Skip to content

Decision Tree and Extreme Learning Machine based Single Image Super-resolution

Notifications You must be signed in to change notification settings

Rainyfish/DELM-SR

Repository files navigation

DELM-SR


SR-demo.m

SR-demo is the test function of DELM-SR. SR-demo has four input parameters, which can be introduced in details as follow:

  • lambda--------The regularization parameter of the ELM in each leaf nodes.
  • NumNodes ---- The number of hidden layer nodes.
  • path----------The path of trained models.
  • DataName------The testing dateset (Set5, Set14).
  • pattern-------The type of test images('png','tmp').

You can run SR_demo by using SR_demo(0.01,200,'model/model_91','Set14','.bmp') as a sample.


Trained models

You can get the trained models of DELM-SR and DELM-SR+ from BaiduYun

Test Datasets

We perform extensive quantitative and qualitative comparisons with current state-of-the-art SR algorithms on 5 datasets: Set5, Set14, BSDS100, Urban100, and Manga109. Set5 and Set14 are included. All the testing_datasets can be download from Link or BaiduYun.

Slide

Decision Tree and Extreme Learning Machine base Single Image Super-resolution_temp.pptx is the slide of our method. You can also download it from Here

About

Decision Tree and Extreme Learning Machine based Single Image Super-resolution

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published