20 Oct 2016
Multimedia Forensics - Using Mathematics and Machine Learning to Determine and Image's Source and Authenticity
(1:10-1:50 pm ET)
And
Towards Automatic Extraction, Synthesis, and Prediction of Cyber Attack Scenarios
(2:00-2:40 pm ET)
Mark your calendars and come join your friends in the CAE community for a Tech Talk. We are a warm group that shares technical knowledge. CAE Tech Talks are free and conducted live in real-time over the Internet, so no travel is required. You can attend from just about anywhere (office, home, etc.) Capitol Technology University (CTU) hosts the presentations using their online delivery platform (Adobe Connect) which employs slides, VOIP, and chat for live interaction. Just log in as “Guest” and enjoy the presentation(s).
Below is a description of the presentation(s) and logistics of attendance:
Date: Thursday 20 Oct 2016
Time: 1:10-1:50 pm ET
Location: https://capitol.adobeconnect.com/cae_tech_talk/
Just log in as “Guest” and enter your name. No password required.
Title/Topic: Multimedia Forensics - Using Mathematics and Machine Learning to Determine and Image's Source and Authenticity
Audience Skill Level: All Levels
Presenter: Dr. Matt Stamm (Drexel University)
Description:
Digital images play a critical role in today's society. They are frequently used by news agencies during reporting, as evidence during criminal investigations, and as intelligence in many military and defense scenarios. This proves problematic since an information attacker can easily create perceptually realistic forgeries using editing tools such as Adobe Photoshop.
In this talk, I will discuss several multimedia forensic algorithms capable of determining if an image has been manipulated that we have developed at the Multimedia and Information Security Lab (MISL) here at Drexel. Instead of relying on extrinsic security measures such as cryptography, these techniques identify image manipulations and forgeries by exploiting the intrinsic fingerprints left in digital media by editing operations. Additionally, I will discuss how multimedia forensic techniques can determine the source of an image by utilizing traces left in an image by its source camera instead of unreliable or easily falsifiable information sources such as metadata.
Date: Thursday 20 Oct 2016
Time: 2:00-2:40 pm ET
Location: https://capitol.adobeconnect.com/cae_tech_talk/
Just log in as “Guest” and enter your name. No password required.
Title/Topic: Towards Automatic Extraction, Synthesis, and Prediction of Cyber Attack Scenarios
Audience Skill Level: All Levels
Presenter: Dr. Jay Yang (Rochester Institute of Technology)
Description:
Cyber attacks to enterprise networks have moved into an era where both attackers and security analysts utilize complex strategies to confuse and mislead one another. Critical attacks often take multitudes of reconnaissance, exploitations, and obfuscation techniques to achieve the goal of cyber espionage and/or sabotage. The discovery and detection of new exploits, though needing continuous efforts, is no longer sufficient. This talk will discuss some of the recent research efforts that build upon and beyond the conventional intrusion detection systems. Specifically, we will present our approaches on learning and predicting attack patterns based on IDS alerts, extracting emerging attack behaviors using semi-supervised learning, modeling and analyzing attack obfuscation, and simulating cyber attack scenarios for diverse attack behaviors and network configurations. This collection of works aims at automatically recognizing and synthesizing models to represent how cyber attacks may transpire in well protected enterprise networks.
CAE Tech Talks are also recorded
CTU will post a recording of the live presentations on its website:
https://capitol.instructure.com/courses/510/external_tools/66
Contact
CAE Tech Talk events are advertised thru email and posted to the news and calendar section of the CAE community website: www.caecommunity.org
For questions on CAE Tech Talk, please send email to CAETechTalk@nsa.gov
