This compilation and all annotations are copyright © Jean Renard Ward, 1996, 1999, 2002, 2005, 2006
Copyright © Sun Jun 28 13:20:08 EDT 2009

      Main Reference Page      

• [Agui79a] (*p) Agui, Takeshi and Nagahashi, Hiroshi "A Description Method of Handprinted Chinese Characters", IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol PAMI-1 No 1, January 1979

  • Chinese handwriting recognition: features are concatenate, crossing, near, relative location among partial patterns.

• [Agui79b] (*p) Agui, Takeshi and Nagahashi, Hiroshi "A Coding Method of Chinese Characters", IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol PAMI-1 No 4, October, 1979

  • Chain codes, production rules, block codes, to encode Chinese characters as chain-code strings for comparison

• [Aho77] . Aho, A.V. and Ullman, J.D. "Principles of Compiler Design", Addison-Wesley, Reading, Massachusetts 1977

  • Compiler/programming language design

• [Aldefeld80] (*A) Aldefeld, B., Levinson, S.E., and Szymanski "A minimum-distance search technique and its application to automatic directory assistance", Bell Systems Technical Journal, Vol 59 No 8, pp 1343-1356, October 1980

  • Cited in Kahan87: : best method of organizing spelling dictionary for context in character recognition (actual paper uses spoken spelled speech recognition of letters)
  • Context raises 80% on letters to 98.6% on names, 71% to 97.2% on 18000 name entries

• [Ali77] . Ali, F. and Pavlidis, T. "Syntactic recognition of handwritten numerals", IEEE Transactions on Systems, Man, and Cybernetics, Vol SMC-7, 1977, pp 537-541

• [Anderson76] (*A) Anderson, Robert H. "Intelligent Terminals: A Potential Solution to Complexity in User-Network Interfaces", Proceedings of ASIS Annual Meeting, San Francisco, California, October 4-9, 1976

  • NTIS citation index

• [Apsey78] (*p) Apsey, R. "Human Factors of Constrained Hand-print for OCR", IEEE Transactions on Systems, Man, and Cybernetics, Vol SMC-8 No 4, pp 292-296, April 1978

  • Constraints on handwriting using various "pre-printing" guides: dots, boxes, shapes, lines

• [Arakawa78] (*p) Arakawa, K., Odata, K., and Masuda, T. "On-line recognition of hand-written characters -- Alphanumeric, hiragana, katakana, kanji", Proc. 4th IJCPR, Kyoto, Japan, November 7-10, 1978, pp 810-812. Published later as (Arakawa82)

• [Arazi78] (*p) Arazi, B. "Handwriting Identification by Means of Run-Length Measurements", IEEE Transactions on Systems, Man, and Cybernetics, Vol SMC-7 No 12, pp 878-ff, December 1977

  • off-line signature verification, study of vertical-line vs. horizontal-line run-length to do signature verification

• [Armstrong79] (*p) Armstrong, William W. and Gecsei, Jan "Adaptation Algorithms for Binary Tree Networks", IEEE Transactions on Systems, Man, and Cybernetics, Vol SMC-9 No 5, May 1979, pp 276-ff

  • adaptive tree networks, heuristic responsibility, with specialized sub-trees

• [Azure77] (*) Azure, Leo L. "Electronic Notebook for Use In Data Gathering, Formatting, and Transmitting System", United States Patent 4,016,542, assigned to Azurdata, Inc., Richland, WA, April 5, 1977

  • Portable hand-held data-entry device, with radio link to a central computer.
  • Uses a simple keyboard, does not refer to hand-written input

• [Babcock77] (*p) Babcock, R.T. "Simulation Method of Feature Selection for Unconstrained Hand-printed Characters", M.S. Thesis, Department of Electrical Engineering and Computer Science, June 1977, Massachusetts Institute of Technology, Cambridge, Massachusetts

  • 2-Z recognition: Blesser's and Shillman's group at Massachusetts Institute of Technology

• [Badler79] (*) Badler, Norman I. and Smoliar, Stephen, W. "Digital Representations of Human Movement", Computing Surveys, Vol 11 No 1, March 1979

  • Translation of Labanotation dance notation to computer graphics of human movement

• [BakerJ80] (*p) Baker, Janet MacIver, Dialog Systems Inc. "Brief Status Summary for Automatic Speech Recognition at the start of the 80's", SAE Technical Paper Series 800195, Soc. of Automotive Engineers, 400 Commonwealth Drive, Warrendale, Pennsylvania 15096

• [Baecker76] (*) Baecker, Ronald "A conversational extensible system for the animation of shaded images", Proc. of 3rd Annual Conference on Computer Graphics and Interactive Techniques, Philadepha, PA, pp 32-39.

  • Pen/lightpen system for interactive animation. Images, and their motion, sketched by user free-hand
  • Citation to Alan Kay presentation at 1972 ACM conference

• [Baecker79a] (*p) Baecker, Ronald, Buxton, William, and Reeves, William "Towards Facilitating Graphical Interaction: Some Examples from Computer-Aided Musical Composition", Proc. of 6th Man-computer Communications Conference, Ottawa CA, May 29-30, 1979

• [Baecker79b] (*p) Baecker, Ronald "Digital Video Display System and Dynamic Graphics", Computer Graphics, Vol 13 No 2, August 1979

• [Baecker80a] (*p) Baecker, R. "Towards an Effective Characterization of Graphical Interaction", in Methodology of Interaction, Guedj et al, editors, North-Holland Publishing Company, 1980, pp 127-ff

  • User-interface: best user interface is device dependent, not device independent
  • Broad list of abstract qualities (type of feedback, etc.) of good user interfaces
  • Gesture-recognition for musical scores input

• [Baecker80b] (*p) Baecker, Ronald "Human-computer interactive systems: A State-of-the-art review", in Processing of Visible Language 2, Kolers, Paul A., Wrolstad, Merald E., and Bouma, Herman, editors, Plenum Press, New York and London, 1980, pp 423-443

  • User-interface review: trainable handwriting character recognizers: recognizer must be accurate and rapid
  • Gesture-recognition for musical scores input

• [Barnard76] (*p) Barnard, P. and Wright, P. "The Effects of Spaced Character Formats on the Production and Legibility of Handwritten Names", Ergonomics, Vol 19 No 1, Jan 1976, pp 81-92

  • (re box-less) wider character spacing makes stuff less readable
  • Boxed input should match normal writing size (handwriting constraints)

• [Barrow77] . Barrow, H.G., Tenenbaum, J.M., Bolles, R.C., and Wolf, H.C. "Parametric correspondence and chamfer matching: Two new techniques for image processing", Proceedings of 5th International Joint Conference on Artificial Intelligence, August 1977, pp 659-663

  • Burr81

• [Bartram78] . Bartram, D.J. "Post-iconic visual storage: Chunking in the reproduction of briefly displayed visual patterns", Cognitive Psychology, Vol 10, 1978, pp 324-355

  • Cited in IchikawaS84
  • Is this where Buxton got his "chunking" ideas?

• [BeckerP77] (*A) Becker, Peter W. "Recognition of Patterns Using the Frequiencies of Occurrence of Binary Words", Springer Verlag, Wien and New York, second revised edition

  • first edition 1968, previously doctoral thesis

• [Bernstein76] . Bernstein, M.I. "Interactive Systems Research: Interim Report to the Director, Advanced Research Projects Agency, for the Period 16 September 1975 to 15 March", 1976, System Development Corporation, Santa Monica, California, Report No SDC-TM-5243/005/00, 15 April 1976

  • NTIS citation index
  • Speech understanding system

• [Berson77] . Berson, Tom "Dynamic Handwriting Recognition by Computer", Ph.D. Thesis, University of London, 1977

• [Berthod79] (*p) Berthod, M. and Maroy, J.P. "Learning in Syntactic Recognition of Symbols Drawn on a Graphic Tablet", Computer Graphics and Image Processing, Vol 9, 1979, pp 166-182

  • The main problem in handwriting recognition is wide range of variability
  • Big practical problem of on-line character recognition (vs OCR) is large deformations in handwriting
  • Statistical approaches: complex classifying scheme and meaningless features
  • Problem of statistical recognition is features do not correspond to any visual entity
  • Says on-line recognition should allow added new symbols (writable icons for commands) (?)
  • Character recognition needs techniques adapted to characters, not perception of drawings in general
  • Character recognition should use human-meaningful concepts
  • "chain codes" of straight line, plus/minus curve, pen-lift, cusp
  • Two-level chain code, decision tree like our BLRTs, nodes (but adaptive?)
  • Gives a grammar of permissible sequences for a sort of chain code

• [Berthod80] . Berthod, M. and Ahyan, S. "On line cursive script recognition: A structural approach with learning", Proceedings of the 5th International Conference on Pattern Recognition, Miami Beach, Florida, December, 1980, pp 723-725

  • Cited in Bozonivic89
  • Bozinovic89 cites this as doing adaptive/training handwriting recognition, but with syntax of letter formation and writing dynamics for strokes

• [Bezdek77] . Bezdek, J.C. and Castelaz, P. "Prototype classification and feature selection with fuzzy sets", IEEE Transactions on Systems, Man, and Cybernetics, Vol SMC-7 No 2, 1977, pp 87-92

  • Biswas81

• [Bjorklund77] . Bjorklund, C.M. "Syntactic analysis and description of stroke-based shapes", Proceedings of IEEE Pattern Recognition and Image Processing Conference, Troy, New York, June 6-8, 1977, pp 198-202

• [Blesser76] (*p) Blesser, B., Kuklinski, T.T., and Shillman, R.J. "Empirical Tests for Feature Selection Based on a Psychological Theory of Character Recognition", Pattern Recognition, Vol 8, pp 77-85, 1976

  • How we pick functional attributes

• [Boldridge77] . Boldridge, A.G. and Freund, R.W. "Personal Identification Apparatus", United States Patent 4,035,768, assigned to Veripen Incorporated, New York, New York, July 12, 1977

  • Signature verification using pressure (actually, tip force)

• [Bolt80] (*p) Bolt, R.A. "'Put-That-There': Voice and Gesture at the Graphics Interface", SIGGRAPH '80 Proceedings, published as Computer Graphics, Vol 14 No 3, July 1980, pp 262-270

  • User interface combining pointing and voice recognition

• [Borning79] (*) Borning, Alan "ThingLab -- A Constraint-Oriented Simulation Laboratory", Ph.D. thesis, CS Dept., Stanford University, published as SSL-79-3 July 1979

  • editing of graphical drawings, no gesture recognition

• [Bouchard80] . Bouchard, D.C. and Toussaint, G.T. "Heuristic search methods for efficient use of dictionary information in text recognition", School of Computer Science, McGill University, Technical Report SOCS 80.5, May, 1980

