Journal of Spectral Imaging,   Volume 9   Article ID a6   (2020)

Peer reviewed Paper

A modified Cuckoo Search algorithm based optimal band subset selection approach for hyperspectral image classification

  • Shrutika S. Sawant
  • Manoharan Prabukumar
  • Sathishkumar Samiappan
School of Information Technology and Engineering, Vellore Institute of Technology, Vellore, India
[email protected]
 https://orcid.org/0000-0002-1532-947X
 Search for papers by this author
Geosystems Research Institute, Mississippi State University, Starkville, Mississippi State, USA
[email protected]
 https://orcid.org/0000-0002-8443-883X
 Search for papers by this author
 Corresponding Author
School of Information Technology and Engineering, Vellore Institute of Technology, Vellore, India
[email protected]
 https://orcid.org/0000-0002-8127-653X
 Search for papers by this author

Band selection is an effective way to reduce the size of hyperspectral data and to overcome the “curse of dimensionality” in ground object classification. This paper presents a band selection approach based on modified Cuckoo Search (CS) optimisation with correlation-based initialisation. CS is a popular metaheuristic algorithm with efficient optimisation capabilities for band selection. However, it can easily fall into local optimum solutions. To avoid falling into a local optimum, an initialisation strategy based on correlation is adopted instead of random initialisation to initiate the location of nests. Experimental results with Indian Pines, Salinas and Pavia University datasets show that the proposed approach obtains overall accuracy of 82.83 %, 94.83 % and 91.79 %, respectively, which is higher than the original CS algorithm, Genetic Algorithm (GA), Particle Swarm Optimisation (PSO) and Gray Wolf Optimisation (GWO).

Keywords: hyperspectral image, band selection, Cuckoo Search optimisation, correlation-based initialisation

Metrics

Downloads:

162

Full-Text HTML Views:

18

Abstract Views:

746