Optimal Level Curves and Global Minimizers of Cost Functionals in Image Segmentation

We propose a variational framework for determining global minimizers of rough energy functionals used in image segmentation. Segmentation is achieved by minimizing an energy model, which is comprised of two parts: the first part is the interaction between the observed data and the model, the second is a regularity term. The optimal boundaries are the curves that globally minimize the energy functional. Our motivation comes from the observation that energy functionals are traditionally complex, for which it is usually difficult to precise global minimizers corresponding to best segmentations. Therefore, we focus on basic energy models, which global minimizers can be characterized. None of the proposed segmentation models captures all the important scene variables but may be useful to get an insight into objects, surfaces or parts of objects in the scene. In this paper, we prove that the set of curves that minimizes the cost functionals is a subset of level lines, i.e. the boundaries of level sets of the image. For the completeness of the paper, we present a fast algorithm for computing partitions with connected components. It leads to a sound initialization-free algorithm without any hidden parameter to be tuned. We illustrate the performance of our algorithm with several examples on both 2D biomedical and aerial images, and synthetic images.     

Variational Image Binarization and its Multi-Scale Realizations

A variational approach for image binarization is discussed in this paper. The approach is based on the interpolation of surface. This interpolation is computed using edge points as interpolating points and minimizing an energy functional which interpolates a smooth threshold surface. A globally convergent Sequential Relaxation Algorithm (SRA) is proposed for solving the optimization problem. Moreover, our algorithm is also formulated in a multi-scale framework. The performance of our method is demonstrated on a variety of real and synthetic images and compared with traditional techniques. Examples show that our method gives promising results.This research is partially supported by HKBU Faculty Research Grant FRG/02-03/II-04 and NSF of China Grant. C.S. Tong received a BA degree in Mathematics and a Ph.D. degree (on Mathematical Modelling of Intermolecular Forces) both from Cambridge University. After graduation, he joined the Signal and Image Processing division of GEC-Marconis Hirst Research Centre as a Research Scientist, working on image restoration and fractal image compression. He then moved to the Department of Mathematics at Hong Kong Baptist University in 1992, becoming Associate Professor since 2002.He is a member of the IEEE, a Fellow of the Institute of Mathematics and Its Application, and a Chartered Mathematician. His current research interests include image processing, fractal image compression, and neural networks. Yongping Zhang received the M. S. degree from Department of Mathematics at Shaanxi Normal University, Xian, China, in 1988 and received the Ph.D. degree from The Institute of Artificial Intelligence and Robotics at Xian Jiaotong University, Xian, China, in 1998.In 1988 he joined Department of Mathematics at Shaanxi Normal University, where he became Associate Professor in July 1987. He held postdoctoral position at Northwestern Polytechnic University during the 1999–2000 academic years. Currently he is a research associate in the Bioengineering Institute at the University of Auckland, New Zealand. His research interests are in Computer Vision and Pattern Recognition, and include Wavelets, Neural Networks, PDE methods and variational methods for image processing. Nanning Zheng received the M.S. degree from Xian Jiaotong University, Xian, China, in 1981 and the Ph.D. degree from Keio University, Japan, in 1985. He is an academician of Chinese Engineer Academy, and currently a Professor at Xian Jiaotong University. His research interest includes Signal Processing, Machine Vision and Image Processing, Pattern Recognition and Virtual Reality.This revised version was published online in June 2005 with correction to CoverDate     

一种利用灰度信息和二值信息的手写字符分割

在字符识别系统中,字符的有效分割是识别的关键。该文描述了一种两阶段的手写字符分割方法。根据字符倾斜角度将原字符串图像划分出若干个形状不规则的区域,每个区域含有一条分割线;针对现有分割方法的某些局限性,提出采用字符图像灰度信息和二值信息相结合的方式来求取分割区域中的分割路径。实验结果表明了该方法对于手写字符分割的有效性。     

Texture Analysis by Accurate Identification of Simple Markovian Models

A more accurate identification (estimation of parameters) of simple Markov-Gibbs random field models of images results in a better segmentation of specific multimodal images and realistic synthesis of some types of natural textures. Identification algorithms for segmentation are based in part on a novel modification of an unsupervised learning algorithm published first in Cybernetics and Systems Analysis (Kibernetika i Sistemnyi Analiz) almost four decades ago. A texture synthesis algorithm uses an identified model for selecting a characteristic geometric shape of and a placement grid for texture elements sampled from a training image utilized for the identification.__________Translated from Kibernetika i Sistemnyi Analiz, No. 1, pp. 37–49, January–February 2005.     

