Liver Hydatid CT Image Segmentation Using Smoothed Bayesian Classification Method and Modified Parametric Active Contour Model

Liver hydatid disease is a common parasitic disease in farm and pastoral areas, which seriously influences people's health. Based on CT imaging features of this disease, an iterative approach for liver segmentation and hydatid lesion extraction simultaneously is proposed. In each iteration, our algorithm consists of two main steps: 1) according to the user-defined pixel seeds in the liver and hydatid lesion, Gaussian probability model fitting and smoothed Bayesian classification are applied to get initial segmentation of liver and lesion; 2) the parametric active contour model using priori shape force field is adopted to refine initial segmentation. We make subjective and objective evaluation on the proposed algorithm validity by the experiments of liver and hydatid lesion segmentation on different patients' CT slices. In comparison with ground-truth manual segmentation results, the experimental results show the effectiveness of our method to segment liver and hydatid lesion.

Application of Active Contour Model in Tracking Sequential Nearshore Waves

In the present study, a generalized active contour model of gradient vector flow is combined with the video techniques of Argus system to delineate and track sequential nearshore wave crest profiles in the shoaling process, up to their breaking on the shoreline. Previous applications of active contour models to water wave problems are limited to controllable wave tank experiments. By contrast, our application in this study is in a nearshore field environment where oblique images obtained under natural and varying condition of ambient light are employed. Existing Argus techniques produce plane image data or time series data from a selected small subset of discrete pixels. By contrast, the active contour model produces line image data along continuous visible curves such as wave crest profiles. The combination of these two existing techniques, the active contour model and Argus methodologies, facilitates the estimates of the direction wave field and phase speeds within the whole area covered by camera. These estimates are useful for the purpose of inverse calculation of the water depth. Applications of the present techniques to Hsi-tzu bay where a beach restoration program is currently undertaken are illustrated. This extension of Argus video techniques provides new application of optical remote sensing to study the hydrodynamics and morphology of a nearshore environment.

Extraction of coastline in high-resolution remote sensing images based on the active contour model

While executing tasks such as ocean pollution monitoring,maritime rescue,geographic mapping,and automatic navigation utilizing remote sensing images,the coastline feature should be determined.Traditional methods are not satisfactory to extract coastline in high-resolution panchromatic remote sensing image.Active contour model,also called snakes,have proven useful for interactive specification of image contours,so it is used as an effective coastlines extraction technique.Firstly,coastlines are detected by water segmentation and boundary tracking,which are considered initial contours to be optimized through active contour model.As better energy functions are developed,the power assist of snakes becomes effective.New internal energy has been done to reduce problems caused by convergence to local minima,and new external energy can greatly enlarge the capture region around features of interest.After normalization processing,energies are iterated using greedy algorithm to accelerate convergence rate.The experimental results encompassed examples in images and demonstrated the capabilities and efficiencies of the improvement.

Active contour model based measurement of distance of material serial sections

The measurement of thickness of material removed between serial sections is a crucial step of three-dimensional reconstruction. Active contour model is an efficient method for contour detection of objects on an image. Based on the segmentation of the FeAl/ZrO2 composite image by using adaptive threshold, the gradient vector flow (GVF) snake was used to detect the contour of the indent. The horizontal diagonal length and the vertical diagonal length of the indent contour were acquired by measuring the distance from the uppermost snaxel to the lowermost snaxel and that from the leftmost snaxel to the rightmost snaxel respectively. Then the final diagonal length was gotten by averaging the vertical diagonal length and the horizontal diagonal length. The Vickers indenter was made by a square pyramidal-shaped diamond with opposite faces at an angle of 136°, so the geometrical relation was established between the thickness of material removed between two successive serial sections and the difference of diagonal length on the two serial sections. Based on the relation, the thickness of material removed between two successive serial sections was calculated using the two successive diagonals.

Target Region Location Based on Texture Analysis and Active Contour Model

Traditional texture region location methods with Gabor features are often limited in the selection of Gabor filters and fail to deal with the target which contains both texture and non-texture parts.Thus,to solve this problem,a two-step new model was proposed.In the first step,the original features extracted by Gabor filters are applied to training a self-organizing map(SOM) neural network and a novel merging scheme is presented to achieve the clustering.A back propagation(BP) network is used as a classifier to locate the target region approximately.In the second step,Chan-Vese active contour model is applied to detecting the boundary of the target region accurately and morphological processing is used to create a connected domain whose convex hull can cover the target region.In the experiments,the proposed method is demonstrated accurate and robust in localizing target on texture database and practical barcode location system as well.

