Comparing supervised and unsupervised learning in image denoising
Hanyun Wang · Applied and Computational Engineering · 2023
Recent studies on unsupervised learning have attracted people's increasing attention. In particular, Deep learning has developed rapidly in recent years. With the development of media images, people's demand for image noise reduction is increasing, and the requirements are becoming more and more strict. The traditional methods used for image noise reduction are far from meeting people's requirements, and people are eager to find a more efficient image noise reduction technology. In recent years, the technology of using a convolutional neural network for image noise reduction has become more and more skilled. This paper explores the reliability of image noise reduction technology using a convolutional neural network as an autoencoder, and whether good performance is maintained without using clean images. The article aims to compare the performance with supervised learning and unsupervised learning by deep learning in image denoising.