The integrity and origin of digital images are critical in our connected world, particularly within sensitive or proprietary contexts. Traditional methods for establishing photo provenance, such as geotagging and watermarking, face challenges when it comes to unwanted clandestine or unauthorized photography. This research proposes a novel solution using radio frequency (RF) signal injection to embed imperceptible watermarks and geolocation data directly into the images captured by CCD and CMOS optical sensors. The proposed research will test RF transmitters designed to induce humanly imperceptible, verifiable errors within the image's pixel data, which serve as both watermarks for copyright management and geotags for determining the origin of photographs. If successful, this research will not only detect and deter the spread of deepfakes purporting to be of real events but will also enhance the attribution of copyright for photographed intellectual property and assist in geotagging sensitive photographs taken within range of the appropriate transmitting satellites—for example, images captured by a kidnapper.

In environments where conventional RF shielding is impractical due to its interference with image capture, this method offers a discreet yet reliable alternative for embedding traceable information. Beyond the protection of sensitive information, this technology has the potential to revolutionize the verification process for digital media, providing a robust tool against the proliferation of deepfakes. By embedding a cryptographic RF watermark and a geotagged token, we can validate the authenticity of images—such as those depicting public figures at verified locations—thereby reinforcing the truth in our digital narrative. This research will explore a range of device designs and their effectiveness in creating durable, nonintrusive watermarks that can withstand attempts at removal or alteration.