分片常值灰度图像的快速分割

PC模型是一个著名的基于区域的活动轮廓模型,它实际上是利用水平集方法解决分片常值灰度图像的分割问题。提出一个以偏微分方程形式表达的新模型,它可以看成是PC模型的一种改进。实验显示：新模型能够实现分片常值灰度图像的快速分割,同时迭代次数对初始轮廓的大小和位置不敏感。     

Discrete image modelling for piecewise constant segmentation by artificial neural networks

The problem of an image best approximation within the class of piecewise constant functions is considered. This allows a simpler data representation with a lower number of grey levels while retaining all information relevant to the particular application considered. The approximant can be found by solving a segmentation problem. The search for a solution is solved efficiently by training an artificial neural network (ANN) on a suitable set of templates by a standard procedure. The samples of the training alphabet fit the signal's local behaviour in the homogeneous image subregions and in the regions crossed by the edges. Therefore the original image domain is partitioned into disjoint 2D intervals (tiling), and for each one of them, the network selects the alphabet element closest to the corresponding image component. The main motivation of this work consists in devising a methodology suitable for real-time applications; indeed, the ANN tool is attractive for a hardware implementation.     

迭代分水岭和脊检测的图像分割

对具有边界不明显和亮度非均匀特征的图像,如金相学图像,使用传统方法进行分割难以得到令人满意的分割效果.因此,提出了一种新的图像分割算法--基于迭代过程的分水岭改进算法.该算法运用脊来限定区域生长的范围,采用双阈值方法选择种子,而脊叠加为分水岭算法中的最高吃水线.为了解决过分割的问题,利用贝叶斯分类规则反复地进行区域块的合并.实验结果表明,在无需调整太多参数的情况下,该算法可以有效地实现图像分割.     

Multiple piecewise constant with geodesic active contours (MPC-GAC) framework for interactive image segmentation using graph cut optimization

This paper proposes an improved variational model, multiple piecewise constant with geodesic active contour (MPC-GAC) model, which generalizes the region-based active contour model by Chan and Vese, 2001 [11] and merges the edge-based active contour by Caselles et al., 1997 [7] to inherit the advantages of region-based and edge-based image segmentation models. We show that the new MPC-GAC energy functional can be iteratively minimized by graph cut algorithms with high computational efficiency compared with the level set framework. This iterative algorithm alternates between the piecewise constant functional learning and the foreground and background updating so that the energy value gradually decreases to the minimum of the energy functional. The k-means method is used to compute the piecewise constant values of the foreground and background of image. We use a graph cut method to detect and update the foreground and background. Numerical experiments show that the proposed interactive segmentation method based on the MPC-GAC model by graph cut optimization can effectively segment images with inhomogeneous objects and background.     

基于修正的分段模糊吉伯斯随机场模型的图像分割

模糊随机场模型在解决多值模糊分割方面主要存在算法的稳定性和效率问题。针对这些不足,提出一种简单、方便有效的多值模糊分割新算法--修正的分段模糊吉伯斯分割算法。该算法利用修正的模糊C均值来提供良好的初始分类,结合传统的二值模糊算法来完成对复杂多值图像的快速、精确分割。实验表明,该修正算法比传统的随机场模型有更好的图像分割能力,能较好地解决目前多值模糊分割算法所面临的稳定性和效率问题。     

多相Chan-Vese模型的直接对偶方法

多相图像分割的变分模型采用水平集函数定义不同区域的特征函数,其极值问题需要迭代求解一系列动态演化方程,计算效率低。较快的方法是对离散的二值标记函数凸松弛后设计对偶方法或Split Bregman方法,并结合阈值化技术得到分割结果。提出一种无需凸松弛和阈值化的快速分割方法—直接对偶方法(DDM)。DDM利用二值标记函数的二值特性,并根据KKT条件得到原变量的二值解析解和对偶变量的简单迭代格式。该方法首先应用到两相Chan-Vese模型,然后拓展到多相Chan-Vese模型。实验结果表明,DDM比梯度降方法、对偶方法和Split Bregman方法分割效果好、计算效率高。     

A model of angiogenesis by hybrid systems with delay on the piecewise constant part

An important approach towards understanding the cancer dynamics is the modeling of angiogenesis process. There have been several attempts to model this process. Among them angiogenesis models with time delays, caused by the physical distance between the tumor and the vessel, are the most realistic ones. Recent studies have suggested that those delays can cause oscillatory behavior in the angiogenesis process. In this work we employed piecewise linear hybrid systems with delay on the piecewise constant part. Our approach is based on piecewise linearization of the system behavior where the delays occur at threshold crossings and state transitions. Piecewise linear systems with a single threshold for each variable are useful in approximating and modeling the dynamical systems especially when the model might need to be calibrated by the observations. Therefore, we used piecewise linear systems where the delays are introduced in piecewise constant part of the equations. Our approach allows tractable approximation of the angiogenesis process with possible advances of incorporating more variables, involving the effect of some possible external inputs, and possible adjustment or correction of parameters by observations.     

