This compilation and all annotations are copyright © Jean Renard Ward, 1996, 1999, 2002, 2005, 2006
Copyright © Sun Jul 6 06:46:37 EDT 2008
Contact: jrward@alum.mit.edu

Mirrors of this page can be found at:

• http://users.erols.com/rwservices/pens/biblio90.html
• http://65.78.28.198:81/pens/biblio90.html
• http://rwservices.no-ip.info:81/pens/biblio90.html

      Main Reference Page      

• [Abramovitz89] (*A) Abramovitz, R.G., Anderson Jr., R.L., Busse, F.L., Celarier, S.A., Poublan, J-Y. H., and Shorthill, L.R. "Optical Character Reader", United States Patent 4,809,351, February 28, 1989, assigned to SABA Technologies, Incorporated, Beaverton, Oregon.

  • Hardware design of a hand-held scanner for optical (typed) character recognition

• [AccessDevelopment90] () Access Development Corporation "FormLink product literature", July, 1990. Address: 2000 Commonwealth Avenue, Auburndale, Massachusetts, 02166, 617-332-7200.

  • Forms package, stylus-keyboard electronic-paper user interface, suitable for adding handwriting recognition

• [ActiveBook89] (*p) Active Book Ltd., Cambridge England "Masterstroke: the notebook computer", London Times, September 7, 1989, p 34.

  • Portable touch-screen/handwriting product, with character recognition
  • Active Book: Dr. Herman Hauser, founder: Olivetti financing

• [Adolf89] (*p) Adolf, Jurine "An Expert System for the Selection of Input Devices", Proceedings of Human Factors Society 33rd Annual Meeting, October 16-20, 1989, Denver Colorado, pp 335-339.

  • Says on-line recognition inferior to almost all other user-interfaces for text input (comparision with Linus)

• [AdvancedProducts89] (*p) Advanced Products and Technologies "Voice Computer product announcement", IEEE Computer, September 1989, p 98.

  • Portable voice (!) recognition and output computer product

• [Agilis89a] (*p) Agilis Corporation "product information on the Agilis System (TM) of Handheld Workstations", June 12, 1989. 1101 San Antonio Road, Mountain View, California 94043-1008.

  • Portable touch-screen/handwriting product, but no character recognition or electronic ink user-interface: radio net, Unix/OS-2/DOS

• [Agilis89b] () BYTE "Agilis Hand-Held Workstations: Computing Power in the Field", BYTE Magazine, Augist 1989, pp 91-94.

• [Atari90] (*p) Atari Corporation "Portfolio hand-held personal computer", Sunnyvale, California, 94089-1302, 1990.

• [Autosig90] (*p) Autosig Systems Inc. "Sign/on product information", 1990, PO Box 16050, Irving Texas 75016, 214-258-8033.

  • Formerly Signify / Sign-on: signature verification product

• [BarkerBA89] (*p) Barker, Barbara A., Hernandez, Irene H., and Machart, Beverly H. "In-line dynamic editing of text or graphics", United States Patent 4,815,029, March 21, 1989, assigned to IBM Corporation.

  • User interface combining free-hand drawing, graphical editing, similar to "Polite" product system by Sheldon Borkin and John Prager, or Xerox Star and Apple MacIntosh with generic functions valid for all types of objects
  • On-line dynamic editing of text or graphics in computer document executing actions on user-selected attributes based on internal knowledge of object type rather than by explicit commands

• [BarkerBA90] (*) Barker, Barbara A.; Hernandez, Irene H.; Machart, Beverly H. "Method for Modifying Intermingled Text Object and Graphic object within an Object Set Individually or Correspondingly", United States Patent 4,974,194, November 27, 1880

  • User-interface for stylus/gesture editing of text, using mark-up gestures

• [BarkerPG89] () Barker, P.G. and Manji, K.A. "Pictorial dialogue methods", International Journal of Man-Machine Studies, Vol 31 No 3, pp 323-347, September 1989.

  • User-interface for electronic books: multi-media pointing a la hypertext with pictures

• [Baum89] () Baru, E. and Haussler, D. "What size net gives valid generalization", in Advances in Neural Information Processing Systems, D.S. Touretzky, editor, Morgan Kaufman, 1989.

  • Martin,GL90b

• [Beringer89] (*p) Beringer, Dennis B. and Bowman, Mary James "Operator Behavioral Biases using High-resolution Touch Input Devices", Proceedings of Human Factors Society 33rd Annual Meeting, October 16-20, 1989, Denver Colorado, pp 320-322.

  • Nothing study of visual parallax and targetting error in integrated tablet/display: 3/4mm error on their hardware at 17 degrees viewing angle: software offset only good for one user in one position, so they suggest letting the user just get used to visual parallax

• [Blatt89a] (*p) Blatt, Louis A. "Human Factors Support of Gesture Recognition", internal report June 16, 1989, Wang Laboratories, 1 Industrial Avenue, Lowell, Massachusetts 01851.

  • Internal Wang report on variations in chain code writing sequences for handwritten alphabetic characters

• [Blatt89] (*p) Blatt, Louis A. "Spreadsheet Gesture Set", internal report February 2, 1989, Wang Laboratories, 1 Industrial Avenue, Lowell, Massachusetts 01851.

  • Internal Wang report on what editing marks subjects used informally on spreadsheets

• [Blatt90] (*p) Blatt, Louis A., and Schell, Alan "Gesture Set Economics for Text and Spreadsheet Editors", unpublished manuscript, March 3, 1990, Wang Laboratories, 1 Industrial Avenue, Lowell, Massachusetts 01851.

  • Gesture recognition UI study on "most natural" gestures for text and spreadsheet editing, resulted in proofreader's marks

• [Borenstein90] (*) Borenstein, Nathaniel S. "Multimedia Applications Development with the Andrew Toolkit", Prentice-Hall, 1990, ISBN 0-13-036633-1

  • Mostly Dialog and other aspects of Andrew UI programmin

• [Bozinovic89] . Bozinovic, R.M. and Srihari, S.N. "Off-Line Cursive Script Word Recognition", IEEE Transactions on Pattern Analysis and Machine Intelligence", Vol 11 No 1, January 1989, pp 68-83.

  • Off-line (vs on line) cursive handwriting recognition. Abstract only.
  • Multi-level analysis of handwriting: points, contours, features, letters, and words
  • Reference-line finding on handwriting, letter segmentation based on local minima or low vertical profiles
  • Topological relationships of contours on handwriting for recognition
  • Lists of features used for cursive handwriting recognition
  • Chain encoding for off-line handwriting cursive script recognition
  • Statistical and syntactic context rules for cursive handwritten words
  • Testing done with two subjects told to do fully-connected script, and to maintain reasonable legibility
  • Said 77% accuracy on whole words with context was good test results (!)
  • Testing of off-line cursive script, but used data taken from tablet digitizer, so no scanning/pixel/gray problems
  • Well-known tenet of pattern recognition practice that for problems with large feature space dimensionality, the sizes of the training and testing sets should be roughly the same
  • Well-known tenet of pattern recognition practice that training and testing sets should not be the same to avoid overly optimistic results

• [BrownNR89] () Brown, Norman R. and Vosburgh, Ann Marie "Evaluating the Accuracy of a Large-Vocabulary Speech Recognition System", Proceedings of Human Factors Society 33rd Annual Meeting, October 16-20, 1989, Denver Colorado, pp 296-300.

  • Recognition performance: simple recognition error (how many wrong): adjusted recognition error (how many wrong after removing errors system "not expected" to get right) out of vocabulary words (also removing errors attributed to bad input: tablet or voice input errors?
  • Use of ontext can ADD errors and reduce recognition accuracy: "drag-along" errors where a "correction" due to context screws up subsequent inputs
  • User interface and recognition %: a (speech) recognition system need very good / easy correction mechanisms for user to correct recognition errors

• [Byrne89] () Byrne, Jani Gabriel "Competitive Evalution in Industry: Some Comments", Proceedings of Human Factors Society 33rd Annual Meeting, October 16-20, 1989, Denver Colorado, pp 423-425.

  • Notes on comparing competitive commercial user interfaces

• [CalComp90] (*p) CalComp "CalComp 2300 Series DrawingBoard Digitizers", CalComp Incorporated, 1990.

• [Canon89] (*) Canon "Ai Note IN-3000 product announcement", CanonVoice, February, 1989 (in Japanese).

  • Small, portable notebook-style handwriting input/recognition (PDA) product from Canon, 1980

• [Carpenter89] () Carpenter, Gail and Grossberg, Stephen "Hierarchical neural approach", Boston University press release, 1989.

• [ChengF89] (*p) Cheng, F., Hsu, W., and Chen, C. "Fuzzy Approach to Solve the Recognition Problem of Handwritten Chinese Characters", Pattern Recognition, Vol 22 No 2, pp 133-141, 1989.

  • Fuzzy set recognition applied to Chinese characters
  • Most troublesome problem in Chinese handwriting is great variation among hand-prints
  • Assumptions about statistical probability density in handwriting recognition are highly dubious
  • It is impractical to deal with variations in handwriting with probability densities
  • A similarity measure between strokes in Chinese character based on angle feature, not length
  • Stroke length not a reliable feature in Chinese characters
  • 96% recognition rate on 881 Chinese characters (160 variations each), once rejects are removed

• [ChowD90] (*) Chow, Doris and Kim, Joonki "Paper Like Interface for Educational Applications", Proceedings fo National Educational Computing Conference, June 20-22, 1989, Boston, Massachusetts, pp. 337..344

  • Part of the paper-like interface PLI projects at IBM: electrostatic/capacitive digitizer (Scriptel) over LCD display, exploration of gesture input. Teaching block-printing to children by direct comparision with prototype shapes. Statest that gesture recognition requires a different recognition strategy. Crossword-puzzle as teaching drill using handwriting recognition. Music editing demonstration modeled after Buxton, but not using same recognition technology.

