Kernel Regression-Based Image Processing ToolBox for MATLAB


Kernel Regression-Based Image Processing ToolBox is a command-line based software package for MATLAB, which was developed at MDSP research laboratory in UCSC. The package is capable of performing several general image processing tasks; for instance,

  1. Image denoising
    • Gaussian noise removal
    • Compression artifact reduction
    • Film grain reduction
    • Salt & pepper noise reduction
  2. Image reconstruction (interpolation + denoising)
    • Image upscale (reconstruction an image from regularly sampled data set)
    • Image reconstruction from an irregularly sampled data set
    • Image fusion
    • Super-resolution

We are demonstrating examples in here.



Relevant publications

  1. Takeda, H., S. Farsiu, and P. Milanfar, "Kernel Regression for Image Processing and Reconstruction." IEEE Transactions on Image Processing, Vol. 16, No. 2, pp. 349-366, February 2007.

  2. Takeda, H., S. Farsiu, and P. Milanfar, "Robust Kernel Regression for Restoration and Reconstruction of Images from Sparse Noisy Data," Proceedings of the International Conference on Image Processing (ICIP), Atlanta, GA, October 2006.

  3. Hiroyuki Takeda, Ph.D. Thesis, "Locally Adaptive Kernel Regression Methods for Multi-dimensional Signal Processing", Electrical Engineering, UC Santa Cruz, September 2010.



The latest version of the software package (ver. 1.2 beta) can be downloaded from here.

This is experimental software. It is provided for noncommercial research purposes only. Use at your own risk. No warranty is implied by this distribution. Copyright © 2007 by University of California.



User's manual

The manual is also downloadable as either PPT or PDF.




This work was supported in part by the US Air Force Grant F49620-03-1-0387.

last update on January 21st, 2011