Abstract: An image input section selects a plurality of sample points from an input image, and passes the density values of the sample points and pixels around them to a feature parameter extractor. The feature parameter extractor calculates the feature parameters of each sample point and passes them to a density modification function generator. The density modification function generator creates a density modification function for each sample point and passes the density modification function data to a density modification function modifier. The density modification function modifier determines a density modification function for all of the pixels, including those other than the sample points, and passes this density modification function data to a processor. The processor modifies image data delivered from the image input section, and passes the modified image data to a modified image output section.

Abstract: An electro-optical imaging array provides compensation for image motion due to variations in scene terrain electronically and with no moving parts. Pixel information representing scene information is transferred through the array in column groups. Each column group has its own charge transfer rate U. Successive images of the scene are generated by the imaging array, and the images are correlated by electronic signal processing circuitry to determine the image displacement of a fixed point in the scene between successive images. The image displacement is used to calculate a residual image velocity U.sub..delta. in each column group. As successive images of the scene are generated, the charge transfer rates U for each column group are updated, whereby U=U.sub.0 -U.sub..delta., where U.sub.0 is the charge transfer rate for the previous exposure, and U.sub..delta. is the residual image velocity in each column group.

Abstract: A method for establishing a relationship between a text image and a transcription associated with the text image uses conventional image processing techniques to identify one or more geometric attributes, or image parameters, of each of a sequence of regions of the text image. The transcription labels in the transcription are analyzed to determine a comparable set of parameters in transcription label sequence. A matching operation then matches the respective parameters of the two sequences to identify image regions that match with transcription regions. The result is an output data structure that minimally identifies image locations of interest to a subsequent operation that processes the text image. The output data structure may also pair each of the image locations of interest to a transcription location, in effect producing a set of labeled image locations. In one embodiment, the sequence of locations of words and their observed lengths in the text image are determined.

Abstract: A first set of input index signals represents a row-column array of quantized pixels; and each input index signal in the first set represents a rectangular non-overlapping quantized pixel group which is aligned to a particular row and a particular column. By repeatedly performing three steps on different pairs of the input index signals in the first set, a second set of index signals is generated wherein a respective index signal exists for each individual pixel in the array and by which the image can be filtered. These three steps are: 1) selecting a pair of input index signals in the first set such that the two quantized pixel groups which the selected index signals represent are adjacent to each other in the array; 2) forming an address signal by combining the two selected input index signals; and 3) reading a memory with the address signal to thereby obtain an output index signal.

Abstract: An optical cutting image which has a maximum portion in the X-axis direction on the screen of an image sensing device and which extends in the X-axis direction away from the maximum portion so as to expand in the Y-axis direction is measured by means of plural windows to be set on both side portions of the image as seen in the Y-axis direction, equations to represent the image lines of both side portions, and the like. The position of a front end point of the image is measured. A first window which is oblong in the X-axis direction is set such that the front end point is contained therein. A position of the center of gravity of image inside the first window is measured. Second and third windows which are oblong in the Y-axis direction are set at a position which is backwardly away by a predetermined length from the center of gravity of image inside the first window in the X-axis direction to measure the position of the center of gravity of image inside each of the windows.