Fujitsu to develop rice paddy parcel maps

Fujitsu to develop rice paddy parcel maps


Tokyo, Japan: Fujitsu in collaboration with the Federation of Hokkaido Agricultural Mutual Aid Association (Hokkaido NOSAI) and the Kamikawa Chuo Agricultural Mutual Aid Association (Kamikawa Chuo NOSAI) will develop maps of rice paddy parcels using proprietary Fujitsu image analysis technology. The technology will utilise satellite images and aerial photographs to recognise the boundaries between rice paddies and the embankments that separate them. From August 9 to October 31, Fujitsu will be conducting a field test to evaluate the technology”s usefulness in confirming the crop area of new members when they join NOSAI, as well as other applications.

As part of the field testing, Fujitsu will employ its proprietary high-precision image analysis technology, which brings together filtering processing for highlighting the contours of an image with an edge extraction technique that can recognise essential shape features. This enables images captured by satellites or aircraft to be used for recognising boundaries between rice paddies and the embankments that separate them. The resulting data can then be used to automatically produce maps of rice paddy parcels. Going forward, by comparing the new maps to rice paddy parcel maps that have been produced by hand, Fujitsu will examine the usefulness of this technology in the operations of agricultural mutual aid associations.

Based on the results of the field test, Fujitsu will aim to develop a commercial service for generating rice paddy parcel maps using its image analysis technology by April 2013. The company will also investigate new applications for the technology, including support for other kinds of farmland parcel mapping beyond rice paddies, as well as applications in areas outside of agricultural mutual aid associations, such as for discovering abandoned arable farmland. Moreover, Fujitsu is working to extend the use of the technology to a wide range of areas and tasks, such as for discovering ways to improve land use efficiency in urban areas.

Source: Fujitsu