A general non-local denoising model using multi-kernel-induced measures

Noises are inevitably introduced in digital image acquisition processes, and thus image denoising is still a hot research problem. Different from local methods operating on local regions of images, the non-local methods utilize non-local information (even the whole image) to accomplish image denoising. Due to their superior performance, the non-local methods have recently drawn more and more attention in the image denoising community. However, these methods generally do not work well in handling complicated noises with different levels and types. Inspired by the fact in machine learning field that multi-kernel methods are more robust and effective in tackling complex problems than single-kernel ones, we establish a general non-local denoising model based on multi-kernel-induced measures (GNLMKIM for short), which provides us a platform to analyze some existing and design new filters. With the help of GNLMKIM, we reinterpret two well-known non-local filters in the united view and extend them to their novel multi-kernel counterparts. The comprehensive experiments indicate that these novel filters achieve encouraging denoising results in both visual effect and PSNR index.     

结合非局部均值的快速FCM算法分割MR图像研究

针对FCM算法分割医学MR图像存在的运算速度慢、对初始值敏感以及难以处理MR图像中固有Rician噪声等缺陷,提出了一种结合非局部均值的快速FCM算法。该算法的核心是首先针对MR图像中存在的Rician噪声,利用非局部均值算法对图像进行去噪处理,消除噪声对分割结果的影响;然后根据所提出的新的自动获取聚类中心的规则得到初始聚类中心;最后将得到的聚类中心作为快速FCM算法的初始聚类中心用于去噪后的图像分割,解决了随机选择初始聚类中心造成的搜索速度慢和容易陷入局部极值的问题。实验表明,该算法能够快速有效地分割图像,并且具有较好的抗噪能力。     

基于非局部低秩和自适应量化约束先验的HEVC后处理算法

经高效视频编解码标准HEVC压缩后的视频在高压缩比、低码率的情况下存在明显的压缩效应。针对该问题,提出了一种基于非局部低秩(Non-local Low-rank,NLLR)和自适应量化约束(Adaptive Quantization Constraint,AQC)先验的HEVC后处理算法。该算法首先构造在最大后验概率框架下的优化问题,然后利用解码后的压缩视频和量化参数QP获取非局部低秩和自适应量化约束先验信息,最后利用split-Bregman迭代算法来解决所提的优化问题,从而有效去除压缩效应,提升重建视频质量。其中,非局部低秩先验通过构建基于相似块聚类的非局部低秩模型来获得;自适应量化约束先验通过联合不同量化参数QP下的约束特性与视频的DCT域块活动性来获得。实验结果表明,在同等码率的情况下,与HEVC标准相比,所提算法在帧内编码模式下可以达到平均0.2597 dB的PSNR提升,在帧间编码模式下可以达到平均0.2828 dB的PSNR提升。     

基于非负邻域嵌入和非局部正则化的单帧图像超分辨率重建算法

单帧图像超分辨率重建是指利用一幅低分辨率图像,通过相应的算法来获取一幅高分辨率图像的技术。提出了一种基于 非负邻域嵌入和 非局部正则化 的单帧图像超分辨率重建算法,以弥补传统邻域嵌入算法的不足。在训练阶段,首先对低分辨率图像预放大2倍,以保证在放大倍数较大时,高、低分辨率图像块之间的邻域关系也能得到较好的保持;在重建阶段,使用非负邻域嵌入来有效地解决近邻数的选取问题;最后利用图像块的非局部相似性构造非局部正则项对重建结果进行修正。实验结果表明,相对于传统算法,本方法的重建结果纹理丰富、边缘清晰。     

一种基于优化参数的非局部均值滤波算法

针对非局部均值滤波算法中难以找到一个全局最优的滤波参数h的问题,给出一种新的该参数的优化方法,并将其应用于传统非局部均值滤波算法的改进。首先基于SUSAN算法提取噪声图像的边缘信息,然后在大量实验的基础上,利用线性回归和非线性回归分析方法建立h与边缘信息、噪声方差之间的优化模型。最后,将基于该优化模型的非局部均值算法应用于多幅图像的去噪处理中。实验结果表明,新算法改善了传统非局部均值算法的去噪性能,取得了良好的滤波效果。     

Numerical approach of cyclic behaviour of 316LN stainless steel based on a polycrystal modelling including strain gradients

A non-local polycrystal approach, taking into account strain gradients, is proposed to simulate the 316LN stainless steel fatigue life curve in the hardening stage. Material parameters identification is performed on tensile curves corresponding to several 316LN polycrystals presenting different grain sizes. Applied to an actual 3D aggregate of 316LN stainless steel of 1200 grains, this model leads to an accurate prediction of cyclic curves. Geometrical Necessary Dislocation densities related to the computed strain gradient are added to the micro-plasticity laws. Compared to standard models, this model predicts a decrease of the local stresses as well as a grain size effect.     

