Database for Amharic Handwritten Recognition

Authors: Fetulhak Abdurahman, Eyob Sisay and Kinde Anlay

Version: 1.0

This is a dataset prepared for handwritten Amharic word recognition. It is prepared by collecting handwritings of Amharic native speakers and writers. The dataset contains 33,672 handwritten Amharic word images with their corresponding labels. From the total 33,672 word images, 12,064 are original handwritten images. The remaining 21,608 are augmented images generated by randomly applying functions such as rotation, shifting, shrinking, expanding, degrading, and applying a varying amount of Gaussian noise and blurring to the original handwritten image. In addition to that, the dataset is prepared in two different word image sizes with 32 x 128 and 64 x 256.