  • Hull,JJ83

• [Brayton79] (*) Brayton, Robert K. "Character Recognition System and Method Multi-Bit Curve Vector Processing", United States Patent 4,177,448, assigned to IBM Corp, Armonk, New York, December 4, 1979

  • Setting up character recognition based on large character training set
  • Shapes are canonicalized to standardized/normalized pieces of curves/segments

• [Bresenham77] (*p) Bresenham, J.E. "A Linear Algorithm for Incremental Digital Display of Circular Arcs", Communications of the A.C.M., Vol 20 No 2, February 1977, pp 100-106

  • Circle-drawing algorithm on bit-mapped displays: cites algorithms for other curves

• [Bridle79] . Bridle, J.S. and Brown, M.D. "Connected word recognition using whole word templates", Proceedings of the Institute for Acoustics, 1979, pp 25-28

  • Cited in Kruskal83

• [Briem79] (*p) Briem, G.S.E. "Wanted: Handwriting That Fits Modern Pens", Visible Language, Vol XIII No 1, 1979, pp 50-62

  • Background on handwriting variability effects from stylus design

• [BrownMK80a] . Brown, M.K and Rabiner, Lawrence R. "An Adaptive, ordered, graph search technique for dynamic time warping for isolated word recognition", IEEE Transactions on Acoustics, Speech and Signal Processing, Vol 30, 1982, pp 535-544

  • Cited in Schomaker86

• [BrownMK80] (*p) Brown, M.K. and Ganapathy, S. "Cursive Script Recognition", 5th International Conference on Character Recognition, 1980, pp 47-51

  • Has a data-base intentionally with lots of variability

• [Burke76] (*p) Burke, Barbara "Reading Writing", M.I.T. Reports on Research, Vol 4 No 3, November 1976

  • Report on work of Shillman, Blesser, and Kuklinski

• [Burr79] . Burr, D.J. "A technique for comparing curves", Proceedings of the IEEE Conference on Pattern Recognition and Image Processing, Chicago, 1979, IEEE New York, pp 271-277

  • Cited in Kruskal83

• [Buxton79] (*) Buxton, William; Sniderman, Richard; Reeves, William; Patel, S.; and Baecker, Ronald "The evolution of the SSP score editing tools", Computer Music Journal, Vol 3, 1979, pp 14-25

  • Reprinted in "Foundations of Computer Music", MIT Press, Cambridge MA 1985, pp 376-402
  • music-editing notation, single/uni-stroke handwriting character recognition
  • Shows other GUIs for music input and editing other than recognition
  • Cited in Wolf87a
  • Canadian equivalent of ANSI74 hand-print standard to constrain handwriting for OCR

• [Canada78] . CSA "Numeric Character Set for Hand-printing", Canadian Standards Association, Standard Z243.34.1-M1978, February 1978

• [Calvert80] (*) Calvert, T.W., Chapman, J., and Patla, A "The Integration of Subjective and Objective Data in the Animation of Human Movement", Conference Proceedings: ACM 0-89791-021-4/80/0700-0198

  • Labanotation

• [Carau78] (*) Carau, F.P. "Easy-to-use, High-Resolution Digitizer Increases Operator Efficiency", Hewlett-Packard Journal, December 1978, pp 2-13

  • Digitizer measuring X and Y sequentially in time causes line to bow due to velocity-related errors. Gives first-order correction. Numerous descriptions of other digitizer position errors, such as non-orthogonal axes, and how to correct for them.

• [Carvey76] . Carvey, P.C. "Electrographic System", United States Patent 3,975,592, assigned to Applicon, Incorporated, Burlington, Massachusetts, August 17, 1976

  • Digitizer using coarse, then fine scanning

• [Cederberg78] . Cederberg, R.L.T. "An iterative algorithm for angle detection on digital curves", Proceedings of 4th International Joint Conference on Pattern Recognition, Kyoto, Japan, Movember 7-10, 1978, pp 576-578

• [ChangP76] (*p) Chang, P. "Efficient Keyboard Layouts for Inputting Chinese Characters", B.S. Thesis, Department of Electrical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts May 1976

  • Cited in Crane77b
  • 117 confusions/ambiguous character in 10000 defined Chinese/Kanji even if every basic stroke has a unique keyboard stroke

• [ChangS77] (*p) Chang, S. and Nagy, G. "Deposit-Slip-First Check Reading", IEEE Transactions on Systems, Man, and Cybernetics, January 1977, pp 64-68

  • Operations in OCR: bank can control deposit slip format, but not checks

• [Chao77] (*p) Chao, Yao-ming "Recognition of handwriting Chinese Characters", 6.362 Student Paper, M.I.T., Fall 1977

  • direction codes / chain-codes, Chinese characters

• [Chen77] (*p) Chen, C.H. "Statistical Pattern Recognition - Review and Outlook", Systems, Man, and Cybernetics Review, Vol 6 No 4, August, 1977, pp 7-8

  • Feature extraction remains key problem in pattern recognition: features must be based on problem at hand
  • Context aids statistical pattern recognition

• [Chinnuswamy80] . Chinnuswamy, P. "Recognition of handprinted Tamil characters", Pattern Recognition, Vol 12, 1980, pp 141-152

  • Loy82 on Indian handwriting script (Tamil)

• [Coffin78] . Coffin, S. "Spatial frequency analysis of block letters does not predict experimental confusions", Perception and Psychophysics, Vol 23 No 1, 1978, pp 69-74

  • Human recognition of letters uses features, not mathematics: Fourier spatial frequency
  • Refers to edge and line detection/recognition in human vision, then
  • Position as something different from low-level features
  • Criticizes studies of human recognition that had visual distraction

• [Computerwoche77] (*) Computerwoche "Online-DE mit dem Bleistift: Datenerfassung als Nebensache", Computerwoche 21/1977, May 20, 1977

  • Quest Automation Datapad "Datapad puts your pencil Online to a computer". Dorset U.K. Handwriting recognition input on a paper form, 32-character LED display, uses Nova 16-bit minicomputer as controller. Statement is that data entry is done just on the side automatically while person is filling out regular paper form.

• [Computerwoche80] (*) Computerwoche "Quest Automation Ltd.: Handschriftleser fuer Direkterfassung", Computerwoche 16/1980, April 18, 1980

  • Quest Automation Datapad/Micropad, newer version of Datapad: 40-character display. Says less than one hour training required. Reference to booth at CeBIT exposition.

• [Cooper76] (*) Cooper, Leon N., and Elbaum, Charles "Information Processing System", United States Patent 3,950,733, assigned to Nestor Associates, New York, New York, April 13, 1976

  • Nestor patent: neural net, adaptive memory

• [Cooper77] . Cooper, Leon N., and Elbaum, Charles "Information Processing System", United States Patent 4,044,243, August 23, 1977, assigned to Nestor Associates, Stonington, Connecticut

  • Nestor Graphics patent

• [Cox76] . Cox, C.H. III and Coueignoux, P. "Concise Letter/Type Font Description: Theory and Computer Implementation", Internal report, Massachusetts Institute of Technology Industrial Liaison Program, Department of Electrical Engineering and Computer Science, November 20, 1976

  • OCR variability based on thinned "skeletons", generatively similar to chain-codes
  • Rules for line thinning, seraphs for typed characters

• [Cox78] . Cox, C., Blesser, B. and Eden, M. "Graphical Context of Printed Characters", Visible Language, Vol XII No 4, Autumn 1978, pp 428-447

  • Generative variability: points out existence of "rules" for how to write, even if not conscious

• [Crane77a] . Crane, H.D., and Savoie, R.E. "An On-Line Data Entry System for Hand-Printed Characters", IEEE Computer, pp 43-50, March 1977

  • Uses ULDR chain codes, similar to BLRT chain codes

• [Crane77b] . Crane, H.D., Ostrem, J.S., Wolf, D.E., and Wang, T.N.C. "A Technique for the Input of Handprinted Chinese Characters Based on Sequential Stroke Recognition", Proceedings of International Computer Symposium 1977, Taipei, Republic of China, Vol 1, pp 246-261

  • Describe drag force, not acceleration, of SRI pen
  • Kanji Chinese has Gaussian distribution of number of strokes, with 11 average
  • 881 characters in basic Kanji defined by Japanese ministry of education
  • Problem focuses on confusions, since did not have a large enough data base to get valid statistical results on recognition accuracy

• [Crane77c] . Crane, H.D, Wolf, D.E., and Ostrem, J.S. "The SRI pen system for automatic signature verification", Symposium Proceedings NBS Trends and Applications 1977, Gaithersburg, May 1977, pp 32-39

  • Liu,CN79
  • Obscured tip on SRI pen
  • SRI dynamic pen design: see Crane75.

• [Crane79a] . Crane, H.D. and Wolf, D.E. "Dynamic Re-creation of Signatures", United States Patent 4,156,911, assigned to Stanford Research Institute, Menlo Park, California, May 29, 1979

  • Signature verification by human eyeball: patent on compressing signature to minimum data without losing detail
  • Use of local extrema/chain codes to store written signatures with minimum memory

• [Crane79b] . Crane, H.D. and Wolf, D.E. "Handwritten signature verification system", United States Patent 4,086,567, April 25, 1978, assigned to Stanford Research Institute, Menlo Park, California

  • Signature verification by looking for landmarks (pen lifts)
  • Hew Crane: compressed-data method of storing signatures for human reading

• [Crane80] . Crane, H.D. and Wolf, D.E. "Signal Train Verification System Using Landmarks", United States Patent 4,190,820, February 26, 1980, assigned to Stanford Research Institute, Menlo Park, California

  • Elastic/rubbery matching of signature to segment into sections to be streatched or contracted for template matching
  • Landmarks in signatures which are invariant features

• [Crawshaw77] . Crawshaw, Martin, and Ottaway, Mary "A Contact-Pencil for Research on Writing", Quarterly Journal of Experimental Psychology, Vol 29, 1977 pp 345-346

  • Stylus using graphite/pencil lead to sense pen down/up.

• [Dasarathy78] . Dasarathy, B.V. and Kumar, K.P.B "Chitra: cognitive handprinted input-trained recursively analyzing system for recognition of alphanumeric characters", International Journal of Computing and Information Science, Vol 7, 1978, pp 253-282

  • Li89

• [DavisLS76] . Davis, Larry S. and Rosenfeld, A. "Applications of relaxation labeling: Spring-loaded template matching", Proceedings of 3rd International Joint Conference on Pattern Recognition, 1976, pp 591-597

  • Davis,LS77

• [DavisLS77] . Davis, Larry S. "Understanding shape: Angles and sizes", IEEE Transactions on Computers, Vol C-26, March 1977, pp 236-242

  • Cited in Teh89

• [DavisLS79] . Davis, Larry S. "Shape Matching using Relaxation Techniques", IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol PAMI-1 No 1, January 1979, pp 60-72

  • Relaxation process for matching shapes (coastlines) by figures of merits on pairs of angles (segment matching)
  • Cited in Baird84: on template matching in feature space using minimum distance

• [DIN77] . Deutsches Institut fuer Normung e.V. "Schrift B fuer die maschinelle optische Zeichenerkennung", DIN 66 009, September, 1977

  • German standard for OCR readable handwritten characters, similar to ANSI standard, but lists several as "not intended for OCR" (too hard to recognize?)