Fast pixel-matching for video object segmentation

Video object segmentation, aiming to segment the foreground objects given the annotation of the first frame, has been attracting increasing attentions. Many state-of-the-art approaches have achieved great performance by relying on online model updating or mask-propagation techniques. However, most online models require high computational cost due to model fine-tuning during inference. Most mask-propagation based models are faster but with relatively low performance due to failure to adapt to object appearance variation. In this paper, we are aiming to design a new model to make a good balance between speed and performance. We propose a model, called NPMCA-net, which directly localizes foreground objects based on mask-propagation and non-local technique by matching pixels in reference and target frames. Since we bring in information of both first and previous frames, our network is robust to large object appearance variation, and can better adapt to occlusions. Extensive experiments show that our approach can achieve a new state-of-the-art performance with a fast speed at the same time (86.5% IoU on DAVIS-2016 and 72.2% IoU on DAVIS-2017, with speed of 0.11s per frame) under the same level comparison. Source code is available at https://github.com/siyueyu/NPMCA-net.     

Efficient numerical integration of an elastic–plastic damage law within a mixed velocity–pressure formulation

This study focuses on numerical integration of constitutive laws in numerical modeling of cold materials processing that involves large plastic strain together with ductile damage. A mixed velocity–pressure formulation is used to handle the incompressibility of plastic deformation. A Lemaitre damage model where dissipative phenomena are coupled is considered. Numerical aspects of the constitutive equations are addressed in detail. Three integration algorithms with different levels of coupling of damage with elastic–plastic behavior are presented and discussed in terms of accuracy and computational cost. The implicit gradient formulation with a non-local damage variable is used to regularize the localization phenomenon and thus to ensure the objectivity of numerical results for damage prediction problems. A tensile test on a plane plate specimen, where damage and plastic strain tend to localize in well-known shear bands, successfully shows both the objectivity and effectiveness of the developed approach.     

On the power of non-local boxes

A non-local box is a virtual device that has the following property: given that Alice inputs a bit at her end of the device and that Bob does likewise, it produces two bits, one at Alice's end and one at Bob's end, such that the XOR of the outputs is equal to the AND of the inputs. This box, inspired from the CHSH inequality, was first proposed by Popescu and Rohrlich to examine the question: given that a maximally entangled pair of qubits is non-local, why is it not maximally non-local? We believe that understanding the power of this box will yield insight into the non-locality of quantum mechanics. It was shown recently by Cerf, Gisin, Massar and Popescu, that this imaginary device is able to simulate correlations from any measurement on a singlet state. Here, we show that the non-local box can in fact do much more: through the simulation of the magic square pseudo-telepathy game and the Mermin-GHZ pseudo-telepathy game, we show that the non-local box can simulate quantum correlations that no entangled pair of qubits can, in a bipartite scenario and even in a multi-party scenario. Finally we show that a single non-local box cannot simulate all quantum correlations and propose a generalization for a multi-party non-local box. In particular, we show quantum correlations whose simulation requires an exponential amount of non-local boxes, in the number of maximally entangled qubit pairs.     

局部熵驱动下的脑MR图像分割与偏移场恢复耦合模型

核磁共振图像受成像机制的影响往往导致图像中含有噪声以及偏移场,使得传统的图像分割方法很难得到较好的分割结果.为此,提出一种基于局部熵的分割与偏移场恢复耦合模型,首先在小邻域内构建基于模糊C均值(FCM)聚类模型的局部统计项并将偏移场信息耦合到模型中,以恢复图像偏移场;其次采用非局部信息来构建邻域正则项,使得模型在降低噪声影响的同时能有效地保留图像结构信息;最后在对局部能量项进行全局积分时引入局部熵信息,使得模型具有各向异性,从而对噪声和偏移场影响更具鲁棒性.实验结果表明,本文方法可以得到较准确的分割和偏移场矫正结果.