Mayachitra is developing advanced cybersecurity tools using deep learning and pattern recognition methods.
Image and Video Forensics
Mayachitra team is developing state-of-the-art methods for media forensics (supported by the DARPA MediFOR program, and in collaboration with researchers at the China Lake Air Warfare Research Center and the University of California, Riverside). The developed methods will be able to differentiate between manipulated and un-manipulated images, and localize the manipulated regions. These methods leverage recent advances in deep learning for image analysis with the team’s expertise in image processing, segmentation and classification.Recent Publication(s):
- T. M. Mohammed, J. Bunk, L. Nataraj, J. H. Bappy, A. Flenner, B. S. Manjunath, S. Chandrasekaran, A. K. Roy-Chowdhury, L. Peterson, "Boosting Image Forgery Detection using Resampling Detection and Copy-move analysis", IS&T International Symposium on Electronic Imaging, 2018 [paper]
- A. Flenner, L. Peterson, J. Bunk, T. M. Mohammed, L. Nataraj, B. S. Manjunath, "Resampling Forgery Detection using Deep Learning and A-contrario analysis", IS&T International Symposium on Electronic Imaging, 2018
- J. H. Bappy, A. K. Roy-Chowdhury, J. Bunk, L. Nataraj, B. S. Manjunath, "Exploiting Spatial Structure for Localizing Manipulated Image Regions", ICCV, 2017 [paper]
- J. Bunk, J. H. Bappy, T. M. Mohammed, L. Nataraj, A. Flenner, B. S. Manjunath, S. Chandrasekaran, A. K. Roy-Chowdhury, L. Peterson, "Detection and Localization of Image Forgeries using Resampling Features and Deep Learning", CVPR Workshop on Media Forensics, 2017 [paper]
New approaches for Malware Detection and Classification
[Jan 2018] Mayachitra, Inc. is proud to announce their web-accessible service MalSee to detect malware binaries
Researchers at Mayachitra are pioneering new ways of detecting and classifying computer malware. A new malware detection/classification system, called MalSee, is deployed as a web-accessible service. Mayachitra’s unique technology empowers robust and fast detection of harmful computer viruses, offering 1000x speedup compared to existing methods. The solution leverages on ideas from signal processing, pattern recognition, and deep learning. This project is supported by a Phase II Navy SBIR funding.
- L. Nataraj, B. S. Manjunath, "SPAM: Signal Processing to Analyze Malware", IEEE SIgnal ProcESSIng MagazInE, 2016 [paper]