RII: Renovating the irregular illumination of digital image archives

• Distortions decrease the quality of the digital pieces and lead to disputation.
• A novel system for handling the over-exposure and under-exposure problem is proposed.
• The simulation is conducted to show the performance.

Digitization is critical for preserving valuable culture archives such as texts and images. Considering the physical characters of digital cameras or scanners and the artificial negligence, some distortions such as under-exposure or over-exposure often appear in the digital form of archives. These distortions decrease the quality of the digital pieces and lead to disputation in some circumstances. Several methods have been used to deal with these illumination problems. These existing methods mainly focus on how to mitigate the under-exposure phenomenon in text-only or text-photo images. Over-exposure cases in which brightness comes from different orientations are not considered. Hence, we propose a novel system for renovating irregular illumination (RII) to handle the over-exposure problem as well as under-exposure distortion. Experimental results show that the processing outcomes of RII can guarantee accurate restoration of the transformed digital pieces. In particular, RII can be extended to improve the uneven light distribution problem for complicated and colorful images.