Survey on CNN based super resolution methods

Super Resolution is a field of image analysis that focuses on boosting the resolution of photographs and movies without compromising detail or visual appeal, instead enhancing both. Multiple (many input images and one output image) or single (one input and one output) stages are used to convert low-resolution photos to high-resolution photos. The study examines super-resolution methods based on a convolutional neural network (CNN) for super-resolution mapping at the sub-pixel level, as well as its primary characteristics and limitations for noisy or medical images.

Download Full-text

CONVOLUTIONAL NEURAL NETWORK BASED DEM SUPER RESOLUTION

ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences ◽

10.5194/isprs-archives-xli-b3-247-2016 ◽

2016 ◽

Vol XLI-B3 ◽

pp. 247-250 ◽

Cited By ~ 1

Author(s):

Zixuan Chen ◽

Xuewen Wang ◽

Zekai Xu ◽

Wenguang Hou

Keyword(s):

Neural Network ◽

High Resolution ◽

Convolutional Neural Network ◽

Super Resolution ◽

Low Resolution ◽

Practical Applications ◽

Interpolation Methods ◽

New Strategy

DEM super resolution is proposed in our previous publication to improve the resolution for a DEM on basis of some learning examples. Meanwhile, the nonlocal algorithm is introduced to deal with it and lots of experiments show that the strategy is feasible. In our publication, the learning examples are defined as the partial original DEM and their related high measurements due to this way can avoid the incompatibility between the data to be processed and the learning examples. To further extent the applications of this new strategy, the learning examples should be diverse and easy to obtain. Yet, it may cause the problem of incompatibility and unrobustness. To overcome it, we intend to investigate a convolutional neural network based method. The input of the convolutional neural network is a low resolution DEM and the output is expected to be its high resolution one. A three layers model will be adopted. The first layer is used to detect some features from the input, the second integrates the detected features to some compressed ones and the final step transforms the compressed features as a new DEM. According to this designed structure, some learning DEMs will be taken to train it. Specifically, the designed network will be optimized by minimizing the error of the output and its expected high resolution DEM. In practical applications, a testing DEM will be input to the convolutional neural network and a super resolution will be obtained. Many experiments show that the CNN based method can obtain better reconstructions than many classic interpolation methods.

Download Full-text

A Super-Resolution Mapping Using a Convolutional Neural Network

IGARSS 2019 - 2019 IEEE International Geoscience and Remote Sensing Symposium ◽

10.1109/igarss.2019.8898537 ◽

2019 ◽

Author(s):

Teerasit Kasetkasem

Keyword(s):

Neural Network ◽

Convolutional Neural Network ◽

Super Resolution ◽

Resolution Mapping ◽

Super Resolution Mapping

Download Full-text

Super-resolution reconstruction of seismic section image via multi-scale convolution neural network

E3S Web of Conferences ◽

10.1051/e3sconf/202130301058 ◽

2021 ◽

Vol 303 ◽

pp. 01058

Author(s):

Meng-Di Deng ◽

Rui-Sheng Jia ◽

Hong-Mei Sun ◽

Xing-Li Zhang

Keyword(s):

Neural Network ◽

High Resolution ◽

Convolutional Neural Network ◽

Super Resolution ◽

Image Features ◽

Image Feature ◽

Reconstruction Method ◽

Low Resolution ◽

Seismic Section ◽

Multi Scale

The resolution of seismic section images can directly affect the subsequent interpretation of seismic data. In order to improve the spatial resolution of low-resolution seismic section images, a super-resolution reconstruction method based on multi-scale convolution is proposed. This method designs a multi-scale convolutional neural network to learn high-low resolution image feature pairs, and realizes mapping learning from low-resolution seismic section images to high-resolution seismic section images. This multi-scale convolutional neural network model consists of four convolutional layers and a sub-pixel convolutional layer. Convolution operations are used to learn abundant seismic section image features, and sub-pixel convolution layer is used to reconstruct high-resolution seismic section image. The experimental results show that the proposed method is superior to the comparison method in peak signal-to-noise ratio (PSNR) and structural similarity (SSIM). In the total training time and reconstruction time, our method is about 22% less than the FSRCNN method and about 18% less than the ESPCN method.

