SVM-based automatic scanned image classification with quick decision capability
Cheng Youn Lu, Jerry Wagner, Brandi Pitta, David J. Larson, Jan Allebach · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 2014
Digital copiers are now widely used. One major issue for a digital copier is copy quality. In order to achieve as high quality as possible for every input document, multiple processing pipelines are included in a digital copier. Every processing pipeline is designed specifically for a certain class of document, which may be text, picture, or a mixture of both as is illustrated by the three examples shown in Fig. 1. In this paper, we describe an algorithm that can effectively classify an input image into its corresponding category. Publisher’s Note: The first printing of this volume was completed prior to the SPIE Digital Library publication and this paper has since been replaced with a corrected/revised version.