Bi-Temporal Satellite Imagery for Terrain Shadow Mitigation in Mountain Forests

This dataset supports research on mitigating terrain shadows in very high-resolution satellite imagery for land cover mapping. It contains WorldView-2 multispectral and WorldView-1 panchromatic images, fused to improve evergreen conifer detection in temperate mixed mountain forests. The underlying research was published in the International Journal of Applied Earth Observation and Geoinformation in October 2024.

Use Cases

• Apply the improved Silva et al. (2018) shadow detection method to identify terrain shadows in mountainous, forested landscapes.
• Implement the quantitative algorithm to differentiate dark and light terrain shadows based on object visibility in shadowed areas.
• Train a 3D U-Net deep learning model to detect evergreen conifers using the bi-temporal fused imagery.
• Evaluate the effectiveness of bi-temporal image fusion for shadow reduction in WorldView-2 and WorldView-1 data.
• Assess the applicability of the fusion approach to other VHR satellite sensors like Pléiades or SkySat, considering image co-registration requirements.

Strengths

• Focuses on very high-resolution (VHR) optical satellite imagery, specifically WorldView-2 and WorldView-1 data.
• Targets a specific and challenging environment: rugged, mountainous temperate mixed forests.
• The methodology is validated in a peer-reviewed article published in October 2024.

Limitations

• The dataset scope and specific file formats, size, and row/column counts are not provided in the input.
• Application to multi-sensor systems beyond WorldView requires careful image co-registration, which may be a technical barrier.
• The geographic extent and specific time range of the satellite imagery are not specified.

Provenance

Source

DANS Data Station Life Sciences Collection, authored by X. Zhu.

Collection Method

Bi-temporal fusion of WorldView-2 multispectral and WorldView-1 panchromatic satellite imagery.

Freshness

The dataset record was last updated on 2026-02-24.

Geography

Temperate mixed mountain forests, potentially related to giant panda habitat (inferred from tags).