Improved JPEG Decompression of Document Images Based on Image Segmentation

  • Authors:

  • Tak-Shing Wong;Charles A. Bouman;Ilya Pollak

  • Affiliations:

  • School of Electrical and Computer Engineering, Purdue University;School of Electrical and Computer Engineering, Purdue University;School of Electrical and Computer Engineering, Purdue University

  • Venue:
  • SSP '07 Proceedings of the 2007 IEEE/SP 14th Workshop on Statistical Signal Processing
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

We propose a JPEG decompression algorithm for document images based on image segmentation. Our segmentation algorithm classifies each JPEG block of the image into one of three classes: text, background, or picture. We develop a different decoding strategy for each class and apply this strategy to every block in this class. Our experiments demonstrate that this approach can improve the quality of JPEG decompression.