Image smoothing with generalized random walks: Algorithm and applications

A novel generalized random walks model based algorithm for image smoothing is presented. Unlike previous image smoothing methods, the proposed method performs image smoothing in a global weighted way based on graph notation, which can preserve important features and edges as much as possible. Based on the new random walks model, input image information and user defined smoothing scale information are projected to a graph, our method calculates the probability that a random walker starting at each pixel node position will first reach one of the pre-defined terminal node to achieve image smoothing, which goes to solving a system of linear equations, the system can be solved efficiently by lots of methods. Theoretical analysis and experimental results are reported to illustrate the usefulness and potential applicability of our algorithm on various computer vision fields, including image enhancement, edge detection, image decomposition, high dynamic range (HDR) image tone mapping and other applications.     

Wavelet domain nonlinear filtering for evoked potential signal enhancement.

A wavelet domain nonlinear filtering method for improving the signal-to-noise ratio (SNR) of the evoked potentials (EP) is proposed. The method modifies the selective filtering technique proposed for edge detection in images by Xu et al. for the case of signals which require a smooth transition at the edge points. It identifies the significant features of a noisy signal based on the correlation between the scales of its nonorthogonal subband decompositions. The signal transition information from interscale correlation coupled with the change in variance around the identified transition region is used to differentiate between noise and the signal. A nonlinear function such as a Gaussian smoothing function applied around the identified edge in the wavelet domain leads to smoothing in the signal space also. Numerical results obtained by applying the proposed nonlinear filtering method on middle latency responses of auditory evoked potentials show that the method is well suited for signal enhancement applications.     

Bayes smoothing algorithms for segmentation of binary images modeled by markov random fields

A new image segmentation algorithm is presented, based on recursive Bayes smoothing of images modeled by Markov random fields and corrupted by independent additive noise. The Bayes smoothing algorithm yields the a posteriori distribution of the scene value at each pixel, given the total noisy image, in a recursive way. The a posteriori distribution together with a criterion of optimality then determine a Bayes estimate of the scene. The algorithm presented is an extension of a 1-D Bayes smoothing algorithm to 2-D and it gives the optimum Bayes estimate for the scene value at each pixel. Computational concerns in 2-D, however, necessitate certain simplifying assumptions on the model and approximations on the implementation of the algorithm. In particular, the scene (noiseless image) is modeled as a Markov mesh random field, a special class of Markov random fields, and the Bayes smoothing algorithm is applied on overlapping strips (horizontal/vertical) of the image consisting of several rows (columns). It is assumed that the signal (scene values) vector sequence along the strip is a vector Markov chain. Since signal correlation in one of the dimensions is not fully used along the edges of the strip, estimates are generated only along the middle sections of the strips. The overlapping strips are chosen such that the union of the middle sections of the strips gives the whole image. The Bayes smoothing algorithm presented here is valid for scene random fields consisting of multilevel (discrete) or continuous random variables.     

Adaptive smoothing: a general tool for early vision

A method to smooth a signal while preserving discontinuities is presented. This is achieved by repeatedly convolving the signal with a very small averaging mask weighted by a measure of the signal continuity at each point. Edge detection can be performed after a few iterations, and features extracted from the smoothed signal are correctly localized (hence, no tracking is needed). This last property allows the derivation of a scale-space representation of a signal using the adaptive smoothing parameter k as the scale dimension. The relation of this process to anisotropic diffusion is shown. A scheme to preserve higher-order discontinuities and results on range images is proposed. Different implementations of adaptive smoothing are presented, first on a serial machine, for which a multigrid algorithm is proposed to speed up the smoothing effect, then on a single instruction multiple data (SIMD) parallel machine such as the Connection Machine. Various applications of adaptive smoothing such as edge detection, range image feature extraction, corner detection, and stereo matching are discussed     

基于矢量尺度的图像滤波

对于医学图像滤波来说,很重要的一点就是滤波后的图像应该尽可能地保留图像中的边缘和细节特征.但通常在滤波过程中,在消除噪声的同时会模糊图像中一些重要的结构信息.在最近几年中,基于尺度的滤波方法已经有效地应用在灰度图像的滤波中.现把基于尺度的方法推广到矢量(彩色)图像的滤波中.在传统滤波方法(矢量中值滤波、基本矢量方向滤波和方向距离滤波)基础上,相应地提出了3种基于球尺度的矢量滤波器.新的滤波方法能根据图像中像素的尺度信息,在图像边缘和细节附近,即区域边界执行较小的平滑,而在区域内部进行较大的平滑,从而能够自适应地控制滤波过程.实验结果表明,所提出的滤波方法与传统滤波方法相比,在消除噪声的同时更能够保留图像中的边缘和细节特征.     