• [CIS89] () CIS: "Geometry Ball product information", CIS Group, Postfach 10 01 80, D-4060 Viersen 1, West Germany, 1989.

  • Six-degree of freedom 3-D pointing mouse device.

• [Clark89] (*) Clark, D.M. "Pattern Recognition Method and Apparatus", United States Patent 4,805,225, February 14, 1989, assigned to The Research Foundation of the State University of New York, Albany, New York.

  • Hierarchical network recognizer, similar to neural nets, to determine the presence of a feature (but how are features weighted?)
  • Cites work on connectionist models for neural nets

• [ClarkAJ90] (*) Clark, Andrew J. "Do you really know who is using your system? - A Survey of Personal Authentication Techniques", Paper to be presented to the British Computer Society, Technology of Software Protection Specialist Group, February 1990

  • Review of authentication techniques: Something Known (password), Something Owned, Optically read cards, Smart cards, Password Generators, Personal Authenticators, Physical Charactgeristics -- Fingerprint, Retinal Scan, Voice Verification. Mentions *dynamic* signature verification as an "involuntary response", cites low hardware reliability of touch-screens: Dr. John Parks of the British Technology Group for signature verification products

• [ComputerS89] () Computer Shopper "Paper-like Interface -- IBM researchers are exploring a way to use computers in which people simply write on a flat surface", Computer Shopper, June 1989, page 169.

  • Press announcement of IBM's electronic ink handwriting recognition product

• [Danis89] () Danis, Catalina M. "Developing Successful Speakers for an Automatic Speech Recognition System", Proceedings of Human Factors Society 33rd Annual Meeting, October 16-20, 1989, Denver Colorado, pp 301-304.

  • Training users to adapt to a particular speech recognition:
  • Problems are too clear at end of words, too mushy at start of words, no word separation, speak too fast
  • Data base /testing: very important to try realistic conditions: realistic conditions include the fact that users adapt to the system

• [Datakech90] () DataKech, Incorporated "Scriptwriter product information", 15331 Manor Village Lane, Rockville, Maryland 20853.

• [Datapro89] () Datapro Research, Manufacturing Information Systems Data Collection report #450 "Handwriting Recognition Systems", January 1989.

  • Datapro report on Linus Write-Top and Data Entry Systems ScriptWriter handwriting recognition products.

• [DavisE89] () Davis, Elliot "Pattern Recognition Error Reduction System", United States Patent 4,802,231, January 31, 1989.

  • Patent abstract only
  • Error reduction by matching patterns against additional templates of patterns otherwise mis-matched.

• [Denker89] () Denker, J.S., Gardner, W.R., Graf, H.P, Henderson, D., Howard, R.E., Hubbard, W., Jackel, L.D., Baird, H.S., and Guyon I. "Neural network recognizer for Hand-written zip code digits", In Advances in neural information processing systems I, D.S. Touretzky, editor. Morgan Kaufman, 1989.

  • Martin,GL90

• [DFM89a] () DFM Business Systems Inc. "DBF unveils 3.5-pound hand-held PC", Travelite 286 Product announcement, Info World, September 11, 1989, vol 11 no 37.

• [DFM89] () DFM Business Systems Inc. "Hand-held CPU for health care uses infrared-link", Travelite 286 Product announcement, Network World, Oct 30, 1989, v6 no 43.

• [Dragon90] () Dragon Systems "DragonDictate product announcement", Newton, Massachusetts, March 1990.

  • Commercial voice/speech recognition product

• [Drumm89] () Drumm, Donald E. "Electronic Light Pointer for a Projection Monitor", United States Patent 4,808,980, February 28, 1989, assigned to Wang Laboratories, Lowell, Massachusetts.

  • Patent on wall-mounted light pen for projection TVs, displays

• [Dynabook89] (*p) Dynabook Technologies "product literature and press release", October 1989. PC/Computing, October 1989.

  • Rumored portable touchscreen product: handwriting recognition?

• [Evergreen90] () Evergreen Technologies "Gestures product announcement", Jeffrey Siegel, Fox Chapel Center, Suite 401, 19751 Frederick Road, Germantown, Maryland 20874, personal correspondence, March 12, 1990.

  • Handwriting (adaptive?) character recognition product, gesture interfaces?

• [Exos90] () EXOS, Inc. "Dexterous Hand Master exoskeleton product information", August 1990. 8 Blanchard Road, Burlington, MA 01803, 617-229-2075.

  • Data-glove, mechanical version, n-dimensional pointing device including Polhemus 3-D digitizer

• [Faggin89] () Faggin, Federico, and Lynch, Gary S. "Brain Learning and Recognition Emulation Circuitry and Method of Recognizing Events", United States Patent 4,802,103, January 31, 1989, assigned to Synaptics, Incorporated, San Jose, California.

  • Neural net patent
  • Successive stages of recognition threshholds set for neural nets learning
  • Includes neural net feature to resolve mis-recognized patterns confused with a pattern already learned
  • Press announcement and description of IBM prototype signature verification system

• [FederalCW89] () FCW "Smart signatures at FED: Net Prototype uses Biometrics, Smart Cards", Federal Computer Week, May 22, 1989, page1.

• [Fellows89] () Fellows, Gerald, L. and Freivalds, Andris "The use of force sensing resistors in ergonomic tool design", Proceedings of Human Factors Society 33rd Annual Meeting, October 16-20, 1989, Denver Colorado, pp 713-717.

  • Pressure/force sensors for hand-tools (drills, hammers, etc.): also stylus?

• [Francik89] () Francik, Ellen and Akagi, Kenichi "Designing a computer pencil and tablet for handwriting", Proceedings of Human Factors Society 33rd Annual Meeting, October 16-20, 1989, Denver Colorado, pp 445-449.

  • Design of WANG Freestyle tablet and stylus: taper, length, comparison of human factors of tablets. Gives samples of types of writing for tablet and stylus design. Freestyle stylus was engineered to look and feel like a pencil, with an "eraser" button (colored pink) on the rear end of the stylus. Tablet digitizer itself was electromagnetic, designed by Hitachi.

• [Fujiwara89] () Fujiwara, A., Takenouchi, M., and Shimizu, M. "Character Recognition Apparatus", United States Patent 4,813,078, March 14, 1989, assigned to Matsushita Electric Industrial Company, Limited, Kadoma, Japan.

  • Optical character recognition for typed Kana using direction codes / chain codes of outer perimeter to compare best matches of lengths of segments

• [Furukoori89] () Furukoori, H. "Image Contour Detecting Apparatus", United States Patent 4,817,173, March 28, 1989, assigned to Kabushiki Kaisha Meidensha, Tokyo, Japan.

  • Patent abstract
  • Curve tracing

• [Gaver89] () Gaver, W.W. "The SonicFinder: An interface that uses auditory icons", Human-Computer Interaction, Vol 4 No 1, 1989, pp 67-94.

  • Cited in Nielsen90

• [Geiser89] () Geiser, Georg "Handprinted Data Entry with a Touch-Sensitive Numeric Keypad", Proceedings of Human Factors Society 33rd Annual Meeting, October 16-20, 1989, Denver Colorado, pp 305-308.

  • Handwriting character recognition using a calculator's 5x5 (3x3 plus spaces between) keypad (similar to some Japanese calculator products): instead of double keying for alphabetics on a small calculator. Claims "very low" error rate: but unknown, does not say what the rate was. Shows number of variants/variations considered is NOT combinatorial.

• [Go90] () Go Corporation, Foster City, California "Go Corp. Gets IBM's Backing For 'Tablet PC'", PC Week, April 9, 1990, page 1.

  • Go Corp. principals: Robert Carr (Aston-Tate), Jerry Kaplan (Lotus), Daniel Lewin (Next).
  • Portable handwriting recognition product

• [Gonzalez89] () Gonzalez, Rafael C. and Jerrera, Juan A. "Apparatus for Reading a License Plate", United States Patent 4,817,166, March 28, 1989, assigned to Perceptics Corporation, Knoxville, Tennessee.

  • Features on license plates are edges, contours, height, width, topline and baseline, structural analysis, templates once the state (and therefor font) are known.

• [Goolsbey89] () Goolsbey, Michael A., Jones, Eugene R., Micka, William F., and Robinson, Thomas, S. "Locating Individual Images in a Field for Recognition or the Like", United States Patent 4,817,186, March 28, 1989, assigned to International Business Machines, Armonk, New York.

  • Hardware to segment separate typed (handwritten?) characters for OCR recognition

• [Graphonomics89] () International Graphonomics Society "Membership directory", April 1989, c/o Department of Experimental Psychology, University of Nijmegen, PO Box 9104, 6500 HE Nijmegen, The Netherlands.

  • Index of researchers in handwriting analysis, signature verification, and character recognition

• [Greenstein89] () Greenstetin, Joel S. and Baijal, Anish "An Investigation of Techniques for Occasional Numeric Data Entry", Proceedings of Human Factors Society 33rd Annual Meeting, October 16-20, 1989, Denver Colorado, pp 310-314.

  • User-interface review of numeric keypad with mouse, multiple menus, sliders, clickers, simultaneous menus.
  • Reference to Quest Automation in 1970's as earlier product

• [GridPad89] () Computergram International "Grid's GridPad handheld takes handprint data entry", Computergram International, October 2, 1989, no 1275.

