Image super-resolution using a dilated convolutional neural network

Lin, Guimin; Wu, Qingxiang; Qiu, Lida; Huang, Xixian

doi:10.1016/j.neucom.2017.09.062

Cited by 78 publications

(44 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In (7), the sigmoid function maps the value of σ j onto [0, 1] and σ j gets a value in the range [σ j,min , σ j,max ]. Similarly, µ j is mapped within the range [λ j,min − bσ j , λ j,max − bσ j ] in (8). The guard gap of bσ j guarantees that the width of the band lie within the limits, [λ j,min , λ j,max ].…”

Section: Network Architecturementioning

confidence: 99%

“…Their value can be any real number. For back-propagation, we can easily calculate partial derivatives needed to optimize µ j and σ j by calculating partial derivatives of (7) and (8) and combining with (5) and 6using the chain rule.…”

Section: Network Architecturementioning

confidence: 99%

“…In the past, super-resolution techniques have been primarily applied to hyperspectral remote sensing for improving the spatial resolution of the HS images by fusing a low spatial resolution HS image with a high spatial resolution panchromatic or MS image of the same scene [5]. However, there have been growing interests in methods, called single image super-resolution, that map low spatial resolution HS image to high spatial resolution image without the use of high spatial resolution panchromatic or MS imagery [6][7][8]. These methods utilize prior knowledge about spatial distribution of HS images, which is learned by training models on large dataset of imagery.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Spectral Super-Resolution with Optimized Bands

2019

View full text Add to dashboard Cite

Hyperspectral (HS) sensors sample reflectance spectrum in very high resolution, which allows us to examine material properties in very fine details. However, their widespread adoption has been hindered because they are very expensive. Reflectance spectra of real materials are high dimensional but sparse signals. By utilizing prior information about the statistics of real HS spectra, many previous studies have reconstructed HS spectra from multispectral (MS) signals (which can be obtained from cheaper, lower spectral resolution sensors). However, most of these techniques assume that the MS bands are known apriori and do not optimize the MS bands to produce more accurate reconstructions. In this paper, we propose a new end-to-end fully convolutional residual neural network architecture that simultaneously learns both the MS bands and the transformation to reconstruct HS spectra from MS signals by analyzing large quantity of HS data. The learned band can be implemented in hardware to obtain an MS sensor that collects data that is best to reconstruct HS spectra using the learned transformation. Using a diverse set of real-world datasets, we show how the proposed approach of optimizing MS bands along with the transformation can drastically increase the reconstruction accuracy. Additionally, we also investigate the prospects of using reconstructed HS spectra for land cover classification.

show abstract

Section: Network Architecturementioning

confidence: 99%

Section: Network Architecturementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Spectral Super-Resolution with Optimized Bands

2019

View full text Add to dashboard Cite

show abstract

“…To solve this problem, Lin et al proposed a dilated CNN to enlarge receptive field size for the image super-resolution [11]. Kim et al [12] proposed Very Deep Super-resolution Convolutional Networks (VDSR) with 20 layers to obtain large receptive field, and learned only the residuals between the LR image and the HR image to accelerate the convergence speed.…”

Section: Introductionmentioning

confidence: 99%

Remote Sensing Imagery Super Resolution Based on Adaptive Multi-Scale Feature Fusion Network

Wang

Yang

et al. 2020

Sensors

View full text Add to dashboard Cite

Due to increasingly complex factors of image degradation, inferring high-frequency details of remote sensing imagery is more difficult compared to ordinary digital photos. This paper proposes an adaptive multi-scale feature fusion network (AMFFN) for remote sensing image super-resolution. Firstly, the features are extracted from the original low-resolution image. Then several adaptive multi-scale feature extraction (AMFE) modules, the squeeze-and-excited and adaptive gating mechanisms are adopted for feature extraction and fusion. Finally, the sub-pixel convolution method is used to reconstruct the high-resolution image. Experiments are performed on three datasets, the key characteristics, such as the number of AMFEs and the gating connection way are studied, and super-resolution of remote sensing imagery of different scale factors are qualitatively and quantitatively analyzed. The results show that our method outperforms the classic methods, such as Super-Resolution Convolutional Neural Network(SRCNN), Efficient Sub-Pixel Convolutional Network (ESPCN), and multi-scale residual CNN(MSRN).

show abstract

“…Recently, deep learning and convolutional neural network (CNN) based methods have emerged as a promising solution for single images SR. These techniques include residual dense network [10], dilated convolutional neural [11], deep convolution neural network [12] and many more. However, in video frames, SR techniques focus on exploiting the temporal and spatial correlations thus it becomes more challenging to perform SR for videos.…”

Section: Introductionmentioning

confidence: 99%

Align-Filter & Learn Video Super Resolution using Deep learning (AFLVSR)

2019

IJRTE

View full text Add to dashboard Cite

In this article, we focused on the super-resolution technique in computer vision applications. During last decade, image super-resolution techniques have been introduced and adopted widely in various applications. However, increasing demand of high quality multimedia data has lead towards the high resolution data streaming. The conventional techniques which are based on the image super-resolution are not suitable for multi-frame SR. Moreover, the motion estimation, motion compensation, spatial and temporal information extraction are the well-known challenging issues in video super-resolution field. In this work, we address these issues and developed deep learning based novel architecture which performs feature alignment, filtering the image using deep learning and estimates the residual of low-resolution frames to generate the high-resolution frame. The proposed approach is named as Align-Filter & Learn Video Super resolution using Deep learning (AFLVSR). We have conducted and extensive experimental analysis which shows a significant improvement in the performance when compared with the state-of-art video SR techniques.

show abstract

Image super-resolution using a dilated convolutional neural network

Cited by 78 publications

References 19 publications

Spectral Super-Resolution with Optimized Bands

Spectral Super-Resolution with Optimized Bands

Remote Sensing Imagery Super Resolution Based on Adaptive Multi-Scale Feature Fusion Network

Align-Filter & Learn Video Super Resolution using Deep learning (AFLVSR)

Contact Info

Product

Resources

About