基于图像分解和相对全变分的图像平滑

图像平滑旨在去除图像中纹理细节信息的同时保留重要的结构边缘,因此如何正确区分二者成了 图像平滑的关键。梯度作为计算图像变化速度的重要指标是区分结构边缘和纹理细节的有效度量,但不同图像 以及同一图像不同区域中的纹理和边缘的梯度差异并非固定不变的。为了能够有效识别结构边缘和纹理细节, 提出了基于图像分解和相对全变分的图像平滑方法。为了扩大结构边缘和纹理细节之间的差异,实现在尽可能 不改变结构边缘的前提下降低纹理细节的梯度,以多方向的梯度为约束对图像进行分解,提取图像的平滑成分。 在特定尺度下,基于图像的区域结构差异,采用相对全变分方法,在保留结构边缘的同时去除该尺度下的纹理 细节。通过迭代优化,不断调整图像区域尺度,实现对不同尺度纹理细节的逐步去除。与现有算法相比,新方 法在有效地去除纹理细节和完整地保留结构边缘方面都具有较好的视觉效果。     

Scale‐aware Structure‐Preserving Texture Filtering

This paper presents a novel method to enhance the performance of structure‐preserving image and texture filtering. With conventional edge‐aware filters, it is often challenging to handle images of high complexity where features of multiple scales coexist. In particular, it is not always easy to find the right balance between removing unimportant details and protecting important features when they come in multiple sizes, shapes, and contrasts. Unlike previous approaches, we address this issue from the perspective of adaptive kernel scales. Relying on patch‐based statistics, our method identifies texture from structure and also finds an optimal per‐pixel smoothing scale. We show that the proposed mechanism helps achieve enhanced image/texture filtering performance in terms of protecting the prominent geometric structures in the image, such as edges and corners, and keeping them sharp even after significant smoothing of the original signal.     

一种基于高阶统计量的图像混和加权滤波方法

平滑噪声和保持边缘细节是对图像滤波两个方面的要求,如何兼顾和平衡二者是图像滤波要解决的核心问题。根据高斯噪声的特点,该文引入高阶统计量并结合空域滤波的模板法描述图像的纹理信息,提出了一种基于高阶统计量分析的图像混和加权滤波方法。文章利用高阶累积量所描述的图像细节复杂程度对模板进行分类,并分别采用相应的滤波方法,最后通过混和加权而得到其估值,从而既较好地保持了图像边缘细节,又有效地滤除了图像噪声。实验结果表明,相对于几种常见的保细节滤波方法,文章介绍的方法能够得到更好的效果。     

A new edge detector integrating scale-spectrum information

This paper presents a new scale space-based method to extract edges in gray level images. The method is based on a novel representation of gray-level shape called the scale-spectrum space. The scale space representation is used to describe an image at different scales. In order to obtain the original image edges, an edge detector is applied to each simplified image on the corresponding scale. At best, some form of compromise among the edges at different scale levels may be sought. To overcome this problem, we present a stability criterion to combine edges obtained at different scales. Usual problems in edge detection such as displacement, redundancy and error are analyzed and solved using a realistic estimation of displacement of points across scale space. The proposed approach suppress the finer details without weakening or dislocating the larger scale edges (the usual problems of edge detection using an isotropic diffusion procedure) in an improved manner compared to anisotropic diffusion procedures because a tuning function is not required. The proposed methodology is biologically inspired by the behavior of visual cortex neurones as well as retinal cells.     

Smoothing: A natural way to detect contour features

This work presents a dominant point detector. The angles of the contour are characterized through local entropy produced by a rotationally symmetric smoothing. The proposed scheme uses a punctual multi-scale approach in which only the candidates are analyzed in higher scales. To preserve the angle-entropy relationship in higher scales, we propose a smoothing kernel which presents special features that ensure its steepness in every scale. It is built from the sum of two Gaussians with different openings resembling center-surround receptive fields. The outputs of the proposed method are confronted to a ground-truth found in the literature, and to popular boundary based corner detectors that used the same set of images. Results reveal that the proposed detector performs extremely well.     

