Image fusion method based on shift-invariant shearlets and stack autoencoder

Abstract

The invention discloses an image fusion method based on shift-invariant shearlets and a stack autoencoder. The implementation steps of the image fusion method includes the following steps that: shift-invariant shearlet transformation is utilized to decompose an image to be fused into a low-frequency subband coefficient and a high-frequency subband coefficient, wherein the low-frequency subband coefficient reflects the basic contour of the image and is fused by using weight averaging, and the high-frequency subband coefficient reflects the edge and texture information of the image. The present invention provides a stack autoencoder feature-based fusion method. According to the stack autoencoder feature-based fusion method, a sliding block division method is adopted to divide a high-frequency subband into different blocks; a stack autoencoder network is trained with the blocks adopted as input; the trained network is adopted to encode the blocks, so that features can be obtained; the features are enhanced through using a spatial frequency, so that an activity measure can be obtained; the fusion of the blocks of the high-frequency subband coefficient is performed through using a principle that the larger numerical value of the activity measure is adopted; after all the blocks are fused, the high-frequency subband can be obtained through using inverse sliding window transformation; and a fused image can be obtained through inverse shift-invariant shearlet transformation. Compared with a traditional fusion method, the fusion method of the invention can better preserve edge and texture information in an original image.

Description

technical field [0001] The invention relates to an image fusion method based on translation invariant shear wave and stacked self-encoding, which is a fusion method in the technical field of image processing and is widely used in military applications and clinical medical diagnosis. Background technique [0002] Due to the limited information contained in a single image, it often cannot meet the practical application. Image fusion is a technology that synthesizes images of the same scene collected by multiple sensors into one image through fusion algorithm processing. The fused image can effectively combine the advantages of multiple images to be fused, which is more suitable for human visual perception. Image fusion was proposed in the 1970s. In recent years, due to the rapid development of multi-sensor technology, image fusion has been widely used in the fields of military reconnaissance, medical diagnosis and remote sensing. [0003] Image fusion methods can be roughly ...

AI Technical Summary

Problems solved by technology

In recent years, the application of deep learning methods in the field of computer vision has achieved great success. Deep learning methods are good at discovering hierarchical features from complex data sets, but due to the lack of sufficient labeled features in the practical application of image fusion Supervised learning methods such as convolutional neural network (CNN) have not yet been applied in image fusion, while stacked autoencoder (SAE), as a deep neural network for unsupervised learning, meets the requirements of image fusion applications. scene

Images