一种面向机载高光谱影像的自动数据质量评价网络  

An Automatic Data Quality Assessment Network for Airborne Hyperspectral Imagery

作　　者：余永安 孙承志 陶进 陈强 刘文博 YU Yong′an;SUN Chengzhi;TAO Jin;CHEN Qiang;LIU Wenbo(Piesat Information Technology Co.,Ltd.,Beijing 100195,China;School of Remote Sensing&Geomatics Engineering,Nanjing University of Information Science and Technology,Nanjing 210044,China)

机构地区：[1]航天宏图信息技术股份有限公司,北京100195 [2]南京信息工程大学遥感与测绘工程学院,江苏南京210044

出　　处：《测绘与空间地理信息》2025年第2期66-69,共4页Geomatics & Spatial Information Technology

基　　金：高分专项航空观测系统项目——高分航空载荷自然资源调查应用示范(04-H30G01-9001-20122)资助。

摘　　要：为了提高影像质量评价的精度和效率,本文构建了一套机载高光谱影像的自动质量评价数据集,并基于此数据集提出了一种基于PP-HGNet(PaddlePaddle High Performance GPU Net)的高光谱影像自动质量评价网络,用于机载高光谱遥感的自动质量评价。实验结果表明,与其他常见的网络结构相比,本方法评价取得了较好的效果,样本分类准确率达到91.82%,多类别平均分类精确度达到93.14%,延迟为2.52 ms,为高光谱数据在资源调查、地学研究和环境监测等领域的应用提供质量保障。In order to improve the accuracy and efficiency of image quality assessment,this paper constructs a set of automatic quality assessment dataset of airborne hyperspectral images,and based on this dataset,proposes a kind of automatic quality assessment network for hyperspectral images based on PP-HGNet(PaddlePaddle High Performance GPU Net),which is used for automatic quality assessment of airborne hyperspectral remote sensing images.The experimental results show that,compared with other common network structures,the assessment of this method achieves better results and performance,with a sample classification accuracy of 91.82%,a multi-category average classification accuracy of 93.14%,and a latency of 2.52 ms,which provides quality assurance for the application of hyperspectral data in the fields of resource investigation,geoscientific research,and environmental monitoring,etc.

关 键 词：自动质量评价 机载高光谱 PP-HGNet 高光谱数据集 

分 类 号：P237[天文地球—摄影测量与遥感] TP751[天文地球—测绘科学与技术]