In context of the rapid development of remote sensing data acquisition and storage methods recently, big data should be processed and analyzed. Conventional image classification methods may not suit in analyzing such big data and thus there is an increasing demand for automated data analysis methods. Application of artificial neural networks is one of the solutions to solve the above issue. This paper proposes a detailed algorithm for training single-layer neural networks and its application for remote sensing image classification. Experimental results and accuracy assessment showed that this method was quite feasible and simple to semi-automate the classification of objects represented on remote sensing images, contributing in establishing and updating national geospatial data.
Full text article
. Л.Н. Чабан (2014). Автоматизированная обработка аэрокосмической информации для картографирования геопространственных данных. Московский государственный университет геодезии и картографии, Москва.