Computers and Electrical Engineering 70 (2018) 931–938
Contents lists available at ScienceDirect
Computers and Electrical Engineering
journal homepage: www.elsevier.com/locate/compeleceng
A multi-level thresholding image segmentation based on an
improved artificial bee colony algorithm
R
Hao Gao
a , b , ∗
, Zheng Fu
a
, Chi-Man Pun
b
, Haidong Hu
c
, Rushi Lan
d , e
a
The Institute of Advanced Technology, Nanjing University of Posts and Telecommunications, Nanjing, China
b
Department of Computer and Information Science, University of Macau, Macau SAR, China
c
Beijing Institute of Control Engineering, Beijing, China
d
School of Computer Science & Engineering, South China University of Technology, China
e
School of Computer Science and Information Security, Guilin University of Electronic Technology, Guilin, China
a r t i c l e i n f o
Article history:
Received 24 September 2017
Revised 19 December 2017
Accepted 20 December 2017
Available online 29 December 2017
Keywords:
Image segmentation
Otsu
Artificial bee colony
Convergence speed
Precise search
a b s t r a c t
As a popular evolutionary algorithm, Artificial Bee Colony (ABC) algorithm has been suc-
cessfully applied into threshold-based image segmentation. Due to its one dimension
search strategy, the convergence speed of ABC is slow and its solution is acceptable but
not precise. For making more fine-tuning search and further enhancing the achievements
on image segmentation, we proposed an Otsu segmentation method based on a new ABC
algorithm. Different from the traditional ABC strategy, our algorithm takes full use of in-
dividuals information which is defined by a focus point and the best point to increase its
accuracy and convergence speed. Furtheremore, we propose an adaptive parameter to ad-
just the search step of individual automatically, which also improves its exploitation ability.
Experimental results on Berkeley segmentation database demonstrate the effectiveness of
our algorithm.
©2017 Elsevier Ltd. All rights reserved.
1.
Introduction
Image segmentation technique is attempts to detect specific parts in an image, which is an important component in im-
age processing, video processing and analysis [1–3] . As an important branch of image segmentation algorithm, thresholding
methods segment a digital image into multiple parts. According to the number of thresholding, they have been divided into
two categories: bi-level and multi-level algorithms. The former means an image should be divided into two subdivisions
which use one grey value to represent its threshold. The multi-level method discriminates several distinct subdivisions from
a digital image with more than one threshold. As a representative threshold-based segmentation method, Otsu [4] has at-
tracted many researchers to do further study. Based on previous findings, the Otsu method can be treated as a maximum
optimization problem. But a traditional exhausted searching method expends too much computational time to endure, es-
pecially on multi-level threshold selection of Otsu [5,6] .
As a population-based algorithm, evolutionary algorithms (EAs), e.g., differential algorithm (DE) and particle swarm op-
timization (PSO), find a potential solution space by employing multiple individuals, which means they could achieve fast
computational ability than the traditional exhausted searching methods [7–12] . As an efficient EA, artificial bee colony (ABC)
R
Reviews processed and recommended for publication to the Editor-in-Chief by Associate Editor Dr. Huimin Lu.
∗
Corresponding author at: The Institute of Advanced Technology, Nanjing University of Posts and Telecommunications, Nanjing, China.
E-mail address: gaohao@njupt.edu.cn (H. Gao).
https://doi.org/10.1016/j.compeleceng.2017.12.037
0045-7906/© 2017 Elsevier Ltd. All rights reserved.