  • BYTE magazine gives award to GridPad product: compare 1985 award to PenPad.

• [GridPad89a] (*p) Grid Systems "Getting Started with GRiDForm", Grid Systems Corporation, September 1989

• [GridPad89b] (*p) Grid Systems "GRiDPAD Program Development Manual", Grid Systems Corporation, September 1989

• [GridPad90a] () BYTE "The BYTE awards (important products and technologies)", Byte magazine, January 1990, vol 15 no1, p. 285. -- GRiD System's GRiDPad, Lotus Magellan, Zenith MinisPort, Poqet PC, etc.

• [GridPad90] () Grid Corporation "GridPAD PadBase+ Training Course", course description, April 11, 1990, provided by Rober Artusy, R2Z Incorporated, 39120 Argonaut Way, Suite 319, Fremont, California 94538.

  • GridPAD user-interface with stylus, training
  • Correspondence includes list of features to be used for punctuation marks, handwriting prototype base forms and shapes

• [Grossberg89] () Grossberg, Steven and Mingolla, Ennio "Neural Networks for Machine Vision", United States Patent 4,803,736, February 7, 1989, assigned to The Trustees of Boston University, Boston, Massachusetts.

  • Patent abstract only
  • Boundary/edge/curve contour, feature contour, and object recognition in machine vision

• [Guo89] () Guo, Z. and Hall, R.W. "Parallel Thinning with Two-Subiteration Algorithms", Communications of the A.C.M., Vol 32 No 3, March, 1989, pp 359-373.

  • Skeletonization/line-thinning using four compass directions, which preserves line connectedness

• [HallRW89] () Hall, R.W. "Fast Parallel Thinning Algorithms: Parallel Speed and Connectivity Preservation", Communications of the A.C.M., Vol 32 No 1, January 1989, pp 123-131.

  • Robert M Harlick is editor for many CA.C.M. OCR line-thinning articles
  • OCR handwriting line thinning algorithm which does NOT eliminate diagonal lines

• [Harnad89] () Harnad, S. "The symbol grounding problem", CNLS Conference on Emergent Computation, Los Alamos, May 1989.

  • Martin,GL90

• [Hayashi89] () Hayashi, Y., Sakata, M., Nakao, T., Ohno, T. and Ohhashi, S. "Alphanumeric character recognition using a connectionist model with the pocket algorithm", (abstract only), Proceedings of International Joint Conference on Neural Networks, Washington, D.C., June 18-22, 1989, Vol II p. 606.

  • OCR for a single font using neural networks

• [Horowitz89] () Horowitz, Steven L. "Method and Apparatus for Generating Size and Orientation Invariant Shape Features", United States Patent 4,803,230, January 31, 1989, assigned to GTX Corporation, Phoenix, Arizona.

  • Patent abstract
  • Character recognition, OCR, statistical decision tree.
  • OCR using features of six bounding rectangles at different angles, translated to direction (chain?) codes
  • OCR using Fourier transforms, autocorrelations, moments, centroid of largest hole and center of character

• [Hsieh90] () Hsieh, A.-J., Kunk, S.-J., Shiau, S.-L., Kao, M.-C., and Chen, J.-W. "An Experimental System for Stroke-number Free On-line Chinese Character Recognition", in Frontiers in Handwriting Recognition, C.Y. Suen, editor, Proceedings of the Conference on Frontiers in Handwriting Recognition, Concordia University, Montreal Canada, April 2-3, 1990, pp 73-86.

  • Taiwanese efforts on both OCR and on-line Chinese character recognition: 5401 characters x 23 people x 10 samples each OCR, 5401 x 23 people (mixed) on-line data base collection.
  • Later came by to visit Wang: part of five-year project, 9 people OCR, 5 on-line, first year is "find out everything there is to know"

• [Hubby89] (*) Hubby, Laurence M. "Flat Panel Display with Integrated Digitzing Tablet", United States Patent 4,873,398, October 10, 1989

  • Optical digitizing tablet integrated with a flat (flat front surface CRT?) display. Light is emitted by the stylus, projected back through the same optical path used for projecting the image on the front surface (?). Cites Scriptel.

• [Hudson90] () Hudson, Henry T. and Newell, G. "Integrating gesture and snapping into a user interface toolkit", in in UIST'90 (1990), ACM, pp 112-122

• [Hull90] () Hull, Jonathan J., Commike, A., and Ho, T.-K. "Multiple Algorithms for Handwritten Character Recognition", in Frontiers in Handwriting Recognition, C.Y. Suen, editor, Proceedings of the Conference on Frontiers in Handwriting Recognition, Concordia University, Montreal Canada, April 2-3, 1990, pp 117-130.

  • OCR for handwritten Zip codes, address blocks
  • Description of IBM signature verification system at Federal Reserve

• [IBM89a] () Federal Computer Week "Smart Signatures at Fed", Federal Computer Week, May 22, 1989, Page 1.

• [IBM89b] () IBM Corp. Research Division, press release "IBM Writing Interface May Replace Typing", spokesperson Gerald Present, September 1989, 914-945-3884.

  • Description of IBM's "paperlike interface" user interface

• [IMCS90] () International Machine Control Systems "MousePen product literature", January 1990. 1332 Vendels Circle, Paso Robles, California 93446. 1-805-239-8976.

  • Stylus-shaped mouse with mechanical trackball: substitute for a tablet?

• [Impedovo90] () Impedovo, S., Castellano, M., Pirlo, G., and Dimauro, G. "An Off-line Writer Identification System Based on a Syntactic Approach", in Frontiers in Handwriting Recognition, C.Y. Suen, editor, Proceedings of the Conference on Frontiers in Handwriting Recognition, Concordia University, Montreal Canada, April 2-3, 1990, pp 53-64.

  • Like signature verification, but using regular hand printing (a nothing paper, looks like they had to change horses from adaptive handwriting recognition in mid-stream)

• [IndustrialEE89] () Industrial Electronic Engineers, Inc. "Pixie Graphic LCD Switch keyboard product literature", 1989, 7740 Lemona Ave, Van Nuys, CA 91409, 818-787-0311.

  • User-interface item: keys with individual graphic LCD displays in the key top for relabeling the keys

• [IshikawaY90] () Ishikawa, Y., and Kakajima, K. "A Real Time Connected Word Recognition System", 10th International Conference on Pattern Recognition, Atlantic City, New Jersey, June 16-21, 1990.

  • Note on Chinese character recognition product internal to Wang Laboratories, but disclosed publicly.

• [ITRI89] () ITRI "Chinese Character Recognition", Wang Laboratories internal memorandum on discussions with Industrial Technology Research Institute (ITRI) in Taiwan, June 20, 1989.

• [Jackel89] () Jackel, L.D. "VLSI Technology and Neural Network Chips", Tutorial Notes No 8, International Joint Conference on Neural Networks, Washington D.C., June 18-22, 1989.

  • Contains examples of neural nets for handwritten numerals recognition in O.C.R.

• [JohnsonJ89] () Johnson, J., Roberts, T.L., Verplank, W., Smith, D.C., Irby, C.H., Beard, M., and Mackey, K. "The Xerox Star: A Restrospective", IEEE Computer, Vol 22 No 9, September, 1989, pp 11-28.

  • User-interface description of the development of the Xerox Star desktop paradigm, now called ViewPoint (originally on Alto computer)
  • Xerox Star original desktop direct-manipulation user-interface intended only for casual users
  • Desktop metaphor for user interface is intended for office applications an publishing, not software developments (tools based)
  • Direct-manipulation desktop user interface (no processing on files per se) is orthogonal to tools-based user interface (UNIX/SunView Windows: any file can be operated on by any tool, even if not sensical)
  • Clean user-interface lies more in Xerox Star details, not in features
  • Mouse pointing input must be handled at low-level by system to prevent jerkiness, jumps, dead time as the system degrades: these (mouse or tablet or whatever) are eye-hand co-ordination devices
  • One-button mouse (or tablet stylus) vs. two-button: fewer button confusion errors, but more increased selection errors
  • Overlapped-windows cause users to spend time adjusting windows so they don't overlap: just use tiled windows to begin with (if the display is 17" large)
  • Xerox Star: cites user-interface work by many others using graphical/sketch input, Xerox View Point

• [JohnstonJP89] () Johnston, J.P. and Russell, D.L. "Information Display and Entry Device", United States Patent 4,814,760, March 21, 1989, assigned to Wang Laboratories, Lowell, Massachusetts.

  • Patent on integrated tablet and display for a dynabook/electronic-ink?

• [Jones89] () Jones, D., Hapeshi, K., and Frankish, C. "Design guidelines for speech recognition interfaces", Applied Ergonomics, Vol 20 No 1, March 1989, pp 47-52.

  • Syntax (context) has profound effects on speech variability, so design user-interface to minimize effects
  • Very important to collect samples (speech) in context for training: phonetic, semantic, and personal stress context affect variability
  • Issues of user-interface design for task-specific, limited speech recognition

• [Jourjine89] () Jourjine, A.N. "Microprocessor Assemblies Forming Adaptive Neural Networks", United States Patent 4,809,193, February 28, 1989.

  • Cellular automaton for neural nets.

• [Jourjine90] () Jourjine, Alexander "Variable capacity transmission medium", United States Patent 4,967,369

• [Katsurada89] () Katsurada, Morihiro "Method of a System for Analyzing Characters", United States Patent 4,811,412, March 7, 1989, assigned to Shapr Kabushiki Kaisha, Osaka, Japan.

