Sailfish Optimizer with Deep Transfer Learning-Enabled Arabic Handwriting Character Recognition

The recognition of the Arabic characters is a crucial task in computer vision and Natural Language Processing fields. Some major complications recognizing handwritten texts include distortion pattern variabilities. So, feature extraction process significant NLP models. If features are automatically selected, it might result unavailability adequate data for accurately forecasting character classes. But, many usually create difficulties due to high dimensionality issues. Against this background, current study develops Sailfish Optimizer with Deep Transfer Learning-Enabled Handwriting Character Recognition (SFODTL-AHCR) model. projected SFODTL-AHCR model primarily focuses on identifying input image. proposed pre-processes image by following Histogram Equalization approach attain objective. Inception ResNet-v2 examines pre-processed produce vectors. Wavelet Neural Network (DWNN) utilized recognize characters. At last, SFO algorithm fine-tuning parameters involved DWNN better performance. performance was validated using series images. Extensive comparative analyses were conducted. method achieved maximum accuracy 99.73%. outcomes inferred supremacy over other approaches.