Skip to content

jbauer7/ShadowTerminator

Repository files navigation

Shadow Terminator:
Shadow terminator detects shadows in images then attemps to remove the 
shadow from the image. 
For more information go to this URL: https://sites.google.com/a/wisc.edu/shadow-terminator/
Authors: Jospeh Bauer, Elliot Busta, Eric Johnson
Other Contributions: RGB2Lab.m and Lab2RGB.m are functions provide by Professor for Homework #2

Running Shadow Terminator:
(1) Run Shadow_Terminator.m script
(2) Example images are located in the Input_Images/ directory
    labled: input1, input2,..., input6
(3) Output images in the Output_Images/ directory

The general method for the shadow detction and removal is as follows:
(1) Convert image to LAB color space
(2) Detect shadow using mean and standard deviation thresholds
(3) Find average luminosity of shadowed and non-shadowed region
(4) Replace pixel luminosity (L channel) of shadowed region with the mean
    from the non-shadow region using mean shift algorithm
(5) Use mean shift on shadow region on the A and B channels to adjust
    chromaticity 
(6) Smooth edges to increase continuity of image
(7) Convert back to RGB color space

About

Shadow Removal Algo

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages