Guided Hyperspectral Image Denoising with Realistic Data

The hyperspectral image (HSI) denoising has been widely utilized to improve HSI qualities. Recently, learning-based HSI denoising methods have shown their effectiveness, but most of them are based on synthetic dataset and lack the generalization capability on real testing HSI. Moreover, there is still no public paired real HSI denoising dataset to learn HSI denoising network and quantitatively evaluate HSI methods. In this paper, we mainly focus on how to produce realistic dataset for learning and evaluating HSI denoising network. On the one hand, we collect a paired real HSI denoising dataset, which consists of short-exposure noisy HSIs and the corresponding long-exposure clean HSIs. On the other hand, we propose an accurate HSI noise model which matches the distribution of real data well and can be employed to synthesize realistic dataset. On the basis of the noise model, we present an approach to calibrate the noise parameters of the given hyperspectral camera. Besides, on the basis of observation of high signal-to-noise ratio of mean image of all spectral bands, we propose a guided HSI denoising network with guided dynamic nonlocal attention, which calculates dynamic nonlocal correlation on the guidance information, i.e., mean image of spectral bands, and adaptively aggregates spatial nonlocal features for all spectral bands. The extensive experimental results show that a network learned with only synthetic data generated by our noise model performs as well as it is learned with paired real data, and our guided HSI denoising network outperforms state-of-the-art methods under both quantitative metrics and visual quality.

     

基于非局部相似联合低秩表示的高光谱图像去噪

高光谱图像(Hyperspectral Images,HSI)在采集过程中常受到多种类型的噪声干扰,会直接影响其在后续应用中的精度,因此HSI的去噪是一项十分重要的预处理过程。低秩表示(Low-Rank Representation,LRR)模型能很好地满足HSI的光谱性质,但该框架下字典的选择尤为重要,在当下仍是一个开放性的问题。同时,典型去噪方法仅考虑了图像的局部相关性,已不能满足去噪要求,非局部相似性在图像中也是不可忽略的。基于LRR,文中提出了一种新的HSI去噪算法。首先,综合考虑噪声的类型,选取具有更全面的噪声判别能力的字典;其次,在对图像分块处理的前提下,通过聚类的方式引入非局部相似信息,将相似的图像块联合起来进行低秩表示。在模拟Indian Pines数据集以及EO-1 Hyperion真实数据集上的实验结果均表明,相较于目前主流的HSI去噪方法,无论是在图像的目视效果还是在模拟数据集的定量评价指标下,所提方法均有显著提升。     

Hyperspectral image classification based on robust discriminative extraction of multiple spectral-spatial features

In hyperspectral image (HSI) processing, the inclusion of both spectral and spatial features, e.g. morphological features, shape features, has shown great success in classification of hyperspectral data. Nevertheless, there exist two main issues to address: (1) The multiple features are often treated equally and thus the complementary information among them is neglected. (2) The features are often degraded by a mixture of various kinds of noise, leading to the classification accuracy decreased. In order to address these issues, a novel robust discriminative multiple features extraction (RDMFE) method for HSI classification is proposed. The proposed RDMFE aims to project the multiple features into a common low-rank subspace, where the specific contributions of different types of features are sufficiently exploited. With low-rank constraint, RDMFE is able to uncover the intrinsic low-dimensional subspace structure of the original data. In order to make the projected features more discriminative, we make the learned representations optimal for classification. With intrinsic information preserving and discrimination capabilities, the learned projection matrix works well in HSI classification tasks. Experimental results on three real hyperspectral datasets confirm the effectiveness of the proposed method.     

基于边信息的高光谱图像恢复模型

在高光谱图像(HSI)恢复中,如何在模型中有效嵌入先验信息和正确建模噪声一直是研究的两个重点.边信息作为一种基于域的先验知识已经在许多方向取得了成功,然而在高光谱去噪领域仍未受到关注.为了将这种领域知识与高光谱恢复模型自然耦合,提出的方法采用双线性映射的方式将边信息链接到表示观测数据潜在低秩结构的底层矩阵,并使用E-3DTV(enhanced 3-D total variation)正则编码了HSI局部平滑先验.此外该方法使用Lp范数进行噪声建模,进一步增强对腐败的鲁棒性.该方法在两个数据集、七种加噪方式下与五种竞争方法在三个数值指标上进行了比较,结果充分反映了提出方法对复杂噪声场景的有效性和鲁棒性.     

