SVH-BD : Synthetic Vegetation Hyperspectral Benchmark Dataset for Emulation of Remote Sensing Images

Chedly Ben Azizi; Claire Guilloteau; Gilles Roussel; Matthieu Puigt

cs.CVMar 30, 2026

SVH-BD : Synthetic Vegetation Hyperspectral Benchmark Dataset for Emulation of Remote Sensing Images

Authors: Chedly Ben Azizi, Claire Guilloteau, Gilles Roussel, Matthieu Puigt

Abstract

This dataset provides a large collection of 10,915 synthetic hyperspectral image cubes paired with pixel-level vegetation trait maps, designed to support research in radiative transfer emulation, vegetation trait retrieval, and uncertainty quantification. Each hyperspectral cube contains 211 bands spanning 400--2500 nm at 10 nm resolution and a fixed spatial layout of 64 \times 64 pixels, offering continuous simulated surface reflectance spectra suitable for emulator development and machine-learning tasks requiring high spectral detail. Vegetation traits were derived by inverting Sentinel-2 Level-2A surface reflectance using a PROSAIL-based lookup-table approach, followed by forward PROSAIL simulations to generate hyperspectral reflectance under physically consistent canopy and illumination conditions. The dataset covers four ecologically diverse regions -- East Africa, Northern France, Eastern India, and Southern Spain -- and includes 5th and 95th percentile uncertainty maps as well as Sentinel-2 scene classification layers. This resource enables benchmarking of inversion methods, development of fast radiative transfer emulators, and studies of spectral--biophysical relationships under controlled yet realistic environmental variability.

Figures & tables

Filename	Description
surf-refl.tif	Hyperspectral surface reflectance cube ( $64\times 64\times 211$ )
traits.tif	Retrieved bio-optical parameters (16 traits)
p5.tif	5th percentile uncertainty map for trait retrievals.
p95.tif	95th percentile uncertainty map for trait retrievals.
quality_scene_classifcation.img	Sentinel-2 scene classification map.
quality_scene_classifcation.hdr	Metadata associated with the classification map.

Table 1: Summary of files provided for each tile

Symbol	Description	Unit	Range
Leaf Biochemical Parameters (PROSPECT–D)
$N$	Leaf structure parameter	–	$[1,\;2.5]$
$C_{ab}$	Leaf chlorophyll a+b	$\mu g\,cm^{-2}$	$[0,\;160]$
$C_{ar}$	Leaf carotenoids	$\mu g\,cm^{-2}$	$[0,\;60]$
$C_{ant}$	Leaf anthocyanins content	$\mu g\,cm^{-2}$	$[0,\;5]$
$C_{brown}$	Brown pigments	–	$[0,\;1]$

Table 2: Parameter ranges for the combined leaf (PROSPECT-D) and canopy (4SAIL) RTM (PROSAIL).

Figure 2: Geographic distribution of the four study regions used for synthetic hyperspectral dataset generation. The regions span diverse climate zones and vegetation types across Europe (France, Spain), Africa (Tanzania), and Asia (India).

Figure 3: Representative examples of emulated hyperspectral image cubes and their corresponding chlorophyll a+b (Cab) maps from the four study regions. For each row, the left column shows the Cab map and the right column shows the associated HSI scene. All scenes are displayed using an RGB composite.

Panel of Figure 4 — Figure 4: Simulated hyperspectral reflectance spectra from the four study regions, color-coded by chlorophyll a+b content ( $C_{ab}$ ).

Explore similar work

A Variational Latent-Space Framework for Uncertainty-Aware Spectral Image Emulation
Mar 23, 2026Chedly Ben Azizi, Claire Guilloteau, Gilles Roussel +1Neural Surrogate Modeling Variational Autoencoders
Univ. Littoral Côte d’Opale, LISIC – UR 4491, 62219, Longuenesse, France
HyperImageNet: A Large-Scale High-Spatial Resolution Hyperspectral Imagery Classification Benchmark
Jul 23, 2026Chuguang Zeng, Jingtao Li, Yinhe Liu +1Remote Sensing Image Segmentation Land Cover Classification
Wuhan University 430072 Wuhan, China
ChronoEarth-492K: A Large Scale and Long Horizon Spatiotemporal Hyperspectral Earth Observation Dataset and Benchmark
May 15, 2026Haozhe Si, Yuxuan Wan, Yuqing Wang +2Geospatial Foundation Models Self-Supervised Time Series Representation Learning
Department of Electrical and Computer Engineering · Siebel School of Computing and Data Science University of Illinois Urbana-Champaign, IL, USA

Mar 23, 2026J/K move · Enter open · S save

A Variational Latent-Space Framework for Uncertainty-Aware Spectral Image Emulation

Chedly Ben Azizi, Claire Guilloteau, Gilles Roussel, Matthieu Puigt

Univ. Littoral Côte d’Opale, LISIC – UR 4491, 62219, Longuenesse, France

Neural Surrogate Modeling Variational Autoencoders Uncertainty Quantification Remote Sensing Hyperspectral Imaging

Open paper AskPDF arXiv

Synthetic spectral image generation is essential for remote sensing simulation and mission design, yet physically based radiative transfer models (RTMs) remain computationally expensive. Existing learning-based emulators reduce this cost, but are mostly deterministic parameter-to-spectrum regressors with limited spatial modeling and uncertainty information. We formulate spectral image emulation as a parameter-conditioned latent-variable problem and propose a variational autoencoder (VAE)-based framework combining nonlinear spectral-image representations, fast inference, and per-pixel uncertainty estimates. The framework is instantiated at spectrum and spatial--spectral levels through two-step VAE pretraining and latent mapping. We evaluate it on PROSAIL-simulated hyperspectral vegetation cubes (211 bands) and real Sentinel-3 OLCI multispectral ocean-colour imagery (21 bands) against classical regression emulators and a deep CNN baseline. Results show that no single architecture is optimal: pixel-to-pixel models perform best on controlled hyperspectral simulations, whereas a fully convolutional VAE is more robust on noisy real observations with missing or contaminated pixels. VAE-based emulators also achieve high throughput for large-scale generation. On Sentinel-3, the spatial--spectral VAE provides predictive intervals closer to empirical errors than pixel-wise neural and classical emulators, although absolute calibration remains incomplete. A look-up-table-based retrieval experiment further shows that reconstruction fidelity alone does not ensure reliable leaf area index and chlorophyll retrieval. Emulators should therefore be evaluated in representative remote-sensing end-use scenarios, not by reconstruction metrics alone.

Abstract

Explore similar work

A Variational Latent-Space Framework for Uncertainty-Aware Spectral Image Emulation

HyperImageNet: A Large-Scale High-Spatial Resolution Hyperspectral Imagery Classification Benchmark

ChronoEarth-492K: A Large Scale and Long Horizon Spatiotemporal Hyperspectral Earth Observation Dataset and Benchmark