Multi-Thresholded Histogram Equalization Based on Parameterless Artificial Bee Colony

Multi-Thresholded Histogram Equalization Based on Parameterless Artificial Bee Colony

Krishna Gopal Dhal (Midnapore College (Autonomous), India), Mandira Sen (Tata Consultancy Services, India), Swarnajit Ray (JBMatrix Technology Pvt. Ltd., India) and Sanjoy Das (University of Kalyani, India)
Copyright: © 2018 |Pages: 19
DOI: 10.4018/978-1-5225-5020-4.ch004

Abstract

This chapter presents a novel variant of histogram equalization (HE) method called multi-thresholded histogram equalization (MTHE), depending on entropy-based multi-level thresholding-based segmentation. It is reported that proper segmentation of the histogram significantly assists the HE variants to maintain the original brightness of the image, which is one of the main criterion of the consumer electronics field. Multi-separation-based HE variants are also very effective for multi-modal histogram-based images. But, proper multi-seaparation of the histogram increases the computational time of the corresponding HE variants. In order to overcome that problem, one novel parameterless artifical bee colony (ABC) algorithm is employed to solve the multi-level thresholding problem. Experimental results prove that proposed parameterless ABC helps to reduce the computational time significantly and the proposed MTHE outperforms several existing HE varints in brightness preserving histopathological image enhancement domain.
Chapter Preview
Top

Introduction

Contrast enhancement methods have been applied for better visual interpretation. Histogram Equalization (HE) is one of the most simple and widely used method for contrast enhancement (Gonzalez, R.C., Woods, R.E. (2002)). Basically HE computes linear cumulative histogram of the original image and dispenses intensity values over its dynamic intensity range. HE based techniques have been used in medical image processing, satellite image processing etc. The method of Traditional Histogram Equalization (Gonzalez, R.C., Woods, R.E. (2002)) is described below:

The method of Traditional Histogram Equalization (Chen, S. D., Ramli, A. R. (2004)) has the following steps:

If the original image 978-1-5225-5020-4.ch004.m01 has total W number of pixels within the dynamic range 978-1-5225-5020-4.ch004.m02, where L is the number of discrete gray levels, then the probability density function 978-1-5225-5020-4.ch004.m03 of intensity level 978-1-5225-5020-4.ch004.m04 of the image is given by:978-1-5225-5020-4.ch004.m05 for978-1-5225-5020-4.ch004.m06(1) Where, 978-1-5225-5020-4.ch004.m07 is the total number of pixels with intensity level 978-1-5225-5020-4.ch004.m08. The plot of 978-1-5225-5020-4.ch004.m09 vs. 978-1-5225-5020-4.ch004.m10 is the histogram of image 978-1-5225-5020-4.ch004.m11 the cumulative density function is defined as (Chen, S. D., Ramli, A. R. (2004)):

978-1-5225-5020-4.ch004.m12
(2)

Traditional HE maps the corresponding image into the total dynamic range 978-1-5225-5020-4.ch004.m13 with the help of the 978-1-5225-5020-4.ch004.m14 as given below:

978-1-5225-5020-4.ch004.m15
(3)

Complete Chapter List

Search this Book:
Reset