结合倒置特征金字塔和U-Net的高光谱图像分类  

Hyperspectral image classification using an inverted feature pyramid network with U-Net

作　　者：程嵩阳 熊玉洁 姚瑶[1] 李庆利 Cheng Songyang;Xiong Yujie;Yao Yao;Li Qingli(School of Electronic and Electrical Engineering,Shanghai University of Engineering Science,Shanghai 201620,China;Shanghai Key Laboratory of Multidimensional Information Processing,East China Normal University,Shanghai 200241,China)

机构地区：[1]上海工程技术大学电子电气工程学院,上海201620 [2]华东师范大学上海市多维度信息处理重点实验室,上海200241

出　　处：《中国图象图形学报》2021年第8期1994-2008,共15页Journal of Image and Graphics

基　　金：国家重点研发计划项目(2019YFC1509202);国家自然科学基金项目(62006150);上海青年科技英才扬帆计划项目(19YF1418400);上海市多维度信息处理重点实验室开放课题基金资助项目(2020MIP001);中央高校基本科研业务费专项资金资助。

摘　　要：目的地物分类是对地观测研究领域的重要任务。高光谱图像具有丰富的地物光谱信息,可用于提升遥感图像地物分类的准确度。如何对高光谱图像进行有效的特征提取与表示是高光谱图像分类应用的关键问题。为此,本文提出了一种结合倒置特征金字塔和U-Net的高光谱图像分类方法。方法对高光谱数据进行主成分分析(principal component analysis,PCA)降维,获取作为网络输入的重构图像数据,然后使用U-Net逐层提取高光谱重构图像的空间特征。与此同时,利用倒置的特征金字塔网络抽取相应层级的语义特征;通过特征融合,得到既有丰富的空间信息又有较强烈的语义响应的特征表示。提出的网络利用注意力机制在跳跃连接过程中实现对背景区域的特征响应抑制,最终实现了较高的地物分类精度。结果分析了PCA降维方法和输入数据尺寸对分类性能的影响,并在Indian Pines、Pavia University、Salinas和Urban数据集上进行了对比实验。本文方法在4个数据集上分别取得了98.91%、99.85%、99.99%和87.43%的总体分类精度,与支持向量机(support vector machine,SVM)等相关算法相比,分类精度高出1%~15%。结论本文提出一种结合倒置特征金字塔和U-Net的高光谱图像分类方法,可以应用于有限训练样本下的高光谱图像分类任务,并在多个数据集上取得了较高的分类精度。实验结果表明倒置特征金字塔结构与U-Net结合的算法能够高效地实现高光谱图像的特征提取与表示,从而获得更精细的分类结果。Objective Terrain classification is an important research task in the field of earth observation using remote sensing technology.The hyperspectral image has rich spectral information;thus,it can be applied to the classification of remote sensing image.With the rapid development of the hyperspectral technology,the hyperspectral remote sensing image processing and analyzing technology has attracted wide attention of academia.The hyperspectral images have dozens or even hundreds of continuous narrow spectral bands compared with the traditional panchromatic band and multi-spectral remote sensing image,which provides detailed spectral and spatial feature information.Accordingly,these images have been widely used in various aspects,such as precision agriculture,city planning,and military defense.Hyperspectral images have high dimensional data,and redundancy and noise exist;thus,transformed data must be utilized for image processing.In the application of hyperspectral image classification,the manner by which to effectively represent the features of hyperspectral image is the most critical step in current studies.In this work,we propose an approach for hyperspectral image classification by using an inverted feature pyramid network and U-Net.Method The dimension of the hyperspectral remote sensing image data is high.Principal component analysis(PCA)method plays a significant role in transforming useful information in the images to the most important k characteristic,thus reducing the amount of data and enhancing the data features.After PCA,the data are segmented and collected by means of sliding window.The surrounding area of each pixel is defined as a patch,which is regarded as the input of the proposed network.The category of the pixel is the ground truth label.In the first stage,U-Net is used to extract spatial features of hyperspectral image at the pixel level.The left side of the network is the contraction path,which corresponds to the encoder part of the classic encoder-decoder.The right side of the network is the e

关 键 词：高光谱图像分类 稀少样本 倒置特征金字塔网络(IFPN) U-Net 特征融合 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]