| 摘要 |
PURPOSE:To perform correction with high accuracy and high reliability by finding the transfer function of MTF(Modulaton Transfer Function) deterioration by flare from the response of a read-in reference chart, and performing the correction to recover MTF by a deconvolution arithmetic operation. CONSTITUTION:A reference chart representing the density distribution of a block-and-white step function is read by a photodeterctor 4, and is stored in reference signal storage memory 10. Furthermore, a reference signal is differentiated by a differentiator 12, and is outputted to a Fourier transformer 15 as the impulse response f(x) of signal deterioration, and is Fourier-transformed, and is stored in buffer memory 16 as the transfer function of signal deterioration. Thence, an original read signal, after being converted to a digital signal, is stored in line memory 11, and is Fourier-transformed by a Fourier transformer 13, and is divided by data in the line buffer memory 16 by a divider 14, and is converted to a spatial area signal by an inverse Fourier transformer 17, and is outputted as image output receiving MTF correction. The deconvolution arithmetic operation can be performed in such way. Thereby, the correction can be surely performed. |
