Multispectral image segmentation using the rough-set-initialized EM algorithm

The problem of segmentation of multispectral satellite images is addressed. An integration of rough-set-theoretic knowledge extraction, the Expectation Maximization (EM) algorithm, and minimal spanning tree (MST) clustering is described. EM provides the statistical model of the data and handles the associated measurement and representation uncertainties. Rough-set theory helps in faster convergence and in avoiding the local minima problem, thereby enhancing the performance of EM. For rough-set-theoretic rule generation, each band is discretized using fuzzy-correlation-based gray-level thresholding. MST enables determination of nonconvex clusters. Since this is applied on Gaussians, determined by granules, rather than on the original data points, time required is very low. These features are demonstrated on two IRS-1A four-band images. Comparison with related methods is made in terms of computation time and a cluster quality measure.     

Efficient edge-preserving algorithm for color contrast enhancement with application to color image segmentation

In this paper, a new and efficient edge-preserving algorithm is presented for color contrast enhancement in CIE Lu^′v^′ color space. The proposed algorithm not only can enhance the color contrast as the previous algorithm does, but also has an edge-preservation effect. In addition, the spurious edge points occurred due to the color contrast enhancement can be well reduced using the proposed algorithm. This is the first edge-preserving algorithm for color contrast enhancement in color space. Furthermore, a novel color image segmentation algorithm is presented to justify the edge-preservation benefit of the proposed color contrast enhancement algorithm. Based on some real images, experimental results demonstrate the advantages of color contrast enhancement, edge-preservation effect, and segmentation result in our proposed algorithm.     

一种新的基于双层PCNN的自适应图像分割算法

提出一种新的基于双层脉冲耦合神经网络（PCNN）的自适应图像分割算法。双层PCNN的前级以简化PCNN模型为基础,获得区域生长的种子;后级采用区域生长机制,征募区域内灰度相似像素,完成前级种子的生长。新算法PCNN的关键参数可自适应更新,避免了传统PCNN参数设置难的问题;区域生长机制强化了PCNN的区域特性。实验结果...     

A hierarchical approach to color image segmentation using homogeneity

In this paper, a novel hierarchical approach to color image segmentation is studied. We extend the general idea of a histogram to the homogeneity domain. In the first phase of the segmentation, uniform regions are identified via multilevel thresholding on a homogeneity histogram. While we process the homogeneity histogram, both local and global information is taken into consideration. This is particularly helpful in taking care of small objects and local variation of color images. An efficient peak-finding algorithm is employed to identify the most significant peaks of the histogram. In the second phase, we perform histogram analysis on the color feature hue for each uniform region obtained in the first phase. We successfully remove about 99.7% singularity off the original images by redefining the hue values for the unstable points according to the local information. After the hierarchical segmentation is performed, a region merging process is employed to avoid over-segmentation. CIE(L*a*b*) color space is used to measure the color difference. Experimental results have demonstrated the effectiveness and superiority of the proposed method after an extensive set of color images was tested.     

A fully unsupervised color textured image segmentation algorithm using weighted mean histograms features

A new integrated feature distribution-based color textured image segmentation algorithm has been proposed. Two novel histogram-based inherent color texture feature extraction methods have been presented. From the histogram features, mean color texture histogram is calculated. Instead of concatenating the feature channels, a multichannel nonparametric Bayesean clustering is employed for primary segmentation. A region homogeneity-based merging algorithm is used for final segmentation. The proposed feature extraction techniques inherently combine color texture features rather then explicitly extracting it. Use of nonparametric Bayesean clustering makes the segmentation framework fully unsupervised where no a priori knowledge about the number of color texture regions is required. The feasibility and effectiveness of the proposed method have been demonstrated by various experiments using color textured and natural images. The experimental results reveal that superior segmentation results can be obtained through the proposed unsupervised segmentation framework.     

ADVISOR 2.1: a user-friendly advanced powertrain simulation using acombined backward/forward approach

ADVISOR 2.1 is the latest version of the National Renewable Energy Laboratory's advanced vehicle simulator. It was first developed in 1994 to support the US Department of Energy hybrid propulsion system program and is designed to be accurate, fast, flexible, easily sharable, and easy to use. This paper presents the model, focusing on its combination of forward- and backward-facing simulation approaches, and evaluates the model in terms of its design goals. ADVISOR predicts acceleration time to within 0.7% and energy use on the demanding US06 to within 0.6% for an underpowered series hybrid vehicle (0-100 km/h in 20 s). ADVISOR simulates vehicle performance on standard driving cycles between 2.6 and 8.0 times faster than a representative forward-facing vehicle model. Due in large part to ADVISOR's powerful graphical user interface and Web presence, over 800 users have downloaded ADVISOR from 45 different countries. Many of these users have contributed their own component data to the ADVISOR library     

