遥感图像的NSCT自适应阈值去噪方法

提出了一种基于非下采样Contourlet变换(NSCT)相结合的遥感图像自适应阈值去噪方法。通过小波估计被噪声污染遥感图像的噪声强弱,根据噪声的强弱以及NSCT的分解特点及系数所在邻域的特性,给出不同尺度不同方向的自适应阈值。仿真实验结果表明,与小波硬阈值、Contourlet硬阈值和基于非下采样Contourlet硬阈值去噪方法比较,该方法在提高了图像的峰值信噪比的同时也减少了Gibbs现象,图像视觉效果也得到明显的改善。     

基于非下采样Contourlet变换和双变量模型的图像去噪

在研究了非下采样Contourlet变换(NSCT)和贝叶斯最大后验估计理论框架下的双变量模型的基础上,该文将二者结合起来,提出了一种新的图像去噪算法。算法在利用变换平移不变性和多方向选择性优点的同时,充分挖掘了图像NSCT系数尺度内和尺度间的双重相关性,并详细阐述了噪声估计方法。仿真结果和分析表明,与当前一些典型的去噪算法相比,该文算法的客观评价指标PSNR和去噪后图像的主观视觉效果都有明显的提高和改善,有效地保持了原图像中的细节和纹理信息。     

基于NSCT域各向异性双变量萎缩图像去噪

提出了一种用各向异性双变量拉普拉斯函数模型去模拟NSCT域的系数的图像去噪算法,这种各向异性双边拉普拉斯模型不仅考虑了NSCT系数相邻尺度间的父子关系,同时满足自然图像不同尺度间NSCT系数方差具有各向异性的特征,基于这种统计模型,文中先推导出了一种各向异性双变量收缩函数的近似形式,然后基于贝叶斯去噪法和局部方差估计将这种新的阈值收缩函数应用于NSCT域,实验结果表明文中提出的方法同小波域BiShrink算法、小波域ProbShrink算法、小波域NeighShrink算法相比,能够有效地去除图像的高斯噪声,提高了图像的峰值信噪比;并较完整地保持了图像的纹理和边缘等细节信息,从而明显改善了图像的视觉效果。     

Speckle noise removal based on structural convolutional neural networks with feature fusion for medical image

Application of convolutional neural networks (CNNs) for image additive white Gaussian noise (AWGN) removal has attracted considerable attentions with the rapid development of deep learning in recent years. However, the work of image multiplicative speckle noise removal is rarely done. Moreover, most of the existing speckle noise removal algorithms are based on traditional methods with human priori knowledge, which means that the parameters of the algorithms need to be set manually. Nowadays, deep learning methods show clear advantages on image feature extraction. Multiplicative speckle noise is very common in real life images, especially in medical images. In this paper, a novel neural network structure is proposed to recover noisy images with speckle noise. Our proposed method mainly consists of three subnetworks. One network is rough clean image estimate subnetwork. Another is subnetwork of noise estimation. The last one is an information fusion network based on U-Net and several convolutional layers. Different from the existing speckle denoising model based on the statistics of images, the proposed network model can handle speckle denoising of different noise levels with an end-to-end trainable model. Extensive experimental results on several test datasets clearly demonstrate the superior performance of our proposed network over state-of-the-arts in terms of quantitative metrics and visual quality.     

Feature-preserving MRI denoising: a nonparametric empirical Bayes approach

This paper presents a novel method for Bayesian denoising of magnetic resonance (MR) images that bootstraps itself by inferring the prior, i.e., the uncorrupted-image statistics, from the corrupted input data and the knowledge of the Rician noise model. The proposed method relies on principles from empirical Bayes (EB) estimation. It models the prior in a nonparametric Markov random field (MRF) framework and estimates this prior by optimizing an information-theoretic metric using the expectation-maximization algorithm. The generality and power of nonparametric modeling, coupled with the EB approach for prior estimation, avoids imposing ill-fitting prior models for denoising. The results demonstrate that, unlike typical denoising methods, the proposed method preserves most of the important features in brain MR images. Furthermore, this paper presents a novel Bayesian-inference algorithm on MRFs, namely iterated conditional entropy reduction (ICER). This paper also extends the application of the proposed method for denoising diffusion-weighted MR images. Validation results and quantitative comparisons with the state of the art in MR-image denoising clearly depict the advantages of the proposed method.     