  • Segment/parse character from right and left, see if they match, if they don't, try templates

• [Katt90] (*) Katt, Spencer F. (pseudonym) "Microsoft ueber alles?", PC Week, April 16, 1990, page 182

  • "Rumor Central": Microsoft conference in Canada, microsoft starting development of Windows-based operating system for handwriting recognition, to compete with GRiD and Go Corporation, estimated release is 12 to 18 months

• [Kawakami90] (*) Kawakami, Yasushi; Obata, Fukue; Makihara, Kayoko; Shiro, Yamada, Morimoto, Yoshinari et al "Data processing device with screen separated into line format area and ata display area", United State Patent 4,897,804, January 30, 1990

  • Text-input recognition area on tablet, data can be moved into regular display area

• [KellyW90] (*) Kelly, William "Sparking Stylus for Acoustic Digitizer", United States Patent 4,891,474, January 2, 1990

  • SAC Science Accessories Corporation sparking stylus for acoustic/sonic tablet digitizer

• [Kensington89] () Kensington Microware, Turbo "Mouse trackball product literature", Kensingon Microware, 1989. 251 Park Avenue South, New York, New York 10010.

  • Mouse/digitizer trackball replacement products

• [Khotanzad90] () Khotanzad, Alireza and Hong, Yaw Hua "Rotation invariant image recognition using features selected via a systematic method", Pattern Recognition, Vol 23 No 10, October 1990, pp 1089-1102.

  • Character recognition using rotation-invariant features - Zernike moments (what about "E" and "W"?)
  • Rotation-invariant OCR: 84% on numeric handprinted characters, 16,550 samples, 99% on "clean" alphabetics
  • Numeric handprinting OCR samples from Recognition Equipment, Inc.
  • Claims 84% correct recognition rate is very effective

• [KimJ88] (*) Kim, Joonki "On-line Gesture Recognition by Feature Analysis", Proceedings of Vision Interface '88, Jun 6-10, Edmonton Convention Center, pp.51..55

  • Gesture / direct manipulation interface to a spreadsheet program (Lotus 123), part of PLI paper-link interface projects at IBM. First stage of recognizer uses chain codes, second stage is custom-programmed for the gesture shapes.

• [Kinnucan89] () Kinnucan, P. "The Versatility of Graphics Tablets", Computer Graphics Review, Vol 4 No 7, July 1989, pp 44-52.

  • Simple-minded review article on digitizer technologies

• [Knight89] () Knight, Leland, W. and Retter, Dale "Datahand: Design, Potential Performance, and Improvements in the Computer Keyboard and Mouse", Proceedings of Human Factors Society 33rd Annual Meeting, October 16-20, 1989, Denver Colorado, pp 450-454.

  • Two handed mouse/keyboard user-interface combination, using the "Datahand" product

• [Koezuka89] () Koezuka, T., Tsukahara, H., Nakashima, M. "Pattern Matching Method and Apparatus", United States Patent 4,805,224, February 14, 1989, assigned to Fujitsu, Limited, Kawasaki, Japan.

  • Video pattern matching recognition using two-dimensional extracted image from frame buffer

• [Kreifeldt89] () Kreifeldt, J.G., Levine, S.L., and Iyengar, C. "Reduced Keyboard Designs using Disambiguation", Proceedings of Human Factors Society 33rd Annual Meeting, October 16-20, 1989, Denver Colorado, pp 441-444.

  • Stenotype-type user interface keyboard design: one key per finger, multiple characters per key

• [Krzyzak90] () Krzyzak, A., Dai, W., and Suen, C.Y. "Unconstrained Handwritten Character Classification Using Modified Backpropagation Model", in Frontiers in Handwriting Recognition, C.Y. Suen, editor, Proceedings of the Conference on Frontiers in Handwriting Recognition, Concordia University, Montreal Canada, April 2-3, 1990, pp 155-166.

• [Kuan89] () Kuan, C-C.L. and Srihari, S.N. "Analysis of handwritten text images using run-length based algorithms", IEEE Transactions on Pattern Analysis and Machine Recognition, to appear, 1989.

  • Taxt90

• [Kulp90] () Kulp, James and John "telephone conversation", April 1990. SuperScript Incorporated, Cambridge, Massachusetts, 617-876-0787.

  • Working on a portable handwriting recognition product (with forms)
  • John Kulp says they are a Symbolics spin-off: CD-ROM portable box for manuals.
  • Did their own elastic matching: not super happy with it (using lots of context helps)

• [Kundu89] () Kundu, Amlan, He, Yang, and Bahl, Paramvir "Recognition of Handwritten Words: First and Second Order Hidden Markov Model Based Approach", Pattern Recognition, vol. 22, no. 3, p. 283, 1989

  • Cited by Marlin Eller, Microsoft Pen Computing group

• [Kurosawa89] (*p) Kurosawa, Y., European Patent 335696 "Pattern recognition apparatus with two-part reference storage", assigned to Toshiba.

  • Two-level OCR recognition using contours for one level, then detailed features for final resolution
  • Performs choice of matching pattern from among number of candidate standards preselected using partial characteristic data.
  • Describes tablet template, side-menu touch buttons for tablet (user-interface)
  • Kurta's cordless stylus/pen for a digitizing tablet

• [Kurta89] (*p) Kurta "Kurta IS/ONE, IS/ADB, and IS/TWO product description", Kurta Corporation, 3007 East Chambers Road, Phoenix, Arizona 85040

  • shows static templates GUI for picking commands in various applications

• [Kuzunuki89] (*) Kuzunuki, Soshiro, Shojima, Hiroshi, Miura, Masaki, Mori, Junko, and Mifune, Toshimi "Real Time Handwritten Character Input System", United States Patent 4,860,372, August 22, 1989, assigned to Hitachi, Limited, Tokyo, Japan.

  • User interface on keyboard emulation with handwriting input: Ward85 was cited as prior art (check US Patent category 382,13).
  • Handwriting user interface: use a Times-square circular buffer for writing in, because written character are bigger than typed font

• [LeCun90] () Le Cun, Y. et al "Constrained Neural Network for Unconstrained Handwritten Digit Recognition", in Frontiers in Handwriting Recognition, C.Y. Suen, editor, Proceedings of the Conference on Frontiers in Handwriting Recognition, Concordia University, Montreal Canada, April 2-3, 1990, pp 145-154.

  • In his notes, claimed neural nets can learn feature on their own: actually, his net structure, pixel resolution, etc. determined the class of features.
  • OCR of handwritten zip codes: segmentation, "bad" samples already done.
  • Other authors: Boxer, B., Denker, J.S., Henderson, D., Howard, R.E., Hubbard, W., Jackel, L.D., and Baird, H.S.

• [LeeS89] () Lee, Seongwhan, and Kim, Jin H. "Unconstrained Seal Imprint Verification using Attributed Stroke Graph Matching", Pattern Recognition, Vol 22 No 6, 1989, pp 653-664.

  • Like signature verification, but for the stamps/seals used in oriental/Chinese countries
  • Spatial distributions of stroke skeletons: recognition independent of size/scale/magnification

• [Legault90] () Legault, R., Suen, C.Y., and Nadal, C. "Classification of Confusing Handwritten Numerals by Human Subjects", in Frontiers in Handwriting Recognition, C.Y. Suen, editor, Proceedings of the Conference on Frontiers in Handwriting Recognition, Concordia University, Montreal Canada, April 2-3, 1990, pp 181-194.

  • Similar to Shillman functional attribute experiments, using old zip-code samples
  • Contains some hypothesis on distinguishing features for handwritten numerals and human recognition

• [Li89a] (*p) Li, H.F., Pao, Derek, and Jayakumar, R. "Improvements and Systolic Implementation of the Hough Transformation for Straight Line Segments", Pattern Recognition, Vol 22 No 6, 1989, pp 697-706.

  • Add contiguity check to Hough transform to find straight line features in a binary image: handles overlapping lines, short lines.

• [Li89b] (*p) Li, H.F., Jayakumar, R., and Youssef, M. "Parallel algorithms for recognizing handwritten characters using shape features", Pattern Recognition, Vol 22 No 6, 1989, pp 641-652.

  • Static/OCR handwriting recognition: features are local shape, start and end points, edge types, and join relations in contours (digits only!)
  • OCR for handwritten digits: 120 samples per digit
  • Handwriting OCR: twelve relations for contours: compare to combinatorial explosion in Ward88? chain codes via systolic array

• [Lien89] (*a) Lien, Yih-Liang L. "Apparatus and Method for Vectorization of Incoming Scanned Image Data", United States Patent 4,817,187, March 28, 1989, assigned to GTX Corporation, Phoenix, Arizona.

  • Patent abstract only.
  • Curve-tracing, curve-follower, scanner algorithm for detecting lines in OCR or image analysis.

• [Litvin89] (*p) Litvin, Y. and Kuklinski, T. "Recognitive Sciences Provides Cost Effective OCR Solutions for OEM's", company profile, Recognitive Sciences, 323 Andover Street, Wilmington, Massachusetts 01887, 1989.

  • Report on OCR work by Kuklinski and Litvin at Recognitive Sciences: see also RecognitiveSciences references.

• [Logitech89] (*) Logitech Corporation "LogiMouse C7 Technical Reference Manual", Logitech Corporation, 1989.

  • Electronic digitizer tablet sold as a mouse replacement: emulates Microsoft Mouse data format, other tablets

• [LuS90] (*) Lu, S.W. and Suen, C.Y. "Heirarchical Attributed Graph Representation and Recognition of Handwritten Chinese Characters", in Frontiers in Handwriting Recognition, C.Y. Suen, editor, Proceedings of the Conference on Frontiers in Handwriting Recognition, Concordia University, Montreal Canada, April 2-3, 1990, pp 87-100.

