Abstract: A method of combined optical mark reading ("OMR") and image processing utilizes a response sheet having a grid of response positions suitable for marking and for processing by OMR and at least one response area for a response of an alphanumeric or pictorial nature, the response area encompassing at least one of the positions suitable for processing by OMR. The response sheet is scanned to detect the presence of data signifying marks at positions in the grid and the grid is sensed for marks capable of detection and reading by OMR. In response to the sensing step, an image in the response area of a response of an alphanumeric or pictorial nature is selectively captured and stored simultaneously with OMR.

Abstract: An optical character recognition system which automatically reads handwritten characters and the like which do not have to be printed in a special format. Recognition tables derived from the pattern bit map and from a skeleton pixel matrix describe the character in terms of the relative position of the pixels along the borders of the character and provide a plurality of recognition strings, one in each table, for the front and rear views of the character and for the shape of the holes in the character which are opened from the top (as in the numeral four) or opened from the bottom (as in the numeral seven). From the recognition tables, the characters are recognized by searching recognition files containing blocks of successions of lines of code corresponding selectively to the codes in the recognition tables.

Abstract: Improvements in bit-image quality in printing systems is achieved by providing several different pattern matching processors. Each pattern matching processor comprises a lookup table of patterns and corresponding replacement image patterns. A source image is modified by replacing areas in the source image that match patterns with the corresponding image patterns. Pattern matching processors are chosen depending on: the type of image being processed, input from the operator (user), feedback from the xerographic processor, and the color of the image. A user interface is provided which allows a user to download template and pattern data for each pattern matching processor. Pattern matching can be used to correct single-bit, half-bit, line-growth, xerographic effects and aliasing errors, and to provide control of print quality to the user.

Abstract: A method and apparatus for differentiating and extracting handwritten annotations and machine printed text in an image. The method provides for the use of morphological operations, preferably at reduced scale, to eliminate for example, the handwritten annotations from an image. A separation mask is produced that, for example, converts all the image pixels corresponding to machine printed text, and none of the image pixels corresponding to handwritten or handprinted annotations. The separation mask is used in conjunction with the original image to produce separate handwritten annotations and machine printed text images. The invention also provides a method and apparatus for identifying the location of specialized type styles such as bold and italic is disclosed. The method erodes a binary image utilizing structuring elements which provide a relatively large number of hits in regions containing the specialized type styles.

Abstract: A method of inspecting a liquid crystal panel and determining the type of defect. Each part of the liquid crystal panel is extracted and inspected by comparison with an extracted reference part. The parts to be inspected may be found based on the location of previously found parts or by examining a plot of the density in the image. Vector data from a part that is known to be defectless is used to extract the reference part. Alternatively, the contour of a reference part may be determined by analyzing the vector data which was obtained for the parts to be inspected. The type of defect is determined by classifying the defective part according to characteristics of the part's image (e.g., brightness of the pixels) as compared with parts having known defects.

Abstract: A system for recognition of characters on a medium. The system includes a scanner for scanning a medium such as a page of printed text and graphics and producing a bit-mapped representation of the page. The bit-mapped representation of the page is then stored in a memory means such as the memory of a computer system. A processor processes the bit-mapped image to produce an output comprising coded character representations of the text on the page. The present invention discloses parsing a page to allow for production of the output characters in a logical sequence, a combination of feature detection methods and template matching methods for recognition of characters and a number of methods for feature detection such as use of statistical data and polygon fitting.

Abstract: A library of L unenhanced images (pixel bit maps) is enhanced for optical character recognition (OCR) with respect to a pre-existing group of G input symbols (pixel bit maps) for creating a library of G recognition enhanced templates (pixel bit maps) of the G input symbols. The enhancement is accomplished by comparing each image of the library with each symbol of the group, and weighting the images with the highest potential for confusion. A primary comparison C* and a secondary comparison C** are identified from the L comparisons within each of the G sets of comparisons. A recognition margin is determined between each pair of identified comparisons C* and C**. The single pair of identified comparisons C* and C** is selected forming the smallest recognition margin M*. The single pair of images I.sub.j * and I.sub.j ** underlying the pair of identified comparisons C* and C** is identified. Certain pixels of the closest pixel image I.sub.j * and the next closest pixel image I.sub.

Abstract: Each processing element has a number of weights for each input connection. These weights are coefficients of a polynomial equation. The use of quadratic nodes permits discrimination between body pixel and edge pixels, in which an intermediate value is present, using a grey scale image. In the training method of the present invention, the middle layer is initially one leaf node which is connected to each output node. The contribution of each leaf node to the total output error is determined and the weights of the inputs to the leaf nodes are adjusted to minimize the error. The leaf node that has the best chance of improving the total output error is then "converted" into a branch node with two leaves. A branch node selected from a pool of trial branch nodes is used to replace the chosen leaf node. The trial branch nodes are then trained by gradient training to optimize the branch error function.

Abstract: In a handy type image scanner, information of an original is observed through an window formed on a case of the image scanner. The handy type image scanner includes a case which encloses a light source for illuminating an original, a reflecting mirror for conducting light reflected from the original to a predetermined direction, and an image sensor for receiving the light from the reflecting mirror so as to input information of the illuminated original. The case includes an upper surface which has a window for observing the information displayed on the original which is being optically read out by the image sensor.

