IMPROVING IJA SKILLS OF AUTISTIC CHILDREN THROUGH MOOD PREDICTION USING DNCNN AND WAVELET TRANSFORM ALGORITHMS
Autism spectrum disorders (ASD) are a group of neurodevelopmental conditions that are characterised by difficulties in social communication as well as repetitive or restricted behaviours and interests. Because autism spectrum disorder (ASD) is becoming more common, it is essential to diagnose patients with ASD in order to provide them with effective treatment and intervention, particularly during early childhood. In this paper, Deep Denoising Convolutional Neural Network (DnCNN) algorithm is used to predict the mood assessment of autistic children. Autism images help to predict the mood of autistic children. Additionally, Discrete Wavelet Transform (DWT) and Non-Decimated Wavelet Transform (NDWT) algorithms are used to filter the autism images for accurate prediction of autistic children behaviours. These algorithms predict the mood of autistic children through statistical value. The proposed algorithm Deep Denoising Convolutional Neural Network (DnCNN) algorithm gives high accuracy about 96%. These algorithms improve the initiation of joint attention (IJA) skills of autistic children.
Autism Spectrum Disorders (ASD); Deep Denoising Convolutional Neural Network (DnCNN) Algorithm, Discrete Wavelet Transform (DWT); Non-Decimated Wavelet Transform (NDWT)