Physics:Photo response non-uniformity

From HandWiki
Short description: Noise in digital sensors


Photo response non-uniformity, pixel response non-uniformity, or PRNU, is a form of fixed-pattern noise related to digital image sensors, as used in cameras and optical instruments. Both CCD and CMOS sensors are two-dimensional arrays of photosensitive cells, each broadly corresponding to an image pixel. Due to the non-uniformity of image sensors, each cell responds with a different voltage level when illuminated with a uniform light source, and this leads to luminance inaccuracy at the pixel level.[1]

Forensic use

The PRNU of a device's imaging sensor is unique to that device. It creates a non-changing, unique noise pattern, which is embedded in the pixels of each photo.[2] PRNU-based identification of a photo's source (i.e., a specific camera) is possible even when the resulting picture has been heavily post-processed and manipulated.[3]

Metrology camera corrections

High-end and metrology camera vendors tend to characterize this non-uniformity during instrument manufacture. The sensor is illuminated with a standardized light source and a two-dimensional table of correction factors is generated. This table is either carried in camera non-volatile memory and dynamically applied to the image on each capture, or ships with the camera to be applied by an external image processing and correcting pipeline.[4]

See also

References

  1. "What is Photo-response Non-uniformity (PRNU) ?". Tucsen. 29 April 2022. https://www.tucsen.com/learning/demystifying-specifications-prnu/. 
  2. Lukáš, J.; Fridrich, J.; Goljan, M (2005). "Digital “bullet scratches” for images". IEEE Int. Conf. on Image Processing 2005 (IEEE): III–65. https://ieeexplore.ieee.org/document/1530329. Retrieved 28 January 2026. 
  3. Chen, M.; Fridrich, J.; Goljan, M. (2007). "Digital imaging sensor identification (further study)". Proc. SPIE 6505, Security, Steganography, and Watermarking of Multimedia Contents IX: 65050P1–65050P13. 
  4. "Shading Correction". Basler. https://docs.baslerweb.com/shading-correction.