• [Donahey76] (*) Donahey, Alvin V. "Character Recognition System and Method", United States Patent 3,996,557, December 7, 1976, assigned to MI-squared Corporation, Columbus, Ohio

  • On-line character recognition using vertical conductive bar zones/region
  • Says does script, but only for separate/discrete characters

• [Donelson78] (*) Donelson, William C. "Spatial Management of Information", ACM Siggraph Conference, 1978, pp 203-209.

  • Gesture (hand-gesture?) user interface for viewing data. Summagraphics digitizer tablet with stylus for writing annotation (electronic ink?), Joypad touchscreen (resistive tablet) for user input? See also Donelson Master's thesis

• [DosterW77] . Doster, W. "Contextual postprocessing system for cooperation with a multiple-choice character recognition system", IEEE Transactions on Computers, Vol C-26, 1977, pp 1090-1101

  • Spelling dictionary look-up for context correction: in separate character OCR (handwriting) Multiple-Choice Single Character Recognition System and Contextual Postprocessing System

• [DosterW80] . Doster, W. and Schuermann, J. "An application of the modified Viterbi-algorithm used in text recognition", Proceedings of 5th International Conference on Pattern Recognition, 1980, pp 853-855

  • In Tappert's bibliography

• [Driscoll77] . Driscoll, R.J. et al "Real-Time Reading of Handwritten Symbols and Applications", Proceedings of the Conferenc eon Digital Processing of Signals in Communications, September 9, 1977, Institute of Electronic and Radio Engineers, pp 293-302

  • Cited in Sklarew patents

• [Druse78] . Druse, B. and Rao, C.V.K. "A matched filtering technique for corner detection", Proceedings of the 4th International Joint Conference on Pattern Recognition, Kyoto, Japan, November 7-10, 1978, pp 642-644

• [Du80] . Du, H.C. and Lee, R.C.T. "Symbolic Gray Code as a Multikey Hashing Function", IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol PAMI-2 No 1, pp 83-90, January 1980

  • Very fast searching using keys (brute-force engineering)

• [DunnRankin78] . Dunn-Rankin, Peter "The Visual Characteristics of Words", Scientific American, Januaray 1978, pp 122-130

  • Word recognition in humans (graphical word context)
  • Other papers describe inter-letter confusion in human perception of English alphabet: compare with Functional Attributes of Shillman

• [Duerr80] . Duerr, B., Haettich, W., Troph, H., and Winkler, G. "A combination of statistical and syntactical pattern recognition applied to classification of unconstrained handwritten numerals", Pattern Recognition, Vol 12, 1980, pp 189-199

  • Cited in Kerrick88

• [Dumas77] . Dumas, G. "A Dual Microprocessor Applications: The Alphabec-75 Data Capture System", in "Microcomputer Design and Applications", Academic Press, New York/San Francisco/London, 1977, pp 301-329

  • Mentions user learning/adaptation/training times and success rate: some learned fast, some never did (!)
  • Xebec/SRI/Crane original character recognition pen product

• [Duvernoy79] . Duvernoy, J. and Charraut, D. "Stability and stationarity of cursive handwriting", Pattern Recognition, Vol 11, 1979, pp 145-154

  • In Tappert's bibliography
  • Handwriting variability in cursive script

• [EerNisse77] . EerNisse, E.P., et al "Piezoelectric Sensor Pen for Dynamic Signature Verification", Conference 1988 Int. Electron Devices Meeting, Washington, D.C., December 1977, pp 473-476

  • Cited in Chainer85a

• [EerNisse78] . EerNisse, E.P., Land, C.E., and Snelling, J.B. "Input Apparatus for Dynamic Signature Verification Systems", United States Patent 4,078,226, assigned to the United States of America, March 7, 1978

  • Signature verification stylus pen using piezoelectric force/pressure sensors to sense X and Y forces/acceleration

• [Ehrich76] . Ehrich, Roger W. "A Readily Computable Decision Rule with Variable Dimensionality", IEEE Transactions on Computers, May, 1976, pp 539-542

  • Bayes and Meyman-Pearson statistical classifiers to select likelihood ratios
  • Features statistically best are generally "specialists" in distinguishing that class

• [Ehrich78] . Ehrich, Roger W. "Handwriting Recognition", Encyclopedia of Computer Science, Vol 9, 1978, Belzer, Holzman and Kent, editors, Dekker, 1978, pp 180-198

• [Elographics80] . Gibson, W., Talmage, J. "Nonplanar Transparent Electrographic Sensor", United States Patent 4,220,815, September 2, 1980, assigned to Elographics, Incorporated, Oak Ridge, Tennessee

  • Resistive-sheet digitizer using small spacing dots

• [Engdahl77] (*) Engdahl, Jean "Data Entry and Decoding System for Scripted Data", United States Patent 4,005,400, January 25, 1977, assigned to Societe Suisse pour l'Industrie Horologere Management Services S.A., Bienne, Switzerland

  • Four-segment panel for character recognition, zone-based recognition for handwriting, using a stylus

• [Engelbrecht76] . Engelbrecht, R.S. "Handwriting Identification Technique", United States Patent 3,962,679, June 8, 1976, assigned to RCA Corporation, New York, New York

  • Signature verification using writing speed/velocity

• [Farag79] (*) Farag, R.F.H. "Word-Level Recognition of Cursive Script", IEEE Transactions on Computers, Vol C-28 No 2, February 1979, pp 172-175

  • Sketch recognition of whole script words vs a table/word dictionary for context
  • Testing involved only a very small sample (ten words)

• [Feucht77] . Feucht, Dennis "Pattern Recognition: Basic Concepts and Implementations", Computer Design, December 1977, pp 57-68

  • Overview of pattern recognition: decision functions, multicategory classifiers (type I: simple, type II: pair-wise, type III: extended to all classes), hyperplanes and weight vectors in feature space, minimum-distance classifiers
  • Clustering techniques: maximin-distance, K-Means algorithms, general cluster-seeking algorithm.
  • Learning-adaptive recognition: perceptron, absolute increment, multicategory perceptron.

• [Fisher76] . Fisher, E. "The use of context in character recognition", COINS Tehnical Report 76-12 (Ph.D. thesis, Computer and Information Sciences Department, University of Massachusetts, Amherst), 1976

  • DosterW77 on spelling correction dictionary? Morse code recognition

• [Foster80] . Foster, D.H. and Mason, R.J. "Irrelevance of local position information in visual adaptation to random arrays of small geometric elements", Perception, Vol 9, 1980, pp 217-221

  • (BLRT chain codes) relative position not as significant as local features in human recognition

• [Fraser80] . Fraser, A. "Comments on 'A Study of Man-Machine Interaction Problems in Character Recognition'", IEEE Transactions on Systems, Man, and Cybernetics, Vol SMC-10 No 9, September 1980, page 589

  • Says Suen is full of it
  • Says variability in writing styles is caused by deep, fundamental human individuality
  • Points out "obvious" testing bias in Suen's paper
  • A graphologist speaks out on graphonomics handwriting research
  • Significant comment on dynamic character recognition by a graphologist
  • "neat", constrained handwriting is not realistic

• [Freeman77] . Freeman, H., and Davis, L.S. "A Corner-Finding Algorithm for Chain-Coded Curves", IEEE Transactions on Computers, Vol C-26, pp 297-303, March 1977

  • Adds "corner" detection to chain codes
  • Source of Barry's original "hi-curve" algorithm for corners?

• [Freeman78a] . Freeman, H. "Application of the generalized coding scheme to map data processing", Proceedings of IEEE Pattern Recognition and Image Processing Conference, Chicago, Illinois, May 31-June 2, 1978, pp 220-226

• [Freeman78b] . Freeman, H. and Saghri, A. "Generalized chain codes for planar curves", Proceedings of the 4th International Joint Conference on Pattern Recognition, Kyoto, Japan, November 7-10, 1978, pp 701-703

• [Frey77] . Frey, P.W. "Chess Skill in Man and Machine", Springer Verlag, New York, 1977

  • Survey of artificial intelligence ideas in computer chess

• [Fu77] . Fu, K.S. "A Brief Review of Pattern Recognition", Systems, Man, and Cybernetics Review, Vol 6 No 4, August, 1977, pp 3-5

  • Equivalence of decision-theoretic/statistical and syntactic pattern recognition

• [Fu80] . Fu, K.S. "Recent development in pattern recognition", IEEE Transactions on Computers, Vol C-29, October, 1980

  • Fu86/Fu83

• [Fujimoto76] . Fujimoto, Y., Kadota, S., Hayashi, S., Yamamoto, M., Yajima, S. and Yasuda, M. "Recognition of Hand-printed Characters by Nonlinear elastic Matching", Proceedings of the 3rd International Joint Conference on Pattern Recognition, November 1976, pp 113-118

  • Quotes Blesser, but says they never built an actual system
  • Uses eight-direction vector/length chain codes for elastic matching
  • Compare to Tappert's paper on elastic matching
  • Goshtasby88 cites that this system used continuity (continuation marks and hooks between characters) to get 99.7% recognition accuracy

• [Fujita76] . Fujita, T., Nakanishi, M., and Miyata, K. "The recognition of Chinese characters (kanji) using time variation of peripheral belt patterns", Proceedings of 3rd International Joint Conference on Pattern Recognition, Coronado, California, November 1976, pp 119-121

• [Gaertner80] . Gaertner, K.P. and Holzhausen, K.P. "Controlling air traffic with a touch sensitive screen", Applied Ergonomics, Vol 11 No 1, 1980, p 17-22

  • Beringer89: user-interface on electronic ink, integrated tablet/display
  • A clamp-on mechanical digitizer
  • Mechanical digitizer

• [Geographics80] . GeoGraphics "GeoGraphics Drafting Board Digitizer", GEOGRAPHICS Incorporated, 1318 Alms Drive, Champaign Ill. 61820, 1986

• [Genesereth79] . Genesereth, M.R. " The Use of Semantics in a Tablet-Based Program for Selecting Parts of Mathematical Expressions", MACSYMA User's Conference, 1979. V.E. Lewis, Ed.

