EDSR (Enhanced Deep Super-Resolution) Matlab port

Version 1.0.4 (151 MB) by manoreken
EDSR (Enhanced Deep Super-Resolution) Single Image Super Resolution Matlab port
757 Downloads
Updated 31 Dec 2025

View License

EDSR (Enhanced Deep Super-Resolution) Single Image Super Resolution Matlab port.
■ Prerequisites ■
Matlab 2025b
Image Processing toolbox
Statistics and Machine Learning toolbox
Deep Learning Toolbox
Parallel Computing Toolbox
■ How to Test ■
Run EDSR_Test.m that calls EDSR_2xSuperResolution.m
Pre-trained network model/params is loaded on the line 5 of EDSR_2xSuperResolution.m
■ How to Perform EDSR Super-Resolution to your image file ■
Input image MyPicture.jpg should be pristine (not blurred) image. EDSR neural net will upscale the image by 2x.
img = imread("MyPicture.jpg"); % 1024x768 input image
imgSR = EDSR_2xSuperResolution(img);
imwrite(imgSR, "MyPicture_2xEDSR.png"); % 2048x1536 image is outputted
■ How to Train the network using Flickr2K dataset ■
Download Flickr2K dataset and place it on
Flickr2K/Flickr2K_HR subdirectory for train data of 2650 images
Run createTrainingSetAll_Flicker2K.m to create HighRes/LowRes image pair mat files
Run EDSR_Train_Flickr2K.m to train and create EDSR_trained_2x_Flickr2K.mat
rename it to EDSR_trained_2x.mat
■ Difference from original EDSR ■
1. Processes only Y component.
2. Only 2x super resolution is implemented.
■ Changelog ■
Version 20251231 1.0.4
・Fix off-by-one error bug on image crop
・Preprocess to select patch images that contains high frequency component.
・PSNR Comparison uses Bilinear method as rule-based upscaling.
Version 20210307 1.0.3
・Updated EDSR_trained_2x.mat : EDSR 32 ResBlock Flickr2K dataset trained file.
Version 20210228 1.0.2
・Updated EDSR_trained_2x.mat to Flickr2K dataset trained file to improve subjective SR quality.
Version 20210224 1.0.1
・Fix typo.
Version 20210223 1.0.0
・Initial release.
■ References ■
Enhanced Deep Residual Networks for Single Image Super-Resolution
http://openaccess.thecvf.com/content_cvpr_2017_workshops/w12/papers/Lim_Enhanced_Deep_Residual_CVPR_2017_paper.pdf
Keras implementation of EDSR
https://github.com/hieubkset/Keras-Image-Super-Resolution
Single Image Super-Resolution Using Deep Learning
(VDSR is implemented using Matlab Deep Learning Toolbox)
https://www.mathworks.com/help/images/single-image-super-resolution-using-deep-learning.html

Cite As

manoreken (2026). EDSR (Enhanced Deep Super-Resolution) Matlab port (https://in.mathworks.com/matlabcentral/fileexchange/87799-edsr-enhanced-deep-super-resolution-matlab-port), MATLAB Central File Exchange. Retrieved .

MATLAB Release Compatibility
Created with R2025b
Compatible with R2025b
Platform Compatibility
Windows macOS Linux
Tags Add Tags
Version Published Release Notes
1.0.4

Version 20251231 1.0.4

・Fix off-by-one error bug on image crop
・Preprocess to select patch images that contains high frequency component.
・PSNR Comparison uses Bilinear method as rule-based upscaling.

Version 20210307 1.0.3

・Updated EDSR_trained_2

1.0.3

ResBlock is increased from 16 to 32

1.0.2

Updated EDSR_trained_2x.mat to Flickr2K dataset trained file to improve subjective SR quality.

1.0.1

Document typo fix

1.0.0