Abstract: An optical pattern recognition method which can be applied to a joint transform correlator operative to effect correlation processing between a plurality of reference images and at least a single object image to produce correlation peaks corresponding to the respective reference images. The joint transform correlator is provided with a feedback system effective to regulate the intensity of the respective reference images according to the corresponding correlation peak value so as to noise-shape the correlation peaks to discriminate most-correlated reference images. The pattern recognition method is carried out in the correlator sequentially by first preparing a plurality of initial groups each containing a given number of individual reference images and carrying out the pattern recognition of the object image with respect to each of the initial groups to discriminate each of the most-associated reference images having a maximum correlation coefficient in each initial group.

Abstract: An image to be smoothed is represented by a matrix of pixels arranged in a plurality of adjacent scan lines. Each pixel under examination is changed to the value of a predetermined number of pixels in a 3.times.3 pixel neighborhood surrounding the pixel under examination. The image pixels are represented by a sequence of 16 bit digital words, each bit representing a corresponding pixel. The bits of three adjacent words on three adjacent scan lines are accumulated in 16 counters, respectively, with the words shifted so that the 9 bits of each of sixteen 3.times.3 neighborhoods are simultaneously accumulated in the sixteen counters. Each counter is constructed so that when a predetermined number of ONEs is accumulated from the 3.times.3 neighborhoods, the most significant counter stage goes to ONE and remains at ONE irrespective of further accumulation. One of the three words is the word to be smoothed.

Abstract: An occurrence of a predefined object in an image is recognized by receiving image data, convolving the image data with a set of predefined functions to analyze the image data into occurrences of predefined elementary features, and examining the occurrences for an occurrence of a predefined combination of the elementary features that is characteristic of the predefined object. Preferably the image data are convolved directly with a first predefined function to determine blob responses, and a second predefined function to determine ganglia responses indicating edges of objects. Then the ganglia responses are convolved with a third predefined function to determine simple responses indicating lines in the image, and the simple responses are combined with the ganglia responses to determine complex responses indicating terminated line segments in the image. A pointing finger, for example, is recognized from the combination of a blob response and a complex response.

Abstract: An image recognition apparatus includes a light source including a plurality of light emitting diodes having different wavelengths, for illuminating an object, a driver for selectively operating the light emitting diodes and adjusting the brightness thereof, an image pick-up device for photoelectrically transferring the light reflected from the object into an image signal, and an image decision device for comparing a reference image signal with the image signal obtained by the image pick-up device and then deciding whether an image corresponding to the obtained image signal is clearer than an image corresponding to the reference image signal, the image decision device making the driver sequentially operates the light emitting diodes and change the brightness thereof in such a manner that the image corresponding to the image signal is clearer than the image corresponding to the reference image signal.

Abstract: An image scanner including a camera portion which is provided with a lens, a CCD sensor and an optical glass component having a uniform thickness and refractive index greater than 1. The optical glass component is movable into and out of an optical path extending between an original and the CCD sensor. By the selective insertion of the optical glass component into the optical path, focussing and defocusing an image of the original is provided at the CCD sensor.

Abstract: A digital image preprocessing method for eliminating background patterns and scenes from digitized gray level document image data. The method accepts multi-gray level image data as input and produces a gray level output image containing a "white" background. Document image backgrounds are eliminated (or suppressed) by computing a separate (and typically different) threshold value for each gray level pixel of the original document image. Input image pixels whose gray level is greater (brighter) than the threshold value are assumed to be background picture data and are modified to be "white" pixels in the output image. The pixel threshold value is determined dynamically pixel-to-pixel based on gray level pixel data observed within the input image. Additionally, the pixel threshold can be automatically controlled to insure retention of written and/or printed information present in the input image.

Abstract: An edge correction circuit and method for enhancing image definition by reducing a conversion time at an region of a luminance signal in a digital image processor, said method including the steps of and the circuitry for: (a) generating a first control signal by detecting a pixel where a image luminance signal starts to change; (b) converting and outputting a pixel level state which is the same as the state before conversion of an edge region while said first control signal is being generated; (c) generating a second control signal by detecting a pixel prior to the luminance signal stabilization; and (d) outputting a pixel level state which is the same as the state after a complete conversion of the edge region which the second control signal generates.

Abstract: An apparatus is capable of reading a character or graphics image on a document at a plurality of reading linear densities. An illuminating device for illuminating the document, a focusing device for focusing the light beam reflected by the document, and a CCD reading sensor for converting the focused light beam into an electric image signal, are integrally assembled together to constitute a carriage device. This carriage device is shifted in a desirable shifting which is determined on the basis of the reading sensitivity of the CCD reading sensor and which corresponds to the linear reading density. The rotation speed of a driving device used for driving the carriage device is reduced to a desirable value by means of a transmission, the conversion ratio of which is determined on the basis of the least common multiple (LCM) of a plurality of reading linear densities. The driving pulses which should be supplied to the driving device can be easily calculated by use of the product of the "reference pulse " .phi.

Abstract: A method of editing image data produced from reconstructed tomographic data comprises the creation and editing of a bit map with bits corresponding to each data element of the image. The bit map permits faster processing of editing commands. The resultant image is displayed from a boolean combination of the edited bit map and the original data elements. Rapid altering of the boolean combination allows "toggling" of the edited image with an unedited image for clarity. The edited bit map may be used to edit a three dimensional reconstruction by substituting pad values for the image data according to the state of the bit map. The bit map may be altered to track certain editing processes such as connectivity or region growing.

