水平方向の運動による流れ画像の点拡がり関数同定について(Ⅰ)
スポンサーリンク
概要
- 論文の詳細を見る
This paper deals with an identification of point spread function (PSF) of noisy images degraded by object motion in horizontal direction. First, assuming a spatial correlation of exponential form and separable variables, the image signal is modeled by a stationary process at the output of a discrete-time linear dynamical system randomly excited. Next, the motion blur is represented by a PSF and a Kalman estimator for noisy images of known PSF is derived. Finally, by examining the relation between an assumed PSF and the theoretical improvement index of the estimator, it is revealed that the gain estimation is most important in PSF identification. This result is very useful for finding a PSF identification method.
- 津山工業高等専門学校の論文
- 1984-10-20