Convolutional Neural Network Based Multi-Focus Image Fusion

In this study, this paper mainly focuses on the use of convolutional neural network (CNN) to improve the clarity of multi-focus image and the fusion effect. First, set up an image picture data set and we convert labels tags to the dataset into binary images. After that, then use the CNN network to train the established datasets. In the end, we need to learn a direct mapping between source and focus map.