Analysis and Techniques of Enhancing the Video Quality of Children with Autism Spectrum Disorder’s Daily Activities

The purpose of this study is to improve the image quality in videos by enhancing their contrast using Histogram Equalization (HE), contrast stretching, and CLAHE techniques. The video was extracted into several frames, and these frames were subsequently processed with image processing methods. The empirical data comprised a primary dataset of 606 video recordings, sourced from therapists and parents of children with Autism Spectrum Disorder (ASD). Mean Squared Error (MSE) and Peak Signal-to-Noise Ratio (PSNR) metrics were used to evaluate the results. The gray level scale of Histogram Equalization (HE), contrast stretching, and CLAHE has MSE values of 99.79, 90.23, and 10.02, respectively, and PSNR values of 17.44 dB, 11.66 dB, and 38.51 dB, respectively. Therefore, CLAHE demonstrated superior performance, yielding the best MSE and PSNR values.