一种由内向外的Active Contour模型图像分割算法

针对主动轮廓模型图像分割,提出一种改进算法。该算法首先选取边界内部点为轮廓初始点,再根据能量与面积的关系,将传统算法中初始点向边界点收缩改进为由初始点向边界点膨胀。算法的实现采用原始模型和改进模型相互交替的贪婪算法,保证了模型的稳定性特点。实验结果表明:这种由内向外的分割算法比传统模型分割效果好及适用范围广;避免了传统模型对凹陷区域分割效果不理想及边界外部复杂环境对分割效果的影响。

基于多层显著性模型的SAR图像舰船目标检测

针对合成孔径雷达图像舰船目标检测问题,提出了一种结合选择机制与轮廓信息的多层显著性目标检测方法。首先,利用非下采样剪切波和频谱残差法进行全局显著性区域提取。其次,提出了一种基于动态恒虚警率的活动轮廓显著性模型,逐步滤除候选区域的虚警,提取目标轮廓,从而实现目标的精确检测。所提方法能够由粗到细地快速捕获目标区域,从而实现高效、高分辨率合成孔径雷达图像舰船检测。最后,在真实SAR数据集进行了测试,与其他经典的舰船检测方法相比,所提算法不仅有效地抑制了海杂波的影响,而且在检测精度上有较大提高。

基于窄带成像及放大内镜的脂质分割方法

一种主要成分是脂质的白色不透明物质(WOS)会覆盖与癌症诊断有关的微观结构,但WOS的形态特征又与肿瘤分级有密切关系。为了给医生提供更多与脂质相关的可用信息,本文对脂质图像的分割方法进行了研究。首先,介绍了基于Retinex框架的脂质图像增强算法,并介绍了反光去除算法。然后,介绍了基于活动轮廓模型的脂质分割方法,该方法从校正后的色调值中提取局部信息,从强度值中提取全局信息,自适应地获得权重因子,并基于初始轮廓来分割脂质区域。最后,基于自研细胞内镜成像系统,设计了仿体实验来验证了该方法的有效性。实验结果表明,该分割方法的像素准确度、灵敏度、Dice系数均高于90%。该方法能够克服照明不均匀、反光等的影响,很好地反映脂质的形状,为医生提供可用的信息。

活动轮廓模型和Contourlet多分辨率分析分割血管内超声图像

针对传统图像分割方法中初始化和鲁棒性两个问题,研究了基于活动轮廓模型和Contourlet多分辨率分析分割血管内超声斑块图像的新方法。该方法运用Contourlet变换将原图像分解为多分辨率低通分量和多分辨率带通分量方向性子带。对低通分量进行模板匹配,确定血管内腔边界和中-外膜边界的初始轮廓;对带通分量方向性子带进行扩散滤波,抑制噪声的同时尽可能保留有用边缘,并结合边界矢量场使轮廓演化得到最终分割结果,从而提高了分割算法的鲁棒性。对100幅仿真图像和120幅实际图像的分割结果表明,相对于传统活动轮廓模型,该方法分割实际图像的平均距离误差提高了3.04 pixel,面积差异百分比提高了6.30%。表明该方法能自动、精确地提取血管的两条边界。

局域化互信息度量的ACM下医学图像的分割

大量医学图像中存在灰度不均匀现象使传统方法很难获得理想的分割结果,针对此问题,将图像的局部统计特征引入互信息度量的分割模型中,考虑不同组织间的方差差异,提出一种局域化互信息度量的活动轮廓模型(ACM),以提高灰度不均匀情况下目标边界识别的精确度。此外,采用一种无需重新初始化的水平集函数规则化方法,演化稳定,收敛速度快。最后,以医学图像分割实验对算法进行了验证,对比实验表明分割和偏移场矫正结果都更精确。

填装材料CT图像中脱粘缺陷识别与深度测量

脱粘缺陷是影响填装材料物理安全性能的重要指标,计算机层析成像(CT)是进行脱粘缺陷检测的有效方法。但填装材料脱粘缺陷紧贴外轮廓、面积小,对比度低,分割时易受其他信息干扰,传统活动轮廓模型难以适用。本研究提出了一种基于SoftMax和正则项的Chan-Vese (SRCV)模型用于填装材料脱粘缺陷分割,系统比较了SRCV模型与多种活动轮廓模型对脱粘缺陷分割效果的差异,利用不同图片分割效果揭示了SRCV模型的抗干扰能力和兼顾全局与局部信息的能力,利用分割曲线的欧氏距离实现脱粘深度测量。研究结果表明,SRCV模型对填装材料脱粘缺陷分割更贴近脱粘边缘,分割曲线更光滑,准确度达99.56%,Dice系数为99.82%,脱粘深度误差不超过6%,特别适用于有大量干扰信息的微小脱粘缺陷的分割,较其他活动轮廓模型具有明显的优势。

基于FVFACM的JV肺结节分割

对几种活动轮廓模型肺结节分割方法进行了改进,使用图像中每一个像素点的模糊隶属度代替传统模型的灰度信息,构造模糊速度函数,将该函数替代传统轮廓模型中基于梯度信息的停止函数。

