Research

We study intelligent information technologies and apply them to various application fields, such as education, tourism, and primary industries.

JST CREST: FishTech for Sustainable Fishery Model

“FishTech” is a new technology which is a collaboration with fishery science, ocean science and informatics. Our goal is to develop FishTech for sustainable fishery which balances economic efficiency with resource management. We develop a new pattern recognition and data assimilation technology which employ domain knowledge of ecology of fish and oceanography, and analyze environmental data acquired in a process of fishing activities. Our technology supports short-term and long-term fishing operation providing suitable fishing spots, oceanographic conditions and fishery management plans.

Details

3D Shape Reconstruction under Participating Media

3D reconstruction is a technique to reconstruct an object’s 3D shape from 2D images captured with cameras. This technique can be utilized for a variety of applications such as self-driving vehicles and AR/VR technology. We develop 3D reconstruction techniques in participating media. For example, the contrast of an image captured under murky water, fog, or smoke is corrupted by scattered light due to suspended particles. 3D reconstruction techniques in such environments enable various applications in the real world.

Details

Smart Tourism

Smart tourism that provides real-time and personal information support to tourists is emerging. In this research, we are studying technologies that can contribute to solving social problems such as over-tourism and disaster prevention by providing appropriate information to tourists by estimating and predicting the condition of them from sensor information such as GPS.

Details

DeSET: High-resolution Bathymetry Data Acquisition via Learning-based Image Super-resolution

This research aims to accelerate the creation of detailed bathymetric charts. In this research, treating bathymetric charts as digital images whose pixel values represent ocean depths, we proposed to use superresolution, a technique to enhance image resolution, to estimate fine bathymetric information from coarse observation data. This approach enables us to make full use of existing data and minimize new observation, thereby realizing efficient mapping of seafloor details.

Details

Pattern Recognition for Learning Analysis

.