Infrared and visible image fusion based on nonlinear enhancement and NSST decomposition

Abstract

In multi-scale geometric analysis (MGA)-based fusion methods for infrared and visible images, adopting the same representation for the two types of images will result in the non-obvious thermal radiation target in the fused image, which can hardly be distinguished from the background. To solve the problem, a novel fusion algorithm based on nonlinear enhancement and non-subsampled shearlet transform (NSST) decomposition is proposed. Firstly, NSST is used to decompose the two source images into low-and high-frequency sub-bands. Then, the wavelet transform (WT) is used to decompose high-frequency sub-bands to obtain approximate sub-bands and directional detail sub-bands. The "average" fusion rule is performed for fusion for approximate sub-bands. And the "max-absolute" fusion rule is performed for fusion for directional detail sub-bands. The inverse WT is used to reconstruct the highfrequency sub-bands. To highlight the thermal radiation target, we construct a nonlinear transform function to determine the fusion weight of low-frequency subbands, and whose parameters can be further adjusted to meet different fusion requirements. Finally, the inverse NSST is used to reconstruct the fused image. The experimental results show that the proposed method can simultaneously enhance the thermal target in infrared images and preserve the texture details in visible images, and which is competitive with or even superior to the state-of-the-art fusion methods in terms of both visual and quantitative evaluations.

SMFD: an end-to-end infrared and visible image fusion model based on shared-individual multi-scale feature decomposition

“…Some methods build upon traditional approaches for decomposition. For instance, Xing et al 44 utilized the non-subsampled shearlet transform for decomposition, Gao et al 45 employed the dual-tree complex wavelet transform, and Lu et al 46 used the quadtree decomposition. In feature decomposition based on deep learning, Zhao et al 47 introduced a method based on deep image decomposition (DIDfuse), which is a novel image fusion approach using an encoder to decompose input images into background and detail feature maps, capturing low-frequency and high-frequency information.…”

“…Some methods build upon traditional approaches for decomposition. For instance, Xing et al 44 . utilized the non-subsampled shearlet transform for decomposition, Gao et al 45 .…”

Underwater polarization image fusion based on NSCT

“…Finally, we make a comprehensive comparison from six evaluation metrics: information entropy, 23 enhancement measure evaluation (EME), 24 contrast, 25 average gradient, 26 standard deviation, 27 and running time. These six indexes reflect the information richness, detail performance, overall performance, sharpness of the image, image information distribution, and algorithm complexity.…”