Fusion of Infrared and Visible Images Using Fast Global Smoothing Decomposition and Target-Enhanced Parallel Gaussian Fuzzy Logic.

Zhisheng Wang,Chaowei Duan,Changda Xing,Yiliu Liu

doi:10.3390/s22010040

Zhisheng Wang, Chaowei Duan + Show 2 more

Open Access

https://doi.org/10.3390/s22010040

Copy DOI

Abstract

As a powerful technique to merge complementary information of original images, infrared (IR) and visible image fusion approaches are widely used in surveillance, target detecting, tracking, and biological recognition, etc. In this paper, an efficient IR and visible image fusion method is proposed to simultaneously enhance the significant targets/regions in all source images and preserve rich background details in visible images. The multi-scale representation based on the fast global smoother is firstly used to decompose source images into the base and detail layers, aiming to extract the salient structure information and suppress the halos around the edges. Then, a target-enhanced parallel Gaussian fuzzy logic-based fusion rule is proposed to merge the base layers, which can avoid the brightness loss and highlight significant targets/regions. In addition, the visual saliency map-based fusion rule is designed to merge the detail layers with the purpose of obtaining rich details. Finally, the fused image is reconstructed. Extensive experiments are conducted on 21 image pairs and a Nato-camp sequence (32 image pairs) to verify the effectiveness and superiority of the proposed method. Compared with several state-of-the-art methods, experimental results demonstrate that the proposed method can achieve more competitive or superior performances according to both the visual results and objective evaluation.

Highlights

As is well known, infrared (IR) imaging is playing an increasingly significant role in various ground object identification cases, such as camouflage recognition and hidden targets [1]
In order to verify the effectiveness and superiority of our multi-scale decomposition using the fast global smoother (MFGS) fusion method, a significant amount of experiments are conducted to compare the proposed method with nine state-of-the-art fusion methods including non-subsampled contourlet transform (NSCT) [33], HyMSD [23], CSR [14], GTF [15], VSMWLS [24], Convolutional neural networks (CNN) [55], DLVGG [40], ResNet [41], and TE
For making a fair comparison, the experimental parameters of NSCT are set in the light of [57], and the experimental setup of HyMSD, CSR, GTF, VSMWLS, CNN, DLVGG, ResNet, and TE are set as the original papers, respectively

Summary

Introduction

As is well known, infrared (IR) imaging is playing an increasingly significant role in various ground object identification cases, such as camouflage recognition and hidden targets [1]. IR images can reveal the thermal radiation difference of diverse objects, which can well distinguish the targets from their backgrounds. IR images typically have inferior detail textures and low-definition backgrounds. The visible imaging technology is able to record the reflected lights of objects. The visible images can provide more considerable texture details and far greater clarity than IR images. It tends to be affected by foul weather. To acquire sufficient information for accurate scenario analysis, users usually require to serially analyze multiple images with different imaging forms of a scene. No doubt analyzing multi-modality images of a scene, one by one, brings some problems (e.g., needing more time and more work) to users. The fused image can provide an enhanced vision of a scene and preserve the useful features of source

Methods

Results

Conclusion