Image fusion method and apparatus based on potentially low rank representation and NSST

Abstract

The invention provides an image fusion method and device based on potential low rank representation and NSST. A source image is discomposed into multi-scale and multi-direction by an NSST including asmall-size shearing wave filter. Because NSST eliminates the sampling operation in the decomposition phase, it has translation invariance. At the same time, in the phase of directional localization,localized small-scale shearing filter can avoid spectrum aliasing phenomenon, and make the image decomposition and reconstruction effect better. For the low frequency component, LatLRR can extract thesalient features from the data robustly, and is more robust to noise, so it can identify the salient objects and regions in the image precisely. For the high frequency component, because the averagegradient represents the change of image gray value, it can reflect the details of image edge and texture, so the average gradient operator can better express the image features and achieve better fusion effect.

Description

technical field [0001] The present invention relates to the field of image fusion, in particular to an image fusion method and device based on latent low-rank representation and NSST Background technique [0002] The development of infrared and visible light image fusion technology is largely to meet the development of modern military reconnaissance technology. The visible light image is a reflection image with many high-frequency components, which can reflect the details of the scene under a certain illuminance. Visible light images (that is, low-light images) have low contrast; infrared images are radiation images, and the gray scale is determined by the temperature difference between the target and the background, which cannot reflect the real scene. There are deficiencies in using visible light or infrared images alone. For these two complementary images, image fusion technology can effectively synthesize and explore their feature information, enhance scene understanding...

AI Technical Summary

Problems solved by technology

The document "Adaptive fusion method of visible light and infrared images based on non-subsampled shearlet transform and fast non-negative matrix factorization [J], Infrared Physics & Technology, 2014, 67: 161–172" proposed a method using NSST and fast non-negative matrix factorization (FNMF) fusion method, which uses non-negative matrix to guide the fusion of low-frequency coefficients, but the brightness of the fused image is dark, and many texture details of visible light are lost; the literature "A fusion algorithm for infrared and visible images based onsaliency analysis and non-subsampled Shearlet transform[J], Infrared Physics & Technology, 2015, 73: 286–297" proposes a method based on the combination of saliency detection and NSST, which uses saliency detection to integrate infrared target information into optical In the image, but the fusion effect of background information needs to be improved; the document "Visible and infrared image fusion using NSST and deep Boltzmann machine [J], Optik, 2018, 157: 334–342" proposes a method based on deep Boltzmann machine and The method of combining NSST, the idea is relatively new, but the application of deep learning in the field of image fusion is not yet mature, so it needs to be improved; in the document "Afusion algorithm for infrared and visible images based on RDU-PCNN and ICA-bases in NSST domain [J], Infrared Physics & Technology, 2016, 79: 183–190 "combined PCNN and NSST to process IR and VI image fusion. Although PCNN has a bionic mechanism, the fusion image introduces artifacts and has hazy image edges; literature "Technique for gray-scale visual light and infrared image fusion based on non-subsampled shearlet transform, Infrared Phys ics&Technology, 2014, 63:110–118" proposed a fusion framework based on regional average energy (RAE) and local directional contrast (LDC) in the NSST domain, which has good timeliness, but the fused image loses some important infrared salient information

Images