Abstract: Apparatus for detecting etch defects in an object having a plurality of openings by generating an image signal of the object representing light intensity values of a plurality of pixels, processing the image signal to form data signals representing light intensity values of groups of pixels, filtering the data signals to remove signals representing the ends of the openings, and pairing two data signals when the groups of pixels represented thereby represent an etch defect.

Abstract: Apparatus for detecting etch defects in an object having a plurality of openings by generating an image signal of the object representing light intensity values of a plurality of pixels, processing the image signal to form data signals representing light intensity values of groups of pixels, filtering the data signals to remove signals representing the ends of the openings, and pairing two data signals when the groups of pixels represented thereby represent an etch defect.

Abstract: Apparatus for determining at least one edge of an object having a plurality of openings with respective edges. An imaging device generates an image signal of the object to represent light intensity values of a plurality of pixels, the image signal being processed to estimate a linear relationship derived from the light intensity values of a predetermined number of the pixels. A determining device determines whether the linear relationship includes at least one light intensity value corresponding to an edge.

Abstract: A pattern searching method using neural networks and correlation. This method combines the quickness and adaptiveness of neural networks with the accuracy of the mathematical correlation approach. Images are divided into small sub-images which are presented to the trained neural network. Sub-images that may contain the pattern or partial pattern are selected by the neural network. The neural network also provides the approximate location of the pattern, therefore the selected sub-images can be adjusted to contain the complete pattern. Desired patterns can be located by measuring the new sub-images' correlation values against the reference models in a small area. Experiments show that this superior method is able to find the desired patterns. Moreover, this method is much faster than traditional pattern searching methods which use only correlation.

Abstract: Apparatus for detecting etch defects in an object having a plurality of openings by generating an image signal of the object representing light intensity values of a plurality of pixels, processing the image signal to form data signals representing light intensity values of groups of pixels, filtering the data signals to remove signals representing the ends of the openings, and pairing two data signals when the groups of pixels represented thereby represent an etch defect.

Abstract: A lighting control apparatus for illuminating a moving variable-speed web for inspection with a camera is provided. The apparatus is comprised of a detecting device for detecting the positioning of the web; a lighting device, aligned with the camera, for emitting a light to illuminate the web; a calculating device for calculating the speed of the web; and a control device which controls the emission of light by the lighting device as a function of the speed of the web.

Abstract: A lighting control apparatus for illuminating a moving variable-speed web for inspection with a camera is provided. The apparatus is comprised of a detecting device for detecting the positioning of the web; a lighting device, aligned with the camera, for emitting a light to illuminate the web; a calculating device for calculating the speed of the web; and a control device which controls the emission of light by the lighting device as a function of the speed of the web.

Abstract: A neural network designed to recognize a particular character is supplied with initial tap weights for a first hidden node which are an image of the character to be recognized. The additive inverse of this set of weights is used as the tap weights for a second hidden node. A third node, if used, is initialized with random noise. The network is then trained with back propagation.

Abstract: A pattern searching method using neural networks and correlation. This method combines the quickness and adaptiveness of neural networks with the accuracy of the mathematical correlation approach. Images are divided into small sub-images which are presented to the trained neural network. Sub-images that may contain the pattern or partial pattern are selected by the neural network. The neural network also provides the approximate location of the pattern, therefore the selected sub-images can be adjusted to contain the complete pattern. Desired patterns can be located by measuring the new sub-images' correlation values against the reference models in a small area. Experiments show that this superior method is able to find the desired patterns. Moreover, this method is much faster than traditional pattern searching methods which use only correlation.

Abstract: A process for converting characters arranged circularly, as for example about the center hole of a compact disk, into a linear arrangement. Points are assigned to locations on the circular arrangement. These points are mapped to a linear arrangement. The number of points is selected to be greater than those of the original image to enhance the resolution of the resulting image. The location of the points is stored in an address array. The values of the pixels in the original image are then copied to a target array. The pixel values are then converted to binary values serving as input to a recognition neural network and a verification neural network.