Noise reduction and feature extraction based on low-rank representation and pairwise constraint preserving for hyperspectral images

ABSTRACT

Feature extraction (FE) methods play a central role in the classification of hyperspectral images (HSIs). However, all traditional FE methods work in original feature space (OFS), OFS may suffer from noise, outliers and poorly discriminative features. This paper presents a feature space enriching technique to address the problems of noise, outliers and poorly discriminative features which may exist in OFS. The proposed method is based on low-rank representation (LRR) with the capability of pairwise constraint preserving (PCP) termed LRR-PCP. LRR-PCP does not change the dimension of OFS and only can be used as an appropriate preprocessing procedure for any classification algorithm or DR methods. The proposed LRR-PCP aims to enrich the OFS and obtain extracted feature space (EFS) which results in features richer than OFS. The problems of noise and outliers can be decreased using LRR. But, LRR cannot preserve the intrinsic local structure of the original data and only capture the global structure of data. Therefore, two additional penalty terms are added into the objective function of LRR to keep the local discriminative ability and also preserve the data diversity. LRR-PCP method not only can be used in supervised learning but also in unsupervised and semi-supervised learning frameworks. The effectiveness of LRR-PCP is investigated on three HSI data sets using some existing DR methods and as a denoising procedure before the classification task. All experimental results and quantitative analysis demonstrate that applying LRR-PCP on OFS improves the performance of the classification and DR methods in supervised, unsupervised, and semi-supervised conditions.     

高光谱图像空谱特征提取综述

由于高光谱图像包含了丰富的光谱、空间和辐射信息，且具有光谱接近连续、图谱合一的特性，可用于地质勘探、精细农业、生态环境、城市遥感以及军事目标检测等领域的目标精准分类与识别。对高光谱图像进行空谱特征提取是遥感领域的研究热点和前沿课题之一。传统空谱特征提取方法对高光谱图像分类的计算量和样本需求小、理论可解释性好、抗噪声能力强，但应用于分类的精度受限于特征来源；基于深度学习的高光谱图像空谱特征提取方法虽然计算量和样本需求大，但是由于深层空谱特征的表达能力更好，可以大幅度提高分类器的性能。为了便于对高光谱图像空谱特征提取领域进行更深入有效的探索，本文系统综述了相关研究进展。首先，概述了空间纹理与形态学特征提取、空间邻域信息获取及空间信息后处理等传统高光谱空谱特征提取方法的原理，对大量的已有工作进行了梳理、分析与总结。然后，从深度空谱特征提取角度出发，介绍了当前流行的卷积神经网络、图卷积神经网络及跨场景多源数据模型的结构特点及研究进展，分析、评价了基于深度学习的网络模型对高光谱图像空谱特征提取的优势及问题所在。最后，对该研究领域的未来相关发展提出建议并进行了展望。     

结合全局上下文与融合注意力的干涉相位去噪

目的 干涉相位去噪是合成孔径雷达干涉测量（interferometric synthetic aperture radar,InSAR）技术中的关键环节,其效果对测量精度具有重要影响。针对现有的干涉相位去噪方法大多关注局部特征以及在特征提取方面的局限性,同时为了平衡去噪和结构保持两者之间的关系,提出了一种结合全局上下文与融合注意力的相位去噪网络GCFA-PDNet(global context and fused attention phase denoising network)。方法 将干涉相位分离为实部和虚部依次输入到网络,先从噪声相位中提取浅层特征,再将其映射到由全局上下文提取模块和融合注意力模块组成的特征增强模块,最后通过全局残差学习生成去噪图像。全局上下文提取模块能提取全局上下文信息,具有非局部方法的优势;融合注意力模块既强调关键特征,又能高效提取隐藏在复杂背景中的噪声信息。结果 所提出的方法与对比方法中性能最优者相比,在模拟数据结果的平均峰值信噪比（peak signal to noise ratio, PSNR）和结构相似性（structural similarity,...     

Image denoising based on improved bidimensional empirical mode decomposition thresholding technology

In this paper, a novel image denoising methodology based on improved bidimensional empirical mode decomposition and soft interval thresholding technique is proposed. First, a noise compressed image is constructed. Then, the noise compressed image is decomposed by means of bidimensional empirical mode decomposition (BEMD) into a series of intrinsic mode functions (IMFs), which are separated into signal-dominant IMFs and noise-dominant IMFs using a similarity measure based on ?2-norm and a probability density function, and a soft interval thresholding technique is used adaptively to remove the noise inherent in noise-dominant IMFs. Finally, a denoised image is reconstructed by combining the signal-dominant IMFs and the denoised noise-dominant IMFs. The performance of the proposed denoising method is evaluated by using multiple images with different types of noise, and results from the proposed method are compared with those of other conventional methods in various noisy environments. Simulation results demonstrate that the proposed denoising method outperforms other denoising methods in terms of peak signal-to-noise ratio, mean square error and energy of the first IMF.