自适应多尺度分块压缩感知算法

目的 基于小波域的多尺度分块压缩感知重构算法忽略了高频信号在重构过程中的作用,丢失了大量的边缘与细节信息。针对上述问题,提出一种自适应多尺度分块压缩感知算法,不仅合理利用低频信息还充分利用图像的高频信息,在图像细节复杂度提高的情况下保证图像重构质量的提高。方法 首先进行3层小波变换,得到一个低频信号和9个高频信号,分别进行小波逆变换后分成大小相同互不重叠的块,对低频部分采用2维邻块边缘自适应加权滤波的方法进行处理,对高频部分采用纹理自适应分块采样,最后利用平滑投影Landweber（SPL）算法对其进行重构。结果 与已有的分块压缩感知算法、基于边缘和方向的分块压缩感知算法和基于纹理和方向的分块压缩感知算法相比,本文算法在不同的采样率下,性能均有所提升,代表细节信息的高频信号得到充分重建,改进的算法所得到的重建图像具有较高的分辨率,尤其对细节较为丰富的图像进行重建后具有较高的峰值信噪比;2维邻块边缘自适应加权滤波有效的去除了重建图像的块效应,且重建时间平均减少了0.3 s。结论 将三层小波变换后的高频分量作为纹理部分,利用自适应多尺度分块重建出图像的轮廓与边缘;将低频分量直接视为平坦部分,邻块边缘自适应加权滤波重建出图像细节,不仅充分利用了图像的高低频信息,还减少了平坦块检测过程,使得重建时间有效缩短。经实验验证,本文算法重建图像质量较好,尤其是对复杂图像明显消除了块效应,边缘和纹理细节较清晰。因此主要适用于纹理细节较复杂的人脸图像、建筑图像和遥感图像等。     

基于脊波融合的射线图像增强算法

提出一种基于脊波变换的射线图像增强算法,根据射线成像的特点,对射线成像系统采集信号做分段灰度变换,得到多幅图像,每幅图像含有被测工件的某种细节,再将这些图像分别做有限脊波变换,对得到的变换系数进行融合,再对融合后的系数进行有限脊波逆变换从而得到增强了的射线图像。在融合中低频系数采用基于区域方差和邻域像素相关性分析的融合策略,高频部分采用脊波变换系数绝对值最大的作为融合的高频系数,此法可以将来自不同图像的特征与细节融合在一起并且可以有效地抑制噪声。文中对融合图像质量进行了对比评价,实验结果表明,这种方法能够有效地提高图像的清晰度,在保留图像微小细节方面获得满意的结果。     

Scale-space and edge detection using anisotropic diffusion

A new definition of scale-space is suggested, and a class of algorithms used to realize a diffusion process is introduced. The diffusion coefficient is chosen to vary spatially in such a way as to encourage intraregion smoothing rather than interregion smoothing. It is shown that the `no new maxima should be generated at coarse scales' property of conventional scale space is preserved. As the region boundaries in the approach remain sharp, a high-quality edge detector which successfully exploits global information is obtained. Experimental results are shown on a number of images. Parallel hardware implementations are made feasible because the algorithm involves elementary, local operations replicated over the image     

Multiresolution color image segmentation

Image segmentation is the process by which an original image is partitioned into some homogeneous regions. In this paper, a novel multiresolution color image segmentation (MCIS) algorithm which uses Markov random fields (MRF's) is proposed. The proposed approach is a relaxation process that converges to the MAP (maximum a posteriori) estimate of the segmentation. The quadtree structure is used to implement the multiresolution framework, and the simulated annealing technique is employed to control the splitting and merging of nodes so as to minimize an energy function and therefore, maximize the MAP estimate. The multiresolution scheme enables the use of different dissimilarity measures at different resolution levels. Consequently, the proposed algorithm is noise resistant. Since the global clustering information of the image is required in the proposed approach, the scale space filter (SSF) is employed as the first step. The multiresolution approach is used to refine the segmentation. Experimental results of both the synthesized and real images are very encouraging. In order to evaluate experimental results of both synthesized images and real images quantitatively, a new evaluation criterion is proposed and developed     

基于混合频谱信号编码的网格纹理平滑

网格纹理平滑技术要求既能保持模型大尺度结构特征又能去除模型小尺度纹理.然而当模型小尺度纹理与噪声相差较大时,大多数网格光顺算法会将网格纹理识别为特征加以保持,而无法有效将其去除;现有的基于谱分析的网格光顺方法尽管能有效去除网格纹理,但又无法同时保持模型大尺度结构特征.为解决该问题,本文提出一种基于混合频谱信号编码的低通...     

