czla / Deblurring

Nonblind Deblurring Real Image/图像去模糊

Nonblind Deblurring Real Image

---

1. Discription

This project deblurring real image(e.g. image with motion blur and out-of-focus blur) quickly with PSF.

1.1 Motion Deblur

For motion deblur, it uses Lucy-Richardson algorithm to estimate Point-Spread-Function(PSF). There are three main parameters to set: len(i.e. blur length), theta(i.e. blur angle) and IterNum(number of iteration times).

• Result on real image

1.2 Out-of-focus Deblur

For out-of-focus deblur, it uses Wiener filter algorithm. There are three main parameters to set:radius(i.e. the radius of Point-Spread-Function(PSF)), smooth(smooth factor to control K) and dering('On' means do suppress ringing effect while other means do not.

• Result on real image

2. Usage

1. Add the directory to the matlab path;
2. Run 'Demo_motion_blur.m' to perform motion deblur;
3. Run 'Demo_out_of_focus_deblur.m' to perform out-of-focus deblur.