Dyslexia Discernment in children based on Handwriting images using Residual Neural Network
Atmakuri Sasidhar, G. Kranthi Kumar, Koneru Yoshitha, Nimmakuri Tulasi
Abstract
A neurological ailment called dyslexia is characterized by a lack of precise word comprehension and generally poor writing abilities. Males are more likely to be affected than females, and it affects school-age children, placing them at risk for low academic achievement. The handwriting style makes it simple to spot the dyslexia symptoms that are most frequently present. Many machine learning techniques, including deep learning techniques recently, have been applied to numerous kinds of datasets. The suggested technique makes use of a residual neural network to more precisely identify dyslexia, a neurological disease, using handwritten images. The augmentation procedure, which includes rotating all of the photos, is used to preprocess the dataset of handwriting photographs. Because it can withstand noise, the residual neural network generated superior outcomes than the current techniques.