Si et al. (2026) SPEX: A Vision–Language Model for Land Cover Extraction on Spectral Remote Sensing Images

Identification

• Journal: IEEE Transactions on Geoscience and Remote Sensing
• Year: 2026
• Date: 2026-01-01
• Authors: Dongchen Si, Di Wang, Erzhong Gao, Xiaolei Qin, Zhao Liu, Jing Zhang, Minqiang Xu, Jianbo Zhan, Jianshe Wang, Lin Liu, Bo Du, Liangpei Zhang
• DOI: 10.1109/tgrs.2026.3670308

Research Groups

[List the main research groups, labs, or departments involved in the study.]

Short Summary

This paper introduces SPEX, a novel vision-language model specifically designed for the task of land cover extraction using spectral remote sensing images.

Objective

• To develop and evaluate a vision-language model (SPEX) for accurate land cover extraction from spectral remote sensing images.

Study Configuration

• Spatial Scale: [Description]
• Temporal Scale: [Description]

Methodology and Data

• Models used: SPEX (a Vision-Language Model)
• Data sources: Spectral remote sensing images

Main Results

[Key findings, synthetic and quantitative]

Contributions

• Introduction of SPEX, a new vision-language model tailored for land cover extraction in remote sensing.

Funding

[List projects, programs, and reference codes that funded this research]

Citation

@article{Si2026SPEX,
  author = {Si, Dongchen and Wang, Di and Gao, Erzhong and Qin, Xiaolei and Liu, Zhao and Zhang, Jing and Xu, Minqiang and Zhan, Jianbo and Wang, Jianshe and Liu, Lin and Du, Bo and Zhang, Liangpei},
  title = {SPEX: A Vision–Language Model for Land Cover Extraction on Spectral Remote Sensing Images},
  journal = {IEEE Transactions on Geoscience and Remote Sensing},
  year = {2026},
  doi = {10.1109/tgrs.2026.3670308},
  url = {https://doi.org/10.1109/tgrs.2026.3670308}
}