A Novel Deep Embedding Network for Building Shape Recognition
A Novel Deep Embedding Network for Building Shape Recognition
复制标题
用于构建形状识别的新型深度嵌入网络
DOI：
10.1109/lgrs.2017.2753821
复制
发表时间：
2017-10
影响因子：
4.8
通讯作者：
Su Nan
中科院分区：
文献类型：
--
作者：
Tian Shu;Zhang Ye;Zhang Junping;Su Nan
建筑物形态作为一种重要的结构要素，在城市遥感应用中起着重要的作用。然而，由于建筑结构的高度复杂性和类内差异，建筑形状的描述和识别的能力变得有限，甚至贫困。在这封信中，提出了一种新的深度嵌入网络来构建形状识别，它结合了卷积神经网络（CNN）的无监督特征学习和一种新的三重丢失。具体来说，我们利用CNN强大的辨别能力来学习一种有效的建筑形状表示，以进行形状识别。通过这种深度嵌入网络，可以将高维图像空间映射到低维特征空间，并且深度特征可以有效地减少类内变化，同时增加不同建筑物形状图像之间的类间变化。然后，将得到的深层特征用于建筑物形状识别。该方法包括两个阶段。在第一阶段，对于存储在形状基元库和建筑物形状数据集中的标准建筑物形状图像查询，使用深度嵌入网络提取两组深度特征。在第二阶段，我们将形状识别问题转化为一个特征匹配问题，并采用集合到集合的特征匹配方法来实现最终的建筑物形状识别结果。在VHR-10和UCML数据集上的实验表明了该方法的有效性和准确性。
Building shape, as a key structured element, plays a significant role in various urban remote sensing applications. However, because of high complexity and intraclass variations between building structures, the capability of building shape description and recognition becomes limited or even impoverished. In this letter, a novel deep embedding network is proposed for building shape recognition, which combines the strength of the unsupervised feature learning of convolutional neural networks (CNNs) and a novel triplet loss. Specifically, we take advantage of the strong discriminative power of CNNs to learn an efficient building shape representation for shape recognition. With this deep embedding network, the high-dimensional image space can be mapped into a low-dimensional feature space, and the deep features can effectively reduce the intraclass variations while increasing the interclass variation between different building shape images. Afterward, the derived deep features are exploited for the process of building shape recognition. This method consists of two stages. In the first stage, for standard building shape image queries stored in the shape primitives library and the building shape data set, two sets of deep features are extracted with the deep embedding network. In the second stage, we formulate the shape recognition task into a feature matching problem and the final building shape recognition results can be achieved by set-to-set feature matching method. Experiments on the VHR-10 and UCML data sets demonstrate the effectiveness and precision of the proposed method.
登录
查看更多内容
通过梯度提升随机卷积网络框架进行场景分类
影响因子：
8.2
作者：
Zhang Fan;Zhang Liangpei;Du Bo
通讯作者：
Du Bo
影响因子：
4.8
作者：
Luus, F. P. S.;Salmon, B. P.;Maharaj, B. T. J.
通讯作者：
Maharaj, B. T. J.
通过相对距离比较进行深度特征学习以进行行人重新识别
影响因子：
8
作者：
Ding, Shengyong;Lin, Liang;Chao, Hongyang
通讯作者：
Chao, Hongyang
影响因子：
4.8
作者：
Yimin Yan;Ye Zhang;Shu Tian;Fengjiao Gao
通讯作者：
Yimin Yan;Ye Zhang;Shu Tian;Fengjiao Gao
影响因子：
4.8
作者：
Marmanis, Dimitrios;Datcu, Mihai;Stilla, Uwe
通讯作者：
Stilla, Uwe