About this document

Journal of Information Systems & Operations Management by Ali QasimPouri is a document available to read on EtoBox.

This article proposes a new document image compression scheme based on k-means clustering to decompose images into layers. The foreground and background layers are compressed using JPEG2000, while the mask is compressed with JBIG2. More than 30 resampling filter functions from three families (polynomial, exponential, windowed-sinc) were implemented and analyzed for their effects on compression. Based on analysis of image quality using PSNR and OCR text confidence, conclusions and recommendations are derived

Author

Ali QasimPouri

Language

EN