激光夜视图像分型分割算法研究

为解决以往采用关联规则挖掘算法对图像进行分割时,对于夜视图像中灰暗区域中颜色特征以及前景/背景特征的采集能力差,不能有效判定图像的多尺度分型特征,分型分割效果差的问题。研究激光夜视图像分型分割算法。先利用计盒维数估计方法计算激光夜视图像分型维数尺度,通过分型维数尺度获取激光夜视图像的多尺度分型特征值,将利用多尺度分型特征值获取的多尺度分型特征约束与图像颜色约束相结合获取多尺度分型特征数据项,融合该多尺度分型特征数据项与通过图像中相邻区域顶点颜色距离获取的光滑项,并加入自适应比重系数获取能量函数,利用最大流/最小割算法求解能量函数最小值,实现激光夜视图像的分割。实验结果表明,该算法可准确分割激光夜视图像中人物目标特征,分割10幅激光夜视图像准确率以及均匀性测度平均值均在95%以上。     

基于改进的Otsu准则的递归图像分割算法

基于最大类间方差阈值图像分割算法的基本原理,然后结合目标与背景两类之间间距和类内距离对图像分割效果的影响,提出了一种改进的最大类间方差法,运用递归思想局部搜索图像的最佳阈值。这样不但缩短了计算时间,而且具有较好的自适应特点。该算法在图像背景不均匀或者图像的直方图不是简单的单峰、双峰图像的情况下可以进行有效的分割,分割后的图像细节更加丰富,能有效的去除噪声的干扰,有利于分割后的特征提取。本文对理论结果进行了仿真实验,获得了较好的分割效果。     

基于改进的Otsu准则的递归图像分割算法

基于最大类间方差阈值图像分割算法的基本原理,然后结合目标与背景两类之间间距和类内距离对图像分割效果的影响,提出了一种改进的最大类间方差法,运用递归思想局部搜索图像的最佳阈值。这样不但缩短了计算时间,而且具有较好的自适应特点。该算法在图像背景不均匀或者图像的直方图不是简单的单峰、双峰图像的情况下可以进行有效的分割,分割后的图像细节更加丰富,能有效的去除噪声的干扰,有利于分割后的特征提取。本文对理论结果进行了仿真实验,获得了较好的分割效果。     

基于区域生长的OCT图像分层算法

OCT(光学相干层析技术)自从1991年出现以来,在多个领域展现出了应用前景,尤其是在眼科疾病的检测方面起了重要作用。视网膜的分层,厚度检测就是利用OCT技术所得图像,对眼底的各种疾病的预防和治疗给出依据。然而,由于散斑噪声,低对比度,不规则的形态结构等原因,使得视网膜的分层检测存在不小的难度。为了解决这些问题,提高电脑对OCT图像处理的能力,文中提出了一种分为两步的图像分割算法。第一步是通过标记的分水岭算法进行初次分割;第二步是利用不同的区域特征值进行最优化合并,以获取需要的边界信息。实验结果表明,对于平滑的部分,分层有相对较好的结果,面对病变区域的处理仍有可改进的部分。     

基于形变模型的图像分割技术综述

基于形变模型的图像分割技术是近年来兴起的一种新型图像分割方法,已有相当广泛的研究。该技术为如何有效地从图像中分割出不规则对象及自然对象指出了一条佳径。该文简要介绍基于形变模型图像分割技术的基本原理和发展历程。按技术发展的线索介绍各种典型的形变模型表示形式,提出各种表示形式的优缺点,分析基于形变模型的图像分割的各种技术所存在的缺点,并建议了可能的研究方向。     

距离选通激光成像空间定位模糊C均值聚类分割法

针对距离选通激光成像对比度低、照度不均、图像模糊的特点,提出了一种基于空间定位的模糊C均值聚类方法（SPFCM）对目标进行分割。传统的模糊C均值聚类法存在以下缺点:一是需要预先获得目标分类数量,自适应性较差;二是对空间信息不敏感,导致目标轮廓不完整以及错误分类。针对上述缺陷,文中对传统算法进行了改进,引入了初定位的概念,首先利用最大类间方差法（Otsu 法）和数学形态学工具对子目标进行初步定位,再将其形心方位信息和灰度信息融合到聚类过程中,以较短的迭代过程实现不同目标的归类。实验结果证明基于空间定位的模糊C均值聚类法可以完整、有效地对距离选通激光图像进行提取分割,处理时间优于传统FCM。     