Download Full-text

CONVOLUTIONAL NEURAL NETWORK BASED DEM SUPER RESOLUTION

ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences ◽

10.5194/isprsarchives-xli-b3-247-2016 ◽

2016 ◽

Vol XLI-B3 ◽

pp. 247-250 ◽

Cited By ~ 6

Author(s):

Zixuan Chen ◽

Xuewen Wang ◽

Zekai Xu ◽

Wenguang Hou

Keyword(s):

Neural Network ◽

High Resolution ◽

Convolutional Neural Network ◽

Super Resolution ◽

Low Resolution ◽

Practical Applications ◽

Interpolation Methods ◽

New Strategy

DEM super resolution is proposed in our previous publication to improve the resolution for a DEM on basis of some learning examples. Meanwhile, the nonlocal algorithm is introduced to deal with it and lots of experiments show that the strategy is feasible. In our publication, the learning examples are defined as the partial original DEM and their related high measurements due to this way can avoid the incompatibility between the data to be processed and the learning examples. To further extent the applications of this new strategy, the learning examples should be diverse and easy to obtain. Yet, it may cause the problem of incompatibility and unrobustness. To overcome it, we intend to investigate a convolutional neural network based method. The input of the convolutional neural network is a low resolution DEM and the output is expected to be its high resolution one. A three layers model will be adopted. The first layer is used to detect some features from the input, the second integrates the detected features to some compressed ones and the final step transforms the compressed features as a new DEM. According to this designed structure, some learning DEMs will be taken to train it. Specifically, the designed network will be optimized by minimizing the error of the output and its expected high resolution DEM. In practical applications, a testing DEM will be input to the convolutional neural network and a super resolution will be obtained. Many experiments show that the CNN based method can obtain better reconstructions than many classic interpolation methods.

Download Full-text

Super-Resolution Land Cover Mapping Based on the Convolutional Neural Network

Remote Sensing ◽

10.3390/rs11151815 ◽

2019 ◽

Vol 11 (15) ◽

pp. 1815 ◽

Cited By ~ 7

Author(s):

Jia ◽

Ge ◽

Chen ◽

Li ◽

Heuvelink ◽

...

Keyword(s):

Neural Network ◽

Remote Sensing ◽

Land Cover ◽

Convolutional Neural Network ◽

Super Resolution ◽

Fine Scale ◽

Remote Sensing Images ◽

Resolution Mapping ◽

Super Resolution Mapping ◽

Land Cover Maps

Super-resolution mapping (SRM) is used to obtain fine-scale land cover maps from coarse remote sensing images. Spatial attraction, geostatistics, and using prior geographic information are conventional approaches used to derive fine-scale land cover maps. As the convolutional neural network (CNN) has been shown to be effective in capturing the spatial characteristics of geographic objects and extrapolating calibrated methods to other study areas, it may be a useful approach to overcome limitations of current SRM methods. In this paper, a new SRM method based on the CNN (SRMCNN) is proposed and tested. Specifically, an encoder-decoder CNN is used to model the nonlinear relationship between coarse remote sensing images and fine-scale land cover maps. Two real-image experiments were conducted to analyze the effectiveness of the proposed method. The results demonstrate that the overall accuracy of the proposed SRMCNN method was 3% to 5% higher than that of two existing SRM methods. Moreover, the proposed SRMCNN method was validated by visualizing output features and analyzing the performance of different geographic objects.

Download Full-text

Dynamic Convolutional Neural Network For Image Super-resolution

Journal of Advanced College of Engineering and Management ◽

10.3126/jacem.v3i0.18808 ◽

2018 ◽

Vol 3 ◽

pp. 1

Author(s):

Anil Bhujel ◽

Dibakar Raj Pant

Keyword(s):

Neural Network ◽

High Resolution ◽

Convolutional Neural Network ◽

Super Resolution ◽

Low Resolution ◽

Scaling Factors ◽

Resolution Image ◽

High Resolution Image ◽

Image Super Resolution ◽

Up Scaling

Single image super-resolution (SISR) is a technique that reconstructs high resolution image from single low resolution image. Dynamic Convolutional Neural Network (DCNN) is used here for the reconstruction of high resolution image from single low resolution image. It takes low resolution image as input and produce high resolution image as output for dynamic up-scaling factor 2, 3, and 4. The dynamic convolutional neural network directly learns an end-to-end mapping between low resolution and high resolution images. The CNN trained simultaneously with images up-scaled by factors 2, 3, and 4 to make it dynamic. The system is then tested for the input images with up-scaling factors 2, 3 and 4. The dynamically trained CNN performs well for all three up-scaling factors. The performance of network is measured by PSNR, WPSNR, SSIM, MSSSIM, and also by perceptual.Journal of Advanced College of Engineering and Management, Vol. 3, 2017, Page: 1-10

Download Full-text