Automated system for restoration of low-resolution document and text images

Restoration of document and text images has become increasingly important i n many areas of electronic imaging. This paper presents an automated system to restore low-resolution document and text images. If makes use of resolu tion expansion to enhance low-resolution images for optical character recog nition accuracy as well as to improve the quality of degraded images. Sever al approaches have been proposed in the past for resolution expansion such as linear interpolation and cubic spline expansion. The proposed system imp lements a bimodal-smooth-average (BSA) scoring function as an optimal crite rion for image quality. The BSA approach is very different from existing me thods in the sense that it uses three measures: bimodal, smooth, and averag e to produce a scoring function as an optimal criterion. Its idea is to cre ate for a given image a strongly bimodal image with smooth regions in both the foreground and background, while allowing for sharp discontinuities at the edges. Then the desired resolution-expanded image is obtained by solvin g a nonlinear optimization problem subject to a constraint that the average of expanded resolution must be equal to the original unexpanded resolution . The system can be used to restore both binary and grayscale images as wel l as video frames. Its capability is demonstrated experimentally to be quan titatively and qualitatively superior to standard interpolation methods. (C ) 2001 SPIE and IS&T.