Topology optimization of steady Navier-Stokes flow via a piecewise constant level set method

This paper presents a piecewise constant level set method for the topology optimization of steady Navier-Stokes flow. Combining piecewise constant level set functions and artificial friction force, the optimization problem is formulated and analyzed based on a design variable. The topology sensitivities are computed by the adjoint method based on Lagrangian multipliers. In the optimization procedure, the piecewise constant level set function is updated by a new descent method, without the needing to solve the Hamilton-Jacobi equation. To achieve optimization, the piecewise constant level set method does not track the boundaries between the different materials but instead through the regional division, which can easily create small holes without topological derivatives. Furthermore, we make some attempts to avoid updating the Lagrangian multipliers and to deal with the constraints easily. The algorithm is very simple to implement, and it is possible to obtain the optimal solution by iterating a few steps. Several numerical examples for both two- and three-dimensional problems are provided, to demonstrate the validity and efficiency of the proposed method.     

融合局部和全局高斯概率信息的图像分割模型

在现有的活动轮廓中,LBF模型、LIF模型和LGDF模型是著名的基于区域的模型。虽然能分割灰度不均匀的图像,但对活动轮廓的初始化和噪声较为敏感。针对该问题,提出一种融合全高斯和局部高斯概率信息的活动轮廓模型。首先由全局高斯模型的全局灰度拟合力和局部高斯模型的局部灰度拟合力的一个线性组合来构造水平集演化力,然后引入这两个拟合力的动态权重以达到该模型的灵活性,实验结果表明,该模型能分割灰度不均的图像,且允许灵活的轮廓初始化,抗噪声性强。     

Analysis and recognition of piecewise constant texture images

New methods of recognition of texture-valued images have been proposed and analyzed. These methods derive from the well-known morphological approach to analysis and recognition [1, 2] of intensity-valued images. In order to convert texture characteristics into numeric intensity values, texture detectors have been developed. The proposed algorithms are tested in application to the problem of digit recognition. Sergey O. Evsegneev. Born 1979. Graduated from the Faculty of Physics, Moscow State University, in 2003. Postgraduate student at the same faculty. Scientific interests: image analysis and recognition.     

Combinatorial Optimization of the piecewise constant Mumford-Shah functional with application to scalar/vector valued and volumetric image segmentation

Front propagation models represent an important category of image segmentation techniques in the current literature. These models are normally formulated in a continuous level sets framework and optimized using gradient descent methods. Such formulations result in very slow algorithms that get easily stuck in local solutions and are highly sensitive to initialization.In this paper, we reformulate one of the most influential front propagation models, the Chan-Vese model, in the discrete domain. The graph representability and submodularity of the discrete energy function is established and then max-flow/min-cut approach is applied to perform the optimization of the discrete energy function. Our results show that this formulation is much more robust than the level sets formulation. Our approach is not sensitive to initialization and provides much faster solutions than level sets. The results also depict that our segmentation approach is robust to topology changes, noise and ill-defined edges, i.e., it preserves all the advantages associated with level sets methods.     

Tracking of non-square nonlinear continuous time systems with piecewise constant model predictive control

The paper presents a model predictive control (MPC) algorithm for continuous-time, possibly non-square nonlinear systems. The algorithm guarantees the tracking of asymptotically constant reference signals by means of a control scheme were the integral action is directly imposed on the error variables rather than on the control moves. The plant under control, the state and control constraints and the performance index to be minimized are described in continuous time, while the manipulated variables are allowed to change at fixed and uniformly distributed sampling times. The algorithm is used to control a continuous fermenter where the manipulated variables are the dilution rate and the feed substrate concentration while the controlled variable is the biomass concentration.     

