This repository contains a Jupyter notebook demonstrating time series classification for crop identification using deep learning techniques - MLP, LSTM and 1D-CNN . The dataset used in this demonstration is a subset of the MiniTimeMatch dataset.
- Data exploration
- EDA analysis
- MLP implementation
- LSTM implementation
- 1D CNN implementation
Each observation in the dataset consists of a time series of 62 observations taken across 10 spectral bands of Sentinel-2 for 19 different classes in France. Each observation corresponds to spectral measurements aggregated over a land parcel. The observations are labeled with the crop found in the parcel.
- Nyborg, J., Pelletier, C., Lefèvre, S., & Assent, I. (2022). TimeMatch: Unsupervised cross-region adaptation by temporal shift estimation. ISPRS Journal of Photogrammetry and Remote Sensing, 188, 301-313.
- Painblanc, F., Chapel, L., Courty, N., Friguet, C., Pelletier, C., & Tavenard, R. (2023, September). Match And Deform: Time Series Domain Adaptation through Optimal Transport and Temporal Alignment. In Joint European Conference on Machine Learning and Knowledge Discovery in Databases (pp. 341-356). Cham: Springer Nature Switzerland.
The dataset used in this demonstration can be found at: Dataset Source