Automatic Clustering of Hyperspectral Images Using Qutrit Exponential Decomposition Particle Swarm Optimization

Hyperspectral images contain one-dimensional spectral data and two-dimensional spatial data. A novel automatic clustering algorithm using qutrit exponential decomposition based particle swarm optimization is proposed for automatically determining the number of optimized cluster segments in hyper-spectral images. The proposed algorithm is compared with the classical particle swarm optimization and exponential decay particle swarm optimization algorithms for establishing its efficiency. The quality of segmentation is determined using F score.