Temporal Convolutional Neural Network for the Classification of Satellite Image Time Series
Authors:
- Charlotte PelletierCPCharlotte Pelletier
- Geoffrey I. WebbGWGeoffrey I. Webb
- Francois PetitjeanFPFrancois Petitjean
Preprint
Abstract
New remote sensing sensors acquire now high spatial and spectral Satellite Image Time Series (SITS) of the world. These series of images are a key component of any classification framework to obtain up-to-date and accurate land cover maps of the Earth's soils. More specifically, the combination of the temporal, spectral and spatial resolutions of new SITS enables the monitoring of vegetation dynamics. Although some traditional classification algorithms, such as Random Forest (RF), have been successfully applied for SITS classification, these algorithms do not fully take advantage of the temporal domain. Conversely, deep-learning based methods have been successfully used to make the most of sequential data such as text and audio data. For the first time, this paper explores the use of Convolutional Neural Networks (CNNs) with convolutions applied in the temporal dimension for SITS classification. The goal is to quantitatively and qualitatively evaluate the contribution of temporal CNNs for SITS classification. More precisely, this paper proposes a set of experiments performed on a million Formosat-2 time series. The experimental results show that temporal CNNs are 2 to 3 % more accurate than RF. The experiments also highlight some counter-intuitive results on pooling layers: contrary to image classification, their use decreases accuracy. Moreover, we provide some general guidelines on the network architecture, common regularization mechanisms, and hyper-parameter values such as the batch size. Finally, the visual quality of the land cover maps produced by the temporal CNN is assessed.Keywords
All Related Versions
This publication has 36 references indexed in Scilit:
- The UCR time series archiveIEEE/CAA Journal of Automatica Sinica, 2019
- Deep learning for time series classification: a reviewData Mining and Knowledge Discovery, 2019
- How much does multi-temporal Sentinel-2 data improve crop type classification?International Journal of Applied Earth Observation and Geoinformation, 2018
- Sentinel-2 cropland mapping using pixel-based and object-based time-weighted dynamic time warping analysisRemote Sensing of Environment, 2018
- Assessing the robustness of Random Forests to map land cover with high resolution satellite image time series over large areasRemote Sensing of Environment, 2016
- The great time series classification bake off: a review and experimental evaluation of recent algorithmic advancesData Mining and Knowledge Discovery, 2016
- Optical remotely sensed time series data for land cover classification: A reviewISPRS Journal of Photogrammetry and Remote Sensing, 2016
- A meta-analysis of remote sensing research on supervised pixel-based land-cover image classification processes: General guidelines for practitioners and future researchRemote Sensing of Environment, 2016
- Time series classification with ensembles of elastic distance measuresData Mining and Knowledge Discovery, 2014
- The use of the Normalized Difference Water Index (NDWI) in the delineation of open water featuresInternational Journal of Remote Sensing, 1996