期刊名称：ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
印刷版ISSN：2194-9042
电子版ISSN：2194-9050
出版年度：1992
卷号：XXIX Part B3
页码：529-533
出版社：Copernicus Publications
摘要：Classification of remotely sensed data with artificial neural networks is called neuro-classification. Artificial neural networkshave shown great potential in classification of remotely sensed data. The amount of data used for training a neural networkaffects accuracy and efficiency of the neural network classifier. A neural network was trained separately with 5%, 10%, 15%,and 20% image data from a LANDSAT Thematic Mapper scene, which was acquired 29 July 1987. At a risk level of 5%, theresults showed that (a) classifiers NN-5% (neuro-classification with 5% of the image data used for training), NN-lO%, andNN-15% did not differ from one another, (b) classifiers NN-15% and NN-20% did not differ from each other, but (c)classifiers NN-5% and NN-lO% differed from classifier NN-20%. The training rates were reduced by more than 10seconds/cycle as we increased the percentage of the image data for training a neural network. Ten percent image data areneeded to adequately train a neural network classifier, the classifier provides satisfactory performance.
