DocumentCode
3695506
Title
Histogram based image enhancement for non-uniformly illuminated and low contrast images
Author
Teck Long Kong;Nor Ashidi Mat Isa
Author_Institution
Imaging and Intelligent Systems Research Team (ISRT), School of Electrical and Electronic Engineering, Universiti Sains Malaysia, 14300 Nibong Tebal, Penang, Malaysia
fYear
2015
fDate
6/1/2015 12:00:00 AM
Firstpage
586
Lastpage
591
Abstract
Non-uniform illumination and low contrast are commonly encountered in low light level night vision equipment and mammogram. This study presents a new method to enhance the non-uniformly illuminated and low contrast images. The proposed method first derives a bright enhancer and a dark enhancer based on the resultant image produced by a mean filter. By applying enhancer on the image, the dark region would be brighter and bright region would be darker. This will improves the non-uniformly illuminated image. Both qualitative and quantitative analyses show that the proposed method has successfully produced better performance in term of measure of enhancement as compare to three state-of-art methods namely Histogram Equalization, Contrast-Limited Adaptive Histogram Equalization and Adjust Image Intensity Values which available in Matlab Image Processing Toolbox.
Keywords
"Histograms","Brightness","Face","Lighting","Adaptive equalizers","MATLAB","Image processing"
Publisher
ieee
Conference_Titel
Industrial Electronics and Applications (ICIEA), 2015 IEEE 10th Conference on
Type
conf
DOI
10.1109/ICIEA.2015.7334179
Filename
7334179
Link To Document