Abstract: A lossless data compression circuit compares a new data string with a set of comparison data, and produces a sequence of codewords representing a sequence of successive, non-overlapping substrings of the new data string. A shift register stores and shifts the comparison data until all of the characters in the comparison data have been compared with the new data string. A composite reproduction length circuit finds the maximum length string within the set of comparison characters matching substrings of characters in the new data string beginning at each position in the new data string. The composite reproduction length circuit produces a multiplicity of data pairs, one for each position of the new data string. Each data pair comprises a maximum length value, corresponding to the maximum length matching comparison string found for the new data substring starting at the corresponding position, and a pointer value denoting where the maximum length matching comparison string is located in the comparison data.

Abstract: Image processing elements forming an image processor for automatic inspection of circuits. Each processing element for performing a thinning algorithm passes along a data bit and a flag bit to each succeeding processing element. The flag bits permit a conditional deletion of a data bit. Within each processing element a feedback bit is used to indicate when a previous bit has been deleted. The skeletonized image is reviewed for identification of circuit geometries and counting the number of separate blobs. A report is generated in response to the identified geometry and blob count.

Abstract: The image forming apparatus includes a receiver for successively receiving image data from an external device, a first memory for storing a basic font pattern, a second memory including a first memory area for storing a modified font pattern obtained by modifying the basic font pattern stored in the first memory, and a second memory area for storing bit image data, a generator for successively generating bit image data from the image data received by the receiver, using the basic font pattern stored in the first memory and the modified font pattern stored in the first memory area of the second memory, a first processor for expanding the bit image data generated by the generator into the second memory area of the second memory. The bit image data generated by the generator is expanded into the second memory area of the second memory. The second memory area, may be used up during the generated bit image expansion.

Abstract: A system and method for recognizing magnetic ink character records (MICR) by establishing a centerline between the first and last peaks of a scanned magnetic-electrical read out, which centerline is used to establish the location of the major peaks. Then the peak magnitudes and peak locations are compared with those same parameters in predetermined template characters and scored according to error deviation. Algorithmic circuitry will find the lowest scored template and recognize the scanned character as the character identified with the lowest scoring template.

Abstract: A color image processing apparatus wherein a color code is generated by a color code generating device in a state where a region to be assigned to chromatic color or achromatic color is changed according to the density gradient of color reading signals, and a document discrimination is executed by counting this color code for each pixel by a document discriminating device. A region to be assigned to achromatic color or chromatic color is changed in accordance with the kinds of images, and simultaneously, a different image processing is carried out according to the kind of an image.

Abstract: A method and system for recognizing an individual unknown character and a document in which, for black pels, one at a time, a vector is generated indicating what is seen from that pel in each of a plurality of directions in terms of white or black pels. That vector is compared to a table created from a large experience, generating a probability for each character in the character set that the vector from the pel on the unknown character is that character. As this process is repeated for other black pels in the unknown character, the probabilities associated with the various pels for each respective character are multiplied together to create a probability for the unknown character. An optional skeletonization of the character prior to recognition assists in the recognition by making the character only one pel width for a stroke, rather than a variable quantity.

Abstract: An image processing method generates a density concurrent matrix. An original image is shifted for generating a shifted image, with pixels in the shifted image corresponding to (conceptually overlapping) pixels in the original image. Bit strings are formed by combining a pixel density value of a pixel from the original image with a density value of a corresponding pixel from the shifted image. In an example, two 4-bit density values are combined into a single 8-bit string. Bit strings formed from pixel pairs are counted to give the number of occurrences of density pairs in the original and shifted image.

Abstract: A document processor for processing a document image which includes both a table formed by a plurality of ruled lines, and characters to generate vector data defining the ruled lines comprises an image input device for reading an image of the document, a display for displaying the document image and a processor for processing the document image. The processor corrects a skew of the input document image, detects a successive black pixel area included in the skew-corrected image, detects a character area based on a size of a rectangle which circumscribes the successive black pixel area, and converts the image to a second document image having characters deleted and including only the ruled lines.

Abstract: A flat panel display device includes a face plate made of a transparent material, a back plate positioned parallel to the face plate, and a wall member extending between the face plate and back plate to define an airtight housing. An anode is provided on a inner surface of the face plate, a fluorescent layer is provided in association with the anode, and a cathode is provided in association with an inner surface of the back plate. A plurality of struts, made of an electrically conductive screen printed powdery material, are tightly held between the back plate and the face plate, such that an electric charge accumulated between the anode and cathode is discharged by a leakage current flowing through the struts.

Abstract: A calibration system for electronically-generated images includes a calibration utility 22 for modifying tone scale and/or color reproduction so that a reproduced print 10 will faithfully match a monitor image 16. An input memory 19 provides a stored image for processing by application software 20 until an acceptable monitor image 16 is obtained. If a hard copy print 20 made from the processed image does not match the monitor image 16, a printer tone scale algorithm 28 and a printer color algorithm 30 in the calibration utility 22 are invoked to match the print to the monitor. An operator adjusts tone scale controls 32 and color controls 34, which provide values for calibrating a tone scale look-up table 49 and a color matrix 48 in a printer 14. Interactive feedback is provided by inversely modifying a tone scale look-up table 44 and a color matrix 46 used in processsing the display signal.

Abstract: A line sensor apparatus is adapted to read a character or pattern image on a document. In this apparatus, an illumination device for illuminating the document, a optical lens for focusing the light beam reflected by the document, a CCD line reading sensor for photoelectrically converting the light beam focused by the optical lens into an electric signal, and a holding member for holding the CCD line sensor and the optical lens are assembled as one body in such a manner as to constitute a carriage device. A uniformly radiated heat-radiating plate is in contact with the entire botttom surface of a support for the CCD line sensor, so that the heat generated by the CCD line sensor is radiated from the heat-radiating plate. Further, the CCD line sensor is kept in tight contact with the holding member, so that the heat generated by the CCD line sensor is also radiated from the holding member.