On PSF for lined motion blurred image

Point-spread function (PSF) plays a key role in image restoration. As a result, prior knowledge and poseteriori estimation of PSF is usually used to restore image. By analyzing and establishing the degraded mathematical model of lined blurred image and studying the calculating method of PSF, this paper gives a new way to identify the motion blur direction from the blurred image by direction derivation method and an approach to calculate the motion blur length. Experiments show that by the methods given above we get satisfactory results. And we also experiment and compare the inverse filtering results with the Wiener filtering results with/without noise respectively.