Litcius/Paper detail

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

G. Fathima, P. Kavitha, Revathi Vaithiyanathan, S Rosenblum, G Dror, P Antrassi, I Perrone, A Cuzzocrea, A Accardo, S Mayes, S Frye, R Breaux, S Calhoun, T Asselborn, T Gargot, L Kidziski, W Johal, D Cohen, C Jolly, P Dillenbourg, Y Tao, B Rapp, P Samodro, S Sihwi, C Lopez, C Hemimou, B Golse, L Douret, C Taleb, L Sulem, C Mokbel, M Khachab, P Drotr, M Dobe, L Deschamps, C Gaffet, S Aloui, J Boutet, V Brault, E Labyt, G Dimauro, V Bevilacqua, L Colizzi, D Pierro, M Gazda, M Hire, P Drotr, J Mucha, J Mekyska, Z Galaz, M Zanuy, K Ipina, V Zvoncak, T Kiska, Z Smekal, L Brabenec, I Rektorova, T Asselborn, M Chapatte, P Dillenbourg, R Lamba, T Gulati, K Dhlan, A Jain, X Sun, P Wu, S Hoi, S Wan, S Goudos, T Haas, C Schubert, M Eickhoff, H Pfeifer, Y Ren, C Zhu, S Xiao

2021International journal of intelligent engineering and systems19 citationsDOIOpen Access PDF

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 of dysgraphia classification based on experts manually classify the dysgraphia that requires more cost and time. Few researches apply the machine learning and deep learning techniques for the classification of dysgraphia and existing models have the limitations of overfitting problems in the training process. In this research, the Non-Discrimination Regularization in Rotational Region Convolutional Neural Network (NDR-R2CNN) is proposed to improve the efficiency of dysgraphia classification. The balancing parameters are introduced in the loss function to balance the class in the training and eliminate the features to reduce the overfitting problem. The collected children's handwriting data were used to evaluate the performance of the proposed NDR-R2CNN model. The proposed NDR-R2CNN model has the advantages of effective feature analysis and non-discrimination word analysis. The NDR-R2CNN model has the accuracy of 98.2 % and the SMOTE+SVM method has 90.2 % accuracy in Dysgraphia classification. The result shows that the NDR-R2CNN model has 98.2 % accuracy and the existing CNN has the accuracy of 94.2 %.

Topics & Concepts

DysgraphiaConvolutional neural networkComputer scienceRegularization (linguistics)Artificial intelligencePattern recognition (psychology)Speech recognitionLinguisticsReading (process)DyslexiaPhilosophyEducational Technology and AssessmentOptical Systems and Laser TechnologyHandwritten Text Recognition Techniques
Dysgraphia Classification based on the Non-Discrimination Regularization in Rotational Region Convolutional Neural Network | Litcius