作者:Qin, FC (Qin, Fachao)[ 1,2 ] ; Guo, JM (Guo, Jiming)[ 2 ] ; Sun, WD (Sun, Weidong)[ 3 ]
卷: 8 期: 3 页: 204-213
DOI: 10.1080/2150704X.2016.1258128
出版年: 2017
摘要
A series of deep learning algorithms have recently shown excellent performances in many different fields. Deep models are usually generated by stacking similar modules, e.g., restricted Boltzmann machines (RBMs), and they are especially suitable for discriminating complex objects through the use of 'big data'. However, object-oriented classification (OOC) for polarimetric synthetic aperture radar (PolSAR) imagery is based on homogeneous regions instead of pixels, which results in a degraded performance for the deep models, as the data volume is inadequate. To solve this problem, we adopt an RBM as the module, and use it to construct an adaptive boosting (AdaBoost) model instead of a stacked deep model, to carry out OOC for PolSAR imagery. The experimental results demonstrate that the proposed model is superior to the stacked RBM model and the other common methods for OOC.
作者信息
通讯作者地址: Qin, FC (通讯作者)
China West Normal Univ, Land & Resources Coll, Nanchong, Peoples R China. |
地址:
[ 1 ] China West Normal Univ, Land & Resources Coll, Nanchong, Peoples R China | |
[ 2 ] Wuhan Univ, Sch Geodesy & Geomat, Wuhan, Peoples R China | |
[ 3 ] Wuhan Univ, State Key Lab Informat Engn Surveying Mapping & R, Wuhan, Peoples R China |
电子邮件地址:larcqfc@163.com
出版商
TAYLOR & FRANCIS LTD, 2-4 PARK SQUARE, MILTON PARK, ABINGDON OR14 4RN, OXON, ENGLAND
类别 / 分类
研究方向:Remote Sensing; Imaging Science & Photographic Technology
Web of Science 类别:Remote Sensing; Imaging Science & Photographic Technology
文献信息
文献类型:Article
语种:English
入藏号: WOS:000390574500001
ISSN: 2150-704X
eISSN: 2150-7058
期刊信息
Impact Factor (影响因子): 1.487