Unsupervised Satellite Image Time Series Clustering Using Object-Based Approaches and 3D Convolutional Autoencoder

Autor: Maria Trocan, Ekaterina Kalinicheva, Jérémie Sublime
Přispěvatelé: Institut Supérieur d'Electronique de Paris (ISEP), Laboratoire d'Informatique de Paris-Nord (LIPN), Institut Galilée-Université Sorbonne Paris Cité (USPC)-Centre National de la Recherche Scientifique (CNRS)-Université Sorbonne Paris Nord
Jazyk: angličtina
Rok vydání: 2020
Předmět:
010504 meteorology & atmospheric sciences
Computer science
satellite image time series
0211 other engineering and technologies
ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION
02 engineering and technology
unsupervised learning
01 natural sciences
[INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI]
3D convolutional network
Segmentation
lcsh:Science
Cluster analysis
Representation (mathematics)
Image resolution
021101 geological & geomatics engineering
0105 earth and related environmental sciences
autoencoder
clustering
segmentation
business.industry
[INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV]
Pattern recognition
Object (computer science)
Autoencoder
General Earth and Planetary Sciences
Unsupervised learning
lcsh:Q
Satellite Image Time Series
Artificial intelligence
business
Zdroj: Remote Sensing; Volume 12; Issue 11; Pages: 1816
Remote Sensing
Remote Sensing, MDPI, 2020, Advanced Machine Learning for Time Series Remote Sensing Data Analysis), 12 (11), ⟨10.3390/rs12111816⟩
Remote Sensing, MDPI, In press, ⟨10.3390/rs12111816⟩
Remote Sensing, Vol 12, Iss 1816, p 1816 (2020)
ISSN: 2072-4292
DOI: 10.3390/rs12111816
Popis: International audience; Nowadays, satellite image time series (SITS) analysis has become an indispensable part of many research projects as the quantity of freely available remote sensed data increases every day. However, with the growing image resolution, pixel-level SITS analysis approaches have been replaced by more efficient ones leveraging object-based data representations. Unfortunately, the segmentation of a full time series may be a complicated task as some objects undergo important variations from one image to another and can also appear and disappear. In this paper, we propose an algorithm that performs both segmentation and clustering of SITS. It is achieved by using a compressed SITS representation obtained with a multi-view 3D convolutional autoencoder. First, a unique segmentation map is computed for the whole SITS. Then, the extracted spatio-temporal objects are clustered using their encoded descriptors. The proposed approach was evaluated on two real-life datasets and outperformed the state-of-the-art methods.
Databáze: OpenAIRE
Externí odkaz:
Nepřihlášeným uživatelům se plný text nezobrazuje