  • Citation by Soiffer Kajler on citeseer.nj.nec.com/context/906109/0 mentions use of GUI for circling parts of mathematical expressions, using semantic information

• [Gilmore79] . Gilmore, G.C., Hersh, H., Caramazza, A. and Griffin, J. "Multidimensional letter similarity derived from recognition errors", Perception and Psychophysics, Vol 25, 1979, pp 425-431

  • Cited in IchikawaS84

• [Gould78] . Gould, J.D. "How experts dictate", Journal of Experimental Psychology: Human Perception and Performance, Vol 4, 1978, pp 648-661

  • Cited in Martin89

• [Greenaway78] (*) Greenaway, David Leslie "Apparatus and Method for Producing an Electrical Signal Responsive to Handwriting Characteristics", United States Patent 4,122,433, assigned to LGZ Landis and GYR Zug Ag, Zug, Switzerland, October 24, 1978

  • Pen stylus moves across a ridged/quadrangled surface while writing: resulting vibrations give and indication of pressure force and writing velocity

• [Greenberg77] . Greenberg, M.E. "An investigation of handwritten numeral character recognition", M.S. Thesis, Department of Electrical Engineering and Computer Science, August 1977, Massachusetts Institute of Technology, Cambridge, Massachusetts

  • Throws out data from IEEE Database 1.2.4 as "poorly written"

• [Groan78] . Groan, F.C.A. and Verbeek, P.W. "Freeman-code probabilities of object boundary quantized contours", Computer Vision, Graphics, and Image Processing, Vol 7, 1978, pp 391-402

  • Cited in Teh89

• [Grossman78] . Grossman, Robert "Hybrid pressure/force transducers give robots a sense of touch", EDN, August 5, 1978, pp 29-31

• [GroverDJ79] . Grover, D.J. "Graphics Tablets - A Review", Displays, July 1979, pp 83-93, IPC Business Press, Great Britain

  • Complete classification of methods of graphical/tablet interaction
  • Broad overview of digitizer tablet technologies

• [Guberman76a] . Guberman, S. and Litvin, Y. "Implementation of an Algorithm for Recognition of Handwritten Letters", Institute of Applied Mathematics, Academy of Sciences, USSR, TR-22, Moscow USSR (in Russian), 1976

  • Goodale83

• [Guberman76b] . Guberman, S. and Rozentsveig, V. "An Algorithm for Reading Handwritten Letters", (Automation and Remote Control), Vol 37, pp 751 ff

  • Goodale83

• [Haberman76] . Haberman, W. and Fejfar, A. "Automatic Identification of Personnel through Speaker and Signature Verification -- System Description and Testing", Proceedings of Carnahan Conference on Crime Countermeasures, May 1976, pp 23-30

• [HallPAV80a] . Hall, P.A.V. and Dowling, G.R. "Approximate string matching", Computing Surveys, Vol 12 no 4, December 1980, pp 381-402

  • Cited in Srihari83: on context/spelling correction

• [Hanaki76] . Hanaki, Shin-Ichi, Temma, Tsutomu and Yoshida, Hiroshi "An On-line Character Recognition Aimed at a Substitution for a Billing Machine Keyboard", Pattern Recognition, Vol 8, pp 63-71, January 1976

  • User-interface: early (1976) Nippon Electric prototype Katakana/Hiragana/Romanji product for data-entry: features chosen by "human intuition"
  • Describes what appears to be oversampling, resampling, noise filter on digitizing tablet (Telemail product)
  • Resampling of handwritten tablet points to be minimum number of feature points for reconstruction/recognition: also used for data reduction/compression of line drawings
  • Character segmentation by stroke time-out, writing in boxes, recognition result
  • Shows shading pattern for boxes to write in
  • 98% recognition rate, using decision tree node
  • Refers to handwriting character segmentation based on recognition result/plausibility

• [Hanson76] . Hanson, A.R., Riseman, E.M., and Fisher, E. "Context in Word Recognition", Pattern Recognition, Vol 8, 1976, pp 35-45

  • Low character error rates can give high word error rates
  • Use context on words to correct errors on single characters
  • "many theoretical systems are impractical in application"
  • Cites use of "clear text" in context: but many applications are not (data entry)
  • Tri-gram statistical Markovian context probabilities, not linguistic grammar context
  • Cites problems of small training sets (98 characters per class)
  • High (context) correction rates introduce many substitution errors
  • The number of substitution errors is about the same as reject recognition
  • Context after the recognition is done, not before

• [Hattich78] . Hattich, W., Tropf, H., and Winkler, G. "Combination of statistical and syntactical pattern recognition -- Applied to classification of unconstrained handwritten numerals", Proceedings of Fourth International Conference on Pattern Recognition, Kyoto, Japan, November 7-10, 1978, pp 786-788

  • Cited in Kahan87

• [Herbst76] . Herbst, Noel M., and Morrissey, John H. "Verifier for Handwritten Signatures", United States Patent 3,983,535, assigned to IBM Corporation, September 28, 1976

  • Cited in Chainer85b: for acceleration on signature verification
  • Signature verification using acceleration.

• [Herbst77a] (*) Herbst, Noel Martin and Morrissey, John Henry "Segmentation Mechanism for Cursive Script Character Recognition Systems", United States Patent 4,024,500, May 17, 1977, assigned to IBM Corporation, Armonk, New York

  • Cursive character segmentation using sharp inflection points (zero velocity)
  • Also makes use of baseline and midline (top of lower-case characters) for character segmentation
  • Criticizes "retrospective" character segmentation in script, such as Harmon

• [Herbst77b] . Herbst, N.M., et al "Signature Verification Based on Complete Accelerometry", IBM Technical Disclosure Bulletin, Vol 19 No 12, May 1977, pp 4827-4828

  • Stylus for digitizer, no tablet: uses acceleration (chain codes? compare to SRI pen)

• [Herbst78] . Herbst, N.M. and Liu, C.H. "Signature Verification Method and Apparatus Utilizing both Acceleration and Pressure Characteristics", United States Patent 4,128,829, December 5, 1978, assigned to IBM Corporation Armonk, New York

  • Signature verification: force/pressure and acceleration
  • (Our BLRT chain codes): position features subject to most variability
  • Separate X and Y acceleration for signature verification

• [Herbst79a] . Herbst, N.M., Liu, C.N., and Panissidi, H.A. "Pressure sensing device and transducer arrangement", United States Patent 4,142,175, February 27, 1979, assigned to IBM Corporation

  • Cited in Chainer85b: for acceleration on signature verification

• [Herbst79b] . Herbst, Noel M., Liu, Chao N., and Panissidi, Hugo A. "Pressure Sensing Device and Transducer Arrangement", United States Patent 4,142,175, assigned to International Business Machines Corporation, Armonk, New York, February 27, 1979

  • Signature verification: pressure/force sensing stylus, not sensitive to lateral force angle: combined with acceleration for signature verification

• [Herot76] . Herot, C.F. "Graphical Input through Machine Recognition of Sketches", Computer Graphics, Vol 10 No 2, Summer 1976, pp 97-102

  • Cited in FoleyJD82

• [Hilbrink78] (*) Hilbrink, Johan O. " System for Optically Entering, Displaying and Decoding Handwritten Symbols", United States Patent 4,122,415, September 5, 1978

  • Lightpen and television monitor display, when used for handwriting input and replacing the handwriting input with a generated character

• [Himmel78] . Himmel, David P. "Some Real-World Experiences with Handprinted Optical Character Recognition", IEEE Transactions on Systems, Man, and Cybernetics, Vol SMC-8 No 4, April 1978, pp 288-292

  • How much effort it takes to train people to handwriting constraints (OCR)
  • Shows samples of constrained writing people were actually trained to

• [Hollerbach78] . Hollerbach, J.M. "A Study of Motor Control Through Analysis and Synthesis of Handwriting", PhD Thesis, Massachusetts Institute of Technology Department of Electrical Engineering and Computer Science, August, 1978. Also described in Visible Language, Vol XIII No 3, 1979, pp 252-264

  • Study of stroke directions based on anatomy of muscles
  • Points out elimination of clock-wise movement in fast writing

• [Hong82] . Hong, T.H., Shneier, M., Hartley, R. and Rosenfeld, A "Using pyramids to detect good continuation", University of Maryland, Computer Science TR 1185, 1982

  • Cited in Lipscomb91

• [Hosaka77] . Hosaka, M. and Kimura, F. "An Interactive Geometrical Design System with Handwriting", Proceedings of the 6th IFIPS Congress, Roronto, August 1977, pp 167-171

  • User-interface: handwritten sketches, sketch recognition, gesture input to do drawings
  • Refers to measuring problems of tablet (and of "human behavior") (human factors / variability)
  • Uses second-level fix-up on recognition if first pass is inconclusive
  • Uses chain codes of inflection points, rotations, straight segments, etc.

• [Hoshino77] . Hoshino, Y. "Word recognition apparatus", United States Patent 4,010,445, March 1977

  • Suen79

• [Hubel79] . Hubel, D.H. and Wiesel, T.N. "Brain Mechanisms of Vision", Scientific American, No 241, 1979, pp 150-162

  • Martin,GL90b

• [IBM79] (*) IBM "Object Identification and Local Highlighting on Displays", IBM Technical Disclosure Bulletin, Jan 1979, p 3284.

  • Compare with object highlighting via proximity tablet data of "GUI-handles" in a GUI: mentions getting local highlighting of an object on a display as the stylus goes over it, and returning a ID/handle to the application instead of X/Y coordinates. Mentions equivalent functions from lightpen, tablet, joystick, or other positioning (position input) device.

