Fully automatic segmentation of breast ultrasound images based on breast characteristics in space and frequency domains

Due to the complicated structure of breast and poor quality of ultrasound images, accurately and automatically locating regions of interest (ROIs) and segmenting tumors are challenging problems for breast ultrasound (BUS) computer-aided diagnosis systems. In this paper, we propose a fully automatic BUS image segmentation approach for performing accurate and robust ROI generation, and tumor segmentation. In the ROI generation step, the proposed adaptive reference point (RP) generation algorithm can produce the RPs automatically based on the breast anatomy; and the multipath search algorithm generates the seeds accurately and fast. In the tumor segmentation step, we propose a segmentation framework in which the cost function is defined in terms of tumor?s boundary and region information in both frequency and space domains. First, the frequency constraint is built based on the newly proposed edge detector which is invariant to contrast and brightness; and then the tumor pose, position and intensity distribution are modeled to constrain the segmentation in the spatial domain. The well-designed cost function is graph-representable and its global optimum can be found. The proposed fully automatic segmentation method is applied to a BUS database with 184 cases (93 benign and 91 malignant), and the performance is evaluated by the area and boundary error metrics. Compared with the newly published fully automatic method, the proposed method is more accurate and robust in segmenting BUS images.     

A marker-based watershed method for X-ray image segmentation

Digital X-ray images are the most frequent modality for both screening and diagnosis in hospitals. To facilitate subsequent analysis such as quantification and computer aided diagnosis (CAD), it is desirable to exclude image background. A marker-based watershed segmentation method was proposed to segment background of X-ray images. The method consisted of six modules: image preprocessing, gradient computation, marker extraction, watershed segmentation from markers, region merging and background extraction. One hundred clinical direct radiograph X-ray images were used to validate the method. Manual thresholding and multiscale gradient based watershed method were implemented for comparison. The proposed method yielded a dice coefficient of 0.964 ± 0.069, which was better than that of the manual thresholding (0.937 ± 0.119) and that of multiscale gradient based watershed method (0.942 ± 0.098). Special means were adopted to decrease the computational cost, including getting rid of few pixels with highest grayscale via percentile, calculation of gradient magnitude through simple operations, decreasing the number of markers by appropriate thresholding, and merging regions based on simple grayscale statistics. As a result, the processing time was at most 6 s even for a 3072 × 3072 image on a Pentium 4 PC with 2.4 GHz CPU (4 cores) and 2G RAM, which was more than one time faster than that of the multiscale gradient based watershed method. The proposed method could be a potential tool for diagnosis and quantification of X-ray images.     

分割多发性硬化症白质病灶的新方法

多发性硬化症(MS)是一种慢性的中枢神经系统疾病,其病灶可由常规脑部核磁共振成像(cMRI)进行检测。为提高图像处理的效率,提出了一种自动分割cMRI图像中的MS白质病灶(WML)的新方法。首先将模糊核聚类（KFCM)用于预处理后的T1加权像,得到白质图像;然后利用一个种子点的区域生长处理白质图像,提取出一个二值模板。该模板与对应的T2加权像进行乘积,得到一幅仅包含白质、病灶及背景的图像;最后再次利用KFCM分割图像,得到病灶的核心部分。实验结果表明,所提出的方法能快速、有效地分割出低噪声仿真图像中的WML,且Dice相似性系数平均值在80%以上。     

一种新的粘连字符图像分割方法

针对监控画面采样图像中数字的自动识别问题,提出一种新的粘连字符图像分割方法。该方法以预处理后二值图像的连通状况来判定字符粘连的存在,并对粘连字符图像采用上下轮廓极值法确定候选粘连分割点,以双向最短路径确定合适的图像分割线路。仿真实验表明,该方法能有效解决粘连字符图像的分割问题。     

A diffusion stick method for speckle suppression in ultrasonic images

This paper presents a diffusion stick method for speckle suppression in ultrasonic images. An asymmetric stick filter kernel is firstly defined by decomposing the rectangle neighborhood into line segments of variable orientations. Then, the weighted sum of averages along each stick is used to produce the filtered images. Implemented in an iteration scheme, our method works as a pseudo-diffusion process, where the diffusivity is controlled by a normalized variance function. Experiments of synthetic and real images show that the diffusion stick technique performs effectively in suppressing speckle noise, preserving resolvable structures and enhancing linear features.     

