A deep learning method for image super-resolution based on geometric similarity

Lü, Jian; Hu, Weidong; Sun, Yi

doi:10.1016/j.image.2018.10.003

Cited by 21 publications

(6 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The PSNR found to be improved as compared to above-mentioned techniques although similar to [3]. But the SSIM results of [2] are quite variable with respect to input image. For images with variation in textures, SSIM is found highest but for smooth images it is quite lesser as compared with existing methods [3], [25].…”

Section: A Techniques For Single Image Super Resolution (Sisr)mentioning

confidence: 69%

“…A survey [1] generalizes the single image SR imaging model as, L ( x , y:c ) =W *(B *(H (m ,n:c ) )) +h (2) , where L & H are input LR & output HR images respectively. W is a warping function and B is a blurring kernel.…”

Section: A Techniques For Single Image Super Resolution (Sisr)mentioning

confidence: 99%

“…In [2], LR to HR reconstruction occurs gradually using 11 layered deep convolution neural network (CNN) with segregation of layers as 3 input layers, 4 spatial transform network (STN) layers and last 4 de-convolution layers. The process [2] involves three steps: 1. Similar patches are collected & fed to the CNN.…”

Section: A Techniques For Single Image Super Resolution (Sisr)mentioning

confidence: 99%

“…In more simpler terms, many super resolution techniques have evolved as & when a need for it arisen, viz. in area of face recognition [2] with surveillance applications [2], [3], [4], satellite imaging [5], [6] with an objective of recognizing special image parts like watery bodies, forest areas which is required in military applications, Diagnosis of diseases in medical science [7], [8]. In short, depending on application areas the usability of a particular algorithm or technique changes.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A New Optimized Recurrent Feedback Deep Convolutional Neural Net for Image Super Resolution

Borse*¹,

Patil²

2019

IJRTE

View full text Add to dashboard Cite

Now-a-days many applications dealing with visual content need to access underlying details in the image or video of interest. For instance, detailing is required to take life critical decisions for further action plans by a doctor. Clarity and structural information are some of the aspects of detailing. It can be achieved by cost effective software solution like super resolution reconstruction of an image. Super resolution (SR) deals in increasing resolution of an image to make it more clear and valid for use. Many SR techniques exist with variable goals to achieve. With this intension a new technique for preserving structural information in the reconstruction process is proposed. The system extends a deep convolution neural network by adding a new optimization layer at the end of network activation layer. This new layer maintains permissible error threshold in the acquired signal and tries to improve the signal by feeding back latest reconstructed frame. The proposed system shows noticeable improvement in structural similarity of reconstructed images as compared with the ground truth.

show abstract

Section: A Techniques For Single Image Super Resolution (Sisr)mentioning

confidence: 69%

Section: A Techniques For Single Image Super Resolution (Sisr)mentioning

confidence: 99%

Section: A Techniques For Single Image Super Resolution (Sisr)mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A New Optimized Recurrent Feedback Deep Convolutional Neural Net for Image Super Resolution

Borse*¹,

Patil²

2019

IJRTE

View full text Add to dashboard Cite

show abstract

“…Nowadays, there are promising studies on using deep convolutional networks [7], [8] to perform single image super-resolution (SISR), where the goal is to recover one * Corresponding author's e-mail: romanukevadimv@gmail.com high-resolution image from one low-resolution image. SISR is challenging because high-frequency image content typically cannot be recovered from the low-resolution image.…”

Section: Analysis Of the Backgroundmentioning

confidence: 99%

An Improvement of the VDSR Network for Single Image Super-Resolution by Truncation and Adjustment of the Learning Rate Parameters

Romanuke

2019

Applied Computer Systems

View full text Add to dashboard Cite

A problem of single image super-resolution is considered, where the goal is to recover one high-resolution image from one low-resolution image. Whereas this problem has been successfully solved so far by the known VDSR network, such an approach still cannot give an overall beneficial effect compared to bicubic interpolation. This is so due to the fact that the image reconstruction quality has been estimated separately by three subjective factors. Moreover, the original VDSR network consisting of 20 convolutional layers is apparently not optimal by its depth. This is why here those factors are aggregated, and the network performance is deemed by a single estimator. Then the depth is tried to be decreased (truncation) along with adjusting the learning rate drop factor. Finally, a plausible improvement of the VDSR network is confirmed. The best truncated network, performing by almost 3.2 % better than bicubic interpolation, occupies less memory space and is about 1.44 times faster than the original VDSR network for images of a medium size.

show abstract

Paradigm shifts in super-resolution techniques for remote sensing applications

Rohith¹,

Kumar²

2020

Vis Comput

View full text Add to dashboard Cite

A deep learning method for image super-resolution based on geometric similarity

Cited by 21 publications

References 38 publications

A New Optimized Recurrent Feedback Deep Convolutional Neural Net for Image Super Resolution

A New Optimized Recurrent Feedback Deep Convolutional Neural Net for Image Super Resolution

An Improvement of the VDSR Network for Single Image Super-Resolution by Truncation and Adjustment of the Learning Rate Parameters

Paradigm shifts in super-resolution techniques for remote sensing applications

Contact Info

Product

Resources

About