     

Tripartite-structure transformer for hyperspectral image classification

Hyperspectral images contain rich spatial and spectral information, which provides a strong basis for distinguishing different land-cover objects. Therefore, hyperspectral image (HSI) classification has been a hot research topic. With the advent of deep learning, convolutional neural networks (CNNs) have become a popular method for hyperspectral image classification. However, convolutional neural network (CNN) has strong local feature extraction ability but cannot deal with long-distance dependence well. Vision Transformer (ViT) is a recent development that can address this limitation, but it is not effective in extracting local features and has low computational efficiency. To overcome these drawbacks, we propose a hybrid classification network that combines the strengths of both CNN and ViT, names Spatial-Spectral Former(SSF). The shallow layer employs 3D convolution to extract local features and reduce data dimensions. The deep layer employs a spectral-spatial transformer module for global feature extraction and information enhancement in spectral and spatial dimensions. Our proposed model achieves promising results on widely used public HSI datasets compared to other deep learning methods, including CNN, ViT, and hybrid models.     

A unified model of noise estimation,band rejection,and de-noising for hyperspectral images

As hyperspectral images (HSIs) often suffer from various kinds of degradation, HSI de-noising becomes a challenging task, which can improve not only the visual appearance but also the performance of subsequent applications. Since the noise level in each band is commonly not uniform, many methods of de-noising all bands equally may fail in some cases. Therefore, this study proposes a bilayer model for HSI noise estimation, band rejection, and de-noising, which is in a unified framework of low-rank representation (LRR). Based on channel-dependent noise estimation, the first layer is used to make the noise level in each band more uniform and perform an efficient band rejection. Then in the second layer, a further de-noising that can deal with the mixed noise in HSIs is performed. Both simulated and real data sets are used for experiments, which demonstrate that the proposed model can achieve better results than several other state-of-the-art methods in HSI de-noising.     

Hybrid Multi-Strategy Aquila Optimization with Deep Learning Driven Crop Type Classification on Hyperspectral Images

Hyperspectral imaging instruments could capture detailed spatial information and rich spectral signs of observed scenes. Much spatial information and spectral signatures of hyperspectral images (HSIs) present greater potential for detecting and classifying fine crops. The accurate classification of crop kinds utilizing hyperspectral remote sensing imaging (RSI) has become an indispensable application in the agricultural domain. It is significant for the prediction and growth monitoring of crop yields. Amongst the deep learning (DL) techniques, Convolution Neural Network (CNN) was the best method for classifying HSI for their incredible local contextual modeling ability, enabling spectral and spatial feature extraction. This article designs a Hybrid Multi-Strategy Aquila Optimization with a Deep Learning-Driven Crop Type Classification (HMAODL-CTC) algorithm on HSI. The proposed HMAODL-CTC model mainly intends to categorize different types of crops on HSI. To accomplish this, the presented HMAODL-CTC model initially carries out image preprocessing to improve image quality. In addition, the presented HMAODL-CTC model develops dilated convolutional neural network (CNN) for feature extraction. For hyperparameter tuning of the dilated CNN model, the HMAO algorithm is utilized. Eventually, the presented HMAODL-CTC model uses an extreme learning machine (ELM) model for crop type classification. A comprehensive set of simulations were performed to illustrate the enhanced performance of the presented HMAODL-CTC algorithm. Extensive comparison studies reported the improved performance of the presented HMAODL-CTC algorithm over other compared methods.     

Densely connected deep random forest for hyperspectral imagery classification

In very recent years, deep learning based methods have been widely introduced for the classification of hyperspectral images (HSI). However, these deep models need lots of training samples to tune abundant parameters which induce a heavy computation burden. Therefore, most of these algorithms need to be accelerated with high-performance graphics processing units (GPU). In this paper, a new deep model–densely connected deep random forest (DCDRF) is proposed to classify the hyperspectral images. This model is composed of multiple forward connected random forests. The DCDRF has following merits: 1) It obtains satisfactory classification accuracy with a small number of training samples, 2) It can be run efficiently on the central processing unit (CPU), 3) Only a few parameters are involved during the training. Experimental results based on three hyperspectral images demonstrate that the proposed method can achieve better classification performance than the conventional deep learning based methods.     

