Super-resolution of retinal images using multi-kernel SVR for IoT healthcare applications

Abstract

Internet of Things (IoT) healthcare is one of the most popular areas of research due to the rapid development in information and communication technologies. IoT system focusing on human vision would be an ideal solution for the people in developing countries to have adequate medical attention. This paper proposes a hybrid architecture for IoT healthcare to process the retinal images captured using smartphone fundoscopy. The proposed super-resolution (SR) algorithm for retinal images use multi-kernel support vector regression (SVR) to improve the quality of the captured images. The experimental results with respect to the peak-signal-to-noise ratio (PSNR) and mean squared error (MSE) show that the proposed super-resolution approach for retinal images performs better when compared to the state-of-art algorithms. Further, the hybrid architecture helps the ophthalmologists in efficient diagnosis by providing high resolution retinal images.