Self-Attention for Raw Optical Satellite Time Series Classification (2019-10-23T00:00:00.000000Z)

TL;DR

This work compares recent deep learning models on crop type classification on raw and preprocessed Sentinel 2 data and qualitatively shows how self-attention scores focus selectively on few classification-relevant observations.

Authors

Marc Rußwurm

5 papers

Marco Körner

4 papers

TL;DR

Authors

References95 items

Satellite Image Time Series Classification With Pixel-Set Encoders and Temporal Self-Attention

A new fully convolutional neural network for semantic segmentation of polarimetric SAR imagery in complex land cover ecosystem

Current status of Landsat program, science, and applications

The Role and Need for Space-Based Forest Biomass-Related Measurements in Environmental Management and Policy

Deep learning based multi-temporal crop classification

Time-Space Tradeoff in Deep Learning Models for Crop Classification on Satellite Multi-Spectral Image Time Series

Kongressvorschau 2019

Temporal Convolutional Neural Network for the Classification of Satellite Image Time Series

$M^3\text{Fusion}$: A Deep Learning Architecture for Multiscale Multimodal Multitemporal Satellite Data Fusion

The UEA multivariate time series classification archive, 2018

The UCR time series archive

DuPLO: A DUal view Point deep Learning architecture for time series classificatiOn

Deep learning for time series classification: a review

UMAP: Uniform Manifold Approximation and Projection

Tune: A Research Platform for Distributed Model Selection and Training

Performance evaluation and hyperparameter tuning of statistical and machine-learning models using spatial data

Learning Spectral-Spatial-Temporal Features via a Recurrent Convolutional Neural Network for Change Detection in Multispectral Imagery

Massively Parallel Hyperparameter Tuning

UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction

Multi-Temporal Land Cover Classification with Sequential Recurrent Encoders

When Deep Learning Meets Metric Learning: Remote Sensing Image Scene Classification via Learning Discriminative CNNs

Google Earth Engine: Planetary-scale geospatial analysis for everyone

Classification of time-series images using deep convolutional neural networks

Incremental Dual-memory LSTM in Land Cover Prediction

Land Cover Classification from Multi-temporal, Multi-spectral Remotely Sensed Imagery using Patch-Based Recurrent Neural Networks

Temporal Vegetation Modelling Using Long Short-Term Memory Networks for Crop Identification from Medium-Resolution Multi-spectral Satellite Images

Efficient Attention using a Fixed-Size Memory Representation

Attention is All you Need

Remote Sensing Image Scene Classification: Benchmark and State of the Art

Time series classification from scratch with deep neural networks: A strong baseline

Semantic Segmentation of Earth Observation Data Using Multimodal and Multi-scale Deep Networks

Learning Rotation-Invariant Convolutional Neural Networks for Object Detection in VHR Optical Remote Sensing Images

Dense Semantic Labeling of Subdecimeter Resolution Images With Convolutional Neural Networks

Learning a Transferable Change Rule from a Recurrent Neural Network for Land Cover Change Detection

Fully Convolutional Networks for Dense Semantic Labelling of High-Resolution Aerial Imagery

An Object-Based Paddy Rice Classification Using Multi-Spectral Data and Crop Phenology in Assam, Northeast India

Multi-Scale Convolutional Neural Networks for Time Series Classification

Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning

Combining satellite data for better tropical forest monitoring

Production of a Dynamic Cropland Mask by Processing Remote Sensing Image Series at High Temporal and Spatial Resolutions

Deep Residual Learning for Image Recognition

End-to-end attention-based large vocabulary speech recognition

An Empirical Exploration of Recurrent Network Architectures

Feature Selection of Time Series MODIS Data for Early Crop Classification Using Random Forest: A Case Study in Kansas, USA

Improvement and expansion of the Fmask algorithm: cloud, cloud shadow, and snow detection for Landsats 4–7, 8, and Sentinel 2 images

Comparison of support vector machine, artificial neural network, and spectral angle mapper algorithms for crop classification using LISS IV data

Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift

A support vector machine to identify irrigated crop types using time-series Landsat NDVI data

Adam: A Method for Stochastic Optimization

Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling

Land Cover Classification of Landsat Data with Phenological Features Extracted from Time Series MODIS NDVI Data

Fully convolutional networks for semantic segmentation

Sequence to Sequence Learning with Neural Networks

Very Deep Convolutional Networks for Large-Scale Image Recognition

ImageNet Large Scale Visual Recognition Challenge

Derivation of temporal windows for accurate crop discrimination in heterogeneous croplands of Uzbekistan using multitemporal RapidEye images

Continuous change detection and classification of land cover using all available Landsat data

ImageNet classification with deep convolutional neural networks

Crop type mapping using spectral-temporal profiles and phenological information

SUPPORT VECTOR MACHINE CLASSIFICATION OF OBJECT-BASED DATA FOR CROP MAPPING, USING MULTI-TEMPORAL LANDSAT IMAGERY

Spatial cross-validation and bootstrap for the assessment of prediction rules in remote sensing: The R package sperrorest

Comparison of support vector machine, neural network, and CART algorithms for the land-cover classification using limited training data points

Object-based cloud and cloud shadow detection in Landsat imagery

Object-based crop identification using multiple vegetation indices, textural features and crop phenology

Scikit-learn: Machine Learning in Python

Detecting trends in forest disturbance and recovery using yearly Landsat time series: 1. LandTrendr — Temporal segmentation algorithms

Per-Field Irrigated Crop Classification in Arid Central Asia Using SPOT and ASTER Data

Intercomparison, interpretation, and assessment of spring phenology in North America estimated from remote sensing for 1982–2006

ImageNet: A large-scale hierarchical image database

Free Access to Landsat Imagery

A recent greening of the Sahel—trends, patterns and potential causes

TIMESAT - a program for analyzing time-series of satellite sensor data

National level forest monitoring and modeling in Canada

Long Short-Term Memory

Measuring phenological variability from satellite imagery

Learning long-term dependencies with gradient descent is difficult

Integrating a robust option into a multiple regression computing environment