2009
DOI: 10.1016/j.patcog.2008.10.012
|View full text |Cite
|
Sign up to set email alerts
|

Evaluation of incremental learning algorithms for HMM in the recognition of alphanumeric characters

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
2

Citation Types

0
7
0

Year Published

2009
2009
2019
2019

Publication Types

Select...
6
1
1

Relationship

0
8

Authors

Journals

citations
Cited by 19 publications
(8 citation statements)
references
References 21 publications
0
7
0
Order By: Relevance
“…An incremental learning method is promising to overcome the shortcomings found in traditional machine learning approaches. Incremental learning method consists of techniques to enable classifiers to gather more information from unseen data but do not forget old information, without having to access previously learned data [18].…”
Section: Systemmentioning
confidence: 99%
See 1 more Smart Citation
“…An incremental learning method is promising to overcome the shortcomings found in traditional machine learning approaches. Incremental learning method consists of techniques to enable classifiers to gather more information from unseen data but do not forget old information, without having to access previously learned data [18].…”
Section: Systemmentioning
confidence: 99%
“…The incremental learning method we applied in our system is ensemble training (ET) and ensemble learning (EL) [18]. ET consists of independently do the learning of each of the observation sequences from the training set so that each sequence generates an HMM.…”
Section: Systemmentioning
confidence: 99%
“…Adding a number of artificial samples to training sets is an effective method of improving the predictive capability of PIRS, including virtual data generation approaches, which are frequently used in pattern recognition, where prior knowledge obtained about the particular subject is used to create virtual samples. These virtual data points can be used to improve generalisation [10]. Niyogi says, ‘In certain function learning contexts, the framework of virtual examples is equivalent to imposing prior knowledge as a regulariser’ [11].…”
Section: Introductionmentioning
confidence: 99%
“…Synthesis methods have been proposed for the generation of virtual samples of various poses and expressions from a given face image [8,9]. Virtual data were used to improve the performance person-independent systems [10]. In summary, the novel contributions made by this paper are as follows: † Employing different mappings for discriminant analysis, dimensionality reduction and style transfer in order to increase the generalisation ability of the recognition system based on dynamic modelling of facial expression.…”
Section: Introductionmentioning
confidence: 99%