Multi-illuminant color reproduction for electronic cameras via CANFIS neuro-fuzzy modular network device characterization

Abstract

We describe color reproduction and correction of images captured by electronic cameras under multiple illumination (or lighting) conditions, relating to color device characterization for enhancing the quality of color in the obtained images. In particular, we highlight a very practical use of neuro-fuzzy modular network coactive neuro-fuzzy inference systems (CANFIS) models for this application, and discuss their strengths and weaknesses compared with other adaptive network models (e.g., multilayer perceptron (MLP)) as well as conventional lookup-table-type (TRC-matrix) methods. Our in-depth investigation based on comprehensive numerical tests with a wide variety of illumination/lighting data (180 sources of illumination) shows that the "neuro-fuzzy CANFIS with MLP local experts" possesses a remarkable generalization/approximation capacity, even under a very restricted condition where only four-illuminant data sets were permitted to be used for optimization because of efficient practical implementation subject to an industrial setting.

Color Calibration of Scanners for Scanner‐Independent Grain Grading

“…Many color correction algorithms have been reported in the literature for color device calibration using mathematical formulation (Vrhel and Trussell 1999a), statistical modeling (Hu and Mojsilovic 2000;Sanchez and Binefa 2000), and neural networks (Cho and Kang 1995;Vrhel and Trussell 1999b). Some researchers combined neural networks with either genetic algorithms (Watanabe et al 2001) or fuzzy logic (Mizutani and Nishio 2002) for improved performance. However, histogram comparison remains the most commonly used technique for comparing color similarity of images due to its compact representation and low complexity (Hashizume et al 1998;Yining et al 2001;Zhang et al 2001).…”