GDMQI

Generalized Dynamic Morphological Quotient Image

By Sanghun Lee (Google Scholar, Website)

Yonsei Univ.

Introduction

This repository contains a Matlab implementation of "Multiscale morphology based illumination normalization with enhanced local textures for face recognition" (http://dx.doi.org/10.1016/j.eswa.2016.06.039) (pdf)

Note

• We converted the original version of GDMQI written in C++ (using OpenCV) into this Matlab version.
• It is not optimized version.
• Only 'Cropped version of Extended Yale-B database' was tested.

How to Use

1. Download and unzip 'Cropped version of Extended Yale-B database' (Link)
2. Open 'main.m' file
3. Modify DB_LOCATION to specify the database location, (ex. DB_LOCATION = 'D:\Database\YaleB';)
4. run

Citation

If you use GDMQI in your research, please cite:

@article{lee2016multiscale,
	title={Multiscale morphology based illumination normalization with enhanced local textures for face recognition},
	author={Lee, Sanghun and Lee, Chulhee},
	journal={Expert Systems with Applications},
	volume={62},
	pages={347--357},
	year={2016},
	publisher={Elsevier}
}