RECORD
Toulouse Hyperspectral Data Set: a benchmark data set to assess semi-supervised spectral representation learning and pixel-wise classification techniques
Computer Vision and Pattern Recognition · Submitted 15 Nov 2023
abstract compiled · 2026-09-23 · no language-model review
The authors' main claim Therefore, we release in this paper the Toulouse Hyperspectral Data Set that stands out from other data sets in the above-mentioned respects in order to meet key issues in spectral representation learning and classification over large-scale hyperspectral images with very few labeled pixels.
The compiled record
The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.
Open the compiled record
The abstract, sentence by sentence
Airborne hyperspectral images can be used to map the land cover in large urban areas, thanks to their very high spatial and spectral resolutions on a wide spectral domain. While the spectral dimension of hyperspectral images is highly informative of the chemical composition of the land surface, the use of state-of-the-art machine learning algorithms to map the land cover has been dramatically limited by the availability of training data. To cope with the scarcity of annotations, semi-supervised and self-supervised techniques have lately raised a lot of interest in the community. Yet, the publicly available hyperspectral data sets commonly used to benchmark machine learning models are not totally suited to evaluate their generalization performances due to one or several of the following properties: a limited geographical coverage (which does not reflect the spectral diversity in metropolitan areas), a small number of land cover classes and a lack of appropriate standard train / test splits for semi-supervised and self-supervised learning. Therefore, we release in this paper the Toulouse Hyperspectral Data Set that stands out from other data sets in the above-mentioned respects in order to meet key issues in spectral representation learning and classification over large-scale hyperspectral images with very few labeled pixels. Besides, we discuss and experiment self-supervised techniques for spectral representation learning, including the Masked Autoencoder, and establish a baseline for pixel-wise classification achieving 85% overall accuracy and 77% F1 score. The Toulouse Hyperspectral Data Set and our code are publicly available at https://www.toulouse-hyperspectral-data-set.com and https://www.github.com/Romain3Ch216/tlse-experiments, respectively.
background contribution method result interpretation main claim
| # | sentence | role | the authors' own work | states the main conclusion | strength |
|---|---|---|---|---|---|
| s0 | Airborne hyperspectral images can be used to map the land cover in large urban areas, thanks to their very high spatial and spectral resolutions on a wide spectral domain. | background 1.00 | 0.06 | 0.02 | asserted 0.98 |
| s1 | While the spectral dimension of hyperspectral images is highly informative of the chemical composition of the land surface, the use of state-of-the-art machine learning algorithms to map the land cover has been dramatically limited by the availability of training data. | background 0.75 | 0.06 | 0.03 | asserted 1.00 |
| s2 | To cope with the scarcity of annotations, semi-supervised and self-supervised techniques have lately raised a lot of interest in the community. | background 1.00 | 0.04 | 0.03 | asserted 0.99 |
| s3 | Yet, the publicly available hyperspectral data sets commonly used to benchmark machine learning models are not totally suited to evaluate their generalization performances due to one or several of the following properties: a limited geographical coverage (which does not reflect the spectral diversity in metropolitan areas), a small number of land cover classes and a lack of appropriate standard train / test splits for semi-supervised and self-supervised learning. | gap 0.94 | 0.09 | 0.06 | asserted 0.97 |
| s4 | Therefore, we release in this paper the Toulouse Hyperspectral Data Set that stands out from other data sets in the above-mentioned respects in order to meet key issues in spectral representation learning and classification over large-scale hyperspectral images with very few labeled pixels. | contribution 1.00 | 0.92 | 0.45 | asserted 1.00 |
| s5 | Besides, we discuss and experiment self-supervised techniques for spectral representation learning, including the Masked Autoencoder, and establish a baseline for pixel-wise classification achieving 85% overall accuracy and 77% F1 score. | contribution 0.52 | 0.97 | 0.32 | asserted 1.00 |
| s6 | The Toulouse Hyperspectral Data Set and our code are publicly available at https://www.toulouse-hyperspectral-data-set.com and https://www.github.com/Romain3Ch216/tlse-experiments, respectively. | contribution 0.96 | 0.47 | 0.05 | asserted 0.99 |
What was examined, and what was not
- arXiv listing abstract v3
- full text
- figures and tables
- references and citing sentences
- relations to other papers
- P abstract
- S sections
- T body sentences
- C citations
- R cross-paper relations
The paper
Original abstract
Airborne hyperspectral images can be used to map the land cover in large urban areas, thanks to their very high spatial and spectral resolutions on a wide spectral domain. While the spectral dimension of hyperspectral images is highly informative of the chemical composition of the land surface, the use of state-of-the-art machine learning algorithms to map the land cover has been dramatically limited by the availability of training data. To cope with the scarcity of annotations, semi-supervised and self-supervised techniques have lately raised a lot of interest in the community. Yet, the publicly available hyperspectral data sets commonly used to benchmark machine learning models are not totally suited to evaluate their generalization performances due to one or several of the following properties: a limited geographical coverage (which does not reflect the spectral diversity in metropolitan areas), a small number of land cover classes and a lack of appropriate standard train / test splits for semi-supervised and self-supervised learning. Therefore, we release in this paper the Toulouse Hyperspectral Data Set that stands out from other data sets in the above-mentioned respects in order to meet key issues in spectral representation learning and classification over large-scale hyperspectral images with very few labeled pixels. Besides, we discuss and experiment self-supervised techniques for spectral representation learning, including the Masked Autoencoder, and establish a baseline for pixel-wise classification achieving 85% overall accuracy and 77% F1 score. The Toulouse Hyperspectral Data Set and our code are publicly available at https://www.toulouse-hyperspectral-data-set.com and https://www.github.com/Romain3Ch216/tlse-experiments, respectively.
Human review
How the literature uses it
1 arXiv paper cite this in 1 sentence
What it cites
Cites as background 3 papers
-
Self-supervised knowledge distillation takes inspiration from knowledge distillation [ref] by taking a teacher network to supervise a student network that sees different transformations of the input data.
Distilling the Knowledge in a Neural Network · Self-supervised learning: an overview · 0.94
-
Those tasks include rotation self-supervision [ref] , exemplar self-supervision [ref] , contrastive learning [ref] or self-supervised knowledge distillation [ref] .
Unsupervised Representation Learning by Predicting Image Rotations · Self-supervised learning: an overview · 0.99
-
Those tasks include rotation self-supervision [ref] , exemplar self-supervision [ref] , contrastive learning [ref] or self-supervised knowledge distillation [ref] .
Representation Learning with Contrastive Predictive Coding · Self-supervised learning: an overview · 0.99
Discussion
Know something about this paper? Leave a comment, or highlight any passage above to comment on it.