Weakly Supervised Segmentation of Buildings in Digital Elevation Models
Weakly Supervised Segmentation of Buildings in Digital Elevation Models
复制标题
数字高程模型中建筑物的弱监督分割
DOI：
--
复制
发表时间：
2022
影响因子：
4.8
通讯作者：
V. Kindratenko
中科院分区：
文献类型：
--
作者：
A. Soliman;Yifan Chen;Shirui Luo;Rauf Makharov;V. Kindratenko
缺乏高质量的标签数据被认为是训练机器和深度学习（DL）模型的主要瓶颈之一。使用不完整、粗糙或不准确的数据进行弱监督学习是克服训练数据稀缺的另一种策略。我们使用开放访问的Microsoft建筑足迹（MS-BF）数据集中的现有标签数据，训练了一个U-Net模型，用于从高分辨率数字高程模型（DEM）中分割建筑物足迹。使用独立的手动标记的基准进行比较表明弱监督学习的成功，因为模型预测的质量[交集对并集（IoU）：0.876]超过了原始Microsoft数据质量（IoU：0.672）约20%。此外，添加额外的通道，如高程导数，坡度，纵横比和剖面曲率并没有增强弱学习过程，因为模型直接从原始高程数据中学习。我们的研究结果证明了使用现有数据训练DL模型的价值，即使它们是嘈杂和不完整的。
The lack of quality label data is considered one of the main bottlenecks for training machine and deep learning (DL) models. Weakly supervised learning using incomplete, coarse, or inaccurate data is an alternative strategy to overcome the scarcity of training data. We trained a U-Net model for segmenting buildings’ footprints from a high-resolution digital elevation model (DEM), using the existing label data from the open-access Microsoft building footprints (MS-BF) dataset. Comparison using an independent, manually labeled benchmark indicated the success of weak supervision learning as the quality of model prediction [intersection over union (IoU): 0.876] surpassed that of the original Microsoft data quality (IoU: 0.672) by approximately 20%. Moreover, adding extra channels such as elevation derivatives, slope, aspect, and profile curvatures did not enhance the weak learning process as the model learned directly from the original elevation data. Our results demonstrate the value of using existing data for training DL models even if they are noisy and incomplete.