Learning to Combine Spectral Indices with Genetic Programming

Author(s)

Hernandez Albarracin, Juan Felipe ; dos Santos, Jefersson Alex ; Torres, Ricardo da S. ; IEEE

• Document type: Journal article
• Source: 2016 29TH SIBGRAPI CONFERENCE ON GRAPHICS, PATTERNS AND IMAGES (SIBGRAPI); v. N/A, p. 8-pg., 2016-01-01.

Abstract

This paper introduces a Genetic Programmingbased method for band selection and combination, aiming to support remote sensing image classification tasks. Relying on ground-truth data, our method selects spectral bands and finds the arithmetic combination of those bands (i.e., spectral index) that best separates examples of different classes. Experimental results demonstrate that the proposed method is very effective in pixel-wise binary classification problems. (AU)

• FAPESP's process: 13/50169-1 - Towards an understanding of tipping points within tropical South American biomes
• Grantee: Ricardo da Silva Torres
• Support Opportunities: Research Grants - Research Partnership for Technological Innovation - PITE
• FAPESP's process: 13/50155-0 - Combining new technologies to monitor phenology from leaves to ecosystems
• Grantee: Leonor Patricia Cerdeira Morellato
• Support Opportunities: Research Program on Global Climate Change - University-Industry Cooperative Research (PITE)