Increasingly sophisticated algorithms, including trained artificial intelligence methods, are now widely employed to enhance image quality. Unfortunately, these algorithms often produce somewhat hallucinatory results, showing details that do not correspond to the actual scene content. It is not possible to avoid all hallucination, but by modeling pixel value error, it becomes feasible to recognize when a potential enhancement would generate image content that is statistically inconsistent with the image as captured. An image enhancement algorithm should never give a pixel a value that is outside of the error bounds for the value obtained from the sensor. More precisely, the repaired pixel values should have a high probability of accurately reflecting the true scene content.
The current work investigates computation methods and properties of a class of pixel value error model that empirically maps a probability density function (PDF). The accuracy of maps created by various practical single-shot algorithms is compared to that obtained by analysis of many images captured under controlled circumstances. In addition to applications discussed in earlier work, the use of these PDFs to constrain AI-suggested modifications to an image is explored and evaluated.
The errpdf program was developed as a reference implementation of various methods for computing these pixel value error models. It is described in detail in Construction, quality assessment, and applications of pixel value error PDF models, by Henry Dietz. That paper uses names for the algorithms that are consistent with the 250302 release version of errpdf; be warned that the slides presented at the Electronic Imaging 2025 conference on 250204 used somewhat different naming. The released versions of errpdf are available as open source code here:
If you have any comments or suggestions, you can send them to Professor Hank Dietz.
The only thing set in stone is our name.