We considered the issue of color mixing from the viewpoint of gray scale images and investigated the effect of the mixed light. The goal of this study was to improve the quality of the grey scale images by using mixed RGB illumination. The mixing mechanism and image quality were formulized and applied to a machine vision system with an RGB mixer. The RGB mixer was made of RGB LEDs, a power AMP, a mixing chamber, and a bundle of optical fibers. The intensity of the RGB light was adjusted by varying the 12 bit DAC voltage. The optical characteristics of the RGB LEDs were measured by a spectrometer. The color variation of the mixed light was monitored by a chromatic meter. Grey scale images were captured and analyzed under various lighting conditions provided by the mixer. The optimal illumination was determined by the distribution of the grey scale in the captured image. The chromatic diagram showed that various colors can be generated, and the image histogram showed that the mixed light can increase the quality of the grey scale image.

This paper proposes a wafer alignment method using a feedforward neural network (FNN). A wafer is placed on a vacuum table and its misalignment is inspected by a frame grabber. The alignment of the wafer is repeatedly corrected by adjusting its kinematic positions until the misalignment becomes zero. The training set is composed of the misalignment and the xy?? compensation. Fifteen sets of misalignment data were measured by the conventional alignment method. The FNN has four normalized inputs as mark locations in the field of vision and three normalized outputs for the kinematic compensation. The macro and micro steps share the FNN by means of a multiplexer (MUX) and a scaler. The training rule was back-propagation (BP). The proposed method was applied to a wafer manufacturing machine and its alignment performance was compared with that of conventional methods. The alignment time was reduced by the FNN model which was especially efficient in the case of macroscopic alignment.

Mixed-color illumination affects the quality of images in industrial vision system and it is important to optimize color and intensity for image acquisition. This study used simplex search to find the optimal illumination in a short amount of time. A typical color mixer synthesized various color of lights by changing the inputs of RGB power LEDs and passing the lights through an optical system. The image quality under mixed-color illumination was calculated according to the sharpness. For the purpose of optimal illumination using simplex search, a probe network was organized with N + 1probing points for N inputs. The shape of the probe network, simplex, was varied through procedures of extension, contraction, and shrinkage. The inputs of the color mixer were changed until the size of the simplex became smaller than a threshold. The simplex search was tested for commercial semiconductor patterns, and was useful for finding the optimal illumination.

Quick search for optimal illumination using mixed color is proposed, and the possibility for use in industrial vision systems is considered. Multidimensional formulation of the optimum search methods is reviewed to use multiple inputs for various color sources and one output of image sharpness. The voltage inputs are used to adjust the intensity of each color source and the optimal illumination is found by the maximum of the sharpness. The optimum search methods reveal that the number of iterations for the optimal condition in mixed color illumination was under 0.1% the number required by equal section search. The searching paths and convergence were considered in both of the input coordinate and the color coordinate. The results show that the optimum search methods make it possible to find the optimal lighting conditions quickly and accurately.