Infrared and visible light image fusion method and system, computer equipment and application

Abstract

The invention belongs to the technical field of image fusion, and discloses an infrared and visible light image fusion method and system, computer equipment and application, and the method comprises the steps: carrying out super-resolution reconstruction of an original infrared image based on a convolutional neural network to acquire a reconstructed high-resolution infrared image; anisotropicallydiffusing the visible light image and the high-resolution infrared image respectively to obtain a corresponding base layer and a detail layer; obtaining a maximum feature vector by utilizing KL transformation to fuse detail layers, and fusing a base layer by utilizing saliency adaptive extraction; and carrying out linear reconstruction on the fused detail layer and the base layer to obtain a finalfused image. A large number of experiments show that compared with four fusion algorithms, the algorithm provided by the invention has a better fusion effect no matter from subjective judgment or from objective evaluation indexes, and has richer detail information and contour textures while enabling a fused image to have higher definition.

Description

technical field [0001] The invention belongs to the technical field of image fusion, and in particular relates to an infrared and visible light image fusion method, system, computer equipment and application. Background technique [0002] Image fusion technology has always been a research hotspot in the field of computer vision. Image fusion is mainly to generate a single image from the complementary information of multiple images in the same scene according to a certain fusion strategy, so as to enhance the ability to express the target details in the image and facilitate subsequent image processing. At present, image fusion technology has been widely used in surveying and mapping remote sensing [1], face recognition [2], medical imaging [3], military reconnaissance [4], transmission line fault detection [5] and other fields. Infrared images and visible light images are the two most widely studied types of images in image fusion [6]. Infrared images can distinguish the ta...

AI Technical Summary

Problems solved by technology

However, most of the improvement methods proposed for the anisotropic diffusion model are proposed for the diffusion equation itself. In the current algorithm, most of the improved methods for the anisotropic diffusion model are proposed for the diffusion equation or the diffusion function itself. Extraction and utilization are not based on the feature information and texture of the image, ignoring the feature information of the image itself, which may cause loss or blurring of image details

[0005] Through the above analysis, the existing problems and defects of the existing technology are: most of the improvement methods proposed for the anisotropic diffusion model are proposed for the diffusion equation itself, ignoring the characteristic information of the image itself, resulting in the loss of image details or Vague

[0006] The difficulty in solving the above problems and defects is: thermal infrared images are relatively poor in target detail resolution compared to visible light images due to the influence of the thermal infrared sensor’s own characteristics

For this reason, how to improve the detail resolution of thermal infrared images is a difficult problem; although super-resolution algorithms are an effective way to solve the above problems, how to design a super-resolution algorithm with simple model, high precision and fast speed is a difficult problem; In addition, when thermal infrared images and visible light images are fused, how to design an effective fusion strategy so that the fused image has good detail expression ability and meets the human visual cognition standard is a difficult problem

Images