基于异常区域感知的多时相高分辨率遥感图像配准

在对多时相高分辨遥感图像进行配准时,由于成像条件差异,图像间存在的地物变化与相对视差偏移两类典型异常区域会影响配准精度。针对上述配准中存在的问题,提出一种基于异常区域感知的多时相高分辨率遥感图像配准方法,包括粗匹配和精配准两个阶段。尺度不变特征变换（SIFT）算法考虑到尺度空间属性,不同尺度空间提取的特征点在图像中对应不同大小的斑块,高尺度空间提取的特征点对应图像中的大斑点,其对应地物相对稳定、不易发生变化。首先,利用SIFT算法提取高尺度空间特征点完成图像快速粗匹配;其次,利用灰度相关性度量对图像块进行相对偏移量统计分类以感知视差偏移区域,同时结合空间约束条件,确定低尺度空间特征点的有效提取区域以及匹配点搜索范围,完成图像精配准。实验结果表明,将该方法用于多时相高分辨遥感图像配准,可有效抑制异常区域对特征点提取的影响进而提高配准精度。     

基于核与局部信息的多维度模糊聚类图像分割算法

在以聚类分析为背景的图像分割算法中,引入局部信息是为了在保留图像细节的同时尽可能地减少噪声.在模糊C均值算法基础上,提出了一种基于核与局部信息的多维度模糊聚类分析方法来权衡图像中的噪声和细节.该算法引入2个基于局部信息的图像变体,即平滑和锐化处理后的图像,使之与原始图像一起构成多维度的灰度值向量来替换原始单维的灰度值; 再利用核方法提高其鲁棒性; 最后添加一个邻域隶属度差异惩罚项很好地修正和增强了最终的分割效果.在人工合成图片的去噪实验中,所提方法取得了近99%的分割正确率,优于Nystrom归一化分割(NNcut)和基于模糊局部信息C均值(FLICM)算法;同时在自然图片和医学图片的对比实验以及参数调控实验中,展现出了其在处理图像噪声和细节时灵活、稳定、健壮且易于调控的特点.     

基于视觉特征的尺度空间信息量度量

图像的多尺度表示指的是从原始图像出发,导出一系列越来越平滑、简化的图像。这种简化意味着信息的丢失。如果能定量描述每一个尺度中图像的信息,这对于多尺度表示来说有着重要的作用。虽然Sporring等人提出的尺度空间信息熵度量能解决一些问题,但是并不满足从视觉理论和直观的基础上提出的尺度空间信息量度量的基本要求,例如形态不变性等,为此在M arr视觉理论基础上定义了一个新的具有视觉意义的尺度空间信息度量,并在典型的高斯尺度空间中,证明了它确实满足从视觉理论和直观的基础上提出的尺度空间信息量度量的基本要求。数值试验验证了这种定义在视觉上是可靠的,从而为图像尺度的自适应选择提供了一种可靠的方法。     

SEGMENTATION OF NATURAL IMAGES USING ANISOTROPIC DIFFUSION AND LINKING OF BOUNDARY EDGES

We present a segmentation method of natural images that uses an anisotropic diffusion algorithm and a region growing algorithm. We propose a modified version of the anisotropic diffusion algorithm as a precise edge-preserving smoothing technique modified by using boundary edges. We incorporate a linking algorithm for boundary edges based on a directional potential function into the anisotropic diffusion algorithm to improve the ability of edge-preserving smoothing. As a result, unnecessary details of images are effectively smoothed before performing a region growing algorithm. Therefore, the proposed method is suitable for an accurate segmentation of natural images. Several simulated examples are presented that demonstrate the effectiveness of the proposed technique.