art approaches, struggle to produce satisfactory results when facing the cross-time challenge. This highlights the importance of proposing MTGL40-5 to address the limitations of current methods in effectively solving the cross-time challenge. |
| first_indexed | 2024-03-10T23:14:36Z |
| format | Article |
| id | doaj.art-ab344d112d024bca9f9f3f986e171be9 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T23:14:36Z |
| publishDate | 2023-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-ab344d112d024bca9f9f3f986e171be92023-11-19T08:46:14ZengMDPI AGRemote Sensing2072-42922023-08-011517422910.3390/rs15174229MTGL40-5: A Multi-Temporal Dataset for Remote Sensing Image Geo-LocalizationJingjing Ma0Shiji Pei1Yuqun Yang2Xu Tang3Xiangrong Zhang4School of Artificial Intelligence, Xidian University, Xi’an 710071, ChinaSchool of Artificial Intelligence, Xidian University, Xi’an 710071, ChinaSchool of Artificial Intelligence, Xidian University, Xi’an 710071, ChinaSchool of Artificial Intelligence, Xidian University, Xi’an 710071, ChinaSchool of Artificial Intelligence, Xidian University, Xi’an 710071, ChinaImage-based geo-localization focuses on predicting the geographic information of query images by matching them with annotated images in a database. To facilitate relevant studies, researchers collect numerous images to build the datasets, which explore many challenges faced in real-world geo-localization applications, significantly improving their practicability. However, a crucial challenge that often arises is overlooked, named the cross-time challenge in this paper, i.e., if query and database images are taken from the same landmark but at different time periods, the significant difference in their image content caused by the time gap will notably increase the difficulty of image matching, consequently reducing geo-localization accuracy. The cross-time challenge has a greater negative influence on non-real-time geo-localization applications, particularly those involving a long time span between query and database images, such as satellite-view geo-localization. Furthermore, the rough geographic information (e.g., names) instead of precise coordinates provided by most existing datasets limits the geo-localization accuracy. Therefore, to solve these problems, we propose a dataset, MTGL40-5, which contains remote sensing (RS) satellite images captured from 40 large-scale geographic locations spanning five different years. These large-scale images are split to create query images and a database with landmark labels for geo-localization. By observing images from the same landmark but at different time periods, the cross-time challenge becomes more evident. Thus, MTGL40-5 supports researchers in tackling this challenge and further improving the practicability of geo-localization. Moreover, it provides additional geographic coordinate information, enabling the study of high-accuracy geo-localization. Based on the proposed MTGL40-5 dataset, many existing geo-localization methods, including state-of-the-art approaches, struggle to produce satisfactory results when facing the cross-time challenge. This highlights the importance of proposing MTGL40-5 to address the limitations of current methods in effectively solving the cross-time challenge.https://www.mdpi.com/2072-4292/15/17/4229geo-localizationremote sensing satellite imagesgeographic coordinate information |
| spellingShingle | Jingjing Ma Shiji Pei Yuqun Yang Xu Tang Xiangrong Zhang MTGL40-5: A Multi-Temporal Dataset for Remote Sensing Image Geo-Localization Remote Sensing geo-localization remote sensing satellite images geographic coordinate information |
| title | MTGL40-5: A Multi-Temporal Dataset for Remote Sensing Image Geo-Localization |
| title_full | MTGL40-5: A Multi-Temporal Dataset for Remote Sensing Image Geo-Localization |
| title_fullStr | MTGL40-5: A Multi-Temporal Dataset for Remote Sensing Image Geo-Localization |
| title_full_unstemmed | MTGL40-5: A Multi-Temporal Dataset for Remote Sensing Image Geo-Localization |
| title_short | MTGL40-5: A Multi-Temporal Dataset for Remote Sensing Image Geo-Localization |
| title_sort | mtgl40 5 a multi temporal dataset for remote sensing image geo localization |
| topic | geo-localization remote sensing satellite images geographic coordinate information |
| url | https://www.mdpi.com/2072-4292/15/17/4229 |
| work_keys_str_mv | AT jingjingma mtgl405amultitemporaldatasetforremotesensingimagegeolocalization AT shijipei mtgl405amultitemporaldatasetforremotesensingimagegeolocalization AT yuqunyang mtgl405amultitemporaldatasetforremotesensingimagegeolocalization AT xutang mtgl405amultitemporaldatasetforremotesensingimagegeolocalization AT xiangrongzhang mtgl405amultitemporaldatasetforremotesensingimagegeolocalization |