Hu et al. (2025) Transferring visual knowledge in large-scale CNNs facilitates interpretable and cost-effective wetland cover mapping under dynamic environments

Identification

• Journal: Mendeley Data
• Year: 2025
• Date: 2025-12-15
• Authors: Hu, Qiao, Yu, Jiahua, Kang, Yu, Li, Dongxue, Li, Jiating
• DOI: 10.17632/tyrwtcycgk.3

Research Groups

• University of Manitoba
• Zhengzhou University

Short Summary

This study proposes a method utilizing transferred visual knowledge in large-scale Convolutional Neural Networks (CNNs) to achieve interpretable and cost-effective wetland cover mapping, particularly in dynamic environments.

Objective

• To develop an interpretable and cost-effective methodology for wetland cover mapping in dynamic environments by leveraging transferred visual knowledge within large-scale Convolutional Neural Networks (CNNs).

Study Configuration

• Spatial Scale: Not specified in the provided text.
• Temporal Scale: Dynamic environments, specific temporal scale not detailed.

Methodology and Data

• Models used: Large-scale Convolutional Neural Networks (CNNs) with a focus on transferring visual knowledge.
• Data sources: Remote sensing data (implied by "visual knowledge" and "Remote Sensing" category).

Main Results

• The proposed approach, based on transferring visual knowledge in large-scale CNNs, successfully facilitates interpretable and cost-effective mapping of wetland cover.

Contributions

• Introduces a novel application of transferred visual knowledge in large-scale CNNs for wetland cover mapping.
• Enhances the interpretability and cost-effectiveness of wetland mapping solutions, particularly relevant for dynamic environmental conditions.

Funding

• Not specified in the provided text.

Citation

@article{Hu2025Transferring,
  author = {Hu, Qiao and Yu, Jiahua and Kang, Yu and Li, Dongxue and Li, Jiating},
  title = {Transferring visual knowledge in large-scale CNNs facilitates interpretable and cost-effective wetland cover mapping under dynamic environments},
  journal = {Mendeley Data},
  year = {2025},
  doi = {10.17632/tyrwtcycgk.3},
  url = {https://doi.org/10.17632/tyrwtcycgk.3}
}