首页|期刊导航|北京测绘|遥感建筑物提取的边界交并比评测协议

遥感建筑物提取的边界交并比评测协议OA

A boundary IoU evaluation protocol for remote sensing building extraction

中文摘要英文摘要

针对高分辨率遥感影像建筑物提取中现有评测体系以平均精度均值(mAP)和掩码交并比(Mask IoU)为主,对边界精度反映不足的问题,本文提出适配该领域的边界交并比(BIoU)评测协议.该协议由BIoU、95分位豪斯多夫距离(HD95)双指标构成,并采用双口径报告策略.基于武汉大学(WHU)与马萨诸塞(Massachusetts)建筑物数据集,系统对比了实时实例分割模型(YOLOv8-seg)系列变体、U型编解码网络(U-Net)、深度语义分割网络(DeepLabv3+)等8种方法配置在本协议下的表现.结果显示,同域场景下各方法交并比(IoU)接近(0.843~0.851),但HD95揭示了范式级差距.跨域场景下只看一次模型(YOLO)系列保持0.224~0.279的IoU,语义分割模型降至0.054~0.095,同域排序被显著反转.少量目标域样本(占比5%)微调后,语义模型IoU反超YOLO零样本性能.本文协议有效揭示了同域Mask IoU评测的局限,为该领域方法评测与选型提供了新的工具.

In the extraction of buildings from high-resolution remote-sensing imagery,the prevailing evaluation framework relies mainly on mean average precision(mAP)and mask intersection over union(Mask IoU),which provides limited insight into boundary accuracy.To address this shortcoming,this paper proposed a boundary intersection over union(BIoU)evaluation protocol tailored to the remote-sensing building extraction domain.The protocol comprised two metrics—BIoU and the 95th-percentile Hausdorff distance(HD95)—and adopted a dual-caliber reporting strategy.Based on the Wuhan Uni-versity(WHU)and Massachusetts building datasets,it systematically compared the performance of eight method configura-tions under the proposed protocol,including variants of the real-time instance-segmentation model(YOLOv8-seg),the U-shaped encoder-decoder network(U-Net),and the deep semantic segmentation network(DeepLabv3+).The results show that,for in-domain scenarios,the intersection over union(IoU)of all methods is comparable(0.843-0.851),whereas the HD95 reveals a paradigm-level gap.In cross-domain scenarios,the you only look once(YOLO)-series detection algorithms retain an IoU of 0.224-0.279,whereas semantic-segmentation models drop to 0.054-0.095,significantly reversing the in-domain ranking.After fine-tuning with only 5%of target-domain samples,the semantic-segmentation models achieve better IoU performance than the zero-shot YOLO.This demonstrates that the proposed BIoU+HD95 protocol effectively exposes the limitations of relying solely on in-domain Mask IoU for evaluation,offering a new tool for method evaluation and selection in remote-sensing building extraction.

王伯桢

北京建筑大学 测绘与城市空间信息学院,北京 102616

天文与地球科学

建筑物提取实例分割边界交并比(BIoU)评测协议跨域泛化95分位豪斯多夫距离(HD95)实时实例分割模型(YOLOv8-seg)系列变体U型编解码网络(U-Net)深度语义分割网络(DeepLabv3+)

《北京测绘》 2026 (9)

1232-1240,9

10.19580/j.cnki.1007-3000.2026060016

评论