Infrared and visible image fusion with supervised convolutional neural network

An, Wen-Bo An; Wang, Hongmei

doi:10.1016/j.ijleo.2020.165120

Cited by 24 publications

(10 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [32], Li et al have used ResNet and zero-phase component analysis, which achieve the good fusion performances. More generally, many convolutional neural networks (CNN) based methods were proposed for the IR and visible image fusion [33][34][35][36]. Unlike CNN and ResNet models, Ma et al [37] have adopted DDcGAN (Dual-discriminator Conditional Generative Adversarial Network) to attain the fusion outputs with the enhanced targets, which facilitates the understandings of scenes for humans.…”

Section: Related Workmentioning

confidence: 99%

Infrared and Visible Image Fusion Using Truncated Huber Penalty Function Smoothing and Visual Saliency Based Threshold Optimization

et al. 2021

View full text Add to dashboard Cite

An efficient method for the infrared and visible image fusion is presented using truncated Huber penalty function smoothing and visual saliency based threshold optimization. The method merges complementary information from multimodality source images into a more informative composite image in two-scale domain, in which the significant objects/regions are highlighted and rich feature information is preserved. Firstly, source images are decomposed into two-scale image representations, namely, the approximate and residual layers, using truncated Huber penalty function smoothing. Benefiting from the edge- and structure-preserving characteristics, the significant objects and regions in the source images are effectively extracted without halo artifacts around the edges. Secondly, a visual saliency based threshold optimization fusion rule is designed to fuse the approximate layers aiming to highlight the salient targets in infrared images and remain the high-intensity regions in visible images. The sparse representation based fusion rule is adopted to fuse the residual layers with the goal of acquiring rich detail texture information. Finally, combining the fused approximate and residual layers reconstructs the fused image with more natural visual effects. Sufficient experimental results demonstrate that the proposed method can achieve comparable or superior performances compared with several state-of-the-art fusion methods in visual results and objective assessments.

show abstract

Section: Related Workmentioning

confidence: 99%

Infrared and Visible Image Fusion Using Truncated Huber Penalty Function Smoothing and Visual Saliency Based Threshold Optimization

et al. 2021

View full text Add to dashboard Cite

show abstract

“…The feature dimension is not proportional to the features it describes. The increase of the dimension may lead to the increase of the classification time and the invalid dimension reduced accuracy due to interference [15][16].…”

Section: Neuron Modelingmentioning

confidence: 99%

Lung Cancer Detection Considering Convolutional Autoencoder Neural Network

2020

View full text Add to dashboard Cite

Recently, smog has reappeared, which is seriously affecting people's lives and endangering people's health. Under the influence of air pollution such as smog and PM2.5, the number of lung cancer patients in various countries and even the world has shown a significant upward trend. This paper aims to investigate lung cancer detection considering convolutional autoencoder neural networks. After studying the research status at home and abroad, this paper further discusses the existing problems and problems. For big data CT images, an automatic diagnosis model of benign lung nodules based on the established automatic neural network. Variants are saved as examples. Using samples to train a nonlinear network model to achieve objective diagnosis of benign and malignant pulmonary nodules. This method can improve the classification accuracy and classification speed of pulmonary nodules on the basis of avoiding complex algorithms such as feature extraction. By building a neural network model, this work reduces the complexity of the algorithm, improves the overall detection rate of pulmonary nodules, and reduces the false-positive rate and missed-diagnosis rate during testing and verification of large data samples. This provides doctors with an accurate, efficient and convenient way of diagnosis and has a positive effect on the early diagnosis and treatment of lung cancer.

show abstract

“…Li et al [ 14 ] decomposed the source images into the basic part and the detailed part, respectively, and then directly used the weighted average method to fuse the basic part, and used the deep learning framework to extract features for the detailed part, and finally reconstruct the fused image. In addition to extracting the feature information, Wen-Bo An et al [ 15 ] constructed a supervised convolutional network to fully extract the complementary information of infrared and visible images, and the obtained fusion image better retained the details in the original image. In addition, end-to-end image fusion methods are also developing continuously.…”

Section: Introductionmentioning

confidence: 99%

Fusion of Infrared and Visible Images Based on Three-Scale Decomposition and ResNet Feature Transfer

Zhang

et al. 2022

Entropy

View full text Add to dashboard Cite

Image fusion technology can process multiple single image data into more reliable and comprehensive data, which play a key role in accurate target recognition and subsequent image processing. In view of the incomplete image decomposition, redundant extraction of infrared image energy information and incomplete feature extraction of visible images by existing algorithms, a fusion algorithm for infrared and visible image based on three-scale decomposition and ResNet feature transfer is proposed. Compared with the existing image decomposition methods, the three-scale decomposition method is used to finely layer the source image through two decompositions. Then, an optimized WLS method is designed to fuse the energy layer, which fully considers the infrared energy information and visible detail information. In addition, a ResNet-feature transfer method is designed for detail layer fusion, which can extract detailed information such as deeper contour structures. Finally, the structural layers are fused by weighted average strategy. Experimental results show that the proposed algorithm performs well in both visual effects and quantitative evaluation results compared with the five methods.

show abstract

Infrared and visible image fusion with supervised convolutional neural network

Cited by 24 publications

References 24 publications

Infrared and Visible Image Fusion Using Truncated Huber Penalty Function Smoothing and Visual Saliency Based Threshold Optimization

Infrared and Visible Image Fusion Using Truncated Huber Penalty Function Smoothing and Visual Saliency Based Threshold Optimization

Lung Cancer Detection Considering Convolutional Autoencoder Neural Network

Fusion of Infrared and Visible Images Based on Three-Scale Decomposition and ResNet Feature Transfer

Contact Info

Product

Resources

About