  • Chinese/Kanji OCR. Abstract only

• [Mack89] (*) Mack, Robert, and Lang, Kathy "A Benchmark Comparison of Mouse and Touch Interface Techniques for an Intelligent Workstation Windowing Environment", Proceedings of Human Factors Society 33rd Annual Meeting, October 16-20, 1989, Denver Colorado, pp 325-329.

  • Stylus faster than keyboard for pointing (this is news?)

• [Mahach89] (*p) Mahach, Karen Renee "A Comparison of Computer Input Devices", Linus Pen, Mouse, Cursor Keys, and Keyboard", Proceedings of Human Factors Society 33rd Annual Meeting, October 16-20, 1989, Denver Colorado, pp 330-334..

  • Linus electronic ink tablet/display had low accuracy: typing faster than writing: points out that a handwriting user-interface is not really just handwriting, but a lot of menu-picking for data entry, which needs to be studied
  • Looks like a good researcher to contact on electronic ink studies

• [Marshall89] () Marshall, William W. and McWhortor, William F. "Method and Apparatus for Pattern Recognition", United States Patent 4,817,176, March 28, 1989, assigned to William F. McWhortor, Fountain Hills, Arizona.

  • Patent abstract
  • Character recognition from OCR using Fourier transforms, returning confidence percentage goodness based on second-best match.

• [MartinGL89] (*) Martin, Gale L. "The utility of speech input in user-computer interfaces", International Journal of Man-Machine Studies, Vol 30 No 4, April 1989, pp 355-376.

  • Points out that user-interface questions for speech recognition are hard to answer, because so many interaction modalities are involved
  • Points out indirectly (for Votan speech recognition) that the "perfect" recognition of 100% went down over time after training.
  • Grand review of speech recognition in user-interfaces
  • Speech recognition is faster than typing, and it has the advantage of being an additional channel of input besides the hands (haptic?)
  • For a simplified set of commands, speech recognition and mouse clicks equally efficient
  • Report by group at MCC Microelectronics and Computer Technology Corporation on their handwriting/speech recognition system (MAGIC)
  • Cites a large number of studies that show speech recognition is not as good as keyboard input (!!), although this paper says speech is better

• [MartinGL90a] (*) Martin, Gale L., Leow, Wee Kheng, and Pittman, James A. "Function Complexity Effects on Backpropagation Learning", MCC Technical Report ACT-HI-062-90, Microelectronics and Computer Technology Corporation, Human Interface Laboratory, 3500 West Balcones Center Drive, Austin, Texas 78758, January 1990.

  • Neural network report on on-line handwriting recognition
  • Data collection of 8,600 upper-case letters from (from NCR for digits) 110 people on an integrated tablet/display, 200 points/second 1000 points/inch
  • Training sets of 500,2500,6261 sample from different people
  • States one needs tens of thousands of samples of handwriting data base to train a system, more than an adaptable neural net
  • Neural nets need to be designed to detect human features, otherwise not much good

• [MartinGL90b] (*) Martin, Gale L., and Pittman, James A. "Recognizing Hand-Printed Letters and Digits Using Backpropagation Learning", MCC Non-confidential report, to appear in Advances in Neural Information Processings, D.S. Touretzky, editor, Morgan Kaufman, 1990.

  • Cites NCR corporation data-base of hand-printed on-line character shape collection
  • 8,600 on-line handwriting samples from 110 people, 200 points/second at 1000 points/inch: images converted to blurred 15x24 greyscale image for static character recognition
  • Training set of 500 to 6,300 samples, 18 people, separate from test data: human handwriting recognition error rate was 3.4% (compare Neisser60)
  • Gives table of exact trade-offs between reject rate and error rate on handwriting recognition, for digits and letters separately
  • Estimate of number of training samples needed for adaptive recognizer / neural network
  • Credits at MCC/NCR to Dave Rumelhart, Joyce Conner, Janet Kilgore, and Kay Bauer for collection

• [MCC90] () MCC Microelectronics and Computer Technology Corporation "Interactive Tablets", description of work in progress, CHI '90 Conference on Human Factors in Computing Systems, Seattle Washington, April 2-3, 1990. Available from Tom Parish, MCC, PO Box 200195, Austin Texas 78720, 512-343-0978.

  • General description of neural net character recognition, user-interface for handwriting at MCC
  • MCC staff on handwriting/gesture user interface: Tom Parish (general), Jim Pittman and David Chapman (hardware), Gale Martin (recognition), Kent Wittenburg (interpretation), Bill Gooch (Prototypes)
  • Working on mathematical expression user-interface stroke segmentation

• [Meeks90] (*p) Meeks, M.L., and Kuklinski, T.T. "Measurement of Dynamic Digitizer Performance", to be published in Computer Processing of Handwriting, Leedham and Plamandon, editors, World Scientific, Singapore, 1990.

  • Report on tablet characteristics and instrumentation for measuring handwriting
  • Describes test bed for calibrating tablet digitizers with an X/Y table to correct for *fixed* positional errors -- not tilt or velocity related accuracy problems?

• [Microslate89] () Microslate "press release", Microslate, 1989.

  • Rumored touchscreen portable product: handwriting recognition?

• [Microsoft90] (*p) Microsoft Corporation, announcement of "Windows/H", product development, PC Week "Spencer F. Katt" column, April 16 1990., p. 182

  • Microsoft (Lloyd Frink and Marlin Eller) on Windows/H PenWindows user-interface for handwriting recognition: using "Anfa" or "Infa" company

• [Microtouch90] () (no author) "ScreenWrite Touch Screen Performance Specifications", Available from Microtouch Corporation, Microtouch.com

  • Digitizer touch-screen, show calibration tablet for adjusting for regional position errors

• [MingJ90] () Ming, Ji "The Statistical Information Formulation for Noisy Speech Recognition", 10th International Conference on Pattern Recognition, Atlantic City, New Jersey, June 16-21, 1990.

• [Mira89] (*p) Mira Imaging, Inc. "Hyperspace (tm) product literature", 1989. 969 Logan Avenue, Salt Lake City, Utah, 84105.

  • 3-D digitizer software: works with Science Accessories, IsoTrak (Polhemus), Cyberware laser scanner, Topcon.

• [Momenta89] () Momenta Inc "press release", Milpitas California, October 19, 1989.

  • Rumored handwriting recognition portable product

• [Momenta90] (*p) Display Momenta Corporation "What's New in Laptops and Pen Computing", Bill Lempesis of Dataquest, Flat Panel Display News, Vol 1 No 3, May/June 1990.

  • Rumors of Go Corporation, Grid, Momenta: contains a propaganda blurb from Momenta on "four waves of personal computing user-interfaces", leading up to verbal/voice recognition.

• [MoriY89] () Mori, Y. and Yokosawa, K. "Neural networks that learn to discriminate similar kanji characters", in Advances in neural information processing systems I, D.S. Touretzky, editor, Morgan Kaufman, 1989.

  • Martin,GL90

• [Myers90] (*) Myers, B.A., Guise, D., Dannenberg, R.B., Zanden, B.V., Kosbie, D., Pervin, E., Mickish, A. and Marchal, P. "Comprehensive support for graphical, highly-interactive user interfaces: The Garnet user interface development environment", IEEE Computer, Vol 23 No 11, Nov. 1990, pp 71 ..85

  • Constraint-based graphical editor for programming construction

• [Nagasawa89] (*a) Nagasawa, H., Ito, K., Horii, S., and Tanoshima, K. "Optical Character Reader Apparatus", United States Patent 4,799,271, January 17, 1989, assigned to Oki Electric Company, Limited, Tokyo, Japan.

  • Patent abstract only
  • OCR character recognition matching entire words against a dictionary

• [Nakamura89] (*p) Nakamura, Y., Ooi, K., and Yura, K. "Hand-held data input apparatus", United States Patent 4,866,646, September 12, 1989, assigned to Toshiba.

  • Patent on external handwriting recognition product with integrated tablet/display to attach to keyboard port, and voice output
  • User-interface for electronic ink, editing documents
  • Recognizes handwritten character data input through force/pressure sensitive tablet and sequentially stores result of character data

• [Nakatani89] (*a) Nakatani, Munehiro "Image Processing Apparatus", United States Patent 4,817,174, March 28, 1989, assigned to Minolta Camera Kabushiki Kaisha, Osaka, Japan.

  • Patent abstract only
  • Edge contour/curve detection for machine vision: for focusing cameras?

• [NCR90] () NCR Corporation, press release "Workstation Sales Systems to be Bought from NCR", May 15, 1990.

  • Press-release on NCR signature capture / handwriting electronic ink system

• [Negroponte89] (*p) Negroponte, Nicholas "A Personal Perspective: An Iconoclastic View Beyond the Desktop Metaphor", International Journal of Human-Computer Interaction, Vol 1 No 1, 1989, pp 109-113.

  • User-interface: a desktop is NOT complete human communication: hand gestures, winks, tone of voice, context of the situation, etc.

• [Neurogen89] (*p) CAP International, Inc. "Neurogen, Inc.: Reading Numbers with Neural Networks", Market Report, October 2, 1989. Principals: Kupfersmith. One Longwater Circle, Norwell, Massachusetts, 02061, 617-982-9500.

