SAR image denoising via Bayesian wavelet shrinkage based on heavy-tailed modeling

Synthetic aperture radar (SAR) images are inherently affected by multiplicative speckle noise, which is due to the coherent nature of the scattering phenomenon. This paper proposes a novel Bayesian-based algorithm within the framework of wavelet analysis, which reduces speckle in SAR images while preserving the structural features and textural information of the scene. First, we show that the subband decompositions of logarithmically transformed SAR images are accurately modeled by alpha-stable distributions, a family of heavy-tailed densities. Consequently, we exploit this a priori information by designing a maximum a posteriori (MAP) estimator. We use the alpha-stable model to develop a blind speckle-suppression processor that performs a nonlinear operation on the data and we relate this nonlinearity to the degree of non-Gaussianity of the data. Finally, we compare our proposed method to current state-of-the-art soft thresholding techniques applied on real SAR imagery and we quantify the achieved performance improvement.     

SAR image despeckling using heavy-tailed Burr distribution

Multiplicative speckle noise diminishes the radiometric resolution of the synthetic aperture radar (SAR) images and all the coherent images. Speckle removal adds an extra value to an automated SAR image interpretation and analysis. In this paper, dual-tree complex wavelet-transform-based Bayesian method is proposed for despeckling the SAR images. In each subband, the reflectance and noise of the logarithmically transformed wavelet coefficients are modeled using heavy-tailed Burr and zero-mean Gaussian distributions. The closed-form expression for the shape parameter of Burr distribution is derived by employing the Mellin transform. The resultant complex-free quadratic maximum a posteriori solution with suitable shrinkage function yields despeckled SAR images. Extensive experiments are carried out using real SAR images as well as simulated images. The proposed method performs well in terms of equivalent number of looks with 3.5751 dB improvement in homogeneous region1 of Pipe river SAR image, edge preservation with 0.6158 improvement, peak signal to noise ratio of 51.3305 dB, and mean structural similarity index measure of 0.9397 at 0.05 noise variance for synthetically speckled image in comparison to the existing methods and takes averagely 2.3461 times less computing time. The proposed method provides a computationally efficient better speckle reduction in homogeneous regions while still preserving the edge.     

SAR图像滤波的迭代方法

对合成孔径雷达图像迭代滤波处理的效果及其问题进行了分析，并结合统计滤波方法和小波分析方法，提出了一种基于细节补偿的MAP（BDC MAP）迭代滤波算法。实验结果表明，新的迭代滤波算法较好地解决了去噪和细节保留之间的矛盾，而且算法稳定，收敛很好。     

SAR image despeckling via bilateral filtering

Bilateral filtering (BF) can realise both smoothing images and preserving edges, whereas its filtering results are always influenced since its two parameters are difficult to configure to the optimum. In this reported work, the application of BF is extended to synthetic aperture radar (SAR) image despeckling, and the despeckling evaluation indexes, including the equivalent number of looks and the edge save index, are used to estimate the parameters. After BF with estimated parameters imposed on a normalised SAR image, further processing can achieve both despeckling and edge preservation simultaneously. Experimental results show that the visual quality and evaluation indexes of the proposed algorithm outperform the classical Lee filtering.     

应用变分法的极化SAR图像分类方法研究

提出了一种用于图像分类的变分模型,该模型结合正则化过程,可以较好地保持图像边缘信息,同时可以用于图像恢复.利用模拟和真实SAR图像的分类仿真试验表明,基于变分法的极化SAR图像分类方法不仅能够实现SAR图像的正确分类,克服SAR图像中相干斑噪声的影响,并且算法快速,易于实现.     

红外图像噪声滤波对比度增强算法

针对红外图像噪声大、对比度低的特点,在传统直方图均衡原理的基础上,利用像素邻域高斯分布特征,介绍了一种基于噪声滤波的对比度增强算法.首先,计算原始图像统计直方图;其次,对统计直方图进行噪声滤波:最后.利用构造的加权函数对直方图间距进行均衡处理.改进的对比度增强算法避免了传统直方图均衡对灰度邻近、分布较少的灰度级进行合并,从而保留了图像细节信息.仿真实验表明:该算法在增强图像对比度的同时,有效保留了局部细节信息,具有一定的实用价值.     

