Computing a Non-trivial Lower Bound on the Joint Entropy between Two Images

Abstract

In this report, a non-trivial lower bound on the joint entropy of two non-identical images is developed, which is greater than the individual entropies of the images. The lower bound is the least joint entropy possible among all pairs of images that have the same histograms as those of the given images. New algorithms are presented to compute the joint entropy lower bound with a computation time proportional to S log S where S is the number of histogram bins of the images. This is faster than the traditional methods of computing the exact joint entropy with a computation time that is quadratic in S.

Publication

https://www.osti.gov/biblio/1347338

Kalyan Perumalla
Kalyan Perumalla

Kalyan Perumalla is Founder and President of Discrete Computing, Inc. He led advanced research and development at ORNL and holds senior faculty appointments at UTK, GT, and UNL.

Next
Previous