Handwritten Amharic Character Dataset
Published: 6 September 2019| Version 1 | DOI: 10.17632/92swcchdyy.1
Contributor:
Fetulhak AbdurahmanDescription
I have prepared this dataset by collecting the handwritings of individuals from different perspectives meaning different age ranges, level of education and right and left handed persons. I given them a form prepared to take their handwritings and I developed an algorithm to crop individual characters after scanned copy is given for the algorithm. the size is 28 x 28. it is for the 34 base characters and 27 labialized (two sound characters)
Files
Categories
Machine Learning Algorithm