• [IchikawaT77] . Ichikawa, Tadao, Sakamura, Ken, and Aiso, Hideo "ARES - A memory, capable of associating stored information through relevancy estimation", Proceedings of 1977 National Computer Conference, pp 947-954

  • Seven-direction chain codes/ segment lengths, with an additional approximate string matching function "Lee Distance" from error-correcting codes (in hardware)

• [IkedoT78] (*) Ikedo, Tsuneo "Automatic Coordinate Determining Device", United States Patent 4,088,842, May 19, 1978

  • Electromagnetic digitizer tablet that can handle multiple styli (multi-touch), and a cordless/wireless probe or stylus

• [Impedovo76] . Impedovo, S., Marangelli, B., and Plantamura, V.L. "Real-Time Recognition of Handwritten Numerals", IEEE Transactions on Systems, Man, and Cybernetics, Vol 6 No 2, February 1976, pp 145-148

  • 65% to 90% recognition, numerals only (!)
  • Gives examples of handwriting variations: stroke directions

• [Ito78] . Ito, M.R. and Chui, T.L. "On-line Computer Recognition of Proposed Standard ANSI (USASI) Handprinted Characters", Pattern Recognition, Vol 10, 1978, pp 341-349

  • Claims 98.3% accuracy
  • A non-training system (non-adaptive)
  • It's not just tablet digitizer noise: unsteadiness in user's pen motion
  • Eight-type chain code classifications
  • Straight-line vs curvilinear stroke classes
  • Tests use "intentional variability" in characters by authors
  • Cites mis-recognition due to missing stroke segments (K-R, J-T)
  • Delay in recognition same as time for user to look at screen: not a user-interface/human-factors problem
  • Shows bizarre four-stroke "Z"
  • Shows proposed ANSI hand-print standard included stroke directions, constrained writing styles

• [Jarvis76] . Jarvis, J.F. "Regular expressions as a feature selection language for pattern recognition", Proceedings of the 3rd International Joint Conference on Pattern Recognition, Coronado, California, November 1976, pp 189-192

• [Jarvis77] . Jarvis, J.F. "The line drawing editor: Schematic diagram editing using pattern recognition techniques", Computer Graphics Image Processing, Vol 6, 1977, p. 452-484

  • Nagura83: user interface on editing drawings?

• [Kanal77] . Kanal, Laveen N. "Current Status, Problems and Prospects of Pattern Recognition", Systems, Man, and Cybernetics Review, Vol 6 No 4, August, 1977, pp 9-14

  • Overview of pattern recognition: many theories, few products: term covers too many divergent problem areas.
  • Pattern recognition overview: well-constrained problems, sustained efforts lead to successful performance.
  • Realistic performance: "Identifying animals in a zoo is much different from hunting them in the forest."
  • Overview: spatial/temporal/external context, feature extraction (most publicly undocumented: not published), experiment/training design, clustering, no standard data bases

• [Kaplow80] (*) Kaplow, R. and Molnar, M.K. "Dynamically variable keyboard system", United States Patent 4,202,041, May 6, 1980 assigned to Massachusetts Institute of Technology, Cambridge, Massachusetts, 1980

  • Dynamically variable keyboard, using digitizer and simulated function keys/buttons: simulated devices, compare with simulated devices for Shumer, Fingerworks, Buxton. User-interface where icons/keyboard labels on touch-screen force/pressure-sensitive digitizer change dynamically. Possible citation for Gesture input? Application Generator?

• [Kato80] . Kato, O, Fujita, T., Niwa, M., Morishita, T., Fujii, N., Tanahashi, J. "Handwriting Input System for Japanese", Proceedings of IFIP 80, Tokyo, Japan, October 6-9, 1980, North-Holland publishers, pp 689-694

  • Science Citation Index. Features are stroke-relative position, then candidate selection, then personal training. Claims 98% recognition rate performance on Japanese Kanji/kana/numerics, after training.

• [Kay77a] . Kay, A.C. "Microelectronics and the Personal Computer", Scientific American, September 1977, page 231

  • The original user-interface paper: Dynabook (does NOT mention electronic ink or character recognition)

• [Kay77] (*) Kay, A.C. and Goldberg, A. "Personal Dynamic Media", IEEE Computer, Vol 10 No 3, March 1977 pp 31-41

  • Dynabook did NOT have stylus or handwriting input: mouse only .. popularly assumed to be the original "electronic ink" handwriting paper: Dynabook, but not so.

• [Kegel78] . Kegel, A.G., Giles, J.K., and Ruder, A.H. "Observations on Selected Application of Optical Character Readers for Constrained Numeric Handprint", IEEE Transactions on Systems, Man, and Cybernetics, Vol SMC-8 No 4, April 1978, pp 282-285

  • Most handwriting recognition OCR problems are over-printing, 0-vs-6 ambiguity
  • How well human writers match constraints for various commercial OCR handwriting recognition products

• [Kickert76] . Kickert, Walter J.M. and Koppelaar, Henk "Application of Fuzzy Set Theory to Syntactic Pattern Recognition of Handwritten Capitals", IEEE Transactions on Systems, Man, and Cybernetics, Vol 6 No 2, February 1976, pp 148-ff

• [Kley78] . Kley "Graphical data entry pad", United States Patent 4,079,194, March 14, 1978

  • Lukis87 on digitizer tablets

• [Knowlton77] . Knowlton, K.C. "Computer Display Optically Superimposed on Input Devices", Bell System Technical Journal, Vol 56 No 3, March 1977, pp 367-384

  • User-interface: project image on keyboard to change keys: refers to tablet input possibility; compare with Kaplow

• [Knowlton80] . Knowlton, K. "Progressive Transmission of Grey-Scale and Binary Picture by Simple, Efficient, and Lossless Encoding Schemes", Proceedings of the IEEE, Vol 68 No 7, July 1980, pp 885-896

  • Compression of images

• [Kuipers76] . Kuipers, Jack "Tracking and Determining Orientation of Object Using Coordinate Transformation Means, System and Process", United States Patent 3,983,474, assigned to Polhemus Navigation Sciences, Incorporated, Burlington, Vermont, September 28, 1976

  • From a old list of Polhemus patents
  • Patent on correction tables on stylus angle, orientation, position in a 3-D tablet digitizer

• [Kuipers77] (*) Kuipers, Jack "(no title)", United States Patent 4,017,858, assigned to Pohemus Nagivation Sciences, April 12, 1977

  • Cited in MaedaK87
  • three-dimensional / six dimensional electromagnetic digitizer, using X/Y/Z coils. Determines both position and orientation of stylus, thus able to report not only the position on a (three-dimensional) surface, but also the plane of the surface at that point.

• [Kuklinski77] . Kuklinski, T. "Implementing a Functional Attribute Based Theory of Character Recognition", unpublished manuscript, Pencept, Incorporated, March 21, 1977

  • Kuklinski voting algorithm for an imperfect decision tree based on pair-wise voting

• [Kuklinski79] . Kuklinski, T., PhD Thesis: "Graphical Context as an Aid to Character Recognition", Massachusetts Institute of Technology, Cambridge, Massachusetts, 1979

• [Kwan79] . Kwan, C.C., Pang, L., and Suen, C.Y. "A Comparative Study of some Recognition Algorithms in Character Recognition", Proceedings of International Conference on Cybernetics and Society, pp 530-535, October 1979, IEEE order number CH1424-1/79/0000-0530

  • Boundless variability of hand-print

• [Lambden78] (*) Lambden, Martin Roy "Electrographic Apparatus and HandRest for Use Therewith", United States Patent 4,129,746 assigned to Quest Automation Limited, Stapehill, Wimborne, England, December 12, 1978.

  • Hand rest for pressure-senstive / resistive-film tablet, so that hand does not press on tablet surface

• [Lambden78b] (*) Lambden, Martin Roy "Electrographic Apparatus and Method Of Producing and Electrode Surface Therefor", United States Patent 4,070,544 assigned to Quest Automation Limited, Stapehill, Wimborne, England, January 24, 1978.

  • Description of the resistive-film, plastic pressure sheet digitizer tablet used for the Quest handwriting product

• [Larsen77] . Larsen "Quadrant Check for Signature Verification", IBM Technical Disclosure Bulletin, Vol 20 No 4, September 1977, pp 1538-1539

  • Accelerometer pen/stylus design (like SRI pen), produces chain codes

• [Lew79] (*p) Lew, J.S. "Optimal Designs of Instrumented Pens for Signature Verification", IBM Technical Disclosure Bulletin, Vol 21 No 8, p. 3415, January 1979

  • Variation on SRI (non-digitizer) pen: accelerometer pen stylus design

• [Liljenwall79] (*) Liljenwall, James and Moss, David "Information Entry System", United States Patent 4,139,837, assigned to Creative Ventures, Incorporated, Dayton, Ohio, 1979

  • On-line character recognition on the face of a watch or calculator : single-stroke characters, written with a fingertip
  • Simple on-line character recognition using zones/regions

• [Lin78] (*p) Lin, W.C. and Pun, J.H. "Machine Recognition and Plotting of Hand-Sketched Line Figures", IEEE Transactions on Systems, Man, and Cybernetics, Vol SMC-8 No 1, January 1978, pp 52-57

  • Symbol recognition and graphical editing: "prettyizing" a sketch : shows recognition of single-stroke circuit-diagram symbols
  • "recognition -- seems to be a problem at times since the user cannot have complete control over the input data when sketching on the tablet"

• [Lin79] . Lin, C.N., Herbst, N.M., and Anthony, N.J. "Automatic signature verification: System description and field test results", IEEE Transactions on Systems, Man, and Cybernetics, Vol 9, 1979, pp 35-38

  • Cited in Teulings86a

• [LiuCN79] (*) Liu, C.N., Herbst, N.M., and Anthony, N.J. "Automatic Signature Verification: System Description and Field Test Results", IEEE Transactions on Systems, Man, and Cybernetics, Vol SMC-9 No 1, January 1979, pp 35-38

  • Signature verification system from IBM using force/pressure and accelerometry
  • Signatures: reference signature modified before storage to deal with variability
  • Signature verification: 1.7% Type I (false rejection) and 0.4% Type II (false acceptance) for deliberate (expert) forgers, false acceptance rate of 0.02% for random (amateur) forgers
  • Signature verification: 1% of users had too much variability for the system to work well: solution is administrative
  • Signature hardware problems: baseline drift in force/pressure sensor, noise in force/pressure sensor, lack of reset, non-uniformity of ink cartridges

• [MatsushitaT78] (*x) Matsushita, T. et al "Personal Computer-Aided Composition and Editing System", 3rd USA-Japan Computer Conference, 1978, 10/10-12/78

  • cited for text-editing of handwritten marks, graphical editing

• [Maugh78] . Maugh, Thomas H. II "Holographic Filing: An Industry on the Verge of Birth", Science, Vol 201, August 4, 1978, pp 431-432

  • Cited in Sziklai84: on storing optical images, with signatures

• [McDermott80] . McDermott, J. "R1: An Expert in the Computer Systems Domain", Proceedings of the First Annual National Conference on Artificial Intelligence, 1980, pp 269-271

  • Expert A.I. system to configure orders for DEC VA computers

• [Mihelic77] . Mihelic, F.; Pavesic, Nokola and Gyergyek, Ludvik "Recognition of Writer of Handwritten Texts", Proceedings of the International Conference on Crime Countermeasures, Kentucky University, Lexington, 1977, pp. 237..240

  • Cited by Marlin Eller, Microsoft Pen Computing group

• [Milloy78] (*p) Milloy, D.G. "Comment on recognition and confusion of the lower-case alphabet", Perception and Psychophysics, 1978, Vol 24 No 2, pp 190-191

  • Substitution errors in human recognition of lower-case characters
  • Criticizes defects in everybody's studies of lower-case character confusion matrices for human recognition

