干旱区生态环境知识资源中心(Arid): 基于A-Clenet5的荒漠化草原草种识别与分类

Arid

	基于A-Clenet5的荒漠化草原草种识别与分类
其他题名	Identification and classification of desert grassland species based on A-Clenet5
	刘浩; 杜建民; 皮伟强; 朱相兵; 张锡鹏; 康拥朝
来源期刊	光电子·激光
ISSN	1005-0086
出版年	2019
卷号	30 期号:10 页码:1056-1061
中文摘要	草原退化已经成为人类面临的主要生态问题,其标志之一是草原植被群落结构的改变,而草原草种的高光谱识别与分类是利用遥感进行大面积高精度草原退化监测与治理的基础与前提。由于天然草原草种分布的随机性和遥感图像云系情况的复杂性,草种识别精度低的问题未能得到满意的解决。本研究基于深度学习首次提出A-Clenet5法,在天然草原自然光下采集典型牧草高光谱数据,对数据进行预处理,利用A-Clenet5法进行特征挖掘和数据分类。研究结果表明,该方法对草种识别精度可达到92.18%,满足基于高光谱的草原草种的识别要求,为草原退化高精度遥感监测提供了可能。
英文摘要	Grassland degradation has become a major ecological problem facing human beings.One of signs is the change of grassland vegetation community structure.The hyperspectral recognition and classification of grassland grasses is the basis and premise of remote sensing for large-area and high-precision grassland degradation monitoring and treatment.Due to the randomness of the distribution of natural grassland species and the complexity of the remote sensing image cloud system,the problem of low recognition accuracy of grass species has not been satisfactorily resolved.This study proposes the A-Clent5 method for the first time based on deep learning,collects typical pasture hyperspectral data under natural light in the natural grassland.Then the data is preprocessed,and the feature mining and data classification are performed by the A-Cletet5 method.The research results show that the recognition accuracy of this method is 92.18%,which realizes the identification of grassland typical grass species based on hyperspectral, which provides the possibility of high-precision remote sensing monitoring of grassland degradation.
中文关键词	高光谱 ; 草种分类 ; 深度学习 ; 特征挖掘
英文关键词	A-Clenet5 hyperspectral grass species classification A-Clenet5 deeplearning feature mining
语种	中文
收录类别	CSCD
WOS类目	AGRICULTURE MULTIDISCIPLINARY
WOS研究方向	Agriculture
CSCD记录号	CSCD:6617445
来源机构	内蒙古农业大学
资源类型	期刊论文
条目标识符	http://119.78.100.177/qdio/handle/2XILL650/315795
作者单位	内蒙古农业大学机电工程学院, 呼和浩特, 内蒙古 010018, 中国
推荐引用方式 GB/T 7714	刘浩,杜建民,皮伟强,等. 基于A-Clenet5的荒漠化草原草种识别与分类[J]. 内蒙古农业大学,2019,30(10):1056-1061.
APA	刘浩,杜建民,皮伟强,朱相兵,张锡鹏,&康拥朝.(2019).基于A-Clenet5的荒漠化草原草种识别与分类.光电子·激光,30(10),1056-1061.
MLA	刘浩,et al."基于A-Clenet5的荒漠化草原草种识别与分类".光电子·激光 30.10(2019):1056-1061.