Using Deep learning for image watermarking attack

Digital image watermarking has justified its suitability for copyright protection and copy control of digital images. In the past years, various watermarking schemes were proposed to enhance the fidelity and the robustness of watermarked images against different types of attacks such as additive noise, filtering, and geometric attacks. It is highly important to guarantee a sufficient level of robustness of watermarked images against such type of attacks. Recently, Deep learning and neural networks achieved noticeable development and improvement, especially in image processing, segmentation, and classification. Therefore, in this paper, we studied the effect of a Fully Convolutional Neural Network (FCNN), as a denoising attack, on watermarked images. This deep architecture improves the training process and denoising performance, through which the encoder–decoder remove the noise while preserving the detailed structure of the image. FCNNDA outperforms the other types of attacks because it destroys the watermarks while preserving a good quality of the attacked images. Spread Transform Dither Modulation (STDM) and Spread Spectrum (SS) are used as watermarking schemes to embed the watermarks in the images using several scenarios. This evaluation shows that such type of denoising attack preserves the image quality while breaking the robustness of all evaluated watermarked schemes. It could also be considered a deleterious attack.     

一种新的结合非下采样Contourlet与自适应全变差的图像去噪方法

该文提出了一种新的结合非下采样Contourlet变换(NSCT)和自适应全变差模型的图像去噪方法。首先通过NSCT对含噪图像进行分解,根据高斯比例混合(GSM)模型建立图像模型;然后利用贝叶斯估计进行图像去噪,重构后得到初次去噪图像;最后,结合自适应全变差模型对初次去噪图像进行重构滤波,得到最终的去噪图像。实验结果表明,该方法可以有效地消除图像中的Gibbs伪影及噪声,在去噪图像峰值信噪比(PSNR)和边缘保持性能上都优于已有的算法。     

基于域自适应的红外图像去噪算法

在红外图像去噪任务中,由于真实的红外噪声图像难以大量获取,而使深度学习算法高度依赖于人工合成噪声,无法很好地去除真实的红外噪声。本文提出一种基于域自适应的红外图像去噪算法,包括一个图像转换模块和两个图像去噪模块。首先利用图像转换模块将合成红外噪声图像和真实红外噪声图像相互转换,然后将转换后的图像和原图像作为去噪模块的训练数据,采用一致性损失函数使两个图像去噪模块产生一致的结果,最后将训练后的去噪网络框架用于红外图像去噪任务。实验表明,本文提出的算法与BM3D、DnCNN和ADNet算法相比在合成红外噪声数据集上有更高的指标数值和更好的视觉效果,在真实红外噪声数据集上有同样优秀的去噪效果。证明了该算法具有良好的泛化能力,能够在真实噪声下恢复清晰的红外图像。     

From synthetic to natural — single natural image dehazing deep networks using synthetic dataset domain randomization

Image dehazing methods aim to solve the problem of poor visibility in images due to haze. Techniques proposed for image dehazing in literature focus on image priors, haze lines or data driven statistical models. Variations of the classical methods relying on prior model or haze line model use no-reference image quality metrics to prove their dehazing performance. Recently developed deep learning models rely on huge amounts of hazy, haze-free pairs for training, and uses PSNR and SSIM like image reconstruction metrics to show their performance. These methods perform poorly on no-reference image quality assessments and also dehazes poorly at the depths of the image. These methods though can be optimized for memory usage and are faster. This work presents a deep learning model (Feature Fusion Attention Network) trained on a domain randomized synthetic dataset generated in simulation. The proposed model achieves the highest scores on blind image assessments through the gradient rationing technique for a deep learning-based approach by a significant margin. The images were evaluated on full-reference metrics as well and obtained favorable results. This approach also yields one of the highest edge sharpness obtained after dehazing. The training procedure adopted to obtain significant gains on real-world dehazing, without using any real-world data is also detailed in this paper.     

基于NSCT域的自适应阈值遥感图像去噪方法

提出一种新的基于非下采样Contourlet变换(NSCT)的图像自适应阈值去噪方法.根据NSCT系数所在领域区域的特性,给出不同尺度不同方向的自适应咧值.仿真实验结果表明,通过本文提出的方法能够有效去除遥感图像的高斯噪声,较完整的保持图像的边缘等细节信息.与小波硬阈值、Contourlet硬阈值和基于非下采样Cont...     