• [Moayer76] . Moayer, B. and Fu, K.S. "A tree system approach for fingerprint pattern recognition", IEEE Transactions on Computers, Vol C-25, 1976, pp 262-274

• [MoriS80] . Mori, S. and Suen, C.Y. "A Study of the Zero and Letter Oh Problem", Proceedings of 5th International Conference on Pattern Recognition", December 1980, pp 842-847

  • Cited in Litvin82

• [Morrissey76] (*) Morrissey, J.H. "Electronic Calculator Based on Character Recognition of Input from Stylus Acceleration Dynamics", IBM Technical Disclosure Bulletin, Vol 19 No 7, December 1976, pp 2816-2817

  • Disclosure of US patent 3,983,535 for sigature verification, but says extension to input for handwriting recognition is obvious. On user-interface of writing on an electronic calculator (using chain-codes/acceleration dynamics)

• [Muth77] . Muth, F.E. and Tharp, A.L. "Correcting human error in alphanumeric terminal input", Information Processing Management, Vol 13, 1977, pp 329-337

  • Cited in Srihari83: on context/spelling correction

• [Myers80] . Myers, C.S., Rabiner, Lawrence R., and Rosenberg, A.E. "Performance tradeoffs in dynamic time warping algorithms for isolated word recognition", IEEE Transactions on Acoustics, Speech and Signal Processing, Vol ASSP-28, 1980, pp 622-635

  • Kruskal83

• [Narasimhan80] (*p) Narasimhan, M.A., Devarajan, Venkat, and Rao, K.R. "Simulation of Alphanumeric Machine Print Recognition", IEEE Trans. on Systems, Man, and Cybernetics, Vol SMC-10 No 5, May 1980, pp 270-275

  • Shows performance results in a confusion matrix for alphabet from OCR data, compare with Shillman, who uses a confusion matrix to show which features are distinguishing

• [Naus76] (*p) Naus, M.J., and Shillman, R.J. "Why a Y Is Not a V: A New Look at the Distinctive Features of Letters", Journal of Experimental Psychology: Human Perception and Performance, Vol 2 No 3, 1986, pp 394-400

  • Shillman article on functional attributes

• [Niemann77] (*p) Niemann, Heinrich "Classification of Characters by Man and Machine", Pattern Recognition, Pergamon Press 1977, Vol 9, pp. 173-179

  • Human performance on OCR recognition of handwritten characters compared with machine performance, under various noise conditions: machine works better on typed fonts, human on handwritten characters

• [O'Brien79] . O'Brien, M.T. "A Network Graphical Conferencing System", Rand Corporation, Stanta Monica, California, August 1979 (N-1250-DARPA)

  • Real-time freestyle user interface for electronic ink: a central blackboard/whiteboard conferencing / drawing communally, separate windows for each user for typing (words).

• [Odaka79] (*p) Okada, Jazumi, and Masua, Isao "Some Results on Stroke Order Free On-line Character Recognition", PRL Vol 79 No 29, pp 77-84 (in Japanese)

  • stroke-order independence by grouping strokes which are close together (?)

• [Odaka80] (*p) Okada, Jazumi, and Masua, Isao "Online Recognition of Handwritten Characters by Approximating Each Stroke with Several Points", Trans. IECE, 1980, Vol 63-D, No 2, pp 153-160 (in Japanese)

  • See same-title paper in IEEE SMC, Nov/Dec 1982

• [OkaM78] . Oka, Masatomo and Yasuhara, Makoto "Signature Verification Experiment based on Nonlinear Time Alignment: A Feasibility Study", IEEE Trans on Systems, Man and Cybernetics, Vol SMC-7, No 3, pp 212-216.

  • From Kuklinski's old files
  • Dynamic time-warping / non-linear time alignment via dynamic programming for signature verification

• [Okuda76] . Okuda, T., Tanaka, E. and Kasai, T. "A method for the correction of garbled words based on the Levenshtein metric", IEEE Transactions on Computers, Vol C-25, February, 1976, pp 172-178

  • DosterW77 on spelling dictionary?

• [Pallishusky76] (*) Pallishusky, Silvia Minka "Method and Means for Teaching Alphabet Recognition and Learning Handwriting", United States Patent 3,950,863, April 20, 1976

  • Overlaying transparencies to teach children the order of strokes for learning handwrigin
  • Refers to nine basic shapes, or segments, from which all 26 letters (52 with lower case) are constructed

• [Pavlidis77a] . Pavlidis, Theo "Structural Pattern Recognition", Springer-Verlag, New York, 1977

  • Defines pattern recognition as "identification of the ideal which a given object is made after"
  • Definition: classification theory vs boundary theory

• [Pavlidis77b] . Pavlidis, Theo "Comments on Current Perspectives in Pattern Recognition", Systems, Man, and Cybernetics Review, Vol 6 No 4, August, 1977, pp 8-9

  • Just because humans have good pattern recognition performance, and poor arithmetic, does not mean pattern recognition is easier than computer arithmetic

• [Pavlidis77c] . Pavlidis, Theo "Polygonal approximations by Newton's method", IEEE Transactions on Computers, Vol C-26, 1977, pp 800-807

• [Pavlidis80] . Pavlidis, Theo "Algorithms for shape analysis and waveforms", IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol PARM-2, July 1980, pp 301-312

  • Chain codes, corner detection, curvature, Fourier shape, shape analysis for OCR image recognition (segmented pictures)

• [PepperW78] (*) Pepper, William Jr. "Human-Machine Interface Apparatus", United States Patent 4,071,691, January 31, 1978, assigned to Peptek, Incorporated, Bethesda, Maryland

  • Phase-detection digitizer, resistive film responds to finger touch: compare to Scriptel?

• [PepperW78a] (*) Pepper, William Jr. "Human-Machine Interface Apparatus", United States Patent 4,129,747, December 12, 1978

  • Cited by Sklarew88b on user interfaces on touchscreens
  • touch-screen digitizer using phase delay of a signal in a resistive film with capacitive coupling

• [PepperW80] . Pepper, William Jr. "System for producing electric field with predetermined characteristics and edge terminations for resistance planes therefor", United States Patent 4,198,539, April 15, 1980

  • Resistive-film digitizer using non-linear and three-dimensional electric fields, two voltage sources

• [Persoon77] . Persoon, E. and Fu, K.S. "Shape discrimination using Fourier descriptors", IEEE Transactions on Systems, Man and Cybernetics, Vol 7, 1977, pp 171-179

  • Taxt90: Pavlidis80 says 95% on IEEE handwriting recognition database

• [Peterson80] . Peterson, J.L. "Computer Programs for Detecting and Correcting Spelling Errors", Communications of the A.C.M., Vol 23 no 12, December 1980, pp 676-687

  • Cited in Sinha88: on spelling context correction
  • Has many reference on spelling correction, detection, best-match to strings, limits of lack of syntactic and/or semantic context

• [Price79] . Price, K. and Reddy, R. "Matching segments of images", IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol PAMI-1, January, 1979, pp 110-116

  • Burr81

• [Prugh80] . Prugh, R.W. and Fadden, B.J. "Graphic Digitizer", United States Patent 4,206,314, June 3, 1980, assigned to GTCO Corporation, Rockville, Maryland

  • Digitizer with grid as transmitter, loop as the receiver

• [Purcell77] (partial) Purcell, Stephen C. "Understanding hand-printed Algebra for computer tutoring", Massachusetts Institute of Technology, Artificial Intelligence Laboratory Memo No 445, February 1977 (also available as S.M. Thesis)

  • Context in parsing and recognizing mathematical expressions with on-line handwriting recognition
  • "chalkboard languages" - two-dimensional doodles, writing, pictures
  • Points out that overlap is inadequate as the only criterion for parsing/segmentation in handwriting (two-dimensional)
  • Uses a lattice representation of possible stroke groupings/segmentations of characters, showing ambiguities of parsing (vs. of recognition)
  • Features: local geometry (curvature, etc.) not enough: need global geometry (closure of endpoints on "U" vs. "O", etc.)
  • Variability: greater when humanswritein context, vs one character at a time
  • Recognizer mentioned by Alan Kay in email posting on www.squeakfoundation.org, November 2000

• [Rabiner78] . Rabiner, Lawrence R., Rosenberg, A.E., and Levinson, S.E. "Considerations in dynamic time warping for discrete word recognition", IEEE Transactions on Acoustics, Speech, and Signal Processing, Vol ASSP-26, 1978, pp 575-582

  • Kruskal83

• [Rabiner79] . Rabiner, Lawrence R., and Wilpon, Jay G. "Speaker-Independent Isolated Word Recognition for a Moderate Size (54 Word) Vocabulary", IEEE Transactions on Acoustics, Speech, and Signal Processing, Vol ASSP-27 No 6, pp 583-587

  • Discrete word speech recognizer using statistical clustering templates: a carefully-trained speaker-independent limited recognizer can work as well as casually-trained speaker-dependent
  • Speech recognition on a professionally trained actor with very human-clear diction was worse
  • Clustering procedure for voice templates much better than casual enrollment
  • Describes many enrollment/training methods for speaker-independent/dependent recognition system
  • Training/enrollment data was "cleaned up" to remove breathing sounds, clicks, etc.

• [Rabiner80] . Rabiner, Lawrence R. and Schmidt, C.E. "Application of dynamic time warping to connected digit recognition", IEEE Transactions on Acoustics, Speech, and Signal Processing, Vol ASSP-28, 1980, pp 337-388

  • Kruskal83

• [Radice80] . Radice, P.F. "Personal Verification Device", United States Patent 4,234,868, November 18, 1980, assigned to Penwalt Corporation, Philadelphia, Pennsylvania

  • Signature verification using force/pressure biometric only. "force/pressure only" digitizing tablet using piezoelectric or pyro-electric sheet or film

• [RahuelJC80] (*) Rahul, Jean-Claude and Dagnelie, Jean-Paul "United States Patent ", United States Patent 4,225,750, September 30, 1980

  • X/Y electromagnetic grid tablet digitizer operating at 10 Megahertz, modulated at 100 Kilohertz. Cites application as a telewriting (telautograph?) device. Cites to 1973 references for electronic whiteboard/blackboard systems.