一种基于多重分形的SAR图像边缘检测方法

分形维数只能刻画那些具有理想的自相似性的分形体,现实中的许多纹理并不满足这一条件,因此单一的分形维数并不足以描述和刻画SAR图像的纹理,多重分形维数更适合于描述图像的纹理.通过计算原始SAR图像离散点数据的奇异性指数,然后对应每一点奇异性指数计算全局多重分形奇异谱,根据判决准则区分边缘和纹理可以实现SAR图像的边缘检测,实验结果表明,基于多重分形特征的边缘检测算法能够检测到许多局部细节,同时又避免出现不重要的细节,突出了主要的边缘信息,很好地区分出SAR图像的纹理和边缘.     

结合全变分的双边滤波图像去噪方法研究

图像去噪是数字图像处理过程中的一个重要步骤,它将直接影响到图像处理的最终质量。针对传统的全变分(TV)正则化去噪算法容易产生阶梯效应的缺点,利用双边滤波去噪算法在空间域和值域两个方面进行滤波的特点,提出一种结合TV模型的双边滤波方法,该方法能在一定程度上有效地改善阶梯效应。仿真实验结果表明,提出的去噪方法不仅能够获得较好的去噪效果,还能有效地保持图像的边缘特征信息,降噪效果明显。在较高水平噪声情况下,与TV算法相比,该方法针对小尺寸灰度图片(256×256)图像的峰值信噪比(PSNR)提高1.45 d B左右,大尺寸灰度图片(512×512)图像的PSNR提高2.56 d B左右。     

利用最小冗余最大相关和SVM的SAR图像海上溢油识别

近年溢油事故频发,海洋污染日益严重。利用合成孔径雷达（SAR）卫星可以有效跟踪由海上溢油事故导致的油膜扩展情况。利用最大冗余最小相关-支持向量机（mRMR_SVM）算法进行SAR图像溢油识别,为溢油事故决策支持提供重要前提。首先采用mRMR提取最优特征向量集,对输入值进行降维; 然后采用SVM算法解决油膜图像分类问题,同时选择径向基函数（RBF）为核函数;使用训练集训练该模型,调整模型参数;以测试集特征向量作为输入,利用训练好的模型进行溢油识别。实验结果表明,mRMR_SVM模型对SAR图像的油膜和类油膜识别有效,准确率为96.875%.     

基于一维匹配滤波的SAR目标识别

为克服合成孔径雷达(SAR)目标位置和姿态不确定性对识别性能的影响,提出了一种基于一维匹配滤波的SAR目标识别方法。基于姿态估计对目标图像进行预处理,基于一维匹配滤波定义两样本图像间的相似度或距离,并采用最近邻分类器。在MSTAR公开数据集上进行了实验,已知和未知目标姿态情况下的识别率分别达到98.8%和98.2%。实验结果表明,所提出的方法性能优于支持向量机(SVM)方法。     

Speckle filtering of SAR images based on adaptive windowing

Speckle noise usually occurs in synthetic aperture radar (SAR) images owing to coherent processing of SAR data. The most well-known image domain speckle filters are the adaptive filters using local statistics such as the mean and standard deviation. The local statistics filters adapt the filter coefficients based on data within a fixed running window. In these schemes, depending on the window size, there exists trade-off between the extent of speckle noise suppression and the capability of preserving fine details. The authors propose a new adaptive windowing algorithm for speckle noise suppression which solves the problem of window size associated with the local statistics adaptive filters. In the algorithm, the window size is automatically adjusted depending on regional characteristics to suppress speckle noise as much as possible while preserving fine details. Speckle noise suppression gets stronger in homogeneous regions as the window size increases succeedingly. In fine detail regions, by reducing the window size successively, edges and textures are preserved. The fixed-window filtering schemes and the proposed one are applied to both a simulated SAR image and an ERS-1 SAR image to demonstrate the excellent performance of the proposed adaptive windowing algorithm for speckle noise     

基于聚类算法的SAR图像去噪

遥感图像的用途非常广泛,而合成孔径雷达图像是遥感图像中重要的一种,人们能从中提供更多的有用信息,但其固有的相干斑给人们对其信息的提取带来了困难,去除相干斑成为SAR图像信息提取中最重要的一步。在此介绍了以模糊C均值聚类算法为基础,同时结合小波变换,对SAR图像进行去噪。并将实验结果与已有的SAR图像去噪方法进行实验、比较和分析。结果表明,模糊C均值聚类和小波变换的所结合的方法,在SAR图像去除斑噪的领域中,成为一种有效且吸引人的算法。     

基于形态滤波BP算法的SAR图像识别研究

SAR图像识别是航空图像判读工作的一项重要内容。本文针对SAR图像的特点,提出了一种有实用意义的形态滤波神经网络模型及其自适应BP学习算法。通过实验结果可以看出,该算法不仅能适应复杂多样的背景环境,而且能克服在成像中的各种形变。     

