PROBLEM TO BE SOLVED: To provide an image reconstruction processing method and an image processing method that can reduce noises even when spatial correlativity of an error occurring in an image is strong.SOLUTION: Being different from a conventional technique regularizing a differential space of an image, regularization function evaluating a degree of mutual similarity of a weighted addition value along a linear space is used to perform reconstruction processing and prepare a digital image of a plurality of dimensions after the reconstruction processing is performed. That is, processing (regularization process) for regularizing a projection space is performed by incorporating the regularization process to the reconstruction processing. In this case, a line integrated value corresponding to the weighted addition value is obtained along a streak artifact SA occurring along the linear space. The regularization function described in the above is used to perform the reconstruction processing and prepare the digital image of the plurality of dimensions after the reconstruction processing is performed, and thus, noise can be reduced even when spatial correlativity of an error occurring in the image is strong.SELECTED DRAWING: Figure 7COPYRIGHT: (C)2016,JPO&INPIT【課題】画像に生じる誤差の空間的相関性が強い場合においてもノイズを低減させることができる画像再構成処理方法および画像処理方法を提供することを目的とする。【解決手段】従来のような画像の微分空間を正則化する手法と相違して、直線状空間に沿った、重み付き加算値の相互類似度を評価した正則化関数を用いて再構成処理を行って再構成処理後の複数次元のデジタル画像を作成する。つまり、投影空間を正則化する処理(正則化処理)を再構成処理に組み込んで行う。その際に、直線状空間に沿って生じるストリークアーティファクトSAに沿って重み付き加算値に相当する線積分値を求める。上述した正則化関数を用いて再構成処理を行って再構成処理後の複数次元のデジタル画像を作成するので、画像に生じる誤差の空間的相関性が強い場合においてもノイズを低減させることができる。【選択図】図7