不完全K-means聚类与分类优化结合的图像分割算法

为提升K均值聚类的效率及图像分割效果,提出了一种不完全K均值聚类与分类优化结合的图像分割(IKCO)算法。首先,采用简单的方法来进行数据精简及初始中心的确定;然后,根据给出的不完全聚类准则对图像进行聚类分割;最后,对分割结果进行分类优化以提升分割效果。实验结果表明,相对于传统的K均值聚类方法,IKCO算法在进行图像分割时具有很好的分割效率,且分割效果与人类视觉感知具有更高的一致性。     

多相主动轮廓模型的眼底图像杯盘分割

目的 视盘及视杯的检测对于分析眼底图像和视网膜视神经疾病计算机辅助诊断来说十分重要,利用医学眼底图像中视盘和视杯呈现椭圆形状这一特征,提出了椭圆约束下的多相主动轮廓模型,实现视盘视杯的同时精确分割。方法 该算法根据视盘视杯在灰度图像中具有不同的区域亮度,建立多相主动轮廓模型,然后将椭圆形约束内嵌于该模型中。通过对该模型的能量泛函进行求解,得到椭圆参数的演化方程。分割时首先设定两条椭圆形初始曲线,根据演化方程,驱动曲线分别向视盘和视杯方向进行移动。当轮廓线到达视盘、视杯边缘时,曲线停止演化。结果 在不同医学眼底图像中对算法进行验证,对算法抗噪性、不同初始曲线选取等进行了实验,并与多种算法进行了对比。实验结果表明,本文模型能够同时分割出视盘及视杯,与其他模型的分割结果相比,本文算法的分割结果更加准确。结论 本文算法可以精确分割医学眼底图像中的视盘和视杯,该算法不需要预处理,具有较强的鲁棒性和抗噪性。     

结合最大方差比准则和PCNN模型的图像分割

脉冲耦合神经网络（PCNN）模型在图像分割方面有着很好的应用。在各项参数确定的情况下,其分割结果的好坏取决于循环迭代次数的多少,而PCNN模型自身无法实现迭代次数的自动判定。为此提出一种结合最大方差比准则的PCNN迭代次数自动判定算法,用于实现图像的自动分割。算法利用最大方差比准则找到图像的最优分割界限,确定PCNN的迭代次数,获得最优图像分割结果,然后利用最大香农熵准则验证分割结果。实验表明：提出的算法实现了PCNN迭代次数的自动判定,提高了PCNN的迭代速度,运行效率优于基于2D-OTSU和基于交叉熵的自动分割算法,图像分割效果良好。     

基于优化算法与多尺度差分的图像分割方法

为了使图像分割算法能够满足实时性要求,针对Otsu法计算量大的问题,将类间方差进行连续域扩展,推导出速度较快的黄金分割法;应用灰度差分直方图进行分割,能够根据多尺度灰度差分直方图得到候选集,将搜索次数减少到候选集中元素个数,计算量小、速度快;结合两者可以实现图像的快速分割。仿真实验和实际应用表明,该方法不仅分割效率高,而且能够得到很好的分割效果。     

基于惯性因子自适应粒子群和模糊熵的图像分割 *

由于基本粒子群算法存在易陷入局部最优以及过早收敛的缺点 ,使得基于粒子群和模糊熵的图像分割算法难以得到理想的分割效果。针对此问题 ,提出了一种基于惯性因子自适应粒子群和模糊熵的图像分割算法,利用惯性因子自适应粒子群和高斯变异来搜索使模糊熵最大的参数值 ,得到模糊参数的最优组合 ,进而确定图像的分割阈值。通过与其他两种粒子群算法的分割结果进行比较 ,表明该算法取得了令人满意的分割结果 ,算法运算时间较小 ,具有很好的鲁棒性和自适应性。     

基于PSO_KFCM的医学图像分割

在核模糊聚类算法(KFCM)的基础上,提出了一种新的PSO KFCM聚类算法.新算法利用高斯核函数,把输入空间的样本映射到高维特征空间,利用微粒群算法的全局搜索、快速收敛的特点,代替KFCM算法逐次迭代的过程,在特征空间中进行聚类,克服了KFCM对初始值和噪声数据敏感、易陷入局部最优的缺点.通过对医学图像进行分割,仿真实验结果表明,新算法在性能上比KFCM聚类算法有较大改进,具有更好的聚类效果,且算法能够很快地收敛.     

基于小波变异粒子群和模糊熵的图像分割

基于粒子群和模糊熵的图像分割算法用于各种图像分割时,由于基本粒子群算法存在易陷入局部最优以及过早收敛的缺点,使得该算法难以得到理想的分割效果。针对此问题,提出了一种基于小波变异粒子群和模糊熵的图像分割算法,利用小波变异粒子群来搜索使模糊熵最大的参数值,得到模糊参数的最优组合,进而确定图像的分割阈值。通过与其他两种粒子群算法的分割结果进行比较,表明该算法取得了令人满意的分割结果,算法运算时间较小,具有很好的自适应性。