Abstract: A method of magnifying a digital image based on an edge map. The image has many original pixels, with each pixel having a level, and the locations of the pixels having a resolution. The method to magnify the image includes the steps of producing an edge map with many boundaries from the digital image, projecting the edge map onto the digital image, generating one or more additional pixels in locations among the original pixels by manipulating the levels of one or more original pixels without crossing any boundaries set by the edge map, and expanding the distances between the pixels so that the resolution of the locations of the pixels becomes the same as the original pixels. In another preferred embodiment, the original digital image is enlarged to the size of the magnified image by extending proportionally the locations of the original pixels. Then the edge map is produced and projected onto the enlarged image to generate additional pixels.

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 video camera system with autonomous target tracking sets a measuring frame on an object with high accuracy using a hue signal and a brightness signal without being affected by intensity and direction of the light source. A display screen is divided into divided areas and for each divided area, a lightness divided area in which the values of the brightness signal are within prescribed ranges of luminance and a skin color divided area in which the values of the hue signal are within prescribed ranges corresponding to skin color are extracted. Based on the extraction result, the position of the object on the screen determined and the measuring frame is set without being affected by intensity and direction of the light source. Thereafter, the object is continuously centered within the measuring frame.

Abstract: The invention is a universal input device for a computer which can be either used as a point-and-click bar code reader, a bar code scanner, mouse, handwriting input, or text scanner, all from an off-screen medium. The universal input device can be used to read both arbitrary user-defined bar code or standardized bar code. In either case, the bar code may symbolize any alphanumeric character or macro instruction for an operating system which is programmed on the computer. The universal input device uses a single line of linear CCD detectors. In the static mode, the single linear array is used as a point-and-click input device for bar code, provided that the optical length of the bar code is equal to or less than the length of the single linear array. In the linear mode, the bar code may exceed the optical length of the single linear array and the array then moved or scanned over the bar code.

Abstract: A write head magnetizes magnetic-ink characters for a MICR system. A magnetic assembly defines a first pole in a first location and an opposite pole in a second location. A first pole piece and a second pole piece, spaced by a gap therebetween, are provided, the first pole piece disposed adjacent the first location and adapted to conduct a magnetic field from the first pole to the gap. A selectably positionable member is disposed between the second location and the second pole piece, the positionable member being adapted to affect a magnetic field from the second location through the second pole piece to the gap, thereby facilitating selectable alteration of a magnetic field profile across the gap.

Abstract: A method for identifying the film type of a color input medium having picture and film border areas and which is scanned to produce an input image having a multiplicity of color separations having portions corresponding to the picture and border areas, the method including the steps of creating picture histograms of the picture area of the input image, defining a dark level and a bright level for each of the picture histograms and identifying a film type by comparing the dark and bright levels to predetermined thresholds defining predetermined characteristics of a given film type.

Abstract: A method for enhancing lofargram data images utilizing a rule base. The system takes a plurality of sequentially occurring vertical frequency strips of a lofargram data image and computes an estimate for the mean and variance of each vertical frequency strip. A linear transformation utilizing a rule base is then applied to each pixel of the vertical frequency strip to improve the visual clarity of the lofargram data image.

Abstract: Edge detecting masks each having a sensing directivity to a specific direction are arranged with each directivity direction shifted by a given angular interval from the neighboring masks. A sum-of-products computation is performed between weighting elements in each of the so-arranged masks and image data indicating the brightness or tint of respective pixels. A selecting device selects a mask which exhibits the maximum sum-of-products output value among others and also selects two or three more masks before and behind the selected mask according to the interpolation method employed. An estimate computing device performs interpolation using the angles and output values of the selected edge detecting masks to estimate a true edge direction.

Abstract: Facsimile image-scaling and -smoothing method and apparatus are described. They scale a received facsimile image along a given axis to increase the printed image resolution therealong, while smoothing the scaled image to reduce visually objectionable ticks. They do so by analyzing received image data with respect to adjacent received lines having disparate pixel counts and by filling in pixels therebetween in accordance with predefined fill rules that are designed also to smooth the output image to be printed. Various predefined rules are applied to produce intermediate results some or all of which are comprehended by the scaling/smoothing algorithm to produce the final smoothed, scaled output pixel image. Preferably such analyzing and filling in accordance with such rules are performed by a hybrid hardware and firmware subsystem within a receiving facsimile machine or facsimile-receiving printer.

Abstract: An optical imaging surface inspection system and method are provided. The system comprises an illumination source for illuminating an image pattern on a surface to be inspected, a video camera for detecting a portion of the light reflected from the surface and outputting an analog video signal, a dynamic thresholding circuit for converting the analog video signal into a digital representation of the image pattern, and means to compare a known accurate digital representation of the image pattern to the digital representation provided by the dynamic thresholding circuit. The method includes the steps of illuminating the surface, scanning the surface with the video camera, separating the video signal into a black level signal component and a white level signal component, detecting successive peak levels in the white level signal component and outputting a corresponding variable peak voltage. A variable threshold voltage is determined by selecting a percentage of the variable peak voltage.

