当前位置: X-MOL 学术IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Noise-Tolerant Deep Neighborhood Embedding for Remotely Sensed Images With Label Noise
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing ( IF 4.7 ) Pub Date : 2021-02-02 , DOI: 10.1109/jstars.2021.3056661
Jian Kang , Ruben Fernandez-Beltran , Xudong Kang , Jingen Ni , Antonio Plaza

Recently, many deep learning-based methods have been developed for solving remote sensing (RS) scene classification or retrieval tasks. Most of the adopted loss functions for training these models require accurate annotations. However, the presence of noise in such annotations (also known as label noise) cannot be avoided in large-scale RS benchmark archives, resulting from geo-location/registration errors, land-cover changes, and diverse knowledge background of annotators. To overcome the influence of noisy labels on the learning process of deep models, we propose a new loss function called noise-tolerant deep neighborhood embedding which can accurately encode the semantic relationships among RS scenes. Specifically, we target at maximizing the leave-one-out $K$ -NN score for uncovering the inherent neighborhood structure among the images in feature space. Moreover, we down-weight the contribution of potential noisy images by learning their localized structure and pruning the images with low leave-one-out $K$ -NN scores. Based on our newly proposed loss function, classwise features can be more robustly discriminated. Our experiments, conducted on two benchmark RS datasets, validate the effectiveness of the proposed approach on three different RS scene interpretation tasks, including classification, clustering, and retrieval. The codes of this article will be publicly available from https://github.com/jiankang1991 .

中文翻译:

带有标签噪声的遥感图像的耐噪深度邻域嵌入

最近,已经开发了许多基于深度学习的方法来解决遥感(RS)场景分类或检索任务。用于训练这些模型的大多数采用的损失函数都需要准确的注释。但是,由于地理位置/注册错误,土地覆被变化以及注释者的不同知识背景,在大型RS基准归档中无法避免此类注释中存在噪声(也称为标签噪声)。为了克服噪声标签对深度模型学习过程的影响,我们提出了一种新的损失函数,称为噪声容忍深度邻域嵌入,可以准确地编码RS场景之间的语义关系。具体来说,我们的目标是最大程度地避免浪费$ K $ -NN分数,用于揭示特征空间中图像之间的固有邻域结构。此外,我们通过学习潜在的噪点图像的局部结构并以较低的留一法修剪来减少图像的权重$ K $ -NN分数。基于我们新提出的损失函数,可以更稳健地区分类特征。我们在两个基准RS数据集上进行的实验验证了该方法在三种不同的RS场景解释任务(包括分类,聚类和检索)上的有效性。本文的代码可从以下位置公开获得https://github.com/jiankang1991
更新日期:2021-02-26
down
wechat
bug