结合Transformer和多层特征聚合的高光谱图像分类算法  

Hyperspectral Image Classification Method Combining Transformer and Multi-Layer Feature Aggregation

作　　者：陈栋 李明[1] 陈淑文 CHEN Dong;LI Ming;CHEN Shuwen(College of Computer and Information Science,Chongqing Normal University,Chongqing 401331,China;College of Mathematical Sciences,Chongqing Normal University,Chongqing 401331,China)

机构地区：[1]重庆师范大学,计算机与信息科学学院,重庆401331 [2]重庆师范大学,数学科学学院,重庆401331

出　　处：《数据与计算发展前沿》2023年第3期138-151,共14页Frontiers of Data & Computing

基　　金：国家自然科学基金(61877051,61170192);重庆市科委重点项目(cstc2017zdcy-zdyf0366);重庆市教委项目(113143);重庆市研究生教改重点项目(yjg182022)。

摘　　要：[应用背景]近年来,卷积神经网络被广泛应用在高光谱图像分类任务中,并获得了优异的分类性能。然而,卷积神经网络依然存在着较多的局限性。例如,卷积接受域较小、降采样操作会带来空间信息丢失等。[目的]为了解决上述问题,本文提出了一种结合Transformer和多层特征聚合的高光谱图像分类方法(Transformer and Multi-Layer Feature Aggregation Network,TMFANet)。[方法]首先,TMFANet采用二维卷积(2DConv)和三维卷积(3DConv)层提取图像的浅层空间光谱特征。接着,提出了一种基于密集卷积Transformer模块来提取图像的全局特征。然后,提出了一种多层特征聚合模块来提取不同层次的图像特征,最后将提取的抽象特征输送到分类器中进行分类。为了验证TMFANet的有效性,在Indian Pines、Pavia University和Salinas三个公开的数据集上进行了一系列实验。[结果]实验结果表明,本文提出TMFANet的分类性能比其他先进的方法具有更为优异的性能和泛化能力。[Background]In recent years,the convolutional neural network has been widely used in hyperspectral image classification tasks and achieved excellent classification performance.However,the convolutional neural network still has many limitations.For example,the convolution has a small receiving area and the downsampling operation will cause the loss of spatial information.[Objective]To solve the above problems,this paper proposes a hyperspectral image classification method that combines transformer and multi-layer feature aggregation(TMFANet).[Methods]First,TMFANet uses 3D CNN and 2D CNN layers to extract the shallow spatial-spectral features of the image.Then,a method based on the dense convolution Transformer module is proposed to extract the global features of images.Then,a multi-layer feature aggregation module is proposed to extract image features at different levels.Finally,the extracted abstract features are sent to a classifier for classification.To verify the effectiveness of TMFANet,a series of experiments are carried out on three public data sets,Indian pines,Pavia University and Salinas.[Results]The experimental results show that the classification performance of TMFANet proposed in this paper is better than other current methods.

关 键 词：卷积神经网络 图像处理 高光谱图像分类 TRANSFORMER 特征聚合 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]