Image Fusion and Super-Resolution with Convolutional Neural Network

Zhong, Jinying; Yang, Bin; Li, Yuehua; Zhong, Fei; Chen, Zhongze

doi:10.1007/978-981-10-3005-5_7

Cited by 28 publications

(14 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For fair comparison, we compare our proposed image fusion method with the state‐of‐the‐art methods, following [14]: the cross bilateral filter fusion method (CBF) [40], the JSR model [24], the JSR model with saliency detection fusion method (JSRSD) [42], the weighted least square optimisation‐based method (WLS) [41], the ConvSR model [27] and the VGG multi‐layers based method [14]. The methods of CBF and WLS are conventional fusion methods, JSR and JSRSD are shallow‐learning based fusion methods, and ConvSR and VGG are deep‐learning based methods.…”

Section: Resultsmentioning

confidence: 99%

“…However, there are many free parameters that impact the fusion results and have not been well studied. Towards the limited resolution of most source images, Zhong et al [27] introduced a joint image fusion and super‐resolution method, which was depended on a CNN pre‐trained by 90 selected images from the ImageNet database. Although the fusion results have higher resolution than original images, the pre‐trained convolutional filters are used for target recognition instead of image fusion.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Infrared and visible image fusion using a shallow CNN and structural similarity constraint

Xia

Han

et al. 2020

IET image process

View full text Add to dashboard Cite

Section: Resultsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Infrared and visible image fusion using a shallow CNN and structural similarity constraint

Xia

Han

et al. 2020

IET image process

View full text Add to dashboard Cite

“…It divides the input feature map xmk−1 into multiple non-overlapping n×n image blocks by sliding down the sampling window, and then calculates the mean value of the pixels in each image block. Therefore, the output image is reduced by n times in both dimensions [26].…”

Section: Dnn Positioning Modulementioning

confidence: 99%

Deep Learning for Fingerprint-Based Outdoor Positioning via LTE Networks

Lei

2019

Sensors

View full text Add to dashboard Cite

Fingerprint-based positioning techniques are a hot research topic because of their satisfactory accuracy in complex environments. In this study, we adopted the deep-learning-based long-time-evolution (LTE) signal fingerprint positioning method for outdoor environment positioning. Inspired by state-of-the-art image classification methods, a novel hybrid location gray-scale image utilizing LTE signal fingerprints is proposed in this paper. In order to deal with signal fluctuations, several data enhancement methods are adopted. A hierarchical architecture is put forward during the deep neural network (DNN) training. First, the proposed positioning technique is pre-trained by a modified Deep Residual Network (Resnet) coarse localizer which is capable of learning reliable features from a set of unstable LTE signals. Then, to alleviate the tremendous collection workload, as well as further improve the positioning accuracy, by using a multilayer perceptron (MLP), a transfer learning-based fine localizer is introduced for fine-tuning the coarse localizer. The experimental data was collected from realistic scenes to meet the requirement of actual environments. The experimental results show that the proposed system leads to a considerable positioning accuracy in a variety of outdoor environments.

show abstract

Section: Video Scaling Issuesmentioning

confidence: 99%

“…In [13], Image Fusion and Super-Resolution with Convolutional Neural Network adopted to eliminate blurriness and provide sharp images for digital photography. In this process Zhong et al [13] faces pixel level image fusion issues. In [18], 3D Video SuperResolution Using Fully Convolutional Neural Networks has been proposed to sort out redundancy, degradation in quality of fused image and huge data size problems.…”

Section: Video Scaling Issuesmentioning

confidence: 99%

Real-Time Video Scaling Based on Convolution Neural Network Architecture

Safinaz¹,

Kumar²

2017

IJIVP

View full text Add to dashboard Cite

In recent years, video super resolution techniques becomes mandatory requirements to get high resolution videos. Many super resolution techniques researched but still video super resolution or scaling is a vital challenge. In this paper, we have presented a real-time video scaling based on convolution neural network architecture to eliminate the blurriness in the images and video frames and to provide better reconstruction quality while scaling of large datasets from lower resolution frames to high resolution frames. We compare our outcomes with multiple exiting algorithms. Our

show abstract

Image Fusion and Super-Resolution with Convolutional Neural Network

Cited by 28 publications

References 15 publications

Infrared and visible image fusion using a shallow CNN and structural similarity constraint

Infrared and visible image fusion using a shallow CNN and structural similarity constraint

Deep Learning for Fingerprint-Based Outdoor Positioning via LTE Networks

Real-Time Video Scaling Based on Convolution Neural Network Architecture

Contact Info

Product

Resources

About