Authors: Purab Alok Jain, Pranali K. Kosamkar Abstract:The advancement of technology has created a huge scope and re-quirement for image super-resolution. Here, image super-resolution means to convert a low-resolution (LR) image into a high-resolution (HR) image. Super-resolution has many applications worldwide like in medical industries, surveil-lance, satellite photography, the study of the galaxy, etc. Also, COVID-19 is al-so a monstrous threat to earth. Doctors are predicting whether the patient is hav-ing the coronavirus or not via X-Rays and CT-Scans. The X-Rays and CT-Scans sometimes miss little details because of the blurriness of the image. This problem can be overcome by using the image super-resolution technique. In this paper, we proposed a system by which the classification of whether the person has the coronavirus or not becomes very accurate by using the image super-resolution technique. For the image super-resolution technique, we have used Super-Resolution Convolutional Neural Network (SRCNN) and for classifying whether the person is having coronavirus or not a Convolutional Neural Net-work (CNN) is used. Our models are trained and tested on 4 datasets, which are Set5, Set14, covid-chestxray-dataset, and chest-xray-pneumonia. Our results show that after applying super-resolution on the X-Rays or the CT-Scans, the classification of COVID-19 attained an accuracy of 95.31% which is higher if compared to the classification of COVID-19 without image super-resolution that attained an accuracy of 92.19%. These were the results after running the model for 20 epochs. Hence, with the help of the image super-resolution tech-nique, the classification of COVID-19 is much easier and accurate as compared to without the image super-resolution technique.
Open Zoom