Mobile Inundation Hazard map System Based on Underground Positioning with Photo Image Collection

2017 Fiscal Year Final Research Report

• Project/Area Number: 15H02983
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Single-year Grants
• Section: 一般
• Research Field: Social systems engineering/Safety system
• Research Institution: Kansai University
• Principal Investigator: Yasumuro Yoshihiro 関西大学, 環境都市工学部, 教授 (50335478)
• Co-Investigator(Kenkyū-buntansha): 檀 寛成 関西大学, 環境都市工学部, 准教授 (30434822) ; 尾崎 平 関西大学, 環境都市工学部, 准教授 (40351499) ; 窪田 諭 関西大学, 環境都市工学部, 准教授 (60527430)
• Project Period (FY): 2015-04-01 – 2018-03-31
• Keywords: 安心の社会技術 / 情報伝達 / ハザードマップ / 拡張現実感 / ロケーションシステム / 避難誘導

Outline of Final Research Achievements

Recently, inundations in underground structures have been increasing due to frequent localized heavy rainfalls in Japanese urban areas. To provide the information on the predicted inundation risks, location-based services are preferable for individual users. However, GPS does not function in underground spaces. This study proposes an augmented reality (AR) visualization of underground flood situations based on a visual location technique. The proposed method acquires the 3D coordinates of the natural features from prerecorded images. The user’s viewpoint can be estimated by corresponding 3D coordinates and natural features of the user photographed image, solving a Perspective-n-Point (PnP) problem. Then the water surfaces expressed by CG is superimposed on the photograph from the viewpoint of the user. We experimented with an actual railway station located in the underground. The reproducibility of the user’s viewpoints was verified, and view-dependent AR representation was achieved.

Free Research Field

画像工学