基于谱聚类的聚类集成算法

谱聚类是近年来出现的一类性能优越的聚类算法,能对任意形状的数据进行聚类, 但算法对尺度参数比较敏感,利用聚类集成良好的鲁棒性和泛化能力,本文提出了基于谱聚类的聚类集成算法.该算法首先利用谱聚类算法的内在特性构造多样性的聚类成员; 然后,采用连接三元组算法计算相似度矩阵,扩充了数据点之间的相似性信息;最后,对相似度矩阵使用谱聚类算法得到最终的集成结果. 为了使算法能扩展到大规模应用,利用Nystrm采样算法只计算随机采样数据点之间以及随机采样数据点与剩余数据点之间的相似度矩阵,从而有效降低了算法的计算复杂度. 本文算法既利用了谱聚类算法的优越性能,同时又避免了精确选择尺度参数的问题.实验结果表明:较之其他常见的聚类集成算法,本文算法更优越、更有效,能较好地解决数据聚类、图像分割等问题.     

Efficient and effective Querying by Image Content

In the QBIC (Query By Image Content) project we are studying methods to query large on-line image databases using the images' content as the basis of the queries. Examples of the content we use include color, texture, shape, position, and dominant edges of image objects and regions. Potential applications include medical (Give me other images that contain a tumor with a texture like this one), photo-journalism (Give me images that have blue at the top and red at the bottom), and many others in art, fashion, cataloging, retailing, and industry. We describe a set of novel features and similarity measures allowing query by image content, together with the QBIC system we implemented. We demonstrate the effectiveness of our system with normalized precision and recall experiments on test databases containing over 1000 images and 1000 objects populated from commercially available photo clip art images, and of images of airplane silhouettes. We also present new methods for efficient processing of QBIC queries that consist of filtering and indexing steps. We specifically address two problems: (a) non Euclidean distance measures; and (b) the high dimensionality of feature vectors. For the first problem, we introduce a new theorem that makes efficient filtering possible by bounding the non-Euclidean, full cross-term quadratic distance expression with a simple Euclidean distance. For the second, we illustrate how orthogonal transforms, such as Karhunen Loeve, can help reduce the dimensionality of the search space. Our methods are general and allow some false hits but no false dismissals. The resulting QBIC system offers effective retrieval using image content, and for large image databases significant speedup over straightforward indexing alternatives. The system is implemented in X/Motif and C running on an RS/6000.On sabbatical from Univ. of Maryland, College Park. His work was partially supported by SRC, by the National Science Foundation under the grant IRI-8958546 (PYI).     

基于DICOM格式的肝脏肿瘤CT图像分割

为减少数据损失与处理时间,直接从DICOM格式的CT胸腹部图像中分割出肝脏肿瘤。为有效地分割出肿瘤,先设置恰当的窗宽窗位。利用ITK读取图像,用均值曲率流滤波法去除图像上的噪声。对目前几种在图像分割领域应用较多的阈值分割及区域生长分割算法进行了探讨,结合获得的活体CT图像进行实验研究,得到较为满意和有效的结果。实验表明：肝脏肿瘤这一目标区域的面积较小,区域生长分割算法中的“置信连接阈值法”能从胸腹腔CT图像中很好地分割出肝脏肿瘤。     

Parallel Algorithms for Image Histogramming and Connected Components with an Experimental Study

This paper presents efficient and portable implementations of two useful primitives in image processing algorithms, histogramming and connected components. Our general framework is a single-address space, distributed memory programming model. We use efficient techniques for distributing and coalescing data as well as efficient combinations of task and data parallelism. Our connected components algorithm uses a novel approach for parallel merging which performs drastically limited updating during iterative steps, and concludes with a total consistency update at the final step. The algorithms have been coded in S -C and run on a variety of platforms. Our experimental results are consistent with the theoretical analysis and provide the best known execution times for these two primitives, even when compared with machine-specific implementations.     

Isophotes Selection and Reaction-Diffusion Model for Object Boundaries Estimation