多相图像分割Vese-Chan模型连续最大流方法

目的 多相图像分割是图像处理与分析的重要问题,变分图像分割的Vese-Chan模型是多相图像分割的基本模型,由于该模型使用较少的标签函数构造区域划分的特征函数,具有求解规模小的优点。图割（graph cut,GC）算法可将上述能量泛函的极值问题转化为最小割/最大流问题求解,大大提高了计算效率。连续最大流（continuous max-flow,CMF）方法是经典GC算法的连续化表达,不仅具备GC算法的高效性,且克服了经典GC算法由于离散导致的精度下降问题。本文提出基于凸松弛的多相图像分割Vese-Chan模型的连续最大流方法。方法 根据划分区域编号的二进制表示构造两类特征函数,将多相图像分割转化为多个交替优化的两相图像分割问题。引入对偶变量将Vese-Chan模型转化为与最小割问题相对应的连续最大流问题,并引入Lagrange乘子设计交替方向乘子方法（alternating direction method of multipliers,ADMM）,将能量泛函的优化问题转化为一系列简单的子优化问题。结果 对灰度图像和彩色图像进行数值实验,从分割效果看,本文方法对于医学图像、遥感图像等复杂图像的分割效果更加精确,对分割对象和背景更好地分离;从分割效率看,本文方法减少了迭代次数和运算时间。在使用2个标签函数的分割实验中,本文方法运算时间加速比分别为6.35%、10.75%、12.39%和7.83%;在使用3个标签函数的分割实验中,运算时间加速比分别为12.32%、15.45%和14.04%;在使用4个标签函数的分割实验中,运算时间加速比分别为16.69%和20.07%。结论 本文提出的多相图像分割Vese-Chan模型的连续最大流方法优化了分割效果,减少了迭代次数,从而提高了计算效率。     

A geodesic-active-contour-based variational model for short-axis cardiac MR image segmentation

We propose a novel geodesic-active-contour-based (GAC-based) variational model that uses two level-set functions to segment the right and left ventricles and the epicardium in short-axis magnetic resonance (MR) images. For the right ventricle, the myocardial wall is typically very thin and hard to identify using the resolution of existing MR scanners. We propose to use two level sets to identify both the endocardial wall by pushing away one level-set function from another, in the setting of the edge-driven GAC model with a new edge detection function. Existing edge detection functions have strict restrictions on the location of initial contours. We develop a new edge detection function that relaxes this restriction and propose an iterative method that uses a sequence of edge detection functions to minimize the energy of our model successively. Experimental results are presented to validate the effectiveness of the proposed model.     

A novel method for speckle noise reduction and ship target detection in SAR images

It is very difficult to detect small targets when the scattering intensity of background clutter is as strong as the targets and the speckle noise is serious in synthetic aperture radar (SAR) images. Because the scattering of man-made objects lasts for a longer time than that of background clutter in azimuth matching scope, it is much easier for man-made objects to produce strong coherence than ground objects. As the essence of SAR imaging is coherent imaging, the contrast between targets and background clutter can be enhanced via coherent processing of SAR images. This paper proposes a novel method to reduce speckle noise for SAR images and to improve the detected ratio for SAR ship targets from the SAR imaging mechanism. This new method includes the coherence reduction speckle noise (CRSN) algorithm and the coherence constant false-alarm ratio (CCFAR) detection algorithm. Real SAR image data is used to test the presented algorithms and the experimental results verify that they are feasible and effective.     

注意力机制下密集空洞卷积的肺部图像分割

目的 卷积神经网络（convolutional neural network,CNN）在计算机辅助诊断（computer-aided diagnosis,CAD）肺部疾病方面具有广泛的应用,其主要工作在于肺部实质的分割、肺结节检测以及病变分析,而肺实质的精确分割是肺结节检出和肺部疾病诊断的关键。因此,为了更好地适应计算机辅助诊断系统要求,提出一种融合注意力机制和密集空洞卷积的具有编码—解码模式的卷积神经网络,进行肺部分割。方法 将注意力机制引入网络的解码部分,通过增大关键信息权重以突出目标区域抑制背景像素干扰。为了获取更广更深的语义信息,将密集空洞卷积模块部署在网络中间,该模块集合了Inception、残差结构以及多尺度空洞卷积的优点,在不引起梯度爆炸和梯度消失的情况下,获得了更深层次的特征信息。针对分割网络常见的特征丢失等问题,对网络中的上/下采样模块进行改进,利用多个不同尺度的卷积核级联加宽网络,有效避免了特征丢失。结果 在LUNA （lung nodule analysis）数据集上与现有5种主流分割网络进行比较实验和消融实验,结果表明,本文模型得到的预测图更接近于标签图像。Dice相似系数、交并比（intersection over union,IoU）、准确度（accuracy,ACC）以及敏感度（sensitivity,SE）等评价指标均优于对比方法,相比于性能第2的模型,分别提高了0.443%,0.272%,0.512%以及0.374%。结论 本文提出了一种融合注意力机制与密集空洞卷积的肺部分割网络,相对于其他分割网络取得了更好的分割效果。     

