TOP-DOWN RECOGNITION OF HANDPRINTED CHARACTERS BY LINE-WISE PARALLEL OPERATION.

被引:0
|
作者
Yamada, Hiromitsu
Mori, Shunji
机构
来源
Systems, computers, controls | 1979年 / 10卷 / 05期
关键词
D O I
暂无
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
A description is given of a high-speed operation using line-wise parallel operation instructions on the 3-dimensional data arranged in a plane structure. These operation instructions provide high-speed transformation of a 2-dimensional pattern into one-dimensional feature enabling the implementation of a top-down character recognition system that extracts only those features, as needed, during the operation. The system has been tested using handprinted characters. The results of the recognition tests are 98. 2% correct with respect to numerical characters written by 1445 individuals and 97. 1% correct with respect to English letters written by 1383 individuals. All these handprinted characters have been used to train the system. In addition, a microprocessor system ″PULMIC″ is implemented as a low-speed OCR machine which has a processing speed of 20 letters per second.
引用
收藏
页码:20 / 29
相关论文
共 50 条
  • [41] Top-down attention control at feature space for robust pattern recognition
    Lee, SI
    Lee, SY
    BIOLOGICALLY MOTIVATED COMPUTER VISION, PROCEEDING, 2000, 1811 : 129 - 138
  • [42] A study on top-down word image generation for handwritten word recognition
    Ishidera, E
    Nishiwaki, D
    SEVENTH INTERNATIONAL CONFERENCE ON DOCUMENT ANALYSIS AND RECOGNITION, VOLS I AND II, PROCEEDINGS, 2003, : 1173 - 1177
  • [43] A top-down information theoretic word clustering algorithm for phrase recognition
    Wu, Yu-Chieh
    INFORMATION SCIENCES, 2014, 275 : 213 - 225
  • [44] A cortical mechanism for triggering top-down facilitation in visual object recognition
    Bar, M
    JOURNAL OF COGNITIVE NEUROSCIENCE, 2003, 15 (04) : 600 - 609
  • [45] RECOGNITION AND TOP-DOWN GENERATION OF BETA-ACYCLIC DATABASE SCHEMES
    LAKSHMANAN, VS
    CHANDRASEKARAN, N
    MADHAVAN, CEV
    LECTURE NOTES IN COMPUTER SCIENCE, 1984, 181 : 344 - 366
  • [46] Top-Down and Bottom-up Cues for Scene Text Recognition
    Mishra, Anand
    Alahari, Karteek
    Jawahar, C. V.
    2012 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2012, : 2687 - 2694
  • [47] Top-down modulated model for object recognition in different categorisation levels
    Sharifizadeh, Fatemeh
    Ganjtabesh, Mohammad
    Nowzari-Dalini, Abbas
    INTERNATIONAL JOURNAL OF BIO-INSPIRED COMPUTATION, 2021, 18 (01) : 13 - 26
  • [48] A top-down event-driven approach for concurrent activity recognition
    Athanasios S. Voulodimos
    Dimitrios I. Kosmopoulos
    Nikolaos D. Doulamis
    Theodora A. Varvarigou
    Multimedia Tools and Applications, 2014, 69 : 293 - 311
  • [49] OBJECT RECOGNITION WITH TOP-DOWN VISUAL ATTENTION MODELING FOR BEHAVIORAL STUDIES
    Buso, Vincent
    Gonzalez-Diaz, Ivan
    Benois-Pineau, Jenny
    2015 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP), 2015, : 4431 - 4435
  • [50] HANDS IN FOCUS: SIGN LANGUAGE RECOGNITION VIA TOP-DOWN ATTENTION
    Sarhan, Noha
    Wilms, Christian
    Closius, Vanessa
    Brefeld, Ulf
    Frintrop, Simone
    2023 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, ICIP, 2023, : 2555 - 2559