Abstract: A document reader, such as a hand-fed check reader is described herein. The reader includes a transducer elastically supported relative to a circuit board. Components and circuitry to manipulate signals produced by (read head transducer), or applied to (write head transducer), the transducers are supported by the circuit board. The transducer is supported by a spring member fabricated from the circuit board material, where the board is mounted relative to a document path of travel such that the transducer is biased against the document. A drive wheel and mechanism is provided to move the document along the path and bias the document against the transducer in the vicinity of magnetic information on the document.

Abstract: A high speed screening technique is disclosed which can be used to enlarge an image, or when combined with a halftone screen, can be used for high speed screening of stored images. A microcompiler generates customized program code responsive to an input enlargement ratio. The customized program code then performs halftone screening on the input image for the specified desired enlargement ratio. In general, for each different enlargement ratio, a different customized program code is generated. Thus, a parameter of the screening process, the enlargement ratio, which is an input to the microcompiler, is not an input during the run time of the customized program code, thereby permitting the customized program code to run faster since an input parameter has been removed and built into the code itself.

Abstract: Character recognition method and system that identifies an input character as being a unique member of a defined character set. Specifically, a character bit-map of an input character is first generated. Thereafter, a character recognition procedure processes the character bit-map to generate a set of confidence measures one for each of the members of the character set. The confidence measures represent the degree of confidence that the input character corresponds to the members of the character set. A test is then made to determine if the confidence measure with the highest degree of confidence is acceptable. If there is an acceptable confidence measure, the member of the character set with the acceptable confidence measure is reported as the output character. If there is no acceptable confidence measure, a number of characters with the highest confidence measures are identified as candidates.

Abstract: The configuration of a path for motor vehicles is recognized on the basis of image data produced by a television camera or the like. Feature points contained in original image data of the path are determined, and a group of straight lines approximating the array of the feature points are also determined. From the group of straight lines, there are extracted straight lines which are effective to determine boundaries of the path. The extracted straight lines divided into a plurality of line segments by points of intersection between the straight lines. The line segments are then checked against the feature points of the original image data to determine whether the line segments correspond to the boundaries of the path. The original image data may be divided into a plurality of areas, and the above process may be carried out with respect to the image data in each of the areas.

Abstract: A data reading apparatus reads data of dots from a recording sheet on which data is recorded in the form of a mesh pattern, and decodes the read data into binary data. The data reading apparatus has an image sensor to read black/white levels of dots of the mesh pattern using the image sensor and to binarize the read levels by a hardware component or CPU processing according to software. An image can have a high recording density by minimizing a dot size.

Abstract: A hand-held scanner is provided for reading characters from a string of characters recorded on a substrate. The scanner may be held like a pen and moved, in contact with the substrate, along the string of characters. Movement of the scanner is sensed by a sensor. An optical system views a small area of the substrate and an optical detector detects the relative intensity of light reflected from each of several points in the area of view. To facilitate tracking, a portion of the string of characters adjacent to the area of view is clearly visible to the user. The scanner includes memory for storing codes representing individual characters. The scanner may include a computer, connected by a cable, or may be self-contained and battery operated.

Abstract: An optical character reader is characterized in that the reader determines accurately the range occupied by the major portion of each character in each character line. This is accomplished by dividing an image read from a document into character liens and counting the number of picture elements in series with each picture element forming each character on each character line in each direction of a plurality of predetermined directions. Each picture element is provided with a code data indicating the direction having the largest number of picture elements having a character. If the picture elements between two picture elements allotted with a specific code data for the direction along each character line do not have the specific data code, the original code data is replaced with the specific code data. The number of picture elements allotted with the specific code data for each character line are then counted and the position with the largest number of picture elements having a character is detected.

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: An image coding method extracts a contour of a bi-level image, segments the contour into a plurality of segments by fitting a predetermined generation curve on each of the segments, where the predetermined generation curve is described by control points which include end control points and direction control points, the end control points indicate ends of each segment and the direction control points indicate directions of each segment at the two ends of each segment, and codes the control points.

Abstract: An A pattern recognition subsystem responds to an A feature representation input to select A-category-representation and predict a B-category-representation and its associated B feature representation input. During learning trials, a predicted B-category-representation is compared to that obtained through a B pattern recognition subsystem. With mismatch, a vigilance parameter of the A-pattern-recognition subsystem is increased to cause reset of the first-category-representation selection. Inputs to the pattern recognition subsystems may be preprocessed to complement code the inputs.