Metaheuristic Algorithms based Multi-objective Optimization for Image Segmentation

In this paper a multi-threshold image segmentation procedure based on nature-inspired multi-objective optimization is proposed. The Particle Swarm Optimization, Black Hole and GravitationalSearch algorithms were adapted for multi-objective optimization. The Root Mean Square Error and the number of segmented regions were used as optimization criteria. The three procedures were applied for human silhouettes detection in video sequences and the obtained results are compared. Concerning the algorithms performances, the experiments revealed that the results of multi-objective Black Hole algorithm based segmentation are better than those of the other two algorithms, at least for the test images used in this experiment.

[1]  Adam P. Piotrowski,et al.  How novel is the "novel" black hole optimization approach? , 2014, Inf. Sci..

[2]  Abdolreza Hatamlou,et al.  Black hole: A new heuristic optimization approach for data clustering , 2013, Inf. Sci..

[3]  Barbara Caputo,et al.  Recognizing human actions: a local SVM approach , 2004, Proceedings of the 17th International Conference on Pattern Recognition, 2004. ICPR 2004..

[4]  Silviu-Ioan Bejinariu,et al.  Automatic multi-threshold image segmentation using metaheuristic algorithms , 2015, 2015 International Symposium on Signals, Circuits and Systems (ISSCS).

[5]  Silviu-Ioan Bejinariu,et al.  Nature-inspired algorithms based multispectral image fusion , 2016, 2016 International Conference and Exposition on Electrical and Power Engineering (EPE).

[6]  Yongbo Li,et al.  AMOBH: Adaptive Multiobjective Black Hole Algorithm , 2017, Comput. Intell. Neurosci..

[7]  James Kennedy,et al.  Particle swarm optimization , 2002, Proceedings of ICNN'95 - International Conference on Neural Networks.

[8]  Hossein Nezamabadi-pour,et al.  GSA: A Gravitational Search Algorithm , 2009, Inf. Sci..

[9]  Prasanta K. Panigrahi,et al.  Multilevel thresholding for image segmentation through a fast statistical recursive algorithm , 2006, Pattern Recognit. Lett..

[10]  Nizamettin Aydin,et al.  Binary black hole algorithm for feature selection and classification on biological data , 2017, Appl. Soft Comput..

[11]  Xiaodong Li,et al.  DMMOGSA: Diversity-enhanced and memory-based multi-objective gravitational search algorithm , 2016, Inf. Sci..

[12]  R. Luca,et al.  An ontology of human walk for autonomous systems , 2014, 2014 18th International Conference on System Theory, Control and Computing (ICSTCC).

[13]  Ramona Luca.  Statistical analysis of some parameters describing human locomotion , 2015, 2015 E-Health and Bioengineering Conference (EHB).

[14]  Wenguo Li,et al.  Multi-threshold color image segmentation based on region growing , 2010, 2010 IEEE International Conference on Intelligent Computing and Intelligent Systems.

[15]  B. Walczak,et al.  Particle swarm optimization (PSO). A tutorial , 2015 .

[16]  Erik Valdemar Cuevas Jiménez,et al.  A multi-threshold segmentation approach based on Artificial Bee Colony optimization , 2012, Applied Intelligence.

[17]  M. E. H. Pedersen.  Good Parameters for Particle Swarm Optimization , 2010 .

[18]  Max Mignotte,et al.  A multi-objective approach based on TOPSIS to solve the image segmentation combination problem , 2016, 2016 23rd International Conference on Pattern Recognition (ICPR).

[19]  Allan D. Jepson,et al.  Benchmarking Image Segmentation Algorithms , 2009, International Journal of Computer Vision.

[20]  Silviu-Ioan Bejinariu,et al.  Classification Method for Human Locomotion , 2018, 2018 10th International Conference on Electronics, Computers and Artificial Intelligence (ECAI).