Hyperspectral data classification based on flexible momentum deep convolution neural network

Classification is a hot topic in hyperspectral remote sensing community. In the last decades, numerous effort has been concentrate on the classification problem. However, most of the methods accuracy is not high enough due to the fact that they do not extract features in a deep manner. In this paper, a new hyperspectral data classification skeleton based on exponential flexible momentum deep convolution neural network (EFM-CNN) is proposed. First, the fitness of convolution neural network is substantiated by following classical spectral information-based classification. Then, a novel deep architecture is proposed, which is a hybrid of principle component analysis (PCA), improved convolution neural network based on exponential flexible momentum and support vector machine (SVM). Experimental results indicate that the classifier can effectively improve the accuracy with the state-of-the-art algorithms. And compared with homologous parameters momentum updating methods such as adaptive momentum method, standard momentum gradient method and elastic momentum method, on LeNet5 net and multiple neural network, the accuracy obtained of proposed algorithm increases by 2.6% and 6.5% on average respectively.

     

结合PCA及字典学习的高光谱图像自适应去噪方法

高光谱图像各波段图像噪声分布复杂,传统去噪方法难以达到理想效果。针对这一问题,在主成分分析（PCA）的基础上,结合噪声估计和字典学习,提出一种新的高光谱去噪方法。首先,对原始高光谱数据进行主成分变换得到一组主成分图像并根据能量比重将其划分为清晰图像组和含噪图像组;然后,根据任一波段图像的信息,利用奇异值分解（SVD）对图像进行噪声估计,再将得到的噪声估计方法与K-SVD字典学习去噪算法结合,提出一种具备自适应噪声估计特性的字典学习去噪算法,并将其应用于信息量较小的含噪图像组进行去噪处理;最后,按各主成分图像对应的信息量比例进行加权融合得到最终的去噪图像。通过对模拟与实际高光谱遥感图像的实验表明,与PCA、PCA-Bish、PCA-Contourlet三种去噪方法相比,所提方法去噪后图像的峰值信噪比（PSNR）可以提升1~3 dB,且具有更多的细节信息和更好的视觉效果。     

高光谱图像小样本分类的卷积神经网络方法

目的 与传统分类方法相比,基于深度学习的高光谱图像分类方法能够提取出高光谱图像更深层次的特征。针对现有深度学习的分类方法网络结构简单、特征提取不够充分的问题,提出一种堆叠像元空间变换信息的数据扩充方法,用于解决训练样本不足的问题,并提出一种基于不同尺度的双通道3维卷积神经网络的高光谱图像分类模型,来提取高光谱图像的本质空谱特征。方法 通过对高光谱图像的每一像元及其邻域像元进行旋转、行列变换等操作,丰富中心像元的潜在空间信息,达到数据集扩充的作用。将扩充之后的像素块输入到不同尺度的双通道3维卷积神经网络学习训练集的深层特征,实现更高精度的分类。结果 5次重复实验后取平均的结果表明,在随机选取了10%训练样本并通过8倍数据扩充的情况下,Indian Pines数据集实现了98.34%的总体分类精度,Pavia University数据集总体分类精度达到99.63%,同时对比了不同算法的运行时间,在保证分类精度的前提下,本文算法的运行时间短于对比算法,保证了分类模型的稳定性、高效性。结论 本文提出的基于双通道卷积神经网络的高光谱图像分类模型,既解决了训练样本不足的问题,又综合了高光谱图像的光谱特征和空间特征,提高了高光谱图像的分类精度。     

高光谱图像超分辨率重建技术研究进展

不同于传统图像（如灰度图像、RGB图像等）专注于保存目标场景的空间信息,高光谱图像蕴含丰富的空—谱信息,不仅可以保存目标的空间信息,还可以保存具有高可辨性的光谱信息。因此高光谱图像广泛应用于多种计算机视觉和遥感图像任务中,如目标检测、场景分类和目标追踪等。然而,在高光谱图像获取以及重建过程中仍然存在许多问题与瓶颈。如传统高光谱成像仪器在成像过程中通常会引入噪声,且获得的图像往往具有较低的空间分辨率,极大地影响了高光谱图像的质量,对后续数据分析任务造成了极大的困难。近年来,高光谱图像超分辨率重建技术研究得到了极大的发展,现有超分辨率重建方法可以大致分为两类,一类为空间超分辨率重建方法,可以通过直接提升高光谱图像的空间分辨率来获得高质量高光谱图像;另一类为光谱超分辨率重建方法,可以通过提升高空间分辨率图像的光谱分辨率来生成高质量高光谱图像。本文从高光谱图像超分辨率重建领域的新设计、新方法和应用场景出发,通过综合国内外前沿文献来梳理该领域的主要发展,重点论述高光谱图像超分辨率重建领域的发展现状、前沿动态、热点问题及趋势。     