基于非下采样Contourlet域局部高斯模型和MAP的SAR图像相干斑抑制

 对SAR图像应用非对数加性模型,通过研究SAR图像中同质区域在非下采样Contourlet域的分布特性,提出对非下采样Contourlet域中、与同质区域相对应的平稳区域的非对数加性噪声使用高斯分布建模。基于该模型,利用局部滑动窗口和对非对数加性噪声方差的自适应估计,在最大后验准则的基础之上求得真实信号的非下采样Contourlet系数。由于未对图像进行对数变换,本文算法很好地保持了原始图像的辐射特性,相干斑被有效地抑制,均匀区域很少有伪吉布斯效应,同时边缘纹理清晰。无论视觉效果还是客观评价指标本文算法都优于许多现存的抑斑算法。     

Cascaded and Recursive ConvNets (CRCNN): An effective and flexible approach for image denoising

Recently, discriminative learning methods have gained substantial interest in solving inverse imaging problems due to their decent performance and fast inferencing capability. Those methods need separate models for specific noise levels, which in turn require multiple models to be trained to denoise an image. However, images exhibit spatial variant noise which limits the applicability of such methods. In addition, the discriminative learning methods introduce artifacts such as blurring, deblocking, and so forth while denoising an image. To address these issues, we propose a cascaded and recursive convolutional neural network (CRCNN) framework which can cope with spatial variant noise and blur artifacts in a single denoising framework. The CRCNN takes into account down-sampled sub-images for fast inferencing along with the noise level map. We adopt the hybrid orthogonal projection and estimation method on the convolutional layers to improve the generalization capability of the network in terms of non-uniform and spatial variant noise levels. In contrast to the existing methods, the CRCNN framework allows both denoising and deblurring of images using a single framework which preserves the fine details in a denoised image. Extensive experiments have been conducted to validate the effectiveness and flexibility of the CRCNN framework on real as well as synthetic noisy images in comparison to the state-of-the-art denoising methods. The results show that the CRCNN performs effectively on both synthetic as well as spatial variant noise-induced images, thus, proving the practicability of the framework.     

Dual-branch deep image prior for image denoising

In this work, we propose a two-stage denoising approach, which includes generation and fusion stages. Specifically, in the generation stage, we first split the expanding path of the UNet backbone of the standard DIP (deep image prior) network into two branches, converting it into a Y-shaped network (YNet). Then we adopt the initial denoised images obtained with DAGL (dynamic attentive graph learning) and Restormer methods together with the given noisy image as the target images. Finally, we utilize the standard DIP on-line training routine to generate two complementary basic images, whose image quality is quite improved, with the help of a novel automatic iteration termination mechanism. In the fusion stage, we first split the contracting path of the standard UNet network into two branches for receiving the two basic images generated in the previous stage, and obtain a fused image as the final denoised image in a fully unsupervised manner. Extensive experimental results confirm that our method has a significant improvement over the standard DIP or other unsupervised methods, and outperforms recently proposed supervised denoising models. The noticeable performance improvement is attributed to the proposed hybrid strategy, i.e., we first adopt the supervised denoising methods to process the common content of images substantially, then utilize the unsupervised method to fine-tune the specific details. In other words, we take full advantage of the high performance of the supervised methods and the flexibility of the unsupervised methods.     

Deep Learning for OFDM Channel Estimation in Impulsive Noise Environments

Impulsive noise suppression is essential in orthogonal frequency division multiplexing (OFDM) systems, since impulsive noise may cause a serious decline in channel estimation performance. To solve this problem, a channel estimator based on denoising autoencoder-deep neural network (DAE-DNN) is proposed in this paper. The proposed method is based on a data-driven deep learning framework. Firstly, DAE preprocesses signals to learn damaged data and recover the complete signal are used in the presence of impulsive noise. Then, the transmitted data processed by DAE are used to train the DNN in the offline training process. Finally, the estimated channel state information (CSI) is offered by the proposed DNN model in the online working process. The simulation results demonstrate that the proposed method improves OFDM channel estimation performance significantly. As expected, the proposed method has a better performance than existing ones, such as least squares, minimum mean square error and orthogonal matching pursuit algorithms. Moreover, the proposed method is robust under impulsive noise environments.

