Dysgraphia Classification based on the Non-Discrimination Regularization in Rotational Region Convolutional Neural Network

Abstract

Dysgraphia is a handwriting disorder and the classification of dysgraphia in children's handwritten images helps to identify the dysgraphia patient effectively and also prevents low self-esteem. Traditional methods ofdysgraphia classification based on experts manually classify the dysgraphia that requires more cost and time. Fewresearches apply the machine learning and deep learning techniques for the classification of dysgraphia and existingmodels have the limitations of overfitting problems in the training process. In this research, the Non-DiscriminationRegularization in Rotational Region Convolutional Neural Network (NDR-R2CNN) is proposed to improve theefficiency of dysgraphia classification. The balancing parameters are introduced in the loss function to balance theclass in the training and eliminate the features to reduce the overfitting problem. The collected children's handwritingdata were used to evaluate the performance of the proposed NDR-R2CNN model. The proposed NDR-R2CNN modelhas the advantages of effective feature analysis and non-discrimination word analysis. The NDR-R2CNN model hasthe accuracy of 98.2 % and the SMOTE+SVM method has 90.2 % accuracy in Dysgraphia classification. The resultshows that the NDR-R2CNN model has 98.2 % accuracy and the existing CNN has the accuracy of 94.2 %