深度学习算法在颈动脉超声图像斑块分割中的应用研究

近年来,随着计算机技术的进步和数据集的大规模化,越来越多的人把计算机视觉技术应用到超声医学图像中.但在超声图像方面却存在低准确度且不稳定产生的模糊、伪影等使现有算法对模糊、噪声图像误判较高.另外由于病例过多,人为的去检测和识别斑块过于繁琐.为了缓解这些问题,提出了采用inception的网络结构方法快速准确地获取高噪声...     

照度不均的皮肤图像分割方法的研究

目的:针对采用普通的图像分割方法对照度不均的皮肤图像分割效果差的情况,提出了一种改进的图像分割方法.方法:首先将图像去背景,将图像划分为若干子块,接着采用OTSU和迭代阈值算法确定每个子图像块的最优阈值,利用此阈值完成对每个子块的分割,最后将子块合并,同时消除图像中噪声的影响.结果:采用改进的方法很好地实现了对照度不均皮肤图像的分割,其分割成功率可达95.5％.结论:在MATLAB平台上,改进方法能有效地实现对照度不均的皮肤表面图像的分割,并且在定性和定量评价中取得了良好的实验效果.     

基于SUSAN算法和行列均值分割的海面舰船检测

提出一种基于SUSAN检测和行列均值分割的复杂海天背景的红外舰船目标检测算法。首先建立一个简单的舰船轮廓模型,结合SUSAN算子检测得到感兴趣区域ROI,并且感兴趣区域的大小可以由轮廓模型进行一定的控制;然后对感兴趣区域运用改进的行列均值分割得到舰船目标。实验结果表明,提出的算法能正确分割出舰船目标。     

The application of competitive Hopfield neural network to medical image segmentation

In this paper, a parallel and unsupervised approach using the competitive Hopfield neural network (CHNN) is proposed for medical image segmentation. It is a kind of Hopfield network which incorporates the winner-takes-all (WTA) learning mechanism. The image segmentation is conceptually formulated as a problem of pixel clustering based upon the global information of the gray level distribution. Thus, the energy function for minimization is defined as the mean of the squared distance measures of the gray levels within each class. The proposed network avoids the onerous procedure of determining values for the weighting factors in the energy function. In addition, its training scheme enables the network to learn rapidly and effectively. For an image of n gray levels and c interesting objects, the proposed CHNN would consist of n by c neurons and be independent of the image size. In both simulation studies and practical medical image segmentation, the CHNN method shows promising results in comparison with two well-known methods: the hard and the fuzzy c-means (FCM) methods.     

A generalized Levinson algorithm for covariance extension with application to multiscale autoregressive modeling

Efficient computation of extensions of banded, partially known covariance matrices is provided by the classical Levinson algorithm. One contribution of this paper is the introduction of a generalization of this algorithm that is applicable to a substantially broader class of extension problems. This generalized algorithm can compute unknown covariance elements in any order that satisfies certain graph-theoretic properties, which we describe. This flexibility, which is not provided by the classical Levinson algorithm, is then harnessed in a second contribution of this paper, the identification of a multiscale autoregressive (MAR) model for the maximum-entropy (ME) extension of a banded, partially known covariance matrix. The computational complexity of MAR model identification is an order of magnitude below that of explicitly computing a full covariance extension and is comparable to that required to build a standard autoregressive (AR) model using the classical Levinson algorithm.     

An extension of the differential image foresting transform and its application to superpixel generation

The Image Foresting Transform (IFT) is a graph-based framework to develop image operators based on optimum connectivity between a root set and the remaining nodes, according to a given path-cost function. Its applications involve a variety of tasks, such as segmentation, boundary tracking, skeletonization, filtering, among others. The Differential Image Foresting Transform (DIFT) allows multiple IFT executions for different root sets and a same monotonically incremental path-cost function, making the processing time proportional to the number of modified nodes. In this paper, we extend the DIFT algorithm for non-monotonically incremental functions with root-based increases. This proposed extension, called Generalized DIFT (GDIFT), has been successfully used as the core part of some modern superpixels methods with state-of-the-art results. Experimental results show considerable efficiency gains over the sequential flow of IFTs for the generation of superpixels, also avoiding inconsistencies in image segmentation, which could occur with the regular DIFT algorithm.