A Document Scanner Equalization Technique

Abstract

Scanned documents/images blur effect is mostly associated with the optics of the scanning system. To reverse this effect, sharpening filters are usually designed to restore blurred images based on ad-hoc pre-determined levels of sharpness. The filter design process is mostly manual, time-consuming, and could provide inconsistent output results among different scanning devices. In this paper, we propose a technique to automatically design an equalizing FIR image filter based on measured blur characteristics of the scanning device. The Slanted-edge technique is used to measure the spatial frequency response (SFR) of the scanning system. The SFR of the equalizing filter (for both horizontal and vertical directions) is composed of two segments: The first segment is the inverse of the measured SFR up to a desired spatial frequency (usually 1/4 of the sampling frequency), while the other segment provides an arbitrary gradual attenuation of high frequency noise. The point spread function (PSF) of the equalizing filter is then derived using a 2D frequency sampling filter design method. Ultimately, the cascaded frequency response of the scanning system and the equalizing filter should resemble a perfect SFR (i.e. of unity gain up to a desired spatial frequency and attenuation thereafter) when measured by the slanted edge technique. Our experiments show that this automated process can be applied to different document scanning devices to equalize their spatial frequency response resulting in consistent output sharpness levels.