2010/02/10 by Sira Ferradans, Ferradans, Sira, Marcelo Bertalmı́o +4
Computer Science · Engineering · Physics and Astronomy · #Advanced Image Fusion Techniques #Color Science and Applications #Computergraphik (swd: 4010450-3) #Gaussian Mixture Model #High Dynamic Range (swd: 4790204-8) #High Dynamic Range Images #Image Enhancement Techniques #Informatik (DDC 004) #Naka Rushton Formula #Tone Mapping #Tone Mapping (swd: 4815312-6) #Weber-Fechner Contrast
paper · doi:10.20385/1860-2037/9.2012.7
openalex publication_date 2010/02/10 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/01
We present an improvement of TSTM, a recently proposed tone mapping operator for High Dynamic Range (HDR) images, based on a multi-modal analysis. One of the key features of TSTM is a suitable implementation of the Naka-Rushton equation that mimics the visual adaptation performed by the human visual system coherently with Weber-Fechner's law of contrast perception. In the present paper we use the Gaussian Mixture Model (GMM) in order to detect the modes of the log-scale luminance histogram of a given HDR image and then we use the information provided by GMM to properly devise a Naka-Rushton equation for each mode. Finally, we properly select the parameters in order to merge those equations into a continuous function. Tests and comparisons to show how this new method is capable of improving the performances of TSTM are provided and commented, as well as comparisons with state of the art methods.