Record number :
2355546
Title of article :
Modified artificial bee colony based computationally efficient multilevel thresholding for satellite image segmentation using Kapur’s, Otsu and Tsallis functions
Author/Authors :
Bhandari، نويسنده , , A.K. Nanda Kumar، نويسنده , , A. and Singh، نويسنده , , G.K.، نويسنده ,
Issue Information :
روزنامه با شماره پیاپی سال 2015
Pages :
29
From page :
1573
To page :
1601
Abstract :
In this paper, a modified artificial bee colony (MABC) algorithm based satellite image segmentation using different objective function has been presented to find the optimal multilevel thresholds. Three different methods are compared with this proposed method such as ABC, particle swarm optimization (PSO) and genetic algorithm (GA) using Kapur’s, Otsu and Tsallis objective function for optimal multilevel thresholding. The experimental results demonstrate that the proposed MABC algorithm based segmentation can efficiently and accurately search multilevel thresholds, which are very close to optimal ones examined by the exhaustive search method. In MABC algorithm, an improved solution search equation is used which is based on the bee’s search only around the best solution of previous iteration to improve exploitation. In addition, to improve global convergence when generating initial population, both chaotic system and opposition-based learning method are employed. Compared to other thresholding methods, segmentation results of the proposed MABC algorithm is most promising, and the computational time is also minimized.
Keywords :
image segmentation , Multilevel thresholding , Between-class variance , Kapur’s entropy , MABC , ABC , PSO and GA algorithm , Tsallis entropy
Journal title :
Expert Systems with Applications
Journal title :
Expert Systems with Applications
Serial Year :
2015
Link To Document :
بازگشت