PRNU
Photo Response Non-Uniformity — the unique sensor noise fingerprint every camera leaves on its photos.
PRNU stands for Photo Response Non-Uniformity. It's a multiplicative noise pattern caused by tiny manufacturing variations between individual pixels on a camera sensor. No two sensors produce the exact same PRNU — even within the same model line — which makes it a reliable forensic fingerprint.
Forensic detectors extract PRNU by averaging many photos from the same camera and subtracting a denoised version of each. The residual is the sensor fingerprint. AI-generated images have no PRNU because there is no physical sensor — just pixels synthesized by a diffusion or GAN model. The absence of PRNU is one of the strongest signals AI image detectors use.
SynthGuard's image humanizer addresses this with a PRNU injection layer. It samples noise from one of dozens of real camera profiles (Canon EOS R5, Sony A7 IV, iPhone 15 Pro, Pixel 8, …) and imprints it onto the image at the same magnitude a real camera would. The injected PRNU passes statistical tests because the underlying noise is real sensor noise, not synthesized.
The injection is invisible to humans (the noise sits well below the perception threshold) but materially changes the detector's score. Combined with FFT disruption and texture perturbation, it's enough to flip most modern AI images to "likely human" on detectors like Hive, Sightengine, and Illuminarty.
Tools that address PRNU
Related terms
Related reading
PRNU, FFT & Sensor Noise — The Forensics Behind Image Authenticity
Image forensics is a small, mathematically dense field that quietly underpins everything from courtroom exhibits to AI detection startups. Three pillars do most of the heavy lifting: PRNU (the sensor…
How AI Image Detectors Actually Work — A 2026 Technical Guide
AI image detectors look magical from the outside — drop an image, get a percentage, ship the verdict. Inside, they are an assembly of brittle statistical signals stacked on top of each other, each ca…