1,721,423 research outputs found
Deep Learning‐based Semantic Segmentation in Remote Sensing
No full textSemantic segmentation consists of the generation of a categorical map, given an image in which each pixel of the image is automatically assigned a class. Deep learning allows the influence of the pixel's context to be learned by capturing the non-linear relationships between surrounding image features at multiple scales, leading to large improvements in performance and opening up the door to new applications. This chapter explores the use of deep learning-based semantic segmentation in Earth observation imagery and presents in detail three approaches specifically aimed at Earth observation applications.ECE
Deep Learning for the Earth Sciences: A Comprehensive Approach to Remote Sensing, Climate Science, and Geosciences
No full textDeep learning is a fundamental technique in modern Artificial Intelligence and is being applied to disciplines across the scientific spectrum; earth science is no exception. Yet, the link between deep learning and Earth sciences has only recently entered academic curricula and thus has not yet proliferated. Deep Learning for the Earth Sciences delivers a unique perspective and treatment of the concepts, skills, and practices necessary to quickly become familiar with the application of deep learning techniques to the Earth sciences. The book prepares readers to be ready to use the technologies and principles described in their own research.ECE
Deep Domain Adaptation in Earth Observation
No full textWhen applied to new datasets, acquired at different time moments, with different sensors or under different acquisition conditions, deep learning models might fail spectacularly. This is because they have learned from the data distribution observed during training and, as such, do not generalize out of that domain naturally. This chapter introduces methodologies designed to tackle this problem and provide deep learning models able to adapt to new data distributions, i.e. domain adaptation. Domain adaptation works by either adapting the representation to the new data distribution, modifying the inputs or performing smart sampling. But independently of the strategy, they lead to updated models, able to process effectively the new data without needing observation from it (or a very limited amount).ECE
Self-supervised Pre-training Enhances Change Detection in Sentinel-2 Imagery
No full textWhile annotated images for change detection using satellite imagery are scarce and costly to obtain, there is a wealth of unlabeled images being generated every day. In order to leverage these data to learn an image representation more adequate for change detection, we explore methods that exploit the temporal consistency of Sentinel-2 times series to obtain a usable self-supervised learning signal. For this, we build and make publicly available (https://zenodo.org/record/4280482 ) the Sentinel-2 Multitemporal Cities Pairs (S2MTCP) dataset, containing multitemporal image pairs from 1520 urban areas worldwide. We test the results of multiple self-supervised learning methods for pre-training models for change detection and apply it on a public change detection dataset made of Sentinel-2 image pairs (OSCD)
Unsupervised change detection in very high-resolution images: an example along Swiss Railways tracks
No full textChange detection (CD) of Land Use and Land Cover (LULC) is a valuable field of study for resource
monitoring and land planning. The progress of remote sensing and the availability of satellite data
allowed for the development of LULC monitoring through bi-temporal orthoimages processing.
However, the annotation of changes based on geospatial data is a time-consuming and laborious
task. Thus, the automation of this process is in current research and development through the
implementation of CD models.
Change Vector Analysis (CVA) emerged as the first principal model for CD, consisting of pixel-based
image differentiation. The main drawback of pixel-based techniques like CVA is their sensitivity
to radiometric variations and misregistration. Recently, supervised deep-learning-based methods
achieved great success in the CD task due to their ability to extract deep features in a pair of images
and their robustness to inner-class variance. Unsupervised deep-learning-based models still are
significantly less performant than supervised ones but need to be developed in order to avoid being
dependent on the laborious production of labeled datasets for training. In addition, the lack of
labeled CD datasets was pointed out in previous papers. Several binary change datasets filled this
gap, but the lack of semantic datasets, containing information about the nature of changes, still
persists. These datasets are crucial for the training and evaluation of CD models.
We tackle the enunciated issues in two parts.
We propose a very high resolution (VHR) 0.25 m/px semantic CD dataset comprised of 127 paired
tiles of 4000×4000 pixels covering 1 km2. All the geospatial data utilized for the dataset is provided
by Swisstopo and the changes are reported along Swiss railway tracks. In addition, the labeling
is partly automated by a script processing vectorized data analysis on vectorized geospatial data.
It has been assessed that the automated processing was useful and efficient for building change
detection. However, manual labeling is still necessary for the other classes of changes and the
semantic classification of the changes.
The dataset is proposed as a benchmark for two unsupervised models: a CVA acting as a baseline for
the evaluation of a promising novel unsupervised CD deep learning model based on reconstruction
loss (CDRL). CDRL outperformed CVA when setting appropriate parameters, with an F1-score of
0.24 versus 0.19 respectively. However, we struggled to get the same accuracy as the original authors.ECE
Classification of urban structural types with multisource data and structured models
No full textIn this paper, we study the land use distribution of the city of Munich, Germany. We describe the city as a set of Urban Structural Types (UST) related to the type of spatial patterns occurring within regions composed of 200m side cells. To do so, we resort to a set of multimodal descriptors extracted from remote sensing data, a 3D city model and open access vector
information. Based on these descriptors, we train a SVM classifier and apply two structured prediction models to enforce spatial relationships (Markov and Conditional Random fields)
Classification of urban structural types (UST) using multiple data sources and spatial priors
Full text linkRemote sensing and geographic information science offer many possibilities in terms of availability of diverse data. Some products like land cover layers or digital elevation models can be extracted from imagery and enable the realization of 3D city models. Starting from these morphological and geographical sources, an approach is proposed to extract information about urban structure types (UST), i.e. types of urban habitat at the neighborhoodscale. We propose an effective processing chain to describe UST : from the different data sources,
we extract spectral and spatial indices and use them as features in a machine learning process to classify these urban structural types using support vector machine classication (SVM). Moreover, Markov Random Fields (MRF) are used to take into account the spatial distribution of the classe and increase the spatial consistency. This study focuses on the city of Munich and uses as different data
sources the land cover data, the 3D city model, spectral images from LandSat TM 8 and OpenStreetMap (OSM) vector data to characterize UST. The main hypothesis is that we can discriminate among urban structural types by using land cover information, spectral properties and
3D structure: in other words, that an industrial area will not have the same structure nor the same properties as a residential or an agricultural area. The proposed processing chain enables to predict with a precision of 70% the 11 UST. This opens possibilities to describe the urban
footprint of the city, to detect the key areas for urban planification and to better understand the city dynamics
- …
