Image Super-Resolution is a machine learning task where the goal is to increase the resolution of an image, often by a factor of 4x or more, while maintaining its content and details as much as possible. The end result is a high-resolution version of the original image.

We first trained the super-resolution convolutional neural network (SRCNN), which is a deep-learning based super-resolution method.

Deep convolutional neural networks (CNN or DCNN) are the type most commonly used to identify patterns in images and video. DCNNs have evolved from traditional artificial neural networks, using a three-dimensional neural pattern inspired by the visual cortex of animals.

The activation function layer can be thought of as the "brain" of the CNN, where the input is transformed into a meaningful representation of the data. It is a fundamental component. To learn more about activation functions, check out our comprehensive guide to activation functions.