SatImNet: Structured and Harmonised Training Data for Enhanced Satellite Imagery Classification
This work addresses the need for standardized training data in remote sensing, which is incremental as it organizes existing data rather than introducing new methods.
The authors tackled the problem of heterogeneous and non-interoperable training data for satellite imagery classification by proposing SatImNet, a structured and harmonized collection of open training data, and demonstrated its use with convolutional neural networks for classification and segmentation tasks.
Automatic supervised classification with complex modelling such as deep neural networks requires the availability of representative training data sets. While there exists a plethora of data sets that can be used for this purpose, they are usually very heterogeneous and not interoperable. In this context, the present work has a twofold objective: i) to describe procedures of open-source training data management, integration, and data retrieval, and ii) to demonstrate the practical use of varying source training data for remote sensing image classification. For the former, we propose SatImNet, a collection of open training data, structured and harmonized according to specific rules. For the latter, two modelling approaches based on convolutional neural networks have been designed and configured to deal with satellite image classification and segmentation.