Artificial Intelligence for Mineral Exploration Imagery: A Systematic Mapping Review of Data, Tasks, and Methods

作者：Yu Xiao, Chunfang Kong, Kai Xu, Daihe Lyu, Yu Zhou, Jiawei Tian · 发表于：Open Science Framework · 年份：2026 · DOI：10.17605/osf.io/k6wv7 · 研究领域：Data science、Computer science、Geology、Artificial intelligence、Data mining、Remote sensing

This registration specifies the protocol for a systematic mapping review of image-based artificial intelligence and computer vision applications in mineral exploration. The review aims to examine how these technologies are applied across complementary evidence domains, including regional surveys, drilling observations, and laboratory analyses. Rather than directly predicting mineral prospectivity or comparing model performance across heterogeneous geological settings, the review focuses on mapping the relationships among data sources, visual tasks, methodological paradigms, and geological or exploration-related outputs. Eligible studies must present an identifiable visual or image-formatted geoscientific data source, a clearly defined computer-vision-related task, a sufficiently described analytical method, and a demonstrable connection to a mineral exploration activity. Because the included studies may vary substantially in data modality, geological objective, spatial scale, and reporting practices, the review will not apply a single universal risk-of-bias score. Instead, the evidence will be systematically mapped according to visual data sources, computer-vision tasks, analytical methods, and, where reported, application contexts, geological outputs, multisource integration, and dataset or code availability.