The use of remotely sensed data for estimating of rice yield(Poster Session,ICSANE 2011(International Conference on Space, Aeronautical and Navigational Electronics 2011))

概要

論文の詳細を見る
To assess the feasibility of developing an efficient rice cultivation support system which can effectively increase the rice production and also realize environmental conservation in Asia, the estimation of rice yield using remotely sensed data was carried out in Hokkaido, Japan. The multiple linear regression analysis and the projection pursuit regression analysis were adopted to estimate the rice yield from different resolution satellite data about three different rice varieties. The results showed that the predictive error of rice yield derived by the projection pursuit regression analysis was smaller than the value derived by the multiple linear regression analysis. The lowest predictive error was obtained using SPOT5 with 10m resolution data. The creation of yield estimation model for different rice varieties was an effective method to reduce the predictive error.
社団法人電子情報通信学会の論文
2011-10-10