COLORED ILLUMINATION FOR ENHANCING DISCRIMINABILITY IN MACHINE VISION

We describe how to design a colored illumination that maximizes the di scriminability of objects in red-green-blue images and we describe the underlying mathematical theory, This procedure is useful in applicati ons where the illumination can be controlled, e.g., automated inspecti on, Our technique finds the optimal colored illumination in terms of t he camera's spectral response and the objects' spectral reflectance. A n experiment with filters simulating colored illumination on painted c olor patches is reported, This experiment confirms the effectiveness o f our design technique, Approximations for the spectral reflectances a nd the illumination sources are useful for designing real colored illu mination, These approximations were tested successfully on live potato plantlets from tissue culture, These experiments on color patches and potato plantlets demonstrate the usefulness of properly designed colo red illumination in machine vision, (C) 1995 Academic Press, Inc.