Classification algorithm for land use in the giant panda habitat of Jiajinshan based on spatial case-based reasoning

A novel spatial case-based reasoning (SCBR) algorithm achieved superior land use classification accuracy of 95% and a Kappa coefficient of 90% in the Jiajinshan giant panda habitat, outperforming machine learning models like SVM, U-Net, ViT, and Trans-Unet. The study demonstrated significant ecological improvements, particularly in forest protection and restoration, between 2018 and 2022 with consistent accuracy above 80%. Unlike traditional methods, SCBR requires minimal domain knowledge and avoids complex model training, offering a practical tool for monitoring and protecting panda habitats.

Xia, W., Zhang, W., & Wu, S. (2024). Classification algorithm for land use in the giant panda habitat of Jiajinshan based on spatial case-based reasoning. Frontiers in Environmental Science, 12. https://doi.org/10.3389/fenvs.2024.1298327