  • Neurogen claims 99% correct OCR recognition on handwritten numerals, but with 30% rejection rate
  • Report on Neurogen, other handwriting OCR recognition companies for numerals : Nestor, Hect-Nielson Neurocomputer, Siemens, REI/ Recognition Equipment Incorporated, Scan-Optics, Elsag (Italy), Recognitive Sciences, Calera, Datacopy Accutext, Caere OmniPage, Calera TrueScan

• [Nielsen89] (*) Nielsen, Jakob "CHI '89 Trip Report", http://www.useit.com/papers/tripreports/chi89.html

  • Comments on demonstration of the Paper-Like Interface with handwriting recognition by Cathy Wolf et al from IBM, with a "formula editor" (spreadsheet) application. Also a mention of gestural interaction in an application toolkit for prototyping user interfaces from Mark Rosenstein of MCC Human Interface Tool Set and "silocon paper". Several paragraphs on Wang Freestyle, described as an email application with the ability to integrate graphics, spoken comments, and animated writing (miss-called "gestures").

• [Nielsen90] () Nielsen, Jakob "Traditional Dialogue Design Applied to Modern User Interfaces", Communications of the Association for Computing Machinery, Vol 33 No 10, October 1990, pp 109-118.

  • Review of Wang Freestyle, other pen and non-pen UIs: shows postage stamp feature of Freestyle for documents. Says scrolling of text is bad in UI design.

• [Nolan89] (*p) Nolan, Peter R. "Designing Screen Icons: Ranking and Matching Studies", Proceedings of Human Factors Society 33rd Annual Meeting, October 16-20, 1989, Denver Colorado, pp 380-384.

  • Suggested icons for user-interface: symmetric confusions, "confusion matrix" of preferred selections

• [Nunez89] (*a) Nunez, Linda Mensinger "professional resume", for work in on-line handwriting recognition, 1989.

  • Resume for former employee of Nestor, Incorporated, worked on Japanese on-line handwriting recognition in 1987

• [ON90] () ON Technology Inc. "press release", April 24, 1990, Peter B. Miller to start handwriting recognition-base systems software company for Slate computers.

• [Ooi89] () Ooi, K., Sasaki, H., and Ariyoshi, S., European Patent 335233 (number incorrect?) "Character recognition system", July 26, 1989, assigned to Toshiba (abstract only.

  • From patent abstracts
  • Includes device for obtaining coincidence between recognition character candidates and character patterns stored in dictionary memory

• [Ossner90] () Ossner, J. "Transnational symbols: The rule of pictograms and models in the learning process", in "Designing User Interfaces for International Use", J. Nielsen, Ed., Elsevier Science Publishing, Amsterdam, 1990, pp 11-38.

  • Cited in Nielsen90

• [PatrickM89] (*) Patrick, Mark and Sachs, George "X11 Input Excension Protocol Specification", M.I.T. X Consortium - Technical Review - X11R4 Public Release, December 1989

  • Extension to X graphical system standard to support devices beyond system keyboard and system pointer (mouse): event input includes Key, Button, Valuator data in motion events, proximity events, focus, feedbacks

• [PentlandA90] () Pentland, A. and Kuo, J. "The Artist at the Interface", 10th International Conference on Pattern Recognition, Atlantic City, New Jersey, June 16-21, 1990.

• [PeppersNA89a] () Peppers, N.A., Young, J.R., Pierce, G.A., Nishi, H., and Katsuki, K. "Apparatus for Preprocessing of Character Recognition", United States Patent 4,809,344, February 28, 1989, assigned to Nippon Sheet Glass Company, Limited, Osaka, Japan.

  • Segmentation in OCR scanner input for paragraphs, words picture, and individual letters, also doing character normalization.

• [PeppersNA89b] () Peppers, N.A., Young, J.R., Nishi, H., and Katsuki, K. "Page Segmentor", United States Patent 4,817,169, March 28, 1989, assigned to Nippon Sheet Glass Company, Limited, Osaka, Japan (abstract only).

  • Segmenting a document image into parts "with a single feature" as pre-processing before pattern recognition

• [PersonalWriter89] (*p) Personal Writer, Inc. "Personal Writer 10SL/15SL product literature and User Manual", 1900 Avenue of the Stars, Suite 2870, Los Angeles, California 90067 (see also Anatex). Copyright 1987/1988/1989.

  • Previously called Anatex (which see)
  • Adaptive handwriting recognition product, using a spelling dictionary

• [PetersonS90] () Peterson, Scott "patent application", for Simulated Input Device User Interface, Wang Laboratories, as drafted by patent counsel.

  • Patent description for Wang/Guide SID simulated device user-interface, application independent

• [PhillipsM90] (*) Phillips, Mark J; Philbin, Patrick T. and Blesser, Barry A. "Coordinated Determining Device using Spatial Filters", United States Patent 4,963,703, October 16, 1990

  • Electromagnetic digitizing tablet/pen with tilt measurement and stylus tilt (mis-named as parallax) correction: Cited later for prior art on acoustic/ultrasonics tablets

• [Plamondon89] (*) Plamondon, R. and Lorette, G. "Automatic signature verification and writer identification -- the state of the art", Pattern Recognition, Vol 22 No 2, March 1989, pp 107-132.

  • Review of all current commercial and research signature verification systems -- specificallyexcludes patents

• [Plamondon90] () Plamondon, R. and Yergeau, P. "A System for the Analysis and Synthesis of Handwriting", in Frontiers in Handwriting Recognition, C.Y. Suen, editor, Proceedings of the Conference on Frontiers in Handwriting Recognition, Concordia University, Montreal Canada, April 2-3, 1990, pp 167-180.

• [Poqet89] () Poqet Computer Corporation "press release", October 2, 1989.

  • Rumored portable touchscreen product: no character recognition

• [Quam89] () Quam, David L., Williams, George B., Agnew, Jeffery R., and Browne, Patricia C. "An Experimental Determination of Human Hand Accuracy with a DataGlove", Proceedings of Human Factors Society 33rd Annual Meeting, October 16-20, 1989, Denver Colorado, pp 315-319.

  • Pointing accuracy with human hand, but open loop (no coupling to a tablet digitizer/display tracking crosshair, for example).

• [QuerciaV89] (*) Quercia, Valerie and Ro'Reilly, Tim "X Window System User's Guide for X11 R3 and R4 of the X Window System", Three volumens: O'Reilly & Associates, 1989

  • X-Windows documentation: mouse/pointer based windows interface, can be used with a tablet digitizer with a puck with at least three functional buttons, for mouse-button menus. No use with a stylus. Mouse-button menus can be re-defined.
  • (ARCHIVED: See Lucent file)

• [RecognitiveSciences89] (*p) Recognitive Sciences "Recognitive Sciences Provides Cost Effective OCR Solutions for OEMs", company profile, 1989. Principals: Ted Kuklinski and Yuri Litvin. 323 Andover Street, Wilmington, Massachusetts, 01887.

  • Kuklinski's and Yuri's O.C.R for handwritten and embossed characters

• [Sabourin89] () Sabourin, Robert and Plamondon, Re'jean "Segmentation of Handwritten Signature Images: A Structural Approach", unpublished manuscript, July 26, 1989. Laboratoire Scribens, Ecole de technologie Supe'rieure, De'partement de Production Automatise'e, Montre'al, Quebec, Canada.

  • Reviewed for PAMI and Henry Baird: extraction of signatures from scanned documents prior to verification

• [Sabourin90] (*a) Sabourin, Robert and Plamondon, Rejean "Progress in the Field of Automatic Handwritten Signature Verification Systems Using Gray-Level Images", in Frontiers in Handwriting Recognition, C.Y. Suen, editor, Proceedings of the Conference on Frontiers in Handwriting Recognition, Concordia University, Montreal Canada, April 2-3, 1990, pp 1-12.

  • Signature verification with OCR: can tell photocopies from originals

• [SatoK89] () Sato, Koji, Kurakake, Masahiro, and Fujiwaki, Kazuo "Optical Character Reader", United States Patent 4,797,940, January 10, 1989,assigned to Sumitomo Electric Industries, Osaka, Japan (abstract only).

  • Hand-held OCR scanner which can read one line at a time, with hardware to segment into separate characters

• [ScanOptics90] (*) Scan-Optics, Inc. "Easyreader 1720 OCR product", press release, February 1, 1990.

• [Scenario89a] () Scenario, Inc. and Visible Systems Corp. "Easy Touch/Touch Translator", structured UI design tool, PC Week, Dec 4, 1989, vol 6 no 48, p 73.

  • Touch-screen user-interface design tool: Pen-base UI

• [Scenario89] () Scenario Inc "Dynabook product announcement", October 1989, Boston, Massachusetts.

  • Touch-screen portable (sort of) CD-ROM product: no handwriting recognition yet.

• [Schneidawind90] (*) Schneidawind, John "Software pioneers are reunited at hot new Arizona company", Press report, 1990 (publication not known)

  • Bob Frankston and Dan Bricklin join Slate Corporation. Investment by Mitch Kapor

• [SeymourJ90] (*) Seymour, Jim "The "Slate PC" Wave Is About To Break", PC Magazine, October 16, 1990, page 85..86

  • Editorial that pen-computing PCs (like Palm?), as small portable devices making only minor use of handwriting character recognition, would be successful. Mentions GRiD, Go/PenPoint, Momenta. No mention of Microsoft Pen-Windows.

• [Shu89] () Shu, Joseph Shou-Pyng "One-pixel-wide Edge Detection", Pattern Recognition, Vol 22 No 6, 1989, pp 665-673.

  • Edge detection/skeletonization/line-thinning using a 3x3 pixel filter, similar to Sobel edge detector filters, Robert 2x2 edge detectors, combined with heuristic (minimum-cost) search
  • Edge-streak encoding, similar to chain codes for edges

• [Signaware89] () Signaware Corp. "software description", 1989, 300 South Duncan Avenue, Suite 273, Clearwater, Florida 34615, 813-461-4211.

