Novel feature extraction technique for the recognition of handwritten digits
This paper presents an efficient handwritten digit recognition system based on support vector machines (SVM). A novel feature set based on transition information in the vertical and horizontal directions of a digit image combined with the famous Freeman chain code is proposed. The main advantage of...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Emerald Publishing
2017-01-01
|
| Series: | Applied Computing and Informatics |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S221083271500006X |
| _version_ | 1797761717119221760 |
|---|---|
| author | Abdelhak Boukharouba Abdelhak Bennia |
| author_facet | Abdelhak Boukharouba Abdelhak Bennia |
| author_sort | Abdelhak Boukharouba |
| collection | DOAJ |
| description | This paper presents an efficient handwritten digit recognition system based on support vector machines (SVM). A novel feature set based on transition information in the vertical and horizontal directions of a digit image combined with the famous Freeman chain code is proposed. The main advantage of this feature extraction algorithm is that it does not require any normalization of digits. These features are very simple to implement compared to other methods. We evaluated our scheme on 80,000 handwritten samples of Persian numerals and we have achieved very promising results. |
| first_indexed | 2024-03-12T19:18:01Z |
| format | Article |
| id | doaj.art-81b1071f601843cb893fc05d5b0f3d67 |
| institution | Directory Open Access Journal |
| issn | 2210-8327 |
| language | English |
| last_indexed | 2024-03-12T19:18:01Z |
| publishDate | 2017-01-01 |
| publisher | Emerald Publishing |
| record_format | Article |
| series | Applied Computing and Informatics |
| spelling | doaj.art-81b1071f601843cb893fc05d5b0f3d672023-08-02T05:25:11ZengEmerald PublishingApplied Computing and Informatics2210-83272017-01-01131192610.1016/j.aci.2015.05.001Novel feature extraction technique for the recognition of handwritten digitsAbdelhak Boukharouba0Abdelhak Bennia1Faculté des Sciences et de la technologie, Département d’Electronique et de Télécommunications, Université 8 Mai 1945 Guelma, BP 401, Guelma 24000, AlgeriaFaculté des Sciences de la Technologie, Département d’Electronique, Université Constantine 1, AlgeriaThis paper presents an efficient handwritten digit recognition system based on support vector machines (SVM). A novel feature set based on transition information in the vertical and horizontal directions of a digit image combined with the famous Freeman chain code is proposed. The main advantage of this feature extraction algorithm is that it does not require any normalization of digits. These features are very simple to implement compared to other methods. We evaluated our scheme on 80,000 handwritten samples of Persian numerals and we have achieved very promising results.http://www.sciencedirect.com/science/article/pii/S221083271500006XFeature extractionFeature selectionDigit recognitionSupport vector machine |
| spellingShingle | Abdelhak Boukharouba Abdelhak Bennia Novel feature extraction technique for the recognition of handwritten digits Applied Computing and Informatics Feature extraction Feature selection Digit recognition Support vector machine |
| title | Novel feature extraction technique for the recognition of handwritten digits |
| title_full | Novel feature extraction technique for the recognition of handwritten digits |
| title_fullStr | Novel feature extraction technique for the recognition of handwritten digits |
| title_full_unstemmed | Novel feature extraction technique for the recognition of handwritten digits |
| title_short | Novel feature extraction technique for the recognition of handwritten digits |
| title_sort | novel feature extraction technique for the recognition of handwritten digits |
| topic | Feature extraction Feature selection Digit recognition Support vector machine |
| url | http://www.sciencedirect.com/science/article/pii/S221083271500006X |
| work_keys_str_mv | AT abdelhakboukharouba novelfeatureextractiontechniquefortherecognitionofhandwrittendigits AT abdelhakbennia novelfeatureextractiontechniquefortherecognitionofhandwrittendigits |