This paper investigates generic region-based segmentation schemes using area-minimization constraint and background modeling, and develops a computationally efficient framework based on level lines selection coupled with biased anisotropic diffusion. A common approach to image segmentation is to construct a cost function whose minima yield the segmented image. This is generally achieved by competition of two terms in the cost function, one that punishes deviations from the original image and another that acts as a regularization term. We propose a variational framework for characterizing global minimizers of a particular segmentation energy that can generates irregular object boundaries in image segmentation. Our motivation comes from the observation that energy functionals are traditionally complex, for which it is usually difficult to precise global minimizers corresponding to best segmentations. In this paper, we prove that the set of curves that minimizes the basic energy model under concern is a subset of level lines or isophotes, i.e. the boundaries of image level sets. The connections of our approach with region-growing techniques, snakes and geodesic active contours are also discussed. Moreover, it is absolutely necessary to regularize isophotes delimiting object boundaries and to determine piecewise smooth or constant approximations of the image data inside the objects boundaries for vizualization and pattern recognition purposes. Thus, we have constructed a reaction-diffusion process based on the Perona-Malik anisotropic diffusion equation. In particular, a reaction term has been added to force the solution to remain close to the data inside object boundaries and to be constant in non-informative regions, that is the background region. In the overall approach, diffusion requires the design of the background and foreground regions obtained by segmentation, and segmentation of the adaptively smoothed image is performed after each iteration of the diffusion process. From an application point of view, the sound initialization-free algorithm is shown to perform well in a variety of imaging contexts with variable texture, noise and lighting conditions, including optical imaging, medical imaging and meteorological imaging. Depending on the context, it yields either a reliable segmentation or a good pre-segmentation that can be used as initialization for more sophisticated, application-dependent segmentation models.     

Segmentation of stick text based on sub connected area analysis

A new stick text segmentation method based on the sub connected area analysis is introduced in this paper.The foundation of this method is the sub connected area representation of text image that can represent all connected areas in an image efficiently.This method consists mainly of four steps:sub connected area classification,finding initial boundary following point,finding optimal segmentation point by boundary tracing,and text segmentaton.This method is similar to boundary analysis method but is more efficient than boundary analysis.     

Image Parsing: Unifying Segmentation,Detection, and Recognition

In this paper we present a Bayesian framework for parsing images into their constituent visual patterns. The parsing algorithm optimizes the posterior probability and outputs a scene representation as a parsing graph, in a spirit similar to parsing sentences in speech and natural language. The algorithm constructs the parsing graph and re-configures it dynamically using a set of moves, which are mostly reversible Markov chain jumps. This computational framework integrates two popular inference approaches—generative (top-down) methods and discriminative (bottom-up) methods. The former formulates the posterior probability in terms of generative models for images defined by likelihood functions and priors. The latter computes discriminative probabilities based on a sequence (cascade) of bottom-up tests/filters. In our Markov chain algorithm design, the posterior probability, defined by the generative models, is the invariant (target) probability for the Markov chain, and the discriminative probabilities are used to construct proposal probabilities to drive the Markov chain. Intuitively, the bottom-up discriminative probabilities activate top-down generative models. In this paper, we focus on two types of visual patterns—generic visual patterns, such as texture and shading, and object patterns including human faces and text. These types of patterns compete and cooperate to explain the image and so image parsing unifies image segmentation, object detection, and recognition (if we use generic visual patterns only then image parsing will correspond to image segmentation (Tu and Zhu, 2002. IEEE Trans. PAMI, 24(5):657–673). We illustrate our algorithm on natural images of complex city scenes and show examples where image segmentation can be improved by allowing object specific knowledge to disambiguate low-level segmentation cues, and conversely where object detection can be improved by using generic visual patterns to explain away shadows and occlusions.     

基于空间相关性的图像分割算法研究

提出一种充分利用图像的空间相关性来达到高效快速地进行图像分割的新方法。利用均值漂移算法对图像进行分割形成过度分割的区域,并使这些区域保持理想的边缘和空间相关部分,用图结构表示的区域相邻图来代替分割的区域。和K-均值算法的思想一样,迭代循环置信传播算法以其具有收敛速度快的特点被用于最小化开销函数、整合过度分割的区域和获得最终的分割结果。基于分割区域而不是图像像素的图像聚类分割方法可降低噪声敏感性,同时提高图像分割质量。与FCM和MRF算法相比较,该算法在复杂场景图像中显示了更好的分割性能。     

NSCT变换与改进PSO算法在含噪图像分割中的应用

为提高含噪图像的分割效果和分割速度,将非下采样Contourlet变换和粒子群优化算法相结合,提出了一种有效的图像分割方法——IPSOC。该方法首先对待分割图像进行多尺度非下采样Contourlet变换,然后利用其最高级低频系数重构图像,计算重构图像与其均值图像的二维直方图中类间离散度矩阵的迹,并以之作为分割图像的目标函数来搜索最佳分割阈值。为加快阈值搜索速度,以改进的粒子群优化算法作为阈值分割的并行搜索策略,通过对基本粒子群优化算法进行个体及全局最优信息的实时更新,防止粒子停滞操作和阈值保持次数限定搜索终止条件等几个方面的改进,快速有效地获得分割图像。实验结果表明,该方法与基于遗传算法和人工鱼群算法的分割方法相比,明显提高了图像分割速度和分割质量。     

