2014
DOI: 10.1007/s00521-014-1758-y
|View full text |Cite
|
Sign up to set email alerts
|

Use of wavelet-based two-dimensional scaling moments and structural features in cascade neuro-fuzzy classifiers for handwritten digit recognition

Abstract: In this paper, a novel handwritten digit recognition system is proposed. The system consist of feature extraction, feature selection and classification stages. The features of digits are extracted by using the moment-based and structural-based methods. For the moment-based method, wavelet-based two-dimensional scaling moments (2-DSMs), which have uniquely different angular divisions of polar form, are considered. The structural-based features including profiles, intersections of horizontal and vertical straigh… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1

Citation Types

0
2
0

Year Published

2016
2016
2020
2020

Publication Types

Select...
3

Relationship

0
3

Authors

Journals

citations
Cited by 3 publications
(2 citation statements)
references
References 33 publications
(90 reference statements)
0
2
0
Order By: Relevance
“…In [26], the feature selection is based on the wavelet transform that uses 2D scaling moments and various classifiers (support vector machines/classifiers, artificial neural networks, neuro-fuzzy classifiers, and others). The SVM classifier demonstrates the best accuracy of 99.39%, while the neuro-fuzzy one achieves accuracy 98.72%.…”
Section: Datasetsmentioning
confidence: 99%
“…In [26], the feature selection is based on the wavelet transform that uses 2D scaling moments and various classifiers (support vector machines/classifiers, artificial neural networks, neuro-fuzzy classifiers, and others). The SVM classifier demonstrates the best accuracy of 99.39%, while the neuro-fuzzy one achieves accuracy 98.72%.…”
Section: Datasetsmentioning
confidence: 99%
“…The CNN extracts the facial features and classifies these using a classifier for recognition and integration, thus making the algorithm simpler (Yu, Zhao, & Wei, 2007). However, due to its strong fitting capacity, the CNN is prone to such phenomenon that the effect can be improved in training but impaired in practices (Liu & Wang, 2011;Cetişli & Edizkan, 2015). Aiming at the structural defects of CNN, we propose a neural network structure that can automatically expand the number of initial neurons based on preset training errors.…”
Section: Introductionmentioning
confidence: 99%