Arabic Handwriting Classification using Deep Transfer Learning Techniques

Arabic handwriting is slightly different from the of other languages; hence it possible to distinguish written by native or non-native writer based on their handwriting. However, classifying challenging using traditional text recognition algorithms. Thus, this study evaluated and validated utilisation deep transfer learning models overcome such issues. Hence, seven types models, namely AlexNet, GoogleNet, ResNet18, ResNet50, ResNet101, VGG16, VGG19, were used determine most suitable model for handwritten images non-native. Two datasets comprised evaluate validate newly developed classify each model’s output as either foreign (non-native) writers. The training validation sets conducted both original augmented datasets. Results showed that highest accuracy GoogleNet normal datasets, with attained 93.2% data 95.5% in