Automatic bitonal image optimization

Abstract

A method for processing a digital image of a document to provide one or more bitonal digital copies of selected image quality that are optimized for specific purposes. The method is suited for use with such documents as bank checks and commercial forms that have printed or handwritten textual characters overlaid on potentially obscuring backgrounds. The method may provide a clean bitonal digital copy with reduced background optimized for use in optical character recognition or an archival copy with meaningful background retained that is optimized for archiving the original document. A set of bitonal copies of the digital image of a document is derived covering a range of contrast sensitivities. The copies are compared and a numerical measure is provided representing the amount of change in going from one copy to the next. The collection of such numerical measures defines a numerical sequence representing variations in the digital image as the contrast sensitivity is monotonically varied. The image-variation sequence is automatically analyzed to find the region or regions of rapid dropoff, generally signifying that certain unwanted or wanted features of the image are dropping out, and to find regions of more gradual changes. Optimizing values of the contrast sensitivities are selected to fall in select positions with respect to the peak or peaks and tails that are found in the sequence. The optimizing contrast sensitivities are used to generate one or more optimized bitonal digital copies such as a clean image or an archival image.