基于多尺度形态滤波的分水岭图像分割方法

采用多尺度形态滤波器对输入噪声图像及滤波后图像的梯度图像进行平滑,实现了消除噪声、简化图像、保持物体重要轮廓信息的作用．最后,给出一种改进的快速区域合并算法优化分割结果．实验证明,采用文中分割方法可以获得很好的分割结果。     

基于DH-Swin Unet的医学图像分割算法

骨关节疾病自古以来是人类最高发的疾病之一, 随着老龄化的不断加快, 这类疾病日趋广泛, 关节外科医师面临着巨大挑战. 对人体关节的图像分割方法研究可以帮助医生进行临床诊断和治疗, 然而, 由于存在噪声、模糊、对比度低等问题, 医学图像的特征提取比普通图像更具挑战性, 而且目前大多数分割模型在编码器和解码器之间都采用了普通的跳跃连接, 没有注重解决跳跃连接过程中的信息间隙和损失问题. 为解决这些问题, 提出一种基于DH-Swin Unet的医学图像分割算法, 该模型在Swin-Unet模型的基础上, 在跳跃连接中引入密集连接的Swin Transformer块, 并加入混合注意力机制, 来强化网络的特征信息传递. 通过在某三甲医院提供的真实临床数据对所提方法的性能进行评价, 结果表明, 所提出的方法取得了DSC为86.79%、HD为32.05 mm的分割结果, 在关节疾病的临床诊断中具有一定的实用价值.     

量子遗传算法在指纹图像分割中的应用

提出了一种基于量子遗传算法(QGA)的指纹图像分割改进方法.这种方法利用量子遗传算法种群多样性好,收敛速度快的特点,将基于量子遗传算法的阈值分割方法与方向图法相结合对指纹图像进行分割.实验结果表明,QGA在指纹图像阈值分割中的速度和精度优于改进的自适应遗传算法和其它一些传统算法,是一种有效的图像分割方法.     

连通域标记并行算法在多核处理器上的设计和实现

为满足全方位视觉运动目标检测跟踪系统中检测和跟踪实时性的要求,采用多核程序设计和并行处理技术,对系统中连通域标记算法,重新进行了设计和实现。为解决由图像分割造成的连通域的分离,应用边界处的游程码,将两个不同任务块中分割处的连通域合并成一个连通域。最后通过实验证明该并行算法比传统算法在多核处理器上运行更快,更好地解决了多核处理器饥饿问题,让多核处理器发挥了更高的效率。     

基于多阈值融合的图像分割

提出了一种基于知识的多阈值融合图像分割新方法．首先利用一组多阈值分割结果建立连通域生长树．然后判断树叉对应的连通域合并是否合理,为此提出了连通体元、体元生命期、体元体积等概念,结合灰度均匀性定义出通用合并准则．最后将图像各位置的最佳连通域组合为最终图像分割结果．该算法充分利用了目标的灰度和空间属性,对灰度平稳和渐近变化的多目标图像分割非常有效．此外,该算法可以有效融合具体应用的先验知识,具有很高的智能性。     

An Improved Method for Edge Detection and Image Segmentation Using Fuzzy Cellular Automata

Image segmentation is one of the most important and challenging problems in image processing. The main purpose of image segmentation is to partition an image into a set of disjoint regions with uniform attributes. In this study, we propose an improved method for edge detection and image segmentation using fuzzy cellular automata. In the first stage, we introduce a new edge detection method based on fuzzy cellular automata, called the texture histogram, and empirically demonstrate the efficiency of the proposed method and its robustness in denoising images. In the second stage, we propose an edge detection algorithm by considering the mean values of the edges matrix. In this algorithm, we use four fuzzy rules instead of 32 fuzzy rules reported earlier in the literature. In the third and final stage, we use the local edge in the edge detection stage to more accurately accomplish image segmentation. We demonstrate that the proposed method produces better output images in comparison with the separate segmentation and edge detection methods studied in the literature. In addition, we show that the method proposed in this study is more flexible and efficient when noise is added to an image.     

密度聚类法在图像分割中的应用

提出了使用密度聚类法解决图像分割的新思路。首先把数字图像按照点的分布情况建立图像样本数据库,然后利用基于密度聚类法的DBSCAN算法进行图像分割。该算法能找到图像样本比较密集的部分,概括出图像样本相对比较集中的类,并可在带有“噪声”的图像中进行聚类,完成图像分割。文章还针对DBSCAN算法的缺点,提出了DBSCAN算法的改进思路。