A Study of Different Classifier Combination Approaches for Handwritten Indic Script Recognition
Script identification is an essential step in document image processing especially when the environment is multi-script/multilingual. Till date researchers have developed several methods for the said problem. For this kind of complex pattern recognition problem, it is always difficult to decide whic...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2018-02-01
|
| Series: | Journal of Imaging |
| Subjects: | |
| Online Access: | http://www.mdpi.com/2313-433X/4/2/39 |
| _version_ | 1818993668995940352 |
|---|---|
| author | Anirban Mukhopadhyay Pawan Kumar Singh Ram Sarkar Mita Nasipuri |
| author_facet | Anirban Mukhopadhyay Pawan Kumar Singh Ram Sarkar Mita Nasipuri |
| author_sort | Anirban Mukhopadhyay |
| collection | DOAJ |
| description | Script identification is an essential step in document image processing especially when the environment is multi-script/multilingual. Till date researchers have developed several methods for the said problem. For this kind of complex pattern recognition problem, it is always difficult to decide which classifier would be the best choice. Moreover, it is also true that different classifiers offer complementary information about the patterns to be classified. Therefore, combining classifiers, in an intelligent way, can be beneficial compared to using any single classifier. Keeping these facts in mind, in this paper, information provided by one shape based and two texture based features are combined using classifier combination techniques for script recognition (word-level) purpose from the handwritten document images. CMATERdb8.4.1 contains 7200 handwritten word samples belonging to 12 Indic scripts (600 per script) and the database is made freely available at https://code.google.com/p/cmaterdb/. The word samples from the mentioned database are classified based on the confidence scores provided by Multi-Layer Perceptron (MLP) classifier. Major classifier combination techniques including majority voting, Borda count, sum rule, product rule, max rule, Dempster-Shafer (DS) rule of combination and secondary classifiers are evaluated for this pattern recognition problem. Maximum accuracy of 98.45% is achieved with an improvement of 7% over the best performing individual classifier being reported on the validation set. |
| first_indexed | 2024-12-20T20:45:43Z |
| format | Article |
| id | doaj.art-3952485e5d9643cda87a9d33d903dc20 |
| institution | Directory Open Access Journal |
| issn | 2313-433X |
| language | English |
| last_indexed | 2024-12-20T20:45:43Z |
| publishDate | 2018-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Journal of Imaging |
| spelling | doaj.art-3952485e5d9643cda87a9d33d903dc202022-12-21T19:27:02ZengMDPI AGJournal of Imaging2313-433X2018-02-01423910.3390/jimaging4020039jimaging4020039A Study of Different Classifier Combination Approaches for Handwritten Indic Script RecognitionAnirban Mukhopadhyay0Pawan Kumar Singh1Ram Sarkar2Mita Nasipuri3Department of Computer Science and Engineering, Jadavpur University, Kolkata-700032, West Bengal, IndiaDepartment of Computer Science and Engineering, Jadavpur University, Kolkata-700032, West Bengal, IndiaDepartment of Computer Science and Engineering, Jadavpur University, Kolkata-700032, West Bengal, IndiaDepartment of Computer Science and Engineering, Jadavpur University, Kolkata-700032, West Bengal, IndiaScript identification is an essential step in document image processing especially when the environment is multi-script/multilingual. Till date researchers have developed several methods for the said problem. For this kind of complex pattern recognition problem, it is always difficult to decide which classifier would be the best choice. Moreover, it is also true that different classifiers offer complementary information about the patterns to be classified. Therefore, combining classifiers, in an intelligent way, can be beneficial compared to using any single classifier. Keeping these facts in mind, in this paper, information provided by one shape based and two texture based features are combined using classifier combination techniques for script recognition (word-level) purpose from the handwritten document images. CMATERdb8.4.1 contains 7200 handwritten word samples belonging to 12 Indic scripts (600 per script) and the database is made freely available at https://code.google.com/p/cmaterdb/. The word samples from the mentioned database are classified based on the confidence scores provided by Multi-Layer Perceptron (MLP) classifier. Major classifier combination techniques including majority voting, Borda count, sum rule, product rule, max rule, Dempster-Shafer (DS) rule of combination and secondary classifiers are evaluated for this pattern recognition problem. Maximum accuracy of 98.45% is achieved with an improvement of 7% over the best performing individual classifier being reported on the validation set.http://www.mdpi.com/2313-433X/4/2/39Classifier combinationDempster-Shafer theory of evidenceIndic script identificationHistograms of Oriented GradientsModified Log-Gabor filter transformElliptical features |
| spellingShingle | Anirban Mukhopadhyay Pawan Kumar Singh Ram Sarkar Mita Nasipuri A Study of Different Classifier Combination Approaches for Handwritten Indic Script Recognition Journal of Imaging Classifier combination Dempster-Shafer theory of evidence Indic script identification Histograms of Oriented Gradients Modified Log-Gabor filter transform Elliptical features |
| title | A Study of Different Classifier Combination Approaches for Handwritten Indic Script Recognition |
| title_full | A Study of Different Classifier Combination Approaches for Handwritten Indic Script Recognition |
| title_fullStr | A Study of Different Classifier Combination Approaches for Handwritten Indic Script Recognition |
| title_full_unstemmed | A Study of Different Classifier Combination Approaches for Handwritten Indic Script Recognition |
| title_short | A Study of Different Classifier Combination Approaches for Handwritten Indic Script Recognition |
| title_sort | study of different classifier combination approaches for handwritten indic script recognition |
| topic | Classifier combination Dempster-Shafer theory of evidence Indic script identification Histograms of Oriented Gradients Modified Log-Gabor filter transform Elliptical features |
| url | http://www.mdpi.com/2313-433X/4/2/39 |
| work_keys_str_mv | AT anirbanmukhopadhyay astudyofdifferentclassifiercombinationapproachesforhandwrittenindicscriptrecognition AT pawankumarsingh astudyofdifferentclassifiercombinationapproachesforhandwrittenindicscriptrecognition AT ramsarkar astudyofdifferentclassifiercombinationapproachesforhandwrittenindicscriptrecognition AT mitanasipuri astudyofdifferentclassifiercombinationapproachesforhandwrittenindicscriptrecognition AT anirbanmukhopadhyay studyofdifferentclassifiercombinationapproachesforhandwrittenindicscriptrecognition AT pawankumarsingh studyofdifferentclassifiercombinationapproachesforhandwrittenindicscriptrecognition AT ramsarkar studyofdifferentclassifiercombinationapproachesforhandwrittenindicscriptrecognition AT mitanasipuri studyofdifferentclassifiercombinationapproachesforhandwrittenindicscriptrecognition |