Visual Positioning in Indoor Environments Using RGB-D Images and Improved Vector of Local Aggregated Descriptors
Positioning information has become one of the most important information for processing and displaying on smart mobile devices. In this paper, we propose a visual positioning method using RGB-D image on smart mobile devices. Firstly, the pose of each image in the training set is calculated through f...
Main Authors: | , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-03-01
|
Series: | ISPRS International Journal of Geo-Information |
Subjects: | |
Online Access: | https://www.mdpi.com/2220-9964/10/4/195 |
_version_ | 1797540211949830144 |
---|---|
author | Longyu Zhang Hao Xia Qingjun Liu Chunyang Wei Dong Fu Yanyou Qiao |
author_facet | Longyu Zhang Hao Xia Qingjun Liu Chunyang Wei Dong Fu Yanyou Qiao |
author_sort | Longyu Zhang |
collection | DOAJ |
description | Positioning information has become one of the most important information for processing and displaying on smart mobile devices. In this paper, we propose a visual positioning method using RGB-D image on smart mobile devices. Firstly, the pose of each image in the training set is calculated through feature extraction and description, image registration, and pose map optimization. Then, in the image retrieval stage, the training set and the query set are clustered to generate the vector of local aggregated descriptors (VLAD) description vector. In order to overcome the problem that the description vector loses the image color information and improve the retrieval accuracy under different lighting conditions, the opponent color information and depth information are added to the description vector for retrieval. Finally, using the point cloud corresponding to the retrieval result image and its pose, the pose of the retrieved image is calculated by perspective-n-point (PnP) method. The results of indoor scene positioning under different illumination conditions show that the proposed method not only improves the positioning accuracy compared with the original VLAD and ORB-SLAM2, but also has high computational efficiency. |
first_indexed | 2024-03-10T12:56:53Z |
format | Article |
id | doaj.art-31faf78542824298964e4cc0e9194b01 |
institution | Directory Open Access Journal |
issn | 2220-9964 |
language | English |
last_indexed | 2024-03-10T12:56:53Z |
publishDate | 2021-03-01 |
publisher | MDPI AG |
record_format | Article |
series | ISPRS International Journal of Geo-Information |
spelling | doaj.art-31faf78542824298964e4cc0e9194b012023-11-21T11:47:48ZengMDPI AGISPRS International Journal of Geo-Information2220-99642021-03-0110419510.3390/ijgi10040195Visual Positioning in Indoor Environments Using RGB-D Images and Improved Vector of Local Aggregated DescriptorsLongyu Zhang0Hao Xia1Qingjun Liu2Chunyang Wei3Dong Fu4Yanyou Qiao5Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaQihoo 360 Technology Co. Ltd., Chaoyang District, Beijing 100016, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaPositioning information has become one of the most important information for processing and displaying on smart mobile devices. In this paper, we propose a visual positioning method using RGB-D image on smart mobile devices. Firstly, the pose of each image in the training set is calculated through feature extraction and description, image registration, and pose map optimization. Then, in the image retrieval stage, the training set and the query set are clustered to generate the vector of local aggregated descriptors (VLAD) description vector. In order to overcome the problem that the description vector loses the image color information and improve the retrieval accuracy under different lighting conditions, the opponent color information and depth information are added to the description vector for retrieval. Finally, using the point cloud corresponding to the retrieval result image and its pose, the pose of the retrieved image is calculated by perspective-n-point (PnP) method. The results of indoor scene positioning under different illumination conditions show that the proposed method not only improves the positioning accuracy compared with the original VLAD and ORB-SLAM2, but also has high computational efficiency.https://www.mdpi.com/2220-9964/10/4/195visual positioningRGB-D Images3D modelimage retrievalpose estimation |
spellingShingle | Longyu Zhang Hao Xia Qingjun Liu Chunyang Wei Dong Fu Yanyou Qiao Visual Positioning in Indoor Environments Using RGB-D Images and Improved Vector of Local Aggregated Descriptors ISPRS International Journal of Geo-Information visual positioning RGB-D Images 3D model image retrieval pose estimation |
title | Visual Positioning in Indoor Environments Using RGB-D Images and Improved Vector of Local Aggregated Descriptors |
title_full | Visual Positioning in Indoor Environments Using RGB-D Images and Improved Vector of Local Aggregated Descriptors |
title_fullStr | Visual Positioning in Indoor Environments Using RGB-D Images and Improved Vector of Local Aggregated Descriptors |
title_full_unstemmed | Visual Positioning in Indoor Environments Using RGB-D Images and Improved Vector of Local Aggregated Descriptors |
title_short | Visual Positioning in Indoor Environments Using RGB-D Images and Improved Vector of Local Aggregated Descriptors |
title_sort | visual positioning in indoor environments using rgb d images and improved vector of local aggregated descriptors |
topic | visual positioning RGB-D Images 3D model image retrieval pose estimation |
url | https://www.mdpi.com/2220-9964/10/4/195 |
work_keys_str_mv | AT longyuzhang visualpositioninginindoorenvironmentsusingrgbdimagesandimprovedvectoroflocalaggregateddescriptors AT haoxia visualpositioninginindoorenvironmentsusingrgbdimagesandimprovedvectoroflocalaggregateddescriptors AT qingjunliu visualpositioninginindoorenvironmentsusingrgbdimagesandimprovedvectoroflocalaggregateddescriptors AT chunyangwei visualpositioninginindoorenvironmentsusingrgbdimagesandimprovedvectoroflocalaggregateddescriptors AT dongfu visualpositioninginindoorenvironmentsusingrgbdimagesandimprovedvectoroflocalaggregateddescriptors AT yanyouqiao visualpositioninginindoorenvironmentsusingrgbdimagesandimprovedvectoroflocalaggregateddescriptors |