Open AccessHistogram specification with generalised extreme value distribution to enhance retinal images
Author(s) -
Intajag S.,
Kansomkeat S.,
Bhurayatachai P.
Publication year - 2016
Publication title -
electronics letters
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.375
H-Index - 146
eISSN - 1350-911X
pISSN - 0013-5194
DOI - 10.1049/el.2015.3746
Subject(s) - brightness , histogram , grayscale , artificial intelligence , luminance , computer vision , computer science , contrast (vision) , adaptive histogram equalization , pattern recognition (psychology) , gamma correction , histogram equalization , image (mathematics) , mathematics , optics , physics
Histogram specification usually deals with greyscale image enhancements, although, it can apply to colour images; however, generally manipulating only the luminance channel. A method for automatic enhancement of digital colour retina images by specifying all colour bands of the histograms is designed to adjust brightness, contrast, and colour balance for visual diagnosis of age‐related macular degeneration. The proposed algorithm employs a generalised extreme value distribution consisting of three parameters to adjust brightness and colour balance by the location parameter, contrast and tone tuned by the scale, and shape parameters, respectively. The algorithm performance is evaluated with two image datasets: structured analysis of the retina and automated retinal image analysis. The improved results show that the specified histograms based on the generalised extreme value have been provided with important features to characterise the symptoms of a macular degeneration.