Bi-level thresholding using PSO, Artificial Bee Colony and MRLDE embedded with Otsu method
- PDF / 2,106,665 Bytes
- 12 Pages / 595.276 x 790.866 pts Page_size
- 52 Downloads / 163 Views
REGULAR RESEARCH PAPER
Bi-level thresholding using PSO, Artificial Bee Colony and MRLDE embedded with Otsu method Sushil Kumar · Pravesh Kumar · Tarun Kumar Sharma · Millie Pant
Received: 30 June 2012 / Accepted: 24 July 2013 / Published online: 28 August 2013 © Springer-Verlag Berlin Heidelberg 2013
Abstract Image segmentation is required to be studied in detail some particular features (areas of interest) of a digital image. It forms an important and exigent part of image processing and requires an exhaustive and robust search technique for its implementation. In the present work we have studied the working of MRLDE, a newly proposed variant of differential evolution combined with Otsu method, a well known image segmentation method for bi-level thresholding. The proposed variant, termed as Otsu+MRLDE, is tested on a set of 10 images and the results are compared with Otsu method and some other well known metaheuristics. Keywords Otsu method · Thresholding · Image segmentation · MRLDE · PSO · ABC
1 Introduction Image segmentation refers to the partitioning or dividing of a digital image into several smaller parts or segments in order to study a given image in a detailed manner. Its applications vary from satellite imaging (for locating roads, forests etc in a satellite image) to medical imaging (for locating tumours, for analyzing the anatomical structure etc); from machine vision S. Kumar (B) · P. Kumar · T. K. Sharma · M. Pant Department of Applied Science and Engineering, IIT Roorkee, Roorkee, India e-mail: [email protected] P. Kumar e-mail: [email protected] T. K. Sharma e-mail: [email protected] M. Pant e-mail: [email protected]
to fingerprint recognition etc. in fact segmentation can be applied to study in detail any real life problem where digital image can be developed. However, it is considered to be quite a demanding task because of the presence of multiple objects in an image and sometimes due to the intrinsic nature of an image. Segmentation methods available in literature can be broadly classified as: (a) region splitting methods, (b) region growing methods, (c) clustering, and (d) edge and line oriented segmentation methods. Despite the fact that several methods have been suggested in literature for segmentation process, researchers are still trying develop efficient and robust algorithms which can meet effectively the challenges of segmentation. Nature inspired metaheuristics (NIM) because of their exhaustive searching nature have emerged as a popular choice for segmentation of complex images. Some popular NIM used in image segmentation include ant colony optimisation (ACO) [66], PSO [61], bacterial foraging (BF) [53], differential evolution [37], and ABC [64]. Because of the simplicity and stability, thresholding has become one of the most popular image segmentation methods. At present, many thresholding segmentation methods, e.g. Otsu, histogram thresholding, maximal entropic thresholding etc are available in literature. We have considered in this paper bi-level thresholding
Data Loading...