基于自监督约束的双尺度真实图像盲去噪算法

现存的图像去噪算法在处理加性高斯白噪声上已经取得令人满意的效果,然而其在未知噪声强度的真实噪声图像上泛化性能较差.鉴于深度卷积神经网络极大地促进了图像盲去噪技术的发展,针对真实噪声图像提出一种基于自监督约束的双尺度真实图像盲去噪算法.首先,所提算法借助小尺度网络分支得到的初步去噪结果为大尺度分支的图像去噪提供额外的有用信息,以帮助后者实现良好的去噪效果.其次,用于去噪的网络模型由噪声估计子网络和图像非盲去噪子网络构成,其中噪声估计子网络用于预测输入图像的噪声强度,非盲去噪子网络则在所预测的噪声强度指导下进行图像去噪.鉴于真实噪声图像通常缺少对应的清晰图像作为标签,提出了一种基于全变分先验的边缘保持自监督约束和一个基于图像背景一致性的背景自监督约束,前者可通过调节平滑参数将网络泛化到不同的真实噪声数据集上并取得良好的无监督去噪效果,后者则可借助多尺度高斯模糊图像之间的差异信息辅助双尺度网络完成去噪.此外,还提出一种新颖的结构相似性注意力机制,用于引导网络关注图像中微小的结构细节,以便复原出纹理细节更加清晰的真实去噪图像.相关实验结果表明在SIDD,DND和Nam这3个真实基准数据集上,所提的基于自监督的双尺度盲去噪算法无论在视觉效果上还是在量化指标上均优于多种有监督图像去噪方法,且泛化性能也得到了较为明显的提升.     

Bi-directional LSTM with multi-scale dense attention mechanism for hyperspectral image classification

Feature representation has always been the top priority of research in the field of hyperspectral image (HSI) classification. Efficient analysis of those features extracted from HSI massively depends on the way how features are represented. In this paper, we propose a bi-directional long short-term memory network (Bi-LSTM)-based multi-scale dense attention framework, namely MBDA-Net. In this framework, we develop a new multi-scale dense attention module (MCDA) that uses different sizes of convolution kernels to obtain multi-scale features. Then, we perform feature selection by using a multi-layer attention mechanism that assigns different weight coefficients to the extracted multi-scale features. Specifically, we use the bi-directional LSTM to obtain contextual semantic information. The extensive experiments conducted on three hyperspectral datasets demonstrate the effectiveness of our method in identifying hyperspectral images.

     

结合孪生网络和像素配对的高光谱图像异常检测

目的 高光谱遥感中,通常利用像素的光谱特征来区分背景地物和异常目标,即通过二者之间的光谱差异来寻找图像中的异常像元。但传统的异常检测算法并未有效挖掘光谱的深层特征,高光谱图像中丰富的光谱信息没有被充分利用。针对这一问题,本文提出结合孪生神经网络和像素配对策略的高光谱图像异常检测方法,利用深度学习技术提取高光谱图像的深层非线性特征,提高异常检测精度。方法 采用像素配对的思想构建训练样本,与原始数据集相比,配对得到的新数据集数量呈指数增长,从而满足深度网络对数据集数量的需求。搭建含有特征提取模块和特征处理模块的孪生网络模型,其中,特征处理模块中的卷积层可以专注于提取像素对之间的差异特征,随后利用新的训练像素对数据集进行训练,并将训练好的分类模型固定参数,迁移至检测过程。用滑动双窗口策略对测试集进行配对处理,将测试像素对数据集送入网络模型,得到每个像素相较于周围背景像素的差异性分数,从而识别测试场景中的异常地物。结果 在异常检测的实验结果中,本文提出的孪生网络模型在San Diego数据集的两幅场景和ABU-Airport数据集的一幅场景上,得到的AUC （area under the curve）值分别为0.993 51、0.981 21和0.984 38,在3个测试集上的表现较传统方法和基于卷积神经网络的异常检测算法具有明显优势。结论 本文方法可以提取输入像素对的深层光谱特征,并根据其特征的差异性,让网络学习到二者的区分度,从而更好地赋予待测像素相对于周围背景的异常分数。本文方法相对于卷积神经网络的异常检测方法可以有效地降低虚警,与传统方法相比能够更加明显地突出异常目标,提高了检测率,同时也具有较强的鲁棒性。