OHCD_GIETV2

Published: 25 July 2024| Version 1 | DOI: 10.17632/kmf4f95782.1
Contributors:
Rabinarayan Panda, SACHIKANTA DASH,
, Shazid Wahid Khandakhani

Description

Creation of Handwritten Odia Compound Character Dataset by using Annotation, Threshold, Augmented and produce image of equal size data Odia Compound character dataset consists of Annotate data that consists of 87,136 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 15352 • Binary Images consists of 15352 • Inverted Images consists of 15352 For research purpose the dataset is being augmented • Grayscale Images consists of 9994 • Binary Images consists of 8568 • Inverted Images consists of 7180

Files

Steps to reproduce

Creation of Handwritten Odia Compound Character Dataset by using Annotation, Threshold, Augmented and produce image of equal size data Odia Compound character dataset consists of Annotate Odia Compound character which consists of 87,136 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 15352 • Binary Images consists of 15352 • Inverted Images consists of 15352 For research purpose the dataset is being augmented • Grayscale Images consists of 9994 • Binary Images consists of 8568 • Inverted Images consists of 7180

Institutions

Gandhi Institute of Engineering and Technology

Categories

Handwriting Recognition

Licence