EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Lightweight CNN Pipeline for Soil–Vegetation Classification from Sentinel-2: A Methodological Study over Dolj County, Romania

Andreea Florina Jocea, Liviu Porumb, Lucian Necula, Dan Raducanu

Applied Sciences · 14.11.2025 · 15 (22) · 12112

DOI: 10.3390/app152212112 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 14.11.2025. Verfasst von Andreea Florina Jocea und 3 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Ministry of Research, Innovation and Digitization, CCCDI—UEFISCDI.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate land cover mapping is essential for environmental monitoring and agricultural management. Sentinel-2 imagery, with high spatial resolution and open access, provides valuable opportunities for operational classification. Convolutional neural networks (CNNs) have demonstrated state-of-the-art results, yet their adoption is limited by high computational demands and limited methodological transparency. This study proposes a lightweight CNN for soil–vegetation classification, in Dolj County, Romania. The architecture integrates three convolutional blocks, global average pooling, and dropout, with fewer than 150,000 trainable parameters. A fully documented workflow was implemented, covering preprocessing, patch extraction, training, and evaluation, addressing reproducibility challenges common in deep leaning studies. Experiments on Sentinel-2 imagery achieved 91.2% overall accuracy and a Cohen’s kappa of 0.82. These results are competitive with larger CNNs while reducing computational requirements by over 90%. Comparative analyses showed improvements over an NDVI baseline and a favorable efficiency–accuracy balance relative to heavier CNNs reported in the literature. A complementary ablation analysis confirmed that the adopted three-block architecture provides the optimal trade-off between accuracy and efficiency, empirically validating the robustness of the proposed design. These findings highlight the potential of lightweight, transparent deep learning for scalable and reproducible land cover monitoring, with prospects for extension to multi-class mapping, multi-temporal analysis, and fusion with complementary data such as SAR. This work provides a methodological basis for operational applications in resource-constrained environments.

Förderung

  • Ministry of Research, Innovation and Digitization, CCCDI—UEFISCDI

Zitieren

APA

Jocea, A. F., Porumb, L., Necula, L., & Raducanu, D. (2025). A Lightweight CNN Pipeline for Soil–Vegetation Classification from Sentinel-2: A Methodological Study over Dolj County, Romania. Applied Sciences, 15(22), 12112. https://doi.org/10.3390/app152212112

Vancouver

Jocea AF, Porumb L, Necula L, Raducanu D. A Lightweight CNN Pipeline for Soil–Vegetation Classification from Sentinel-2: A Methodological Study over Dolj County, Romania. Applied Sciences. 2025;15(22):12112. doi:10.3390/app152212112

BibTeX
@article{Jocea2025,
  author = {Jocea, Andreea Florina and Porumb, Liviu and Necula, Lucian and Raducanu, Dan},
  title = {A Lightweight CNN Pipeline for Soil–Vegetation Classification from Sentinel-2: A Methodological Study over Dolj County, Romania},
  journal = {Applied Sciences},
  year = {2025},
  volume = {15},
  number = {22},
  pages = {12112},
  doi = {10.3390/app152212112}
}
Kostenlos starten →