Training an Arabic handwriting recognizer without a handwritten training data set

被引：0

作者：

Ahmad, Irfan ^{[1
,2
]}

Fink, Gernot A. ^{[2
]}

机构：

[1] KFUPM, Informat & Comp Sci Dept, Dhahran, Saudi Arabia

[2] TU Dortmund Univ, Dept Comp Sci, Dortmund, Germany

来源：

2015 13TH IAPR INTERNATIONAL CONFERENCE ON DOCUMENT ANALYSIS AND RECOGNITION (ICDAR) | 2015年

关键词：

Handwritten text recognition; hidden Markov models; training data; efficient training; HMM adaptation; OCR; ADAPTATION; MODELS;

D O I：

暂无

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Handwritten text recognition is an active research area in pattern recognition. One of the prerequisites of setting up a handwritten text recognizer is to train them using, mostly, large amounts of labeled training data. In the current paper we report our work on handwritten text recognition using no handwritten training set. We investigate different approaches including, computer generated text in different typefaces as training data, unsupervised adaptation, and using recognition hypothesis on the test sets as training data. Results from handwritten Arabic word recognition task show that the approach is promising with good recognition rates.

引用

页码：476 / 480

页数：5

共 50 条

[21] EFFECTS OF TRAINING FOR MANUSCRIPT HANDWRITING ON GENERALIZATION TO CURSIVE HANDWRITING
LEUNG, EKT
TREBLAS, PV
COOPER, JO
PORTER, JT
BEHAVIOURAL PSYCHOTHERAPY, 1982, 10 (04): : 311 - 323
[22] Recognition of Cursive Arabic Handwritten Text Using Embedded Training Based on Hidden Markov Models
Rabi, Mouhcine
Amrouch, Mustapha
Mahani, Zouhir
INTERNATIONAL JOURNAL OF PATTERN RECOGNITION AND ARTIFICIAL INTELLIGENCE, 2018, 32 (01)
[23] Unpaired training: Optimize the seismic data denoising model without paired training data
Ma, Haitao
Ba, He
Li, Yue
Zhao, Yuxing
Wu, Ning
GEOPHYSICS, 2023, 88 (01) : WA345 - WA360
[24] Arabic Handwriting Data Base for Text Recognition
Ramdan, Jabril
Omar, Khairuddin
Faidzul, Mohammad
Mady, Ali
4TH INTERNATIONAL CONFERENCE ON ELECTRICAL ENGINEERING AND INFORMATICS (ICEEI 2013), 2013, 11 : 580 - 584
[25] Training Simulation with Nothing but Training Data Simulating Performance based on Training Data Without the Help of Performance Diagnostics in a Laboratory
Ludwig, Melanie
Schaefer, David
Asteroth, Alexander
ICSPORTS: PROCEEDINGS OF THE 4TH INTERNATIONAL CONGRESS ON SPORT SCIENCES RESEARCH AND TECHNOLOGY SUPPORT, 2016, : 75 - 82
[26] Influence of Training Data on the Invertability of Neural Networks for Handwritten Digit Recognition
Adler, Antonia
Geierhos, Michaela
Hobley, Eleanor
20TH IEEE INTERNATIONAL CONFERENCE ON MACHINE LEARNING AND APPLICATIONS (ICMLA 2021), 2021, : 730 - 737
[27] Efficient training data reduction for SVM based handwritten digits recognition
Javed, I.
Ayyaz, M. N.
Mehmood, W.
2007 INTERNATIONAL CONFERENCE ON ELECTRICAL ENGINEERING, 2007, : 425 - 428
[28] A study on the effect of varying training set sizes on recognition perfonnance with handwritten Bangla numerals
Roy, K
Chaudhuri, C
Pal, U
Kundu, M
INDICON 2005 PROCEEDINGS, 2005, : 570 - 574
[29] Unsupervised training on a large amount of arabic broadcast news data
Ma, Jeff
Matsoukas, Spyros
2007 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, VOL II, PTS 1-3, 2007, : 349 - +
[30] Using Diversity in Classifier Set Selection for Arabic Handwritten Recognition
Azizi, Nabiha
Farah, Nadir
Sellami, Mokhtar
Ennaji, Abdel
MULTIPLE CLASSIFIER SYSTEMS, PROCEEDINGS, 2010, 5997 : 235 - +

← 1 2 3 4 5 →