A sonar image segmentation algorithm based on quantum-inspired particle swarm optimization and fuzzy clustering
- 68 Downloads
The underwater sonar image has the characteristics of complex background and heavy noise pollution. By using multi-bit quantum system to encode particles, a new sonar image segmentation algorithm based on quantum-inspired particle swarm and fuzzy clustering is proposed. Based on its own optimal particles and global optimal particles, the rotation angle is determined. By calculating the variance of the particle group fitness, the multi-bit quantum revolving gate is used to update the particle position in real time. The output of improved particle swarm optimization is used to initialize the K-mean clustering center to converge to the global optimal solution. Based on the idea of fuzzy membership matrix in FCM, combined with the isolated spatial information characteristics of the noise, the sonar image segmentation and the denoising are carried out. The experimental results show that the proposed algorithm can improve the global search ability of particle swarm optimization effectively. It is better than the quantum fuzzy clustering and quantum genetic algorithm in image segmentation. The analysis results of multiple real underwater sonar images show that the new optimization algorithm has faster convergence speed, better optimizing capacity and better segmentation results for sonar images.
KeywordsSonar image segmentation Particle swarm optimization Quantum-inspired Fuzzy clustering
The paper is supported by National Science foundation of China (61571150, 61872204), Chinese Education Department overseas returnees’ funds, Heilongjiang Provincial Natural Science Fund F2017029 and Heilongjiang Provincial Education Department Surface Scientific Research Project 12521600 and 135109237.
- 6.Ding WP, Wang JD, Guan ZJ (2011) Efficient rough attribute reduction based on quantum frog-leaping co-evolution. Acta Electron Sin 39(11):2597–2603Google Scholar
- 8.Ting LI, Zhang JS, Wang SC et al (2015) Geomagnetic navigation path planning based on quantum particle swarm optimization algorithm. Electron Opt Control 22(7):43–47Google Scholar
- 10.Han B, Zhang PH, Hui XU et al (2013) Region-based image fusion algorithm using bidimensional empirical mode decomposition. Infrared Technol 9:546–550Google Scholar
- 11.Liu Y (2018) A fuzzy classification method for distinguishing haze from cloud based on FCM and MFR. Territ Nat Resour StudyGoogle Scholar
- 12.Zhao XM, Yu LI, Zhao QH (2016) Hidden Markov gaussian random field based fuzzy clustering algorithm for high-resolution remote sensing image segmentation. Acta Electron Sin 44(3):679–686Google Scholar