The localization technology using images has been mainly studied in the robotics field. However, high-performance cameras are installed in personal mobile devices (smart phone, Google Glass, etc.), images are receiving great attention as positioning infrastructure for indoor localization technology.The image based indoor localization technology estimates the user's position with the photographs taken by the user's mobile device using the database constructed by collecting the image information of the indoor environment. In this paper, we constructed the image positioning database in the complex indoor space like the subway station in Seoul of Korea and analyzed the image positioning result. We use 3D LIDAR scanner to collect image data and build a image positioning server based on MPEG7 CDVS. MPEG7 CDVS, a high-performance image search technology standard optimized for mobile devices, is suitable for image based positioning system because it is easy to build and search huge database. The CDVS-based image positioning DB consists of a number of reference images. The reference image uses the image created in the 3D model instead of the actually photographed image. After collecting the depth information and color information of the indoor space using the 3D LIDAR device, a color point cloud 3D model is constructed. In this 3D model, the color information and the coordinate information are projected in two dimensions to generate an image. The generated image is similar to the image taken by the person directly at that position. This makes it easier to create a large number of reference images taken at the correct location. By building the database using the created reference image and matching with the image of the user, more precise position information can be estimated. In this paper, we constructed an image - based positioning system for subway stations in Seoul, which is a complex indoor space. We gathered the indoor space information of the sports complex station of Seoul and constructed the database and confirmed the positioning results.

Analysis of Image based Indoor Positioning Results in Complex Indoor Environment / Jeon, J; Kim, J; Ji, M; Cho, Y; Lingua, A; Chiabrando, F; Di Pietra, V. - ELETTRONICO. - (2017), pp. 414-418. (Intervento presentato al convegno ION 2017 Pacific PNT Meeting tenutosi a Honolulu, Hawaii nel May 1 - 4, 2017) [10.33012/2017.15039].

Analysis of Image based Indoor Positioning Results in Complex Indoor Environment

Lingua, A;Chiabrando, F;Di Pietra, V
2017

Abstract

The localization technology using images has been mainly studied in the robotics field. However, high-performance cameras are installed in personal mobile devices (smart phone, Google Glass, etc.), images are receiving great attention as positioning infrastructure for indoor localization technology.The image based indoor localization technology estimates the user's position with the photographs taken by the user's mobile device using the database constructed by collecting the image information of the indoor environment. In this paper, we constructed the image positioning database in the complex indoor space like the subway station in Seoul of Korea and analyzed the image positioning result. We use 3D LIDAR scanner to collect image data and build a image positioning server based on MPEG7 CDVS. MPEG7 CDVS, a high-performance image search technology standard optimized for mobile devices, is suitable for image based positioning system because it is easy to build and search huge database. The CDVS-based image positioning DB consists of a number of reference images. The reference image uses the image created in the 3D model instead of the actually photographed image. After collecting the depth information and color information of the indoor space using the 3D LIDAR device, a color point cloud 3D model is constructed. In this 3D model, the color information and the coordinate information are projected in two dimensions to generate an image. The generated image is similar to the image taken by the person directly at that position. This makes it easier to create a large number of reference images taken at the correct location. By building the database using the created reference image and matching with the image of the user, more precise position information can be estimated. In this paper, we constructed an image - based positioning system for subway stations in Seoul, which is a complex indoor space. We gathered the indoor space information of the sports complex station of Seoul and constructed the database and confirmed the positioning results.
2017
0-936406-19-4
File in questo prodotto:
File Dimensione Formato  
PNT17-0036.pdf.pdf

non disponibili

Descrizione: Proceedings ION PNT
Tipologia: 2a Post-print versione editoriale / Version of Record
Licenza: Non Pubblico - Accesso privato/ristretto
Dimensione 823.65 kB
Formato Adobe PDF
823.65 kB Adobe PDF   Visualizza/Apri   Richiedi una copia
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/2854814