Cost-effective design for injection molding

It is commonly agreed that a large proportion of the ultimate product cost is determined at product design stage. Therefore, a cost-effective design cannot be obtained unless all cost issues are resolved at early design stage. Therefore, instead of performing cost estimation after design, research presented in this paper aims to provide on-line cost evaluation and advisory to help product designers avoid cost-ineffective design. The objective can be obtained by (1) identifying factors that might affect product cost at each product design stage, (2) developing a design for cost effectiveness methodology that accommodates the concepts of concurrent engineering, and (3) developing a computer-based design for cost effectiveness system based on the proposed methodology. In this research, we focus on injection molding product design due to the advantages of injection molding process, such as high production rates, excellent quality and accuracy of the parts, and very long mold life. This paper first reviews and characterizes the conventional molding product development process with an emphasis on the identification of cost factors. Based on the results of process characterization, a cost model is developed, which depicts the relationships between cost factors and product development activities, as well as their relationships with product geometry. According to the product life cycle activities and the cost model, a design for cost effectiveness process is proposed. The process and the cost model are then employed for the development of a computer-based product design for cost effectiveness as one of the module of an integrated design for injection molding environment.     

改进CV模型图像分割的Split-Bregman方法

水平集方法中的Chan-Vese模型（简称CV模型）对灰度不均匀及边界对比度低的图像的分割效果不够精确,计算效率也不是很高。针对灰度不均匀引入偏差场来修正CV模型中的区域平均灰度并引入核函数来加权能量泛函。针对计算效率低下的问题,在上述基础上得出其全局凸分割模型（Global Convex Segmentation,GCS）,用Split-Bregman迭代求解该模型。实验结果表明：改进后的模型提高了分割精确度和计算效率。     

A coarse-to-fine approach for segmenting melanocytic skin lesions in standard camera images

Melanoma is a type of malignant melanocytic skin lesion, and it is among the most life threatening existing cancers if not treated at an early stage. Computer-aided prescreening systems for melanocytic skin lesions is a recent trend to detect malignant melanocytic skin lesions in their early stages, and lesion segmentation is an important initial processing step. A good definition of the lesion area and its border is very important for discriminating between benign and malignant cases. In this paper, we propose to segment melanocytic skin lesions using a sequence of steps. We start by pre-segmenting the skin lesion, creating a new image representation (channel) where the lesion features are more evident. This new channel is thresholded, and the lesion border pre-detection is refined using an active-contours algorithm followed by morphological operations. Our experimental results based on a publicly available dataset suggest that our method potentially can be more accurate than comparable state-of-the-art methods proposed in literature.     

A reversible watermarking scheme for two-dimensional CAD engineering graphics based on improved difference expansion

Difference expansion is widely used in reversible watermarking for raster images, but it cannot be applied to data with low correlation, due to its large distortion. In this paper, an improvement is made to difference expansion technology to make it suitable for two-dimensional (2D) computer-aided design (CAD) engineering graphics. Based on improved difference expansion (IDE), a reversible watermarking scheme for 2D CAD engineering graphics is proposed. First, all vertices of the 2D CAD engineering graphics are extracted, and then the relative coordinates of these vertices are calculated. Finally, the encrypted watermark is embedded into the scale factor of the relative coordinates by using IDE technology. Experimental results show that the proposed scheme has a large capacity, good imperceptibility, and is robust against operations such as transformation, rotation, and equal scaling.     

Detection of microaneurysms using multi-scale correlation coefficients

This paper presents a new approach to the computer aided diagnosis (CAD) of diabetic retinopathy (DR)—a common and severe complication of long-term diabetes which damages the retina and cause blindness. Since microaneurysms are regarded as the first signs of DR, there has been extensive research on effective detection and localization of these abnormalities in retinal images. In contrast to existing algorithms, a new approach based on multi-scale correlation filtering (MSCF) and dynamic thresholding is developed. This consists of two levels, microaneurysm candidate detection (coarse level) and true microaneurysm classification (fine level). The approach was evaluated based on two public datasets—ROC (retinopathy on-line challenge, http://roc.healthcare.uiowa.edu) and DIARETDB1 (standard diabetic retinopathy database, http://www.it.lut.fi/project/imageret/diaretdb1). We conclude our method to be effective and efficient.     

