Volume IV-4/W3
ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., IV-4/W3, 63-70, 2017
https://doi.org/10.5194/isprs-annals-IV-4-W3-63-2017
© Author(s) 2017. This work is distributed under
the Creative Commons Attribution 4.0 License.
ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., IV-4/W3, 63-70, 2017
https://doi.org/10.5194/isprs-annals-IV-4-W3-63-2017
© Author(s) 2017. This work is distributed under
the Creative Commons Attribution 4.0 License.

  25 Sep 2017

25 Sep 2017

A GEOGRAPHIC DATA GATHERING SYSTEM FOR IMAGE GEOLOCALIZATION REFINING

B. Semaan1,2, M. Servières1, G. Moreau1, and B. Chebaro2 B. Semaan et al.
  • 1UMR 1563 AAU-CRENAU
  • 2Lebanese University

Keywords: Image Geolocalization Refining, Building Detection, Semantic Data Extraction, Geographic Data Collection, Multi-Source Data Analysis

Abstract. Image geolocalization has become an important research field during the last decade. This field is divided into two main sections. The first is image geolocalization that is used to find out which country, region or city the image belongs to. The second one is refining image localization for uses that require more accuracy such as augmented reality and three dimensional environment reconstruction using images. In this paper we present a processing chain that gathers geographic data from several sources in order to deliver a better geolocalization than the GPS one of an image and precise camera pose parameters. In order to do so, we use multiple types of data. Among this information some are visible in the image and are extracted using image processing, other types of data can be extracted from image file headers or online image sharing platforms related information. Extracted information elements will not be expressive enough if they remain disconnected. We show that grouping these information elements helps finding the best geolocalization of the image.