SMW公式による最大エントロピー像再構成アルゴリズムと衛星「ようこう」搭載硬X線望遠鏡への応用

概要

論文の詳細を見る
A new algorithm of maximum entropy that is effective whenever the number of data is much smaller than that of unknowns is proposed using the Sherman-Morrison-Woodbury (SMW) formula of matrix in the scheme of Newton iteration for Lagrangian function minimization. The algorithm has been applied to the image reconstruction of solar flares with the hard X-ray telescope aboard the Satellite Yohkoh for solar observation. Result was good in computational efficiency and reliability with the change of the regularization parameter value in a wide range. For optimizing the parameter value, the generalized cross validation (GCV) that is extended to the nonlinear regularization proved useful, suggesting a weaker regularization in comparison with the criterion of X^2.
大同工業大学の論文