Infrared Weak-Small Targets Fusion Based on Latent Low-Rank Representation and DWT

Xiaozhu Wang,Kai Zhang,Shayi Li,Jie Yan,Jianfei Yin

doi:10.1109/access.2019.2934523

Abstract

For the problem of anti-background interference of weak-small targets in infrared images, target extraction and texture detail processing are key tasks in the image fusion algorithm. The single-band infrared data can not fully reflect image details and contour information. There are texture differences in different bands of data, which makes it difficult to recognize targets. Therefore, it is necessary to fuse dual-band data to identify weak-small targets clearly. To solve these question, in this paper, we propose an effective image fusion framework using Latent Low-Rank Representation (LatLRR) and Discrete Wavelet Transform (DWT). Firstly, all source images are trained as L matrix by LatLRR which is used to extract salient features. And the original images are decomposed into high frequency and low frequency by DWT. Then high frequency parts are fused by maximum absolute value and low frequency parts are fused by weighted-average. On this basis, the training matrix L and high frequency fusion parts are used for contrast modulation fusion. Finally, the fused image is reconstructed by combining the contour parts and feature parts. The experimental results demonstrate that our proposed method achieves state-of-the-art performance in objective and subjective assessment.

Highlights

In infrared dual-band sensor, the infrared dual-band image fusion is an important task
Based on the above questions, we propose a fusion algorithm based on Latent Low-Rank Representation (LatLRR) and Discrete Wavelet Transform (DWT) for infrared dual-band image fusion
High frequency parts are fused by maximum absolute value and low frequency parts are fused by weighted-average, The training matrix and high frequency fusion parts are used for contrast modulation fusion

Summary

INTRODUCTION

In infrared dual-band sensor, the infrared dual-band image fusion is an important task. Because the multi-scale method can extract the edge and texture details of the image, so Huang et al proposed an infrared and visible image fusion method that was based on curvelet transformation and visual attention mechanisms Their model could elevate the signal-to-noise ratio of fused images and highlight dim targets [7]. B. RESEARCH ON INFRARED DUAL-BAND IMAGE FUSION ALGORITHM multi-scale transforms, sparse representation and deep learning methods obtain good fusion performance, but for infrared images fusion, the infrared image is different from the visible image. Sun et al [27] proposed dim small targets detection based on dual-band infrared image fusion They use wavelet transformation to decompose the source images. We select different algorithms to do contrast experiments for infrared weak-small targets dual-band images Through these experimental results, we can find that the LatLRR method is best than the other algorithms. The salient features are automatically extracted from original image so as to produce effective features for recognition

IMAGE DECOMPOSITION THEORY

THE ANALYSIS PRODUCE OF SALIENCY PARTS

EXPERIMENTAL RESULTS AND ANALYSIS

RECONSTRUCTION

CONCLUSION