EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

ECO-DEAU: An Ecologically Constrained Deep Learning Autoencoder for Sub-Pixel Land Cover Unmixing in Arid and Semi-Arid Regions

Leixuan Zhou, Long Li, Dehui Li, Yong Bo, Hang Li, Kai Liu, Shudong Wang

Remote Sensing · 19.03.2026 · 18 (6) · 941

DOI: 10.3390/rs18060941 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 19.03.2026. Verfasst von Leixuan Zhou und 6 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 61 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Foreign Technical Cooperation and Scientific Research Program, Science and Technology Plan Project of Hohho.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Arid and semi-arid regions are critical to terrestrial ecosystems and regional carbon cycle regulation, directly contributing to peak carbon and carbon neutrality goals. However, the fragmented landscapes in these regions pose significant challenges to conventional pixel-based classification, which often struggles with mixed pixel issues and lacks biophysical interpretability. To address these limitations, this study develops an Ecologically Constrained Deep Learning Autoencoder (ECO-DEAU) framework for sub-pixel land cover mapping by integrating biophysical constraints. Specifically, ECO-DEAU employs spectral indices to extract standard spectral signatures for five primary land cover types, which serve as initial weights to guide the autoencoder in estimating fractional abundances. The model was trained across ten representative landscape zones in the Inner Mongolia section of the Yellow River Basin and validated against high-resolution Gaofen-2 data. Results demonstrated that ECO-DEAU yielded an average R2 of 0.687, reaching a maximum R2 of 0.749 in spatially heterogeneous transition zones, representing a substantial improvement over the baseline unconstrained Deep Autoencoder (DEAU). By effectively resolving the blind source separation problem and improving decomposition accuracy, ECO-DEAU serves as a robust tool for addressing mixed pixel challenges in heterogeneous environments, thereby facilitating large-scale, high-resolution carbon sink monitoring.

Förderung

  • Foreign Technical Cooperation and Scientific Research Program
  • Science and Technology Plan Project of Hohho

Zitieren

APA

Zhou, L., Li, L., Li, D., Bo, Y., Li, H., Liu, K., & Wang, S. (2026). ECO-DEAU: An Ecologically Constrained Deep Learning Autoencoder for Sub-Pixel Land Cover Unmixing in Arid and Semi-Arid Regions. Remote Sensing, 18(6), 941. https://doi.org/10.3390/rs18060941

Vancouver

Zhou L, Li L, Li D, Bo Y, Li H, Liu K, et al. ECO-DEAU: An Ecologically Constrained Deep Learning Autoencoder for Sub-Pixel Land Cover Unmixing in Arid and Semi-Arid Regions. Remote Sensing. 2026;18(6):941. doi:10.3390/rs18060941

BibTeX
@article{Zhou2026,
  author = {Zhou, Leixuan and Li, Long and Li, Dehui and Bo, Yong and Li, Hang and Liu, Kai and Wang, Shudong},
  title = {ECO-DEAU: An Ecologically Constrained Deep Learning Autoencoder for Sub-Pixel Land Cover Unmixing in Arid and Semi-Arid Regions},
  journal = {Remote Sensing},
  year = {2026},
  volume = {18},
  number = {6},
  pages = {941},
  doi = {10.3390/rs18060941}
}
Kostenlos starten →