Capsule Feature Pyramid Network for Building Footprint Extraction From High-Resolution Aerial Imagery
Capsule Feature Pyramid Network for Building Footprint Extraction From High-Resolution Aerial Imagery
复制标题
DOI：
10.1109/lgrs.2020.2986380
复制
发表时间：
2021-05
影响因子：
4.8
通讯作者：
Yongtao Yu;Yongfeng Ren;H. Guan;Dilong Li;Changhui Yu;Shenghua Jin;Lanfang Wang
中科院分区：
文献类型：
--
作者：
Yongtao Yu;Yongfeng Ren;H. Guan;Dilong Li;Changhui Yu;Shenghua Jin;Lanfang Wang
建筑物足迹的提取有着广泛的应用。然而，由于大小和形状的差异，遮挡和复杂的场景，它仍然是具有挑战性的，准确地提取建筑物的足迹从航空图像。这封信提出了一个胶囊特征金字塔网络（CapFPN）的建筑足迹提取航空图像。CapFPN利用胶囊的特性，融合不同层次的胶囊特征，能够提取高分辨率、内在的、语义强的特征，有效地提高了逐像素建筑物足迹提取的精度。通过使用带符号的距离图作为地面实况，CapFPN可以提取没有小孔的实心建筑区域。对航空图像数据集的定量评估表明，精度，召回率，交集（IoU），和F-分数分别为0.928，0.914，0.853和0.921，分别获得。与六种现有方法的比较研究证实了CapFPN在准确提取建筑物足迹方面的上级性能。
Building footprint extraction plays an important role in a wide range of applications. However, due to size and shape diversities, occlusions, and complex scenarios, it is still challenging to accurately extract building footprints from aerial images. This letter proposes a capsule feature pyramid network (CapFPN) for building footprint extraction from aerial images. Taking advantage of the properties of capsules and fusing different levels of capsule features, the CapFPN can extract high-resolution, intrinsic, and semantically strong features, which perform effectively in improving the pixel-wise building footprint extraction accuracy. With the use of signed distance maps as ground truths, the CapFPN can extract solid building regions free of tiny holes. Quantitative evaluations on an aerial image data set show that a precision, recall, intersection-over-union (IoU), and F-score of 0.928, 0.914, 0.853, and 0.921, respectively, are obtained. Comparative studies with six existing methods confirm the superior performance of the CapFPN in accurately extracting building footprints.