- Binghamton University Research News - https://discovere.binghamton.edu -

Digital watermarking algorithm wins praise

The Digital Watermarking Alliance (DWA), the international alliance of industry leading organizations delivering valuable digital watermarking solutions, presented its 2010 Best Paper award to a team of electrical and computer engineers from Binghamton University.

Doctoral student Tomas Filler and visiting scholar Jan Judas, under the guidance of Jessica Fridrich, professor of electrical and computer engineering in the Thomas J. Watson School of Engineering and Applied Science, earned the top spot for their paper. According to the alliance, it demonstrated the highest level of originality, creativity, clarity and potential impact on the field of watermarking.

Announced at the IS&T/SPIE Electronic Imaging Conference held Jan. 23 – 27, the winning team received a cash prize of $1,500, and their paper will be publicized in this year’s conference proceedings.

The Binghamton University team’s winning paper, titled “Minimizing Embedding Impact in Steganography using Trellis-Coded Quantization,” proposed and developed a practical algorithm to minimize the average embedding distortion in steganography.

Serving as a modern-day version of writing with invisible ink, steganography enables additional hidden information to be exchanged during regular communication without anyone knowing. In today’s world of technology, this means hiding secret communication data in digital media files.

“A digital picture of a dog may contain an entire PDF document even though the image itself may look completely innocuous to most people,” said Fridrich. “I like to call this research a ‘Swiss-Army Knife’ for steganographers. It’s a very general and elegant framework to build stego systems that work in the best possible way and hide the maximum amount of information, without creating any artifacts that someone might use to find out that secret communication is taking place.”

Fridrich said the research also extends to what is referred to as a “lossy” compression of data, an example of which is the commonly used JPEG image format. Lossy allows a user to store an image in a compact size at the price of slightly distorting it.

“Our paper describes a method that could be used to optimally encode the image so that we could get the smallest sized file possible,” Fridrich said. “And the way to do that is by allowing each pixel in the image to be distorted by a specified amount.”

A leading authority in the field of steganography, Fridrich credits Filler for helping the team achieve this honor. “The research was conceived by Tomas and he deserves all the accolades,” Fridrich said. “He is an unbelievably brilliant student who has made numerous contributions and revolutionized the field.”

This is the second time Fridrich and Filler have received the alliance’s top recognition. They were both members of the 2009 Best Paper award-winning team that was honored for theoretical work on secure capacity of practical stego systems.

Like this article? Please share!