Skip to Content
Find More Like This
Return to Search

Modular error embedding

United States Patent

5,970,140
October 19, 1999
View the Complete Patent at the US Patent & Trademark Office
Los Alamos National Laboratory - Visit the Technology Transfer Division Website
A method of embedding auxiliary information into the digital representation of host data containing noise in the low-order bits. The method applies to digital data representing analog signals, for example digital images. The method reduces the error introduced by other methods that replace the low-order bits with auxiliary information. By a substantially reverse process, the embedded auxiliary data can be retrieved easily by an authorized user through use of a digital key. The modular error embedding method includes a process to permute the order in which the host data values are processed. The method doubles the amount of auxiliary information that can be added to host data values, in comparison with bit-replacement methods for high bit-rate coding. The invention preserves human perception of the meaning and content of the host data, permitting the addition of auxiliary data in the amount of 50% or greater of the original host data.
Sandford, II; Maxwell T. (Los Alamos, NM), Handel; Theodore G. (Los Alamos, NM), Ettinger; J. Mark (Los Alamos, NM)
The Regents of the University of California (Los Alamos, NM)
08/ 646,837
August 2, 1996
This invention was made with Government support under Contract No. W-7405-ENG-36 awarded by the U.S. Department of Energy. The Government has certain rights in the invention.