Semi-automated training approaches for spectral class definition

Autor:	Paul V. Bolstad, T. M. Lillesand
Rok vydání:	1992
Předmět:	Training set Computer science business.industry Training (meteorology) Sampling (statistics) Pattern recognition Stellar classification Maximum likelihood classification Region growing General Earth and Planetary Sciences Artificial intelligence Level ii business Cartography
Zdroj:	International Journal of Remote Sensing. 13:3157-3166
ISSN:	1366-5901 0143-1161
Popis:	A semi-automated approach to spectral training for a maximum likelihood classification is shown to maintain or improve classification accuracy while reducing analyst input five-fold. The semi-automated approach is based on spectral sampling via region growing and training set refinement via transformed-divergence based mergers and deletions. Classification accuracies at the Anderson level II/III in northern Wisconsin were higher for the semi-automated approach in five of six combinations of imagery, analyst, and study area, at significantly reduced training expense.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::fe24edd53dff90d191158ff479697463 https://doi.org/10.1080/01431169208904108 Zobrazit plný text záznamu