OHCD_GIETV1

Published: 23 July 2024| Version 1 | DOI: 10.17632/r9mrt9fw8y.1
Contributors:
,
,
,

Description

Odia Simple character dataset consists of Annotate Simple Odia character which consists of 1, 32,353 examples of image data. The original scanned images from Handwritten Character were sized normalized to fit in a 128x128 pixel box while preserving their aspect ratio. Six files are available: • Grayscale Images consists of 34,138 • Binary Images consists of 34,138 • Inverted Images consists of 34,138 For research purpose the dataset is being augmented • Grayscale Images consists of 9,939 • Binary Images consists of 10,000 • Inverted Images consists of 10,000

Files

Steps to reproduce

Creation of Handwritten Odia Dataset by using Annotation, Threshold, Augmented and produce 1,32,353 image of equal size data The original scanned images from Handwritten Character were sized normalized to fit in a 128x128 pixel box while preserving their aspect ratio. • Grayscale Images consists of 34,138 • Binary Images consists of 34,138 • Inverted Images consists of 34,138 For research purpose the dataset is being augmented • Grayscale Images consists of 9,939 • Binary Images consists of 10,000 • Inverted Images consists of 10,000

Institutions

Gandhi Institute for Technology

Categories

Handwriting Recognition

Licence