A Comparative Study Of Treebased Structure Methods For Handwriting Identification Springer
Handwriting is unique for each individual. Every single word presents the numbers of significant features that can be used for authenticating the author of the writing. The process of identify this significant features is called as feature selection process. Feature selection is an important area in...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Book Chapter |
| Language: | English |
| Published: |
Springer
2014
|
| Subjects: | |
| Online Access: | http://eprints.utem.edu.my/id/eprint/20004/2/Daeng_Mai.pdf http://eprints.utem.edu.my/id/eprint/20004/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Handwriting is unique for each individual. Every single word presents the numbers of significant features that can be used for authenticating the author of the writing. The process of identify this significant features is called as feature selection process. Feature selection is an important area in the machine learning, specifically in pattern recognition which is becoming famous among the researchers. Tree-based structure method is one of the feature selection methods which is able to generate a compact subset of non-redundant features and hence improves interpretability and generalization. However its focus is still limited especially in Writer Identification domain. This paper proposes the role of the tree-based structure method performs in Writer Identification. Several methods of the tree-based structure are selected and performed using image dataset from IAM Handwriting Database. The results of each methods of Writer Identification are also analyzed and compared. The most interesting method will be further explored and adapted in future works. |
|---|