Adaptive grayscale adjustment-based stripe noise removal method of single image

An effective adaptive grayscale adjustment-based stripe noise removal method of single image is presented. In this method, we extract all stripe noise spectra from each frame, and then exclude relatively stationary images by sub-pixel registration to obtain continuously moving image sequences. By accumulating the same frequency spectra of the image sequences, we acquire accurate stripe noise spectra. Using the proportion of each stripe noise spectrum, we calculate the new histogram of the current column image, thereby effectively diminishing all frequency noises. In using the histogram for grayscale processing, we adopt the coefficient weight of the bilateral filtering function. Through intensity and distance factors, this function controls the ratio of the column histograms included in the calculation of the new current column histogram. This prevents the production of artifacts in the proposed method. Experiments demonstrate that our algorithm efficiently removes stripe noise and exhibits better performance than do the other algorithms discussed in literature.     

Unidirectional total variation destriping using difference curvature in MODIS emissive bands

This paper presents a method of unidirectional total variation destriping using difference curvature in MODIS (Moderate Resolution Imaging Spectrometer) emissive bands. First, difference curvature is utilized to extract spatial information at each pixel; and the spatially weighted parameters that constructed by extracted spatial information are incorporated into the unidirectional total variation model to adaptively adjust the destriping strength for achieving a better destriping result and preserving the detail information meantime. Second, the split Bregman iteration method is employed to optimize the proposed model. Finally, experimental results from MODIS emissive bands and comparisons with other methods demonstrate the potential of the presented method for MODIS image destriping.     

Fast Poissonian image segmentation with a spatially adaptive kernel

The variational models with the goal of minimizing the local variation are widely used for the segmentation of the intensity inhomogeneous images recently. Local variation is a good measure for the images corrupted by Gaussian noise. However, in many applications such as astronomical imaging, electronic microscopy and positron emission tomography, Poisson noise often occurs in the observed images. To deal with this kind of images, we develop a novel segmentation model based on minimizing local generalized Kullback–Leibler (KL) divergence with a spatially adaptive kernel. A fast algorithm based on the split-Bregman method is proposed to solve the corresponding optimization problem. Numerical experiments for synthetic and real images demonstrate that the proposed model outperforms most of the current state-of-the-art methods in the present of Poisson noise.     

自适应非凸稀疏正则化下自适应光学系统加性噪声的去除

自适应光学系统可以实时测量并校正波前信息,但是系统中大量的噪声严重影响了系统的探测精度.自适应光学系统中一般为加性噪声,本文提出一种全新的变分处理模型去除加性噪声,该模型采用自适应非凸正则项.非凸正则项在保持图像细节上较凸正则项具有更好的效果,能更好地保持点源目标的完整性.另外,根据不同区域的噪声水平自适应地构建正则化参数,使不同区域的像素点受到不同程度的噪声抑制,可以更好地保持目标的边缘细节.在算法实现上,为了解决非凸正则项收敛性较差的缺陷,采用分裂Bregman算法及增广拉格朗日对偶算法进行计算.实验及数值仿真结果都表明,该方法能够较好地去除系统中的加性噪声,且光斑信号保存得较为完整,处理后的质心探测精度及信噪比较高.     

基于空间自适应非凸正则项全变差相干斑噪声抑制

全变差正则化方法是相干斑噪声抑制研究的热点.非凸正则项能够更好地保持图像的边缘、纹理细节信息;空间自适应正则化参数可以根据像素点所在的区域,合理地控制噪声抑制程度,从而提高噪声抑制效果.本文结合非凸正则项和空间自适应正则化参数提出了一种新的全变差相干斑噪声抑制模型,并且给出了一种模型求解的数值算法.数值试验结果表明该模型能够获得更好的相干斑噪声抑制效果.