ReVQ-VAE: A Vector Quantization-Variational Autoencoder for COVID-19 Chest X-Ray Image Recovery

Nowadays, digital images have a valuable role in our daily life and can be used for a variety of applications like fingerprint recognition, video surveillance, etc. Sometimes, images mainly medical images get infected with attacks due to many reasons such as transmission in a noisy channel. Diagnosing an attacked medical image yields erroneous interpretation. So, to improve it for the right decision, the image must be recovered in advance. Therefore, there is a need for an efficient image recovering technique that helps to deal with medical attacked images. Image recovery is a process to realign the original image from the attacked image. In this paper, we propose an approach for chest X-Ray image recovery based on Vector Quantized Variational Autoencoder (VQ-VAE). The attacked images fed into the model to recover the images and get the original ones. The results of the SSIM show that our proposed model has produced better results than existing models.