Objective Rapid and accurate monitoring of vegetation coverage in mountainous mining areas is a key task in ecological environment research and is crucial for soil and water conservation. The Hutubi mining area in Xinjiang has complex terrain, and the accuracy of existing monitoring methods needs to be improved. It is urgent to identify suitable vegetation indices to achieve precise monitoring; therefore, this area is selected for the research.

Methods Combining UAV visible light images with field survey data, the applicability of ten visible light vegetation indices was analyzed along the elevation gradient. The Otsu threshold method was used to calculate vegetation coverage, and the accuracy of each index was evaluated through correlation analysis with measured values.

Results 1) The suitable vegetation indices varied across areas with low, medium, and high vegetation coverage. In low-coverage areas, the coverage calculated by CIVE and ExG indices showed the highest correlation with measured values, with accuracies of 87% and 85%, respectively. In medium-coverage areas, ExG and RGBVI indices performed better, with correlation accuracies of 86% and 77%, respectively. In high-coverage areas, ExG, RGBVI, and RGR indices performed the best. Among them, ExG achieved an estimation accuracy of 87%, with root mean square error (RMSE) < 0.05. 2) Considering performance in each area, the ExG index showed high stability and applicability under different coverage levels, making it the optimal vegetation index for the study area. 3) By optimizing the identification accuracy of vegetation indices with consideration of topographic factors, it was found that the identification of vegetation indices in medium-coverage areas still showed some misclassification and omission. Further optimization should be carried out by incorporating more topographic factors, and the accuracy could be further improved by introducing multispectral images or machine learning methods in the future.

Conclusions This study clarifies the suitable vegetation indices for areas with different vegetation coverage levels in the study area, provides reliable technical support for the rapid monitoring and management of vegetation in the Hutubi mining area, and offers a reference for the application of UAV images in vegetation monitoring of mountainous mining areas, which is of great significance for soil and water conservation and ecological restoration.