Abstract: A color image processing apparatus comprises an image reading circuit for reading an original image as three color-separated images, a color code generating circuit for generating a color code indicating whether each pixel of the original image is white, black, or chromatic, a color reproducing circuit for generating density data for each pixel corresponding to each of a plurality of recording colors based on the color-separated images, a marker area detecting circuit for detecting a marker on the original image based on the color code generated by the color code generating circuit and for extracting a marker area enclosed by the marker, a density data sampling circuit for sampling density data of pixels on the marker at a sampling point located at a selected distance, corresponding to a predetermined number of pixels, inside of the edge of the marker, and for determining a density data of the marker based on the sample density data, and a density data conversion circuit for converting the density data of pix

Abstract: A vector neural network (VNN) of interconnected neurons is provided in transition mappings of potential targets wherein the threshold (energy) of a single frame does not provide adequate information (energy) to declare a target position. The VNN enhances the signal-to-noise ratio (SNR) by integrating target energy over multiple frames including the steps of postulating massive numbers of target tracks (the hypotheses), propagating these target tracks over multiple frames, and accommodating different velocity target by pixel quantization. The VNN then defers thresholding to subsequent target stages when higher SNR's are prevalent so that the loss of target information is minimized, and the VNN can declare both target location and velocity. The VNN can further include target maneuver detection by a process of energy balancing hypotheses.

Abstract: A pattern inspection apparatus comprises a sensor data input section for inputting a two-dimensional inspected pattern as image data (sensor data) having a multivalued (non-binary) density distribution, design data input section for inputting reference pattern data (reference data) corresponding to the inspected pattern, a compare section for making a comparison between the image data and the reference data to obtain the difference in density therebeween, a minimum compare section for performing spatial differentiation filtering on the distribution of density between the inspected pattern and the reference pattern in different directions and obtaining the minimum of the absolute values of the filtered outputs, and a first defect determining section for detecting a defect on the basis of the minimum obtained by the compare section.