结合高斯滤波和导向滤波的单幅图像去雾算法

针对暗原色去雾后图像偏暗的问题,提出一种基于结合高斯滤波和导向滤波的单幅图像去雾算法。假设图像的最大值作为全局大气光强的初始值,然后使用高斯滤波获取分块图像的局部大气光强,使用最小值滤波获取分块图像的局部透射率。最后使用导向滤波分别对大气光强和透射率再次细化以处理分块后图像的边界跳跃问题。实验测试结果表明,所提算法去雾后的图像明亮度较高,视觉效果自然。     

基于像素滤波和中值滤波的深度图像修复方法

针对Kinect相机采集的深度数据一般有噪声和黑 洞现象,直接应用于场景三维重建系统中效果差的问 题,提出了像素滤波器和中值滤波器相结合的Kinect 深度图像修复方法。首先对一帧图片 上的像素进行搜索, 找到像素值为0的点,以该点为中心,利用其邻域内的像素,定义一个两层的滤波器,根据 像素滤波器的原理 对其进行修复,填充深度图片的黑洞;然后采用中值滤波器,在平滑深度图像的同时保留边 缘信息,去除孤立 噪声点。实验结果表明,本文方法去噪的同时也能对黑洞进行修补,与原始深度图像相比, 空洞明显减少,深度图像质量大大提高。     

Speckle filtering in polarimetric SAR data based on the subspace decomposition

In this paper, a new approach to speckle filtering of synthetic aperture radar (SAR) data is presented. We define a parameter space consisting of two orthogonal subspaces-the signal subspace and the noise subspace. Then, the full polarimetric information from the signal subspace is obtained after speckle filtering. In this way, edges of different kinds of targets are preserved. The effectiveness of this method is demonstrated using the National Aeronautics and Space Administration Jet Propulsion Laboratory airborne L-band polarimetric SAR data.     

基于三边滤波的Retinex图像去雾算法

针对传统Retinex算法采用高斯滤波估计图像的照射分量易产生边缘模糊,不能有效去除脉冲噪声且处理后的图像颜色易失真等问题,提出一种基于三边滤波的Retinex图像去雾算法。该算法利用三边滤波器估计图像的照射分量,三边滤波器继承了双边滤波器既可以有效降低图像加性高斯噪声又可以保持图像边缘细节的特性,同时又解决了双边滤波器与高斯滤波器不能有效滤除脉冲噪声,易产生伪边缘等问题。为验证该算法的有效性,采用5种不同的客观评价参数对处理后的图像进行评价。实验证明,该算法能有效地改善雾天图像的退化现象,提高图像的清晰度。     

基于VGG-16的电商评论图像审核

如今互联网电商平台中用户对于购买商品上传的点评图片质量参差不齐,影响其他用户的购物体验和对于商品质量的判断,电商公司通常通过人工审核来规避这种情况,然而大量的上传图片数据需要大量的人力进行运营审核,针对平台当前审核成本过高的问题,本文设计了一种基于VGG-16卷积神经网络的电商评论图像分类模型,并采用随机梯度下降算法、...     

基于同态滤波的彩色图像光照补偿方法

为了消除光照对彩色图像的影响,提出了一种基于同态滤波的光照补偿方法。在频域内采用同态滤波对图像进行处理,然后将巴特沃斯高通滤波传递函数引入同态滤波器中,设计出一种新的动态巴特沃斯同态滤波器,增强图像在YIQ色彩空间中的亮度分量,并保持图像色度分量不变。在增加图像高频分量的同时,削减低频分量,弥补由光照不足引起的图像质量下降,最终实现对图像的光照补偿。实验结果表明,该方法在弱化光照影响,提高图像质量方面具有良好的性能,是一种比传统高斯高通滤波和直方图均衡更有效的彩色图像光照补偿方法。     

边缘保护滤波的多焦点图像融合算法

基金项目：。摘要：已有的多焦点图像融合算法要么基于不含噪声的源图像假设,要么进行高频滤波时造成边缘特征的破坏;针对这些不足,提出边缘保护滤波的图像融合算法。首先利用局部细节特征的各向异性,建立了能够同时去噪和融合的能量函数,然后采用多焦点特征加权的融合规则,建立了能量函数的迭代求解策略。实验结果表明,算法的残留噪声更少,结构相似度和视觉融合质量更高。