Multi-Focus Image Fusion Based on Improved CNN

Abstract

In order to avoid the limitations of artificial feature extraction, the CNN model is adopted to extract image features by big data-driven adaptive learning, which improves the accuracy of the features. For avoiding the loss of spatial information, an improved CNN model based on up-sampling is proposed, which consists of six layers of superimposed small convolution. The multi-layer design not only expands the receptive field, but also reduces the number of training parameters, and improves the running speed. The fusion method based on improved CNN model is proposed for multi-focus images. The improved CNN model divides the input image into focus region and non-focus region, and form the decision map. According to the decision map optimized by GFF, the focus regions are intergraded by pixel-by-pixel weighted fusion strategy to obtain fusion image. Experimental results show that the fusion results of the proposed method are clear in detail, complete in structure, no distortion in contrast, and no artifacts in the picture. It effectively avoids grayscale discontinuity, artifacts and other problems, and is better than classical methods we selected.