  • Internal Wang memo on signature verification vendors: one in Florida, one in Ireland

• [Simon90] (*p) Simon, J.C. and Baret, O. "Handwriting Recognition as an Application of Regularities and Singularities in Line Pictures", in Frontiers in Handwriting Recognition, C.Y. Suen, editor, Proceedings of the Conference on Frontiers in Handwriting Recognition, Concordia University, Montreal Canada, April 2-3, 1990, pp 23-38.

• [Sklarew90] () Sklarew, Ralph "patent application data", U.S. Patent application on handwriting recognition user-interface.

  • Patent description of prior art for Linus product, Grid patent on user interfaces

• [Sklarew89a] (*) Sklarew, Ralph "Handwritten Keyboardless Entry Computer System", European Patent Application 0 254 561 A2

  • Abstract refers to transparent sensing surface over display, electronik ink, character recognition, and learning mode
  • Title and abstract read almost identically to other Sklarew patents

• [Sklarew90a] (*) Sklarew, Ralph "Handwritten Keyboardless Entry Computer System", United States Patent 4,972,496 assigned to Grid Systems Corporation, Fremont, California, March 24, 1987

  • Abstract refers to transparent sensing surface over display, electronik ink, character recognition, and learning mode
  • Title and abstract read almost identically to other Sklarew patents
  • Transparent front-mounted digitizer tablet: cites advantages (stylus tilt, etc.) over rear-mounted tablets. Refers to trainable recognition for handwritten characters and symbols, but not to macros or gestures.

• [Small90] (*p) Small, Charles H. "Inexpensive software reads printed material", Electronic Design News, January 18, 1990, pp 77-86.

  • Review of OCR scanner companies: Caere Corporation, Inovatic, Olduvai Corporation, Xerox Imaging Systems.

• [Sony90] () Sony Corporation "hand-held computer press release", March, 1990.

  • Hand-held slate computer product from Sony Corporation.

• [Spitz89] (*p) Spitz, Gabriel "The Preferred Pen Activation Force for the Low Resolution Freestyle Tablet", Internal report 14 July, 1989, Wang Laboratories, 1 Industrial Avenue, Lowell, Massachusetts 01851.

  • Internal Wang report on stylus pen force/pressure for handwriting force

• [Spitz90] () Spitz, Gabriel, Akagi, Kenichi "Stylus Button Location", internal report 3 January, 1990, Wang Laboratories, 1 Industrial Avenue, Lowell, Massachusetts 01851.

  • Internal Wang report on barrel-switch/side-button on digitizer tablet stylus

• [Srihari90] () Srihari, S.N. "Reading Unconstrained Handwriting with Bounded Context", in Frontiers in Handwriting Recognition, C.Y. Suen, editor, Proceedings of the Conference on Frontiers in Handwriting Recognition, Concordia University, Montreal Canada, April 2-3, 1990, pp 109-116.

  • Overview of some recognition context problems: project was OCR for handwritten Zip codes, address blocks

• [Stentiford89] (*a) Stentiford, Frederick W.M. "Pattern Recognition System", United States Patent 4,817,171, March 28, 1989, assigned to British Telecommunications public limited company, Great Britain.

  • Patent abstract
  • OCR

• [Stringa90] () Stringa, L. "A Structural Approach to Automatic Primitive Extraction in Hand-printed Character Recognition", in Frontiers in Handwriting Recognition, C.Y. Suen, editor, Proceedings of the Conference on Frontiers in Handwriting Recognition, Concordia University, Montreal Canada, April 2-3, 1990, pp 65-72.

  • Abstract only

• [Suen90] () Suen, C.Y., Nadal, C., Mai, T.A., Legault, R. and Lam, L. "Recognition of Totally Unconstrained Handwritten Numerals Based on the Concept of Multiple Experts", in Frontiers in Handwriting Recognition, C.Y. Suen, editor, Proceedings of the Conference on Frontiers in Handwriting Recognition, Concordia University, Montreal Canada, April 2-3, 1990, pp 131-144.

• [Suzuki90] () Suzuki, T. and Mori, S. "A Thinning Method Based on Cell Structure", in Frontiers in Handwriting Recognition, C.Y. Suen, editor, Proceedings of the Conference on Frontiers in Handwriting Recognition, Concordia University, Montreal Canada, April 2-3, 1990, pp 39-52.

  • Line-thinning / skeletonization

• [Taguchi89] (*) Taguchi, Yoshinori; Yamanami, Tsuguya "Coordinate Input Device with Display", United States Patent 4,845,478, July 4, 1989

  • Patent on Wacom electromagnetic / magnetostrictive digitizer using cordless stylus. Does not disclose sensor or switch for detecting stylus down or contact with surface.

• [Takahashi91] () Takahashi, Toimoichi, and Kishino, Fumio "Hand Gesture Coding Based on Experiments Using a Hand Gesture Interface Device", SIGCHI Bulletin, Vol 23 No 2, April 1991, pp 67-74.

  • DataGlove product used to read hand alphabet gestures for Japanese Kana

• [Tappert89] () Tappert, Charles C. and Jeanty, H.H. "A Study of Several Accuracy-Improvement Methods for a Handwriting Recognition System", IBM Research Report, IBM T.J. Watson Research Center, P.O. Box 704, Yorktown Heights, NY 10598, November 1989.

  • Most promising feature algorithm for improving recognition over elastic matching is pairwise comparison (but NOT tuned by hand!)
  • Features (beyond elastic matching) are heights of endpoints, relative stroke positions (min and max extents), line extension, relative width, etc.
  • Most handwriting recognition error for other systems was transducer/stylus % error, user error (user wrote "wrong")!
  • Cites on study that handwriting on integrated tablet/display (electronic ink) sloppier than on separate tablet, much sloppier than on paper

• [Tappert90a] () Tappert, C.C., Suen, C.Y., and Wakahara, T. "On-line Handwriting Recognition - a Survey", manuscript in preparation, to appear in IEEE Transactions on Pattern Analysis and Machine Intelligence: C.C. Tappert, IBM T.J. Watson Research Center, Yorktown Heights, N.Y 10598.

  • Historical development of elastic matching for script, discrete handwriting character recognition
  • Commercial handwriting input systems listed

• [Tappert90b] () Tappert, Charles C. "Rationale for Adaptive Online Handwriting Recognition", in Frontiers in Handwriting Recognition, C.Y. Suen, editor, Proceedings of the Conference on Frontiers in Handwriting Recognition, Concordia University, Montreal Canada, April 2-3, 1990, pp 13-22.

• [Tappert90c] (*) Tapper, Charles C., Suen, Ching Y,. and Wakahara, Toru "The State of the Art in On-line Handwirting Recognition", IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol 12 No 8, August 1990, pp 787-ff

  • Comprehensive review as of 1989.
  • Gives historial remarks pen-computing in the 1950s, 1960's, 1970's, 1980's

• [Taxt89] () Taxt, Torfinn, Flynn, Patrick J., and Jain, Anil K. "Segmentation of Document Images", IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol 11 No 12, December 1989, pp 1322-1329.

  • Statistical classifier to separate text and background images on documents
  • Document segmentation using contextual Bayes, Besag's, relaxation, Owen and Switzer, versus Haslett's classifiers: which is best varies

• [Taxt90] () Taxt, Torfinn, Olafsdottir, Jorunn B., and Daehlen, Morten "Recognition of Handwritten Symbols", accepts for Journal of Pattern Recognition, 1990.

  • O.C.R. on handwriting with 98.6% and 97.7% accuracy
  • O.C.R. of handwritten map symbols, using several edge detection, spline curve fits, Fourier expansion, and statistical classification (lower-case alphanumerics)
  • O.C.R. of handwritten lower-case alphanumeric using circular graphs

• [Teh89] () Teh, Cho-Huak and Chin, Roland T. "On the Detection of Dominant Points on Digitial Curves", IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol 11 No 8, August 1989, pp 859-872.

  • Scale-space filtering (size-dependent) of small features in machine vision. Keywords: chain code, curve tracing, polygonal approximation, tangential deflection. Points of high curvature are dominant features for pattern recognition. Lots of references on chain codes and curve tracing for machine vision.

• [Termiflex89] (*) Termiflex Corporation "TouchGraf Keyless Keypad: the Next Generation of Flexible Terminals", 316 Daniel Webster Highway, Merrimack, NH 03054, March 1989.

  • Handwriting capture terminal product, separate digitizer built into a hand-held terminal: digitizer/touchpad and display separate

• [Teulings90a] () Teulings, H.-L. "Handwriting movement control - Research into different levels of the motor system", Doctoral Thesis, Nigmeegs Instituut voor Cognitie-onderzoek en Informatietechnologie, Nijmegen, 1990.

• [Teulings90] () Teulings, H.-L. and Schomaker, L.R.B. "A Handwriting Recognition System Based on the Properties and Architectures of the Human Motor System", in Frontiers in Handwriting Recognition, C.Y. Suen, editor, Proceedings of the Conference on Frontiers in Handwriting Recognition, Concordia University, Montreal Canada, April 2-3, 1990, pp 195-ff.

• [Toshiba89] (*p) Toshiba America Information Systems, Inc. "Toshiba PenPC product information", November 3, 1989, 714-455-1707.

  • Pre-announced portable handwriting recognition product: kludged together from Linus, other parts?. DOS 3.3 OS. Mentions mouse emulation with touchscreen digitizer.