     

遥感图像自适应去噪方法研究

遥感图像的获取、传输过程中很容易受到噪声的污染。在研究形态成分分析（MCA）稀疏分解和遥感图像修复方法的基础上,提出了基于MCA稀疏分解的自适应去噪方法和基于图像修复的去噪方法。通过对比其他经典去噪模型,发现前者适合自适应有效去除高斯白噪声,后者对灰度或彩色遥感图像的椒盐噪声能自适应有效去除,且能够同时去除“胡椒”噪声和“盐”噪声,无论是主观视觉效果还是客观量化评价效果都要优于常见模型。     

基于非亚采样Contourlet和SWT的多光谱图像和全色图像的融合算法

该文研究了多尺度几何分析工具非亚采样Contourlet变换(NSCT),提出一种新的全色图像和多光谱图像融合的方法。该方法首先对全色图像和进行过IHS变换的多光谱图像的亮度分量进行NSCT变换,对于二者的低频近似系数再进行平稳小波变换(SWT)并融合,进一步提高融合图像的空间信息量,对于高频细节系数,采用基于局部平均梯度的方法进行融合,经过逆NSCT得到融合图像。实验结果表明,该文提出的方法在保留多光谱图像的光谱信息的同时,增强了融合图像的空间细节表现能力,提高了信息量,并且优于传统的基于IHS变换、小波变换、双树复小波变换及Contourlet变换的融合方法,该方法是有效可行的。     

Robust Deep Age Estimation Method Using Artificially Generated Image Set

Human age estimation is one of the key factors in the field of Human–Robot Interaction/Human–Computer Interaction (HRI/HCI). Owing to the development of deep‐learning technologies, age recognition has recently been attempted. In general, however, deep learning techniques require a large‐scale database, and for age learning with variations, a conventional database is insufficient. For this reason, we propose an age estimation method using artificially generated data. Image data are artificially generated through 3D information, thus solving the problem of shortage of training data, and helping with the training of the deep‐learning technique. Augmentation using 3D has advantages over 2D because it creates new images with more information. We use a deep architecture as a pre‐trained model, and improve the estimation capacity using artificially augmented training images. The deep architecture can outperform traditional estimation methods, and the improved method showed increased reliability. We have achieved state‐of‐the‐art performance using the proposed method in the Morph‐II dataset and have proven that the proposed method can be used effectively using the Adience dataset.     

基于主成分分析与深度神经网络的快速噪声水平估计算法

鉴于从噪声图像分解获得的原生图块集合的协方差矩阵前若干个特征值（按照升序排序）与图像噪声水平值具有强相关性,提出了一种基于主成分分析和深度神经网络的快速噪声水平估计算法.该算法首先选用原生图块集合协方差矩阵前若干个特征值构成刻画图像噪声水平高低的特征矢量,然后在大量有代表性且已标定噪声水平值的噪声图像集合上利用深度神经网络训练预测模型以实现将特征矢量直接映射为噪声水平值,最后为获得更高的预测准确性,采用粗精预测模型相结合的两步预测方式实现.实验表明：文中算法在各个噪声级别上都具有稳定的预测准确性,且执行效率非常高,作为降噪算法的前置预处理模块具有更好的综合优势.     

利用奇异谱分析的深度神经网络语音增强方法

针对现有深度神经网络语音增强方法对带噪语音的去噪能力有限、语音质量提升不高的问题，提出了一种基于奇异谱分析的深度神经网络语音增强方法。通过引入奇异谱分析算法对带噪语音进行预处理，以初步分离得到语音信号与噪声。接着将语音信号与噪声用于深度神经网络模型得训练，以得到性能更优的网络模型，从而使得本文方法具有更好的性能。最后在重建干净语音的环节中，同时使用神经网络估计得到的对数功率谱和带噪语音的对数功率谱，并加入了权重系数，使得本文提出的方法可以适应不同信噪比的情形，有效的去除背景噪声，降低语音信号的失真。本文通过仿真实验验证了该方法的有效性和鲁棒性。      

基于图像退化模型的红外与可见光图像融合方法

基于深度学习的红外与可见光图像融合算法依赖人工设计的相似度函数衡量输入与输出的相似度,这种无监督学习方式不能有效利用神经网络提取深层特征的能力,导致融合结果不理想。针对该问题,该文首先提出一种新的红外与可见光图像融合退化模型,把红外和可见光图像视为理想融合图像通过不同退化过程后产生的退化图像。其次,提出模拟图像退化的数据增强方案,采用高清数据集生成大量模拟退化图像供训练网络。最后,基于提出的退化模型设计了简单高效的端到端网络模型及其网络训练框架。实验结果表明,该文所提方法不仅拥有良好视觉效果和性能指标,还能有效地抑制光照、烟雾和噪声等干扰。