Abstract: The profile of a tool or other object having a flat reference plane is illuminated by interfering beams of coherent light to generate a striped light pattern which at any point deviates from straight lines as a function of the surface spacing from the reference plane. A camera image of the pattern taken at an angle to the beams is processed by an image analysis program which converts the stripes to lines corresponding to light peaks and valleys of the pattern. Tool wear or object profile is determined from the deviation of lines from straight reference lines. The program includes algorithms for identifying peak and valley line segments, determining the segments which are neighbors of other segments, and connecting lines according to a set of rules and constraints which use the data on neighbor segments and the proximity of segment end points.

Abstract: A method and apparatus for identifying and correcting for document skew. Lines of a bitmap are scanned and a variance in the number of ON pixels as a function of skew angle is calculated. Skew of the original document occurs when the variance is a maximum.

Abstract: An image reduction apparatus includes a smoothing unit for smoothing binary image data, a binarization unit for binarizing an output from the smoothing unit, a detection unit for detecting an attribute of the binary image data smoothed by the smoothing unit, a plurality of correction units corresponding to a plurality of attributes, and a selector for selecting one of the correction units in accordance with the attribute detected by the detection unit. An output from the binarization unit is corrected by the correction unit selected by the selector.

Abstract: A handwritten character image normalization technique provides predetermined pixel dimensions and a normalized skew. The skew slope of the input image is calculated. This skew slope is employed to determine the bounds of the smallest parallelogram which completely encloses all of the stroke pixels of the input image. This parallelogram has a first pair of opposed horizontal sides and a second pair of opposed sides having the skew slope. The stroke pixels of this parallelogram are then mapped into the standard size horizontal row and vertical column pixel dimensions using horizontal and vertical scaling factors determined from the parallelogram dimensions and the standard dimensions. This mapping employs a subpixel grid of the normalized pixels. Candidate stroke pixels are identified which correspond to any part of a stroke pixel of the input image.

Abstract: A method for automatically measuring the complexity of a technical illustration for prospective users involves converting (50, 52, 54) the technical illustration into compressed data and measuring (60, 66, 70, 72, 74, 78, 80, 82, 86, 88, 92, 94, 96) conversion of the technical illustration into compressed data, to obtain a performance value (94) corresponding to the complexity of the technical illustration (10). This process includes, for example, the step of processing (50, 52, 54) a technical illustration with image-processing software that processes the technical illustration (10). The process further includes embedding measurement operations (60, 66, 70, 72, 74, 78, 80, 82, 86, 88, 92, 94, 96) at predetermined locations in the software to evaluate a selected task performed by the software during the processing (50, 52, 54) of the illustration. The measurements obtained are used to calculate a performance value (94) indicative of the performance of the software in processing the illustration (10).

Abstract: A method for identifying wheat cultivars is provided which employs an imaging device such as a video camera and a computer to acquire and process images of grain samples and to identify the particular type or cultivar of the wheat by the images obtained thereby. The method involves the steps of acquiring the image of the kernel and germ for each grain, processing the image into digital format, storing data in the computer corresponding to the edge of the kernel and germ, determining the image characteristics of the kernel and germ, and comparing the image characteristics for the sample to a known standard. By accumulating data for a number of grains, calculating statistical information for the sample, and comparing to a known standard, the particular cultivar of the sample may be identified.

Abstract: Transforms such as the DCT are useful for image compression. One close relative of the DCT is preferred for its arithmetic simplicity. A method and apparatus is described whereby the image compression is done with no multiplications while still compatible with a JPEG (Joint Photographic Experts Group) Transform. Other enhancements are made to improve image quality.

Abstract: A forms recognition system having a method which assists an operator in creating an electronic template. The automated method would allow an operator to view an electronic bitmap of a master or blank form on a graphical display, the master form having bounded or semi-bounded regions depicting fields. An operator would display the bitmap, select a single point within a bounded or semi-bounded region on the bitmap with a pointing device, and the coordinates representing the bounded region would be automatically determined using the single point selected by the operator.

Abstract: In a pattern recognition system, input signals are applied to a short term feature representation field of nodes. A pattern from the short term feature representation field selects at least one category node in a category representation field. The selected category then generates a template pattern. With an insufficient match between the input pattern and template pattern, the category selection is reset. Category selection is based on selection weights which are initially set equal to long term memory weights. After reset, however, selections weights are reduced. Reduction is greatest at those nodes where excitation in F.sub.2 was greater prior to reset. The category representation field is of the same form as the field which receives the input and may itself serve as an input to a higher level pattern recognition system.

Abstract: Characters to be recognized and which may be handwritten are each described by a pattern of one bit thresholded pixel data which follows the center line of the character and encoded into strings of digital data signals (bytes) which represent the shape of the pattern from three aspects, namely the top view, the front edge and the rear edge thereof, thereby providing low level recognition information in the form of three digital data strings. From these strings, the characters are recognized by finding high level recognition strings of digital signals which correspond to different characters in the character set to be recognized. This high level recognition is carried on in two steps, namely allocating the high level recognition strings which correspond to the low level strings for each aspect of the character and then finding combinations of high level strings reflecting the same character, thereby accounting for variations in the shapes in which the same character can be written.

Abstract: There is disclosed an image communication apparatus, such as a facsimile apparatus, having the function of displaying the received image. The apparatus is provided with a storage unit and a control unit, and is capable of storing the received image. The control unit is capable of displaying the image information upside down by the unit of a page in order to facilitate the viewing of the image in case the image information is erroneously sent upside down.