J. ICT Res. Appl., Vol. 14, No. 1, 2020, 95-114 95
Received December 12
th
, 2019, 1
st
Revision May 28
th
, 2020, 2
nd
Revision July 23
rd
, 2020, Accepted for
publication August 13
th
, 2020.
Copyright © 2020 Published by IRCS-ITB, ISSN: 2337-5787, DOI: 10.5614/itbj.ict.res.appl.2020.14.2.1
A Global Two-Stage Histogram Equalization Method for
Gray-Level Images
Khaled H Almotairi
Computer Engineering Department, Umm Al-Qura University, 715, Alabidiah, Taif
Street, Makkah, Saudi Arabia
E-mail: khmotairi@uqu.edu.sa
Abstract. Digital image histogram equalization is an important technique in image
processing to improve the quality of the visual appearance of images. However, the
available methods suffer from several problems such as side effects and noise,
brightness and contrast problems, loss of information and details, and failure in
enhancement and in achieving the desired results. Therefore, the Adaptive Global
Two-Stage Histogram Equalization (GTSHE) method for visual property
enhancement of gray-level images is proposed. The first stage aims to clip the
histogram and equalize the clipped histogram based on the number of occurrences
of gray-level values. The second stage adaptively adjusts the space between
occurrences by using a probability density function and different cumulative
distribution functions that depend on the available and missing gray-level
occurrences. Experiments were conducted using a number of benchmark datasets of
images such as the Galaxies, Biomedical, Miscellaneous, Aerials, and Texture
datasets. The results of the experiments were compared with a number of well-
known methods, i.e. HE, AHEA, ESIHE, and MVSIHE, to evaluate the performance
of the proposed method. The evaluation analysis showed that the proposed GTSHE
method achieved a higher accuracy rate compared to the other methods.
Keywords: gray-scale images; histogram equalization; image enhancement; image
processing; images quality; visual appearance.
1 Introduction
The current explosion of the creation and usage of digital images in our lives leads
to the need for more digital image processing techniques to manipulate digital
images using computers [1-3]. Digital images are created using several different
methods ranging from mobile phone cameras for daily usage to satellite cameras
for aerial images and from simple scanners for documents to medical scanners for
the human body, and several more [3,4]. These different methods and conditions
can lead to several undesirable effects, such as blurring, and degraded contrast and
brightness as well as an incorrect balance of color levels [3-6]. Therefore, image
enhancement techniques have attracted the attention of many researchers.
Furthermore, image enhancement techniques aim to improve the quality of the
visual appearance of images that suffer from the previously mentioned problems