Introducing the Boise State Bangla Handwriting Dataset and an Efficient Offline Recognizer of Isolated Bangla Characters

Nishatul Majid; Elisa H. Barney Smith

Introducing the Boise State Bangla Handwriting Dataset and an Efficient Offline Recognizer of Isolated Bangla Characters

Nishatul Majid
, Elisa H. Barney Smith

Electrical Engineering Department

Boise State University

Research output: Contribution to conference › Presentation

8 Downloads (Pure)

Abstract

This introduces the Boise State Bangla Handwriting Dataset , a publicly accessible offline handwriting dataset of Bangla script. This can be found at https://scholarworks.boisestate.edu/saipl/1/

A basic character recognition method is presented where the features are extracted based on zonal pixel counts, structural strokes and grid points with U-SURF descriptors modeled with bag of features.

Benchmarking with this approach on 3 other publicly available Bangla datasets is reported. The highest classification accuracy obtained with a n SVM classifier based on a cubic kernel is 96.8%.

Original language	American English
State	Published - 7 Aug 2018
Event	The 16th International Conference on Frontiers in Handwriting Regocnition (ICFHR 2018) - Duration: 7 Aug 2018 → …

Conference

Conference	The 16th International Conference on Frontiers in Handwriting Regocnition (ICFHR 2018)
Period	7/08/18 → …

EGS Disciplines

Electrical and Computer Engineering

Access to Document

Introducing the Boise State Bangla Handwriting Dataset and an EffFinal published version, 4.13 MBLicense: Unspecified

Cite this

@conference{2f4e1e679bec4acba54bb9b2d608097d,

title = "Introducing the Boise State Bangla Handwriting Dataset and an Efficient Offline Recognizer of Isolated Bangla Characters",

abstract = " This introduces the Boise State Bangla Handwriting Dataset , a publicly accessible offline handwriting dataset of Bangla script. This can be found at https://scholarworks.boisestate.edu/saipl/1/ A basic character recognition method is presented where the features are extracted based on zonal pixel counts, structural strokes and grid points with U-SURF descriptors modeled with bag of features. Benchmarking with this approach on 3 other publicly available Bangla datasets is reported. The highest classification accuracy obtained with a n SVM classifier based on a cubic kernel is 96.8\%. ",

author = "Nishatul Majid and \{Barney Smith\}, \{Elisa H.\}",

year = "2018",

month = aug,

day = "7",

language = "American English",

note = "The 16th International Conference on Frontiers in Handwriting Regocnition (ICFHR 2018) ; Conference date: 07-08-2018",

}

TY - CONF

T1 - Introducing the Boise State Bangla Handwriting Dataset and an Efficient Offline Recognizer of Isolated Bangla Characters

AU - Majid, Nishatul

AU - Barney Smith, Elisa H.

PY - 2018/8/7

Y1 - 2018/8/7

N2 - This introduces the Boise State Bangla Handwriting Dataset , a publicly accessible offline handwriting dataset of Bangla script. This can be found at https://scholarworks.boisestate.edu/saipl/1/ A basic character recognition method is presented where the features are extracted based on zonal pixel counts, structural strokes and grid points with U-SURF descriptors modeled with bag of features. Benchmarking with this approach on 3 other publicly available Bangla datasets is reported. The highest classification accuracy obtained with a n SVM classifier based on a cubic kernel is 96.8%.

AB - This introduces the Boise State Bangla Handwriting Dataset , a publicly accessible offline handwriting dataset of Bangla script. This can be found at https://scholarworks.boisestate.edu/saipl/1/ A basic character recognition method is presented where the features are extracted based on zonal pixel counts, structural strokes and grid points with U-SURF descriptors modeled with bag of features. Benchmarking with this approach on 3 other publicly available Bangla datasets is reported. The highest classification accuracy obtained with a n SVM classifier based on a cubic kernel is 96.8%.

UR - https://scholarworks.boisestate.edu/electrical_facpubs/391

M3 - Presentation

T2 - The 16th International Conference on Frontiers in Handwriting Regocnition (ICFHR 2018)

Y2 - 7 August 2018

ER -

Introducing the Boise State Bangla Handwriting Dataset and an Efficient Offline Recognizer of Isolated Bangla Characters

Abstract

Conference

EGS Disciplines

Access to Document

Other files and links

Fingerprint

Cite this