Document Analysis by Processing JBIG-Encoded Images

Document Type



Techniques are presented to directly process JBIG-encoded document images. Two experimental processing pipelines are designed to evaluate the performance of the methods from the application perspective. They are document segmentation for obtaining the global layout and the form processing system for form type identification and the form dropout. The JBIG coding context is employed to perform horizontal smearing and connected-component detection concurrently in the course of decoding the base layer of the JBIG images. It is shown that, using a simple segmentation algorithm, the global layout is identified 50 times faster compared to the case of processing the full resolution images. In addition, an original solution is presented for form type identification by use of the Hough transform of the JBIG base layer images, thus expediting it by a factor of 16 in the designed form dropout system. Advantages of the compressed domain processing include fast procedures, reduced memory requirements, and the possibility of hardware implementation.


Connected components; Form dropout; Segmentation


Electrical and Computer Engineering | Engineering


Use Find in Your Library, contact the author, or interlibrary loan to garner a copy of the item. Publisher policy does not allow archiving the final published version. If a post-print (author's peer-reviewed manuscript) is allowed and available, or publisher policy changes, the item will be deposited.

UNLV article access