On stochastic finite element method for linear elastostatics by the Taylor expansion

The main aim of the paper is to present an application of the Taylor expansion in formulation and computational implementation of the perturbation-based stochastic finite element method. Random-input parameters as well as all-state functions included in static equilibrium equations are expanded in this approach around their expectations via Taylor series up the order given a priori. It further enables a dual computational approach for determination of probabilistic moments of the state functions—a formation and the solution of increasing order equilibrium equations and, on the other hand, polynomial approximation of deterministic state functions with respect to a given input random parameter. Theoretical and technical details of such methodology are explained also; some elementary engineering application with analytical solution is available to derive explicitly fundamental probabilistic moments of the resulting state function.     

快速稳定的局部二元拟合分割算法

基于局部区域信息的局部二元拟合(LBF)模型在处理弱边界或灰度不均匀的图像分割方面有一定优势,但该方法非常依赖于初始轮廓,不当的初始轮廓不仅会导致分割时间较长,甚至分割失败。针对这一不足,提出一种快速稳定的LBF模型。首先通过添加带有变权系数面积项的LBF模型进行初始分类以获取较好的初始轮廓,然后采用传统的LBF模型对图像进行进一步的分割。实验证明,在保证良好分割效果的前提下,该方法对初始轮廓的选择更加灵活,分割速度明显快于传统的LBF模型。     

Incomplete series expansion for function approximation

We present an incomplete series expansion (ISE) as a basis for function approximation. The ISE is expressed in terms of an approximate Hessian matrix, which may contain second, third, and even higher order “main” or diagonal terms, but which excludes “interaction” or off-diagonal terms. From the ISE, a family of approximation functions may be derived. The approximation functions may be based on an arbitrary number of previously sampled points, and any of the function and gradient values at suitable previously sampled points may be enforced when deriving the approximation functions. When function values only are enforced, the storage requirements are minimal. However, irrespective of the conditions enforced, the approximate Hessian matrix is a sparse diagonal matrix. In addition, the resultant approximations are separable. Hence, the proposed approximation functions are very well-suited for use in gradient-based sequential approximate optimization requiring computationally expensive simulations; a typical example is structural design problems with many design variables and constraints. We derived a wide selection of approximations from the family of ISE approximating functions; these include approximations based on the substitution of reciprocal and exponential intervening variables. A comparison with popular approximating functions previously proposed illustrates the accuracy and flexibility of the new family of approximation functions. In fact, a number of popular approximating functions previously proposed for structural optimization applications derive from our ISE. Based on the similarly named paper presented at the Sixth World Congress on Structural and Multidisciplinary Optimization, Rio de Janeiro, Brazil, May 2005     

A narrow band graph partitioning method for skin lesion segmentation

Accurate skin lesion segmentation is critical for automated early skin cancer detection and diagnosis. In this paper, we present a novel multi-modal skin lesion segmentation method based on region fusion and narrow band energy graph partitioning. The proposed method can handle challenging characteristics of skin lesions, such as topological changes, weak or false edges, and asymmetry. Extensive testing demonstrated that in this method complex contours are detected correctly while topological changes of evolving curves are managed naturally. The accuracy of the method was quantified using a lesion similarity measure and lesion segmentation error ratio, Our results were validated using a large set of epiluminescence microscopy (ELM) images acquired using cross-polarization ELM and side-transillumination ELM. Our findings demonstrate that the new method can achieve improved robustness and better overall performance compared to other state-of-the-art segmentation methods.     

基于语种相似性挖掘的神经机器翻译语料库扩充方法

针对低资源语言机器翻译任务上一直存在的标注数据资源匮乏问题,提出了基于语种相似性挖掘的神经机器翻译语料库扩充方法。首先,将维吾尔语和哈萨克语作为相似语言对并将其语料进行混合;然后,对混合后的语料分别进行字节对编码（BPE）处理、音节切分处理以及基于音节切分的BPE处理,从而深度挖掘哈语和维语的相似性;最后,引入“开始-中部-结束（BME）”序列标注方法对语料中已切分完成的音节进行标注,以消除音节输入所带来的一些歧义。在CWMT2015维汉平行语料和哈汉平行语料上的实验结果表明,所提方法相较于不进行特殊语料处理以及BPE语料处理训练所得普通模型在维吾尔语-汉语翻译上的双语评估替补（BLEU）值分别提升了9.66、4.55,在哈萨克语-汉语翻译上的BLEU值分别提升了9.44、4.36。所提方案实现了维语和哈语到汉语的跨语言神经机器翻译,提升了维吾尔语-汉语和哈萨克语-汉语机器翻译的翻译质量,可应用于维语和哈语的语料处理。     

基于数学形态学的SAR图像分割方法

图像分割是遥感图像处理中很重要的一步。因SAP图像通常带有较强的嗓声，用传统的边缘检测方法效果不理想。作者利用数学形态学开闭运算和混合滤波，可据目标的形状选用算法中的探针，取得了较好的滤波去噪和目标分割的效果。