The particle size of lake sediments integrates important environmental information, and the detection of changes in this variable over time provides important information for understanding ecosystem and sedimentary processes. This code develops and applies a new methodology based on a 1D convolutional autoencoder as the feature extractor and a 1D CNN architecture for regression. The proposed architecture was applied to hyperspectral images of nine lake sediment cores across Canada.
Highlights
-
Application of remote sensing techniques in inland water environmental studies.
-
Laboratory-based hyperspectral imaging of nine lake sediment cores across Canada.
-
Using of convolutional autoencoders feature extraction method.
-
Combination of HSIs and deep learning methods to replace the measurements from laser granulometry.
-
Comparison of CNN regression model with RF regression model.
Paper : Convolutional neural networks for mapping of lake sediment core particle size using hyperspectral imaging
authors: H. Ghanbari, D. Antoniades