• [Reenskaug79] (*) Reenskaug, Trygve "A note on Dynabook requirements", Xerox PARC memo, 22 March 1979: partial copy available on the Internet

  • Automatic layout / structuring of graphical objects with connections: note that it does not mention handwriting recognition or character recognition?
  • Refers to Prokon
  • Anticipates market for Dynabook-like products about 1982-1986, programming in a higher-level language (e.g. APL, Pascal)

• [Rosenfeld78] . Rosenfeld, Azriel "Survey: Picture Processing: 1977", (source not known) 1978

  • Not sure of source: Check Kuklinski's thesis?
  • Large bibliography on picture processing, filtering, edge detection, image compression, character recognition and other kinds of pattern recognition (fingerprint, face, etc.), machine vision

• [Rosenfeld80] . Rosenfeld, A. "Picture processing: 1980", Computer Graphics and Image Processing, Vol 16, May, 1981

  • Fu86/Fu83

• [Rubincam79] (*) Rubincam, David P. "Electronic Book", United States Patent 4,159,417, June 26, 1979

  • Cited in Sklarew88b: on a portable electronic book, with user interface (no computer applications): book text contained on memory card, shown in display. Hinged notebook-like design. Compare with Dynabook by Alan Kay?

• [Sakoe76] . Sakoe, Hiroaki "Automatic Character Recognition Device Employing Dynamic Programming", United States Patent 3,979,722, September 7, 1976, assigned to Nippon Electric Co. Limited, Tokyo Japan

  • Refers to "standard patterns" for template matching (which we is not)
  • Eight-direction chain codes
  • Uses dynamic matching of vector maxima/minima for DCR (BLRT chain codes)
  • BLRT points are absolute, vary distances between for degree of match

• [Sakoe78] (*p) Sakoe, Hiroaki and Chiba, S. "Dynamic programming algorithm optimization for spoken word recognition", IEEE Transactions on Acoustics, Speech and Signal Processing, Vol ASSP-26, 1978, pp 43-49

  • Kruskal83

• [Sakoe79] (*p) Sakoe, Hiroaki "Two-Level DP-Matching -- A Dynamic Programming-Based Pattern Matching Algorithm for Connected Word Recognition", IEEE Transactions on Acoustics, Speech and Signal Processing, Vol ASSP-27, NO 6, 1979, pp 588-ff

• [Seelbach80] . Seelbach, H.E. "Input Device for Input of Alphanumeric Characters into a Computer", United States Patent 4,184,147, January 15, 1980

  • Simple four-grid "character recognition" device: sequence of crossings

• [Sethi77] . Sethi, I.K. and Chatterjee, B. "Machine recognition of constrained handprinted Devanagari", Pattern Recognition, Vol 9, pp 69-75, 1977

  • Biswas81

• [Seuffert77] . Seuffert, P. "An Application of Line and Character Recognition in Cartography", Proceedings of IEEE Conference on Pattern Recognition and Image Processing, pp 337-342, June 1977

  • Goodale83

• [Shillman76a] . Shillman, R. "Automatic Recognition of Thick Stroke Characters", Quarterly Progress Report No 118, Appendix IX, Massachusetts Institute of Technology Research Laboratory of Electronics, July 1976

  • How "legs" appear shorter as they get thicker

• [Shillman76b] . Shillman, R. and Naus, G. "The Distinctive Features of the Letters O and D", Quarterly Progress Report No 118, Appendix X, Massachusetts Institute of Technology Research Laboratory of Electronics, July 1976

• [Shillman76] . Shillman, R., Kuklinski, T. and Blesser, B. "Psychophysical Techniques for Investigating the Distinctive Features of Letters", International Journal of Man-Machine Studies, Vol 8, pp 195-205, 1976

  • Classical letter shapes do not occur frequently in hand-printing
  • Shillman article on functional attributes

• [Shillman77] . Shillman, R.J. and Babcock, R.T. "Preliminary Steps in the Design of Optical Character Recognition Algorithms", unpublished manuscript, Cognitive Information Processing Group, Research Laboratory of Electronics, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, 1977

  • Physical to functional rules (functional attributes) for 2 and Z at Massachusetts Institute of Technology (see Babcock)
  • Example of 2-Z space, one space invading the other
  • Good introductory example of physical to functional rules, SIZE=-1> how to pick the dominant feature for a character

• [Shinghal79a] . Shinghal, R. and Toussaint, G.T. "A bottom-up and top-down approach to using context in text recognition", International Journal of Man-Machine Studies, Vol 11, 1979, pp 201-212

  • Cited in Sinha88: on spelling/dictionary context correction

• [Shinghal79] . Shinghal, R. and Toussaint, G.T. "Experiments in text recognition with modified Viterbi algorithm", IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol PAMI-1, 1978, pp 184-192

  • Cited in Sinha88: on spelling/dictionary context correction

• [Shortliffe76] . Shortliffe, E.H. "Computer-Based Medical Consultation: MYCIN", American Elsevier, New York, 1976

  • Expert A.I. system to diagnose infectious blood diseases
  • Cited in FoleyJD84

• [SIGGRAPH79] . SigGraph "General Methodology and the Proposed Core System", report of the Graphics Standards Planning Committee, in Computer Graphics (Proc. of SIGGRAPH 79), Vol 13 No 3, August 1979

• [Simmons80] (*) Simmons, Robert M. "An On-line Character Recognizer", Interface Age, March 1980, pp 110-114

  • Very simple on-line handwriting character recognition program running on a 1980 Z-80 CP/M system
  • effectively trims ends of stroke by looking for matching chaincode sequence within longer chain-code sequences
  • Zone-based recognizer, using chain of codes for the zones visited: single-stroke characters?

• [SMC78] . (various authors) "Special section on hand-print OCR", IEEE Transactions on Systems, Man, and Cybernetics, Vol SMC-8 No 4, pp 279 ff, April 1978

  • Special section on real-world application experience with hand-print OCR, vs laboratory results

• [Snyder77] . Snyder, Edward J. and Domyan, Stephen L. "Position coordinate determination device", United States Patent 4,018,989, assigned to Summagraphics Corporation, April 19, 1977

  • Lukis87 on digitizer tablets

• [Spanjersberg78] (*) Spanjersberg, A.A. "Experiments with Automatic Input of Handwritten Numerical Data into a Large Administrative System", IEEE Trans. on Systems, Man, and Cybernetics, Vol SMC-8, No 4, April 1978, pp 286-288

  • Shows various contraint methods on handwriting styles (boxes, zones, write around the dots, marked-sense card) to make handwriting more recognizable by machine, gives some performance results

• [Spohrer80] (*p) Spohrer, James C., Brown, Peter F, Hochschild, Peter H., and Baker, James K. "Partial Traceback in Continuous Speech Recognition", IEEE publication 0360-0913/80/000/0036

  • Verbex Corporation, formerly Dialog Systems Inc, Two Oak Park, Bedford MA 01730
  • use of context in continuous speech recognition

• [Srihari76] . Srihari, S. and Zack, G. "Document Image Analysis", Proceedings of 8th International Conference on Pattern Recognition, Paris, France, October 1986, pp 434-436

  • Cited in Baird86a: on document skew alignment angle for OCR

• [Stallings76] . Stallings, W. "Approaches to Chinese Character Recognition", Pattern Recognition, Vol 8, 1976, pp 87-98

• [Stockholm79] . Stockholm, E. "Recognition of a writer as a function of his method of writing", Perception and Motor Skills, Vol 49, 1979, pp 483-488

  • Cited in Teulings86a
  • Teulings86a cites for invariant spatial features in handwriting, SIZE=-1> regardless of writing conditions for subject (blackboard, paper, etc.)

• [Suen77a] . Suen, C.Y. "Alphanumeric Hand-prints with Stroke Directions and Sequences", internal report for Pencept, Incorporated, 39 Green Street, Waltham, Massachusetts 02154, 1977

  • Original collection of writing samples for PenVerter Partners (Pencept)

• [Suen77b] . Suen, C.Y. and Shillman, R.J. "Low Error Rate Optical Character Recognition of Unconstrained Hand-printed Letters Based on a Model of Human Perception", IEEE Transactions on Systems, Man, and Cybernetics, Vol 7 No 6, June, 1977, pp 491-495

  • Suen support (with Shillman) use of psychologically base features
  • Suen claims the computer did better recognition than human on U-V

• [Suen79a] . Suen, C.Y. "A Study on Man-Machine Interaction Problems in Character Recognition", IEEE Transactions on Systems, Man, and Cybernetics, Vol SMC-9 No 11, pp 732-737, November 1979

  • Refers to "users do not follow models, but revert to usual writing"
  • Ergonomics / user-interface problems in handwriting recognition

• [Suen79b] . Suen, C.Y. "n-Gram statistics for natural language understanding and text processing", IEEE Transactions on Pattern Analysis and Machine Intelligence", Vol PAMI-1, 1979, pp 164-172

  • Cited in Sinha88: for statistics on language (one million word samples) to use for context correction
  • Contains several lists of English-language word frequency dictionaries

• [Suen80a] . Suen, C. "Computer Analysis of Handprinted Characters", Proceedings of the 5th International Conference on Pattern Recognition, Miami Beach, Florida, December, 1980, pp 398-401

  • Human recognition success rate is 7.6% substitution error.
  • Examples of handwriting variability from 1500 student enrollment cards, with distribution frequency
  • Constraints: examples of how humans give cursive, lower-case, etc. even when asked specifically to print styles
  • Handwriting recognition by machine with better performance than humans requires understanding of how humans recognize characters
  • Taxonomy of different approaches to handwriting recognition: template matching, correlation; distribution of points, moments, crossings, distances; transformations and series expansions; geometrical and topological features, syntactic and structural analysis
  • Describes Suen's research on distinctive features (functional attributes)
  • Lists of constraints typically tried to reduce handwriting variability

• [Suen80b] (*) Suen, C., Berthold, M. and Mori, S. "Automatic Recognition of Hand-printed Characters - The State of the Art", Proceedings of the IEEE, pp 469-487, Vol 68 No 4, April 1980

  • Survey of handwriting recognition, both static (OCR) and dynamic (on-line). Used as citation in several patents. Follow-up by Tappert and Suen in 1990, by Plamondon in 2000. Specifically mentions decision trees at several points.

• [Suenaga80] (*p) Suenega, Y. and Nagura, M. "A facsimile based manuscript layout and editing system by auxiliary mark recognition", Proceedings of 5th Int Conf on Pattern Recognition, Dec 1-4, 1980, pp 856-858

  • Cited in Kankaanpaa87, Goodale83
  • See also Nagura83, same paper

• [Talmage78] (*) Talmage, John E. and Bates, L. Dexter "Nonplanar Transparent Electrographic Sensor", United States Patent 4,071,689, January 31, 1978

  • Resistive-sheet digitizer using small spacing dots

• [Tappert78a] . Tappert, C.C. and Das, Subratta K. "Memory and Time Improvements in a Dynamic Programming Algorithm for Matching Speech Patterns", IEEE Trans. Acoustics, Speech, and Signal Proc., Vol. ASSP-26, pp 583-586, 1978

  • Cited by Marlin Eller, Microsoft Pen Computing group

• [Tappert78b] . Tappert, C.C., and Kurtzberg, J.M. "Elastic Matching for Handwritten Symbol Recognition", Proc. IBM Int. Conf. Image Processing Pattern Recognition, 1978. Also, IBM Res. Rep. RC9988, May 1983.

