<abstract xmlns="http://www.w3.org/1999/xhtml"><p>The paper describes the process of training a convolutional neural network to segment land into its labelled land cover types such as grass, water, forest and buildings. This segmentation can promote automated updating of topographical maps since doing this manually is a time-consuming process, which is prone to human error. The aim of the study is to evaluate the application of U-net convolutional neural network for land cover classification using countrywide aerial data. U-net neural network architecture has initially been developed for use in biomedical image segmentation and it is one of the most widely used CNN architectures for image segmentation. Training data have been prepared using colour infrared images of Ventspils town and its digital surface model (DSM). Forest, buildings, water, roads and other land plots have been selected as classes, into which the image has been segmented. As a result, images have been segmented with an overall accuracy of 82.9 % with especially high average accuracy for the forest and water classes.</p></abstract>

The paper describes the process of training a convolutional neural network to segment land into its labelled land cover types such as grass, water, forest and buildings. This segmentation can promote automated updating of topographical maps since doing this manually is a time-consuming process, which is prone to human error. The aim of the study is to evaluate the application of U-net convolutional neural network for land cover classification using countrywide aerial data. U-net neural network architecture has initially been developed for use in biomedical image segmentation and it is one of the most widely used CNN architectures for image segmentation. Training data have been prepared using colour infrared images of Ventspils town and its digital surface model (DSM). Forest, buildings, water, roads and other land plots have been selected as classes, into which the image has been segmented. As a result, images have been segmented with an overall accuracy of 82.9 % with especially high average accuracy for the forest and water classes.

Land Cover Classification using Very High Spatial Resolution Remote Sensing Data and Deep Learning

Engineering Research Institute “Ventspils International Radio Astronomy Centre”

Latvian Journal of Physics and Technical Sciences

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.

The paper describes the process of training a convolutional neural network to segment land into its labelled land cover types such as grass, water, forest and...

Land Cover Classification using Very High Spatial Resolution Remote Sensing Data and Deep Learning

Data publikacji: 11 maj 2020

Zakres stron: 71 - 77

DOI: https://doi.org/10.2478/lpts-2020-0009

Słowa kluczowe
land cover classification, neural networks, remote sensing, topographic map

© 2020 R. Ķēniņš, published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.

Land Cover Classification using Very High Spatial Resolution Remote Sensing Data and Deep Learning

Data publikacji: 11 maj 2020

Zakres stron: 71 - 77

DOI: https://doi.org/10.2478/lpts-2020-0009

Słowa kluczoweland cover classification, neural networks, remote sensing, topographic map

© 2020 R. Ķēniņš, published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.

Słowa kluczowe
land cover classification, neural networks, remote sensing, topographic map