Quantifying and Reducing Registration Uncertainty of Spatial Vector Labels on Earth Imagery
Quantifying and Reducing Registration Uncertainty of Spatial Vector Labels on Earth Imagery
复制标题
DOI：
10.1145/3534678.3539410
复制
发表时间：
2022-08
期刊：
影响因子：
--
通讯作者：
Wenchong He;Zhenling Jiang;Marcus Kriby;Yiqun Xie;X. Jia;Da Yan;Yang Zhou
中科院分区：
文献类型：
--
作者：
Wenchong He;Zhenling Jiang;Marcus Kriby;Yiqun Xie;X. Jia;Da Yan;Yang Zhou
鉴于栅格图像特征和具有注册不确定性的不完美矢量训练标签，本文研究了一个深度学习框架，可以量化和减少培训标签的注册不确定性，并同时训练培训神经网络参数。这个问题在广泛的应用中很重要，例如在地球图像上的简化分类或在医学图像上的组织分割，从而注释精确的矢量标签是昂贵且耗时的。但是，由于类标签的矢量表示与图像特征的栅格表示以及对具有不确定标签位置的训练神经网络的需求，问题是具有挑战性的。现有对不确定培训标签的研究通常集中在标签类语义中的不确定性或在像素级别（不是连续矢量）的标签注册不确定性表征。为了填补空白，本文提出了一个新颖的学习框架，该框架明确量化了矢量标签的注册不确定性。我们通过重新估算基于高斯过程的真实矢量标签位置分布的后验来提出一个注册 - 不确定性损失函数，并设计迭代不确定性降低算法。对国家水文数据集精炼中现实世界数据集的评估表明，所提出的方法在注册不确定性估计的性能和分类性能中大大优于几个基准。
Given raster imagery features and imperfect vector training labels with registration uncertainty, this paper studies a deep learning framework that can quantify and reduce the registration uncertainty of training labels as well as train neural network parameters simultaneously. The problem is important in broad applications such as streamline classification on Earth imagery or tissue segmentation on medical imagery, whereby annotating precise vector labels is expensive and time-consuming. However, the problem is challenging due to the gap between the vector representation of class labels and the raster representation of image features and the need for training neural networks with uncertain label locations. Existing research on uncertain training labels often focuses on uncertainty in label class semantics or characterizes label registration uncertainty at the pixel level (not contiguous vectors). To fill the gap, this paper proposes a novel learning framework that explicitly quantifies vector labels' registration uncertainty. We propose a registration-uncertainty-aware loss function and design an iterative uncertainty reduction algorithm by re-estimating the posterior of true vector label locations distribution based on a Gaussian process. Evaluations on real-world datasets in National Hydrography Dataset refinement show that the proposed approach significantly outperforms several baselines in the registration uncertainty estimations performance and classification performance.