The visual reality of a holographic image has improved effectively by utilizing multicolor reconstruction procedure. This fact is applicable to a real-time three-dimensional display for a computer generated hologram (CGH). However, it is quite difficult to generate a CGH for multicolor imaging in real-time because a CGH contains essentially a huge amount of information, and increases further information produced by multiplying the number of primary colors for multicolor imaging. Moreover, the optical system is considerably complicated for the multicolor image reconstruction. In this paper, a new method is presented to reconstruct a three dimensional multicolor image from a CGH. In this method, three sub-holograms to reconstruct the primary color images are sampled respectively for reducing the amount of computation and realizing a simple optical system. Fringe patterns are displayed by only one spatial light modulator (SLM) and color crosstalk images are eliminated by a color filtering system for ensuring that each sub-hologram can be only illuminated by the light with an appropriate color. A multicolor imaging method from a CGH is proposed and also the experimental results are shown.

We modified the adaptive fuzzy classification algorithm (AFC), which allows fuzzy clusters to grow to meet the demands of a given task during training. Every fuzzy cluster is defined by a reference vector and a fuzzy cluster radius, and it is represented as a shape of hypersphere in pattern space. Any pattern class is identified by overlapping plural hyperspherical fuzzy clusters so that it is possible to approximate complex decision boundaries among pattern classes. The modified AFC was applied to recognize handwritten digits, and performances were shown compared with other neural networks.