• [Uncapher89] (*) Uncapher, Kieth "Interview with Kieth Uncapher", Arther L. Norberg, 10 July, 1989. Charles Babbage Institute, Center for the History of Information Processing, University of Minnesota, Minneapolis. Available at http://www.cs.utexas.edu/users/umair/cs370/interviews/oh174ku.rtf

  • Interview with Kieth Uncapher of RAND Corporation: mentions GRAIL system with electronic ink on the Rand tablet. Describes use of tablet for flow-chart and programming text entry. Describes invention of the disc drive (to replace the drum drive) as a result of this project.
  • Mentions a video film showing the Rand tablet and handwriting recognition in operation in the 1960's.
  • Rand tablet invented by Tom Ellis around 1966: also an earlier tablet by Herb Teager at Boston College (Stylalator?). Other work by Gabe Groner, and Bill Sibley.

• [Unser89] () Unser, Michael and Eden, Murray "Mutliresolution Feature Extraction and Selection for Texture Segmentation", IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol 11 No 7, July 1989, pp 717-728.

  • Segmentation of textured images for machine vision: refers to Fisher's multiple linear discriminants and Karhunen-Loe`ve transform

• [VanLeeuwen89] () Van Leeuwen, H.C. "A development tool for linguistic rules", Computer Speech and Language, Vol 3 No 1, January 1989, pp 83-104.

  • Phonology, grapheme-to-phoneme conversion rules for text-to-speech

• [Vosburgh89] () Vosburgh, A.M. and Brown, N.R. "Following the bouncing ball: An evaluation of a self-pacing enrollment interface for a large-vocabulary speech recognition system", Research Report RC-64372, IBM Thomas J. Watson Research Center, Yorktown Heights, New York, 1989.

  • Brown,NR89 on user interface / enrollment for adaptive recognition

• [Wacom89] () Wacom, Inc. "Super Digitizer product information", West 115 Century Road, Paramus NJ 07562, August 1989.

  • Product literature on Wacom force/pressure-sensitive pen stylus and cordless pen stylus tablet.
  • Product literature describing Wacom cordless pen stylus using resonating coil in tablet digitizer.
  • 1986 CIC memo lists a katakana and Roman and kanji system under development.
  • Digitizer tablet with absolute and relative (i.e. mouse) modes.

• [Wacom89a] (*) Wacom, Inc. "Wacom PC Installation and Utilities Manual", Park 80 West Plaaza II, Saddle Brook, NJ 07662, November 1989.

  • Describes emulation and multiple command sets for digitizer controller/driver
  • Digitizer tablet with absolute and relative (i.e. mouse) modes

• [WakimotoK90] () Wakimoto, K., Shima, M., Tanaka, S., and Maeda, A. "An Intelligent User Interface to an Image Database Using a Figure Interpreation Method", 10th International Conference on Pattern Recognition, Atlantic City, New Jersey, June 16-21, 1990.

• [Wang89] (*) Wang Laboratories "The Wang Freestyle System", product literature

  • Wang Freestyle: pen-computing electronic-ink and voice annotation, no character recognition. Used customized Hitachi tablet, with first (?) double-ended stylus with an eraser on the back end. Pure electronic ink and selection application system, with syncronized voice: interesting case of human communication. Standalone Demo diskette (5.25").

• [Wang90a] (*) Wang Laboratories "Freestyle: Wang Freestyle/Light User's Guide, Release 1.21", Wang Laboratories, Lowell Massachusetts

  • Wang Freestyle: pen electronic-ink and voice annotation, no character recognition. Used customized Hitachi tablet, with first (?) double-ended stylus with an eraser on the back end. Describes image-only document (voice/electronic ink/imaging) data base using no handwriting recognition. No use of gestures. Cited by Bill Buxton.

• [Wang90b] (*) Wang Laboratories "Freestyle: Take a look at the personal communication system you already know", Wang Laboratories, Lowell Massachusetts

  • 5.25" floppy demo of Wang Freestyle, with brochure explaining the Freestyle UI: pen electronic-ink and voice annotation, no character recognition.

• [Wang90c] (*) Wang Laboratories "Equipment/Device: Wang Freestyle miniature keyboard ca. 1990", Wang Laboratories

  • Physical device in collection: Small keyboard with pressure switches, suitable for activation with a stylus. Full 102-key keyboard, including function and shift keys. Intended to be used in conjunction with Freestyle tablet digitizer for Freestyle UI. Likely a prototype.

• [WangPSP89] () Wang, P.S.P. "A New Approach for Chinese Handwriting Recognition", Proc. Int. Conf. on Electronic Imaging, Pasadena, pp. 373-379, 1989

  • On-line (D.C.R.) recognition with very little system training/learning for Chinese. Chinese recognition using stroke angles in 22.5 degree ranges (chain codes?)

• [WangJS89] () Wang, J.S. "Electronic text-graphic editing system", United States Patent 4,829,470, May 9, 1989, assigned to IBM Corporation.

  • Text-graphic editing system includes mixed object editor which flows text around into irregularly shaped objects on page (for a mark-up language, like runoff for text composition in documents)

• [WangPSP89] () Wang, P.S.P. and Zhang, Y.Y. "A Fast and Flexible Thinning Algorithm", IEEE Transactions on Computers, Vol 38 No 5, May 1989, pp 741-745.

  • Compares and shows results of different skeletonization/thinning algorithms for character and arbitrary shapes

• [Ward89] (*) Ward, Jean Renard "Testing tablet characteristics", Internal Report, Wang Laboratories, May 1989

  • Technical instructions on testing tablets to reveal bad behaviors for electronic ink, handwriting character recognition
  • Describes differential vs. absolute error, spirograph tests, etc.

• [Ward90] () Ward, Jean Renard "One View of On-going Problems in Handwriting Character Recognition", in Frontiers in Handwriting Recognition, C.Y. Suen, editor, Proceedings of the Conference on Frontiers in Handwriting Recognition, Concordia University, Montreal Canada, April 2-3, 1990, pp 101-108.

• [Ward90a] (*) Ward, Jean Renard and Kabel, Robert "The "How to" of Electronic Ink: Tablet Error Mechanisms and Performance Analysis for Handwriting Capture", Internal report, Wang Laboratories, July 1990

  Descriptions of "bad behaviors" in digitizing tablets, engineering review of the effects, test procedures for showing the bad behaviors

• [Ward91] () Ward, Jean Renard "Pen-Based Products", correspondence in "ACM Forum", Communications of the Association for Computing Machinery, Vol 34 No 12, December 1991, pp 18-19.

  • Reply to Larry Press that Linus was not the first pen computing company

• [Ware89a] (*p) Ware, Colin and Baxter, Curtis "BAT Brushes: On the Uses of Six Position and Orientation Parameters in a Paint Program", Proceedings of 1989 CHI conference, 1989, pp 155-160.

  • User-interface on simulating a brush with a stylus using stylus rotation angle, x/y/z digitizer tablet

• [Ware89b] () Ware, Colin and Jessome, D "Using the Bat: A Six-Dimensional Mouse for Object Placement", to appear in IEEE Computer Graphics and Applications, November 1989.

  • Ware89a

• [Weideman89] () Weideman, W.E., Manry, M.T. and Yau, H.C. "A comparison of nearest neighbor classifier and a neural network for numeric handprint character recognition", IEEE International Conference on Neural Networks, Washington, D.S., 1989.

  • Martin,GL90

• [Wolf89] (*) Wolf, Catherine G., Rhyne, J.R., and Ellozy, H.A. "The paper-like interface", in "Designing and Using Human-Computer Interfaces and Knowledge-Based Systems", G. Salvendy and M.J. Smith, Eds., Elsevier Science Publishers, Amsterdam, 1989, pp 494-501.

  • Description of paper-like interface PLI projects at IBM: electrostatic/capacitive digitizer (Scriptel) over LCD display, exploration of gesture input. Video demonstration shown in IBM lobby. Gestures described as handmarkings for user-interface in spreadsheet (Lotus) and drawing applications. Music editing demonstration modeled after Buxton. Two-dimensional mathematics editing. Note: video demonstration filed under "Lucent".

• [Yamada89] (*p) Yamada, K., Kami, H., Tsukumo, J., and Temma, T. "Handwritten Numeral Recognition by Multi-layered Neural Network with Improved Learning Algorithm", Proceedings of International Joint Conference on Neural Networks, Washington, D.C., June 18-22, 1989, Vol II pp 259-266.

  • OCR for handwritten numerals using neural network. Gives samples of low-quality, but real-world handwriting test data for numerics for OCR.

• [Yamaguchi89] (*A) Yamaguchi, Mikio and Sato, Koji "Optical Character Reader", United States Patent 4,817,185, March 28, 1989, assigned to Samitomo Electric Industries, Limited, Osaka, Japan.

  • Patent abstract only
  • Hand-held scanner for reading price tags

• [Yoshimura91] () Yoshimura, I. and Yoshimura, M. "Off-Line Writer Verification Using Ordinary Characters as the Object", Pattern Recognition, Vol 24 No 9, September 1991, pp 909-915.

  • Writer (not just signature) recognition in Japanese Kanji writing: OCR skeletonization based: claims 99.5% accuracy: uses arcs and angles/vectors

• [Yu90] () Yu, Shiaw-Shian and Tsai, Wen-Hsiang "A new thinning algorithm for gray-scale images by the relaxation technique", Pattern Recognition, Vol 23 No 10, October 1990 pp 1067-1077.

  • Uses chain codes context to preserve line straightness in line thinning: for sketch recognition?
  • Straightness of lines very important in Chinese/Kanji character recognition

• [Zhuang89] () Zhuang, X., Haralick, R.M., and Joo, H. "A Simplex-Like Algorithm for the Relaxation Labeling Process", IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol 11 No 12, December, 1989, pp 1316-1321.

      Main Reference Page