HLJ_MISDataset

HLJ_MISDataset, a large-scale remote sensing instance segmentation dataset specifically designed for mining land occupation analysis. The dataset comprises 2,616 high-resolution images encompassing 3,808 meticulously annotated instances across over 130 mineral types in Heilongjiang Province, China. Our analysis reveals significant challenges including irregular object geometries, heterogeneous background clutter, and substantial intra-category scale disparities.HLJ_MISDataset是一款专为采矿用地占用分析设计的大规模遥感实例分割数据集。该数据集包含2616张高分辨率遥感图像，涵盖中国黑龙江省境内130余种矿物类型对应的3808个经精细标注的实例。经分析可知，该数据集存在多项显著挑战：目标几何形态不规则、背景杂波异质性较强，且类别内部尺度差异显著。