  • Cited by Marlin Eller, Microsoft Pen Computing group

• [Tersoff78] (*) Tersoff, Abraham I. "Man-Machine Considerations in Automatic Handprint Recognition", IEEE Trans. on Systems, Man, and Cybernetics, Vol SMC-8 No 4, April 1978, p 279

  • short review: reliable OCR for hand-print is nowhere in sight

• [Teulings79] . Teulings, H.H.M., and Thomassen, A.J.W.M. "Computer-Aided Analysis of Handwriting Movements", Visible Language, Vol XIII No 3, 1979, pp 218-231

  • Describes digitizer set up to study handwriting: was tablet any good?

• [Thornburg79] (*) Thornburg, David D. "Capacitive Transducer", United States Patetn 4,177,421

  • Capacitive / electrostatic digitizer

• [Thornburg80] (*) Thornburg, David D. "A Low-Cost Alternative to Data Entry Keyboards", Recreational Computing, January 1980, pp 16-18

  • PrestoDigitizer product: seven-zone handwriting feature recognition

• [Thornburg80a] (*) Thornburg, David D. "A Low-Cost Alternative for the Real-Time On-Line Entry of Handprinted Characters", Proceedings of 3rd ACM SIGSMALL Symposium and the first SIGPC symposium on Small Systems

  • PrestoDigitizer product: seven-zone handwriting feature recognition

• [Thornburg80b] (*) Thornburg, David D. "A Low-Cost Device for the Real-Time On-Line Entry of Handprinted Characters", Proceeding of 3rd ACM SIGSMALL Symposium and the first SIGPC symposium on Small Systems, pp 179-183

  • PrestoDigitizer product: seven-zone handwriting feature recognition. stroke sequence and direction information, simple dictionary to identify/recognize the handwritten character. Refers to learning algorithms (training of system?)

• [Thornton79] . Thornton, R.W. "The Number Wheel: A Tablet Based Valuator for Three-dimensional Positioning", Proceedings of SIGGRAPH '79, published as Computer Graphics, Vol 13 No 2, August 1979, pp 102-107

  • Cited in FoleyJD84

• [Toussaint77] . Toussaint, G.T. "The Use of Context in Pattern Recognition", Proceedings of IEEE Computer Society Conference on Pattern Recognition and Image Processing, Renselear Polytechnic Institute, Troy, New York, June 6-8, 1977, pp 1-10

  • Context lies not just in the data, but in the perceiver
  • Syntactical vs statistical use of context
  • Statistical context: huge number crunching to reduce unmanageable probability distribution
  • Types of word-feature context: graphological, phonological, statistical, syntactic, semantic, pragmatic
  • Markov statistical methods for very large context dictionaries
  • Modified Viterbi algorithm (statistical context) not so hot
  • Goshtasby88 cites for using spelling dictionary look-up for context correction

• [Toussaint78a] . Toussaint, G.T. "The use of context in pattern recognition", Pattern Recognition, Vol 10, 1978, pp 189-204

  • Hull,JJ83

• [Toussaint78b] . Toussaint, G.T. and Shinghal, R. "Cluster analysis of English Text", Proceedings of IEEE Conference on Pattern Recognition and Image Processing, Chicalo, IL, 1978, pp 164-172

  • Hull,JJ83 on context dictionary

• [Tversky77] . Tversky, A. "Features of similarity", Psychological Review, Vol 84, 1977, pp 327-352

  • Cited in IchikawaS84: on human recognition

• [Ullman77] . Ullman, J.R. "A binary n-gram technique for automatic correction of substitution, deletion, insertion and reversal error in words", Computer Journal, Vol 20, 1977, pp 141-147

  • Cited in Hull83a: on spelling correction

• [Viviani80] . Viviani, P. and Terzuolo, V. "Space-time invariance in learned motor skills", in "Tutorials in motor behavior", Stelmach, G.E and Requin, J., editors, New York, North Holland, publishers, pp 525-533

  • Cited in Schomaker86
  • Schomaker86 cites that overlearned/over-trained writing strokes (signatures?) show great invariance in relative stroke timing

• [Wakamatsu79] (*) Wakamatsu, Shuichi "Character Writing System", United States Patent 4,144,405, March 13, 1979

  • Method for generating Kanji characters making use of the sequential order of writing parts of the characters with an input unit
  • Shows tablet input of handwritten Kanji characters, but does not refer to character recognition per se

• [Wallace76] (*) Wallace, Victor L. "The Semantics of Graphic Input Devices", "Proceedings of SIGGRAPH/SIGPLAN Conference on Graphics Languages, published as Computer Graphics, Vol 10 No 1, April 1976, pp 61-65

  • Mentions virtual devices: relevant to Schumer patent claims, though not cited in that case
  • Virtual devices (for semantic purposes) are Pick, Button, Keyboard, Locator, Valuator: specifically refers to character recognition for keyboard input

• [Waltz78] (*) Waltz, David L. "On the Interdependence of Language and Perception", U. Penn Technical Report T78-1020

  • Partly an essay on the ambiguity of perception as a function of context

• [Ward76] . Ward, J.R., Nelson, G.E., Desch, S.H., and Kaplow, R. "Two New Strategies for Computer-Assisted Language Instruction (CALI)", Foreign Language Annals, Vol 9 No 1, February 76, pp 28-37

  • CAI Computer-aided instruction, grammatical/syntax analysis of German sentences: Gordon Eugene Nelson, Jean Renard Ward, Robert Kaplow (see Kaplow patent for UI item)

• [WhiteGM78] . White, G.M. "Dynamic programming for the Viterbi algorithm, and low cost speech recognition", Proceedings of the 1978 IEEE International Conference on Acoustics, Speech, and Signal Processing, pp 413-417

  • Kruskal83

• [Wing78] . Wing, A.M. "Response time in handwriting", in "Information processing in motor control and learning", New York: Academic Press, G.E. Stelmach, editor, 1978, pp 153-172

  • Cited in Teulings86
  • Teulings86 cites this that up/down stroke pairs may be basic grapheme unit of writing

• [Wing79] (*p) Wing, A.M. "Variability in Handwritten Characters", Visible Language, Vol XIII No 3, 1979, pp 283-298

  • Confusable forms of script b-f, i-r
  • Printing slower than script, but gets faster with practice
  • People have same writing style on blackboard and on paper

• [WrightS78] . Wright, Jr., S.J., Anderson, P.T., and Grimes, R.S. "Multi-Modal Data Input/Output Apparatus and Method Compatible with Bio-Engineering Requirements", United States Patent 4,070,649, January 24, 1978

  • Character recognition using a four segment panel, diagonally
  • Reference similar to BLRT chain codes

• [Yacyk78] (*) Yacyk, J. "Man-Machine Considerations in Automatic Handprint Recognition", IEEE Transactions on Systems, Man, and Cybernetics, Vol SMC-8 No 4, pp 279-282, April 1978

  • Constraints do not work well on handpriting in practice: Users tend to omit serifs from O and S in ANSI74 standard of constrained writing styles. Review of OCR for ANSI-standard handprints, shows Universal Subset standard neographic shapes for hand-printed characters and symbols.
  • Tersoff paper is on same page

• [YamamotoK78] (*p) Yamamoto, K. and Mori, S. "Recognition of hand-printed characters by outermost point method", Proceedings of Fourth International Joint Conference on Pattern Recognition, Kyoto, Japan, November 7-10, 1978, pp 794-796

  • Cited in Kahan87

• [YamamotoK80] . Yamamoto, K. and Mori, S. "Recognition of Hand-printed Characters by an Outermost Point Method", Pattern Recognition, Vol 12., pp 229-236, March 1980

  • Similar to BLRTs chain codes, but uses 16 directions, not 4
  • Refers to dis-ambiguation of Katakana and Roman characters

• [Yasuhara76] (*p) Yasuhara, M. "On Distinctive Features in Character Recognition -- Some Evidence from Reaction-time Measurements", Working Paper 102, Cognitive Information Processing Group, Research Laboratory of Electronics, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, August 15, 1976

  • Mentions lack of consensus on what features to use in any given pattern recognition problem

• [Yasuhara77] (*p) Yasuhara, M. and Oka, M. "Signature Verification Experiment Based on Nonlinear Time Alignment: A Feasibility Study", IEEE Transactions on Systems, Man, and Cybernetics, Vol SMC-7 No 3, pp 212-216, March 1977

• [Yasuhara77a] (*p) Yasuhara, M. and Yasumoto, Yasuhiko "Measurement of Handwriting Pressure - Pressure-Senstivie Pen", Annual Report of RLCS

  • Avoid feature extraction in signatures by time-template matching

• [Yasuhara78] (*p) Yasuhara, M. and Kuklinski, T. "Category boundary effect for grapheme perception", Perception and Psychophysics, Vol 23 No 2, 1978, pp 97-104

  • Discrimination function peaks at the category boundary, just as speech does
  • Boundary classification for character recognition
  • Ted Kuklinski and Yasuhara: says say "physical attribute" and "functional attribute", since "feature" is used sloppily
  • Shows ambiguity of feature "leg" along character trajectory, instead of character along feature trajectory
  • Cultural bias: l vs r in speech by Japanese, and by Americans (extension to characters: European 7 vs F (?))

• [Yoshida78] (*) Yoshida, Masumi; Iwata, Kiyoshi; Yamamoto, Eiichiro; Masui, Takeshi; and Kabuyama, Yukikazu "Pattern Recognition Processing System", United States Patent 4,105,998, assigned to Fujitsu Limited, Kawasaki, Japan, August 18 1978

  • Recognizing handwritten characters by their intersections within a rectangle with horizontal and vertical reference lines; used the distances and positions of the intersection points, such as to get at the internal "open" space within a character

• [Yoshida80] . Yoshida, M. et al "Recognition System for Design Chart Drawn on Section Paper", Proceedings of 5th International Joint Conference on Pattern Recognition, 1980, pp 127-130

  • Cited in Kato80: user-interface for prettying up drawings

• [YouKC79] . You, K.C. and Fu, K.S. "A syntactic approach to shape recognition using attributed grammars", IEEE Transactions on Systems, Man, and Cybernetics, Vol SMC-9, 1979

• [Zadeh80] (*) Zadeh, L.A. "Fuzzy sets", Citation Classics, Number 47, November 24, 1980

  • Retrospective note on the origin of fuzzy set theory in 1963 by Zadeh L.A. and Desoer, C.A.

      Main Reference Page