Praveen Cyriac, doctoral student of the Department of Information and Communication Technologies (DTIC) at UPF, won the best paper award for his work “Optimized Tone Curve for In-Camera Image Processing” at the “Image Quality and System Performance” symposium of the Electronic Imaging 2016 conference held in San Francisco (USA) from 14 to 18 February. Other authors of the paper are David Kane and Marcelo Bertalmío, of the IP4EC (Image Processing for Enhanced Cinematography) Research Group of the DTIC. 

As explained by Bertalmío, associate professor at the DTIC and coordinator of the IP4EC group, “our methodological approach is based on recent knowledge of neuroscience and visual perception, developing a type of software that emulates these processes of the human visual system and applying them to images captured by a standard camera”.

The values captured by the sensors of digital cameras are transformed according to the sensitivity of the human visual system. The paper by Cyriac et al. proposes a transformation of the image capable of accurately reproducing the detail and contrast of the original scene.

The computational complexity of the method is extremely low, so the camera operates almost instantaneously and is able, with use, to predict the settings defined by the user. The result “is a natural final image, without any halos, colours or spurious artifacts”, Cyriac explains.

With the appropriate adjustments, the new method developed by the research team can also be applied to video.

Reference work:

P. Cyriac, Mr. Kane and Mr. Bertalmío, “Optimized Tone Curve for In-Camera Image”, Proceedings of IS&T Electronic Imaging Conference, San Francisco (USA), 14-18 February, 2016.