局域傅立叶变换实现三色光栅黑白编码图象的彩色恢复

A Neural Network Model Constructed by RBF Function

本文提出一种从编码黑白胶片恢复真彩图象的数字局域解码技术和方法，首次实现了用计算机设备作编码片的解码。从离散数字抽样角度解释了黑白胶片光栅编码彩色图象记录技术，分析了光栅编码方法，根据局域变换思想建立起局域傅立叶变换带通滤波算子，实现从黑白编码胶片还原由Ｒ、Ｇ、Ｂ构成的原彩色像。该方法不需要傅里叶逆变换，只需一次傅里叶变换就可完成彩色恢复的任务。

A new neural network model constructed by the RBF Function is presented.Not only is the normal classification of the multipatterns carried out,but also is the classification of the multiobjects with rotation invariance done.The computer simulation of classifying the three kinds planes with rotation invariance is accomplished.The results state that this model is very good for the multiobjects classification with the rotation invariance.One of the optoelectronic implementation is presented in this paper.