SciELO - Scientific Electronic Library Online

 
vol.105 issue9-10 author indexsubject indexarticles search
Home Pagealphabetic serial listing  

South African Journal of Science

Print version ISSN 0038-2353

Abstract

PEREA, A.J.; MERONO, J.E.  and  AGUILERA, M.J.. Application of Numenta® Hierarchical Temporal Memory for land-use classification. S. Afr. j. sci. [online]. 2009, vol.105, n.9-10, pp. 370-375. ISSN 0038-2353.

The aim of this paper is to present the application of memory-prediction theory, implemented in the form of a Hierarchical Temporal Memory (HTM), for land-use classification. Numenta®HTM is a new computing technology that replicates the structure and function of the human neocortex. In this study, a photogram, received by a photogrammetric UltraCamD® sensor of Vexcel, and data on 1 513 plots in Manzanilla (Huelva, Spain) were used to validate the classification, achieving an overall classification accuracy of 90.4%. The HTM approach appears to hold promise for land-use classification.

Keywords : memory-prediction theory; NuPIC®; UltraCamD® sensor; Hierarchical Temporal Memory.

        · text in English     · pdf in English