AI-Driven Imaging System Protects Authenticity
May 30, 2019 | NYU Tandon School of EngineeringEstimated reading time: 2 minutes
To thwart sophisticated methods of altering photos and video, researchers at the NYU Tandon School of Engineering have demonstrated an experimental technique to authenticate images throughout the entire pipeline, from acquisition to delivery, using artificial intelligence (AI).
In tests, this prototype imaging pipeline increased the chances of detecting manipulation from approximately 45 percent to over 90 percent without sacrificing image quality.
Determining whether a photo or video is authentic is becoming increasingly problematic. Sophisticated techniques for altering photos and videos have become so accessible that so-called "deep fakes"—manipulated photos or videos that are remarkably convincing and often include celebrities or political figures—have become commonplace.
Pawel Korus, a research assistant professor in the Department of Computer Science and Engineering at NYU Tandon, pioneered this approach. It replaces the typical photo development pipeline with a neural network - one form of AI - that introduces carefully crafted artifacts directly into the image at the moment of image acquisition. These artifacts, akin to "digital watermarks," are extremely sensitive to manipulation.
"Unlike previously used watermarking techniques, these AI-learned artifacts can reveal not only the existence of photo manipulations, but also their character," Korus said.
The process is optimized for in-camera embedding and can survive image distortion applied by online photo sharing services.
The advantages of integrating such systems into cameras are clear.
"If the camera itself produces an image that is more sensitive to tampering, any adjustments will be detected with high probability," said Nasir Memon, a professor of computer science and engineering at NYU Tandon and co-author, with Korus, of a paper detailing the technique. "These watermarks can survive post-processing; however, they're quite fragile when it comes to modification: If you alter the image, the watermark breaks," Memon said.
Most other attempts to determine image authenticity examine only the end product - a notoriously difficult undertaking.
Korus and Memon, by contrast, reasoned that modern digital imaging already relies on machine learning. Every photo taken on a smartphone undergoes near-instantaneous processing to adjust for low light and to stabilize images, both of which take place courtesy of onboard AI. In the coming years, AI-driven processes are likely to fully replace the traditional digital imaging pipelines. As this transition takes place, Memon said that "we have the opportunity to dramatically change the capabilities of next-generation devices when it comes to image integrity and authentication. Imaging pipelines that are optimized for forensics could help restore an element of trust in areas where the line between real and fake can be difficult to draw with confidence."
Suggested Items
ASMC 2024 to Showcase AI, Smart Manufacturing and Sustainability to Advance Chip Industry Manufacturing Expertise
03/27/2024 | SEMIMore than 125 experts will offer insights into the latest semiconductor manufacturing strategies and methodologies as hundreds of industry stakeholders gather at the 35th annual SEMI Advanced Semiconductor Manufacturing Conference (ASMC 2024), May 13-16 in Albany, New York.
IPC APEX EXPO 2024: LPKF—Debunking Depaneling Industry Perceptions
03/27/2024 | Nolan Johnson, I-Connect007In this audio interview, listen to Jake Benz discuss advances in laser depaneling at LPKF. Thanks to advances in laser technology, perceptions about laser depaneling are changing from a low-speed, specialized process to a high volume process suitable for production manufacturing. Benz elaborates on some of the development and engineering that went into creating their latest, most capable depaneling machines.
Seeking Employment: Meet Gary Turner
03/25/2024 | Barry Matties, I-Connect007Meet Gary Turner, a recent graduate from the University of Texas at Dallas with a bachelor’s degree in mechanical engineering and a master’s in material science and engineering. He is currently seeking employment in the industry. The following interview will allow you to learn about Gary and see if he might be a good candidate for a position you are looking to fill.
The Challenges, Opportunities, and Future Specialties of PCB Design
03/19/2024 | Stephen V. Chavez, Siemens EDAWhat were once specialties have become more generalized over time—PCB designers must learn about design automation, signal integrity (SI), electromagnetic compatibility (EMC), complex high-speed design, mechanical design, and manufacturability/producibility. Design engineers must learn about layout, simulation, and supply chains—and in their place new specialties have emerged, like multi-gigabit SerDes channel design, advanced manufacturing, IoT, and multi-physics system verification.
Siemens, NVIDIA Expand Collaboration on Generative AI for Immersive Real-time Visualization
03/19/2024 | PRNewswireSiemens announced that it will deepen its collaboration with NVIDIA to help build the industrial metaverse. Siemens is bringing immersive visualization powered by new NVIDIA Omniverse Cloud APIs to the Siemens Xcelerator platform, driving increased use of AI-driven digital twin technology.