A Neuro-Fuzzy Approach for Compensating Color Backlight Images.
Journal
Neural Processing Letters
Journal Volume
23
Journal Issue
3
Pages
273-287
Date Issued
2006
Author(s)
Abstract
This paper presents a neuro-fuzzy approach for compensating exposure in the case of backlighting, regardless of the position of objects. To achieve the compensation effect, the fuzzy C-means algorithm is first used to extract features from a backlight image. Then these extracted features are presented to a trained artificial immune system based neuro-fuzzy system (AISNFS) to estimate the amount of compensation. Finally, the estimated amount of compensation incorporated with a compensation equation is used to enhance the intensity component of the backlight image to produce a compensated image. Several backlight images were used to test the performance of the algorithm. © Springer 2006.
SDGs
Type
journal article