基于L_(1)范数混合主动轮廓的河流SAR图像分割

为解决现有主动轮廓模型难以准确分割河流SAR图像的问题,提出一种基于L_(1)范数的混合主动轮廓模型。首先,计算轮廓曲线内外区域像素灰度的中值作为区域拟合中心,以抑制SAR图像中干扰区域对其准确性的影响;然后,利用L_(1)范数构建新的能量约束项并在模型能量泛函中引入边缘指示函数,进一步提升模型的分割性能;最后,将基于L_(1)范数的中值和均值能量约束项结合起来并添加额外的区域拟合中心约束项,以提高模型的整体稳定性。针对实际河流SAR图像进行分割试验,结果表明,与现有分割方法相比,本文模型能更准确、稳定地分割河流SAR图像。

基于加权混合SPF主动轮廓的SAR图像水陆分割

针对现有主动轮廓模型(Active Contour Model,ACM)无法进行准确高效的SAR(Synthetic Aperture Radar)图像水陆分割的难题,提出一种基于加权混合符号压力(Signed Pressure Force,SPF)主动轮廓的新模型.该模型将中值拟合中心和均值拟合中心结合起来,建立新的混合符号压力函数,以抑制干扰边缘的影响.通过设计基于绝对中位差的自适应权重,实现自动调节内外拟合中心的比例系数,以更合理地描述灰度差异.此外,计算平滑后图像梯度信息构建边缘引导函数并引入到SPF函数中,以进一步提高分割性能和演化效率.在实际SAR图像上进行分割实验,结果表明:与相关实验模型相比,提出模型在保持高效的同时能获得更加准确的分割结果,其中Dice相似系数(Dice Similarity Coefficient,DSC)至少提高6%.

基于多尺度显著性检测的SAR图像海岸线检测

SAR图像岸线检测是SAR近岸海洋目标检测的一项重要环节。提出了一种基于多尺度显著性检测的SAR图像海岸线检测方法,用以检测低对比度SAR图像的海岸线。首先将SAR图像依据尺度系数进行多尺度变换,然后对于每种尺度图像进行谱残差法显著性检测,得到一系列显著性子图;而后应用NSCT变换融合各显著子图得到最终显著图,将显著图代入到活动轮廓模型中进行检测,得到检测结果。SAR图像实验结果表明,该方法相比于传统方法,提升了对于低对比度图像的适应性,增强了抗噪性能。

基于LoG和因子分解的集成ACM的亚实性肺结节分割

针对亚实性肺结节(SSN)边缘模糊、结构复杂、区域亮度不均等问题,传统的活动轮廓模型(ACM)方法难以达到高精度的分割结果。因此,提出一种优化的LoG算子和因子分解结合的集成ACM分割算法。用高斯拉普拉斯算子(LoG)处理肺实质图像,对算子进行优化并构建能量项,以加强边界并增强区域亮度;以像素值作为描述特征构建因子分解能量项,使曲线演变到目标边缘;将LoG能量项和因子分解能量项集成到LGIF模型中对SSN分割。实验结果证实,该算法模型对SSN的分割更有效。

基于活动轮廓模型的SAR图像海岸线检测

SAR图像岸线检测是SAR近岸海洋目标检测的一项重要环节。该文提出一种SAR图像海岸线检测方式,该方法基于自适应混合活动轮廓模型,旨在对灰度不均匀SAR图像的海岸线进行检测。首先对SAR图像进行显著性检测的预处理,然后将预处理后的结果代入混合活动轮廓模型中进行检测。该文采用改良后的CV模型和LIF模型并以其各自的自适应参数为基础进行融合,从而增强整个模型的识别性能。从SAR图象试验的结果来看,这种方式相较于传统的CV模型技术,对灰度分布不均一的SAR图像具有更强的适用性和计算速度上的优化。

融合U-Net 3+和ACM的医学图像分割模型

活动轮廓模型方法不需要训练集且能较好利用边缘信息,但对初始轮廓较敏感,在处理复杂背景图像时分割不够精确。U-Net 3+网络可以分割更复杂的医学图像,但需要大量的人工标记,且模型的特征提取机制导致其在非典型边界特征的决策时通常是不准确的。因此,针对训练集较小的医学图像,提出了一种融合卷积神经网络和活动轮廓模型的医学图像自动分割模型。模型通过U-Net 3+网络获得目标先验信息,使用先验信息构造拟合能量项,并融合到活动轮廓模型中约束曲线演化。在皮肤镜病变和胸部X光片图像上测试,该模型的分割精度高于单独使用U-Net 3+网络和活动轮廓模型的分割结果。

基于超像素分割算法的景观边缘提取仿真

由于遥感图像景观边缘具有复杂非线性结构,受光照与场景影响,使得图像边缘存在极多样化纹理和噪声,导致景观边缘难以捕捉。于是提出基于超像素分割算法的景观边缘提取方法。采用超像素分割算法,将图像映射至CIE-LAB(International Commission on Illumination-Lightness-L*A*B)色彩空间中,利用最小化梯度,将超像素置于梯度对应区域中,获取标准化度量,将水平与垂直梯度加入特征矢量中,组建景观弱边缘超像素分割模型,分割景观弱边缘超像素,利用高斯滤波器,模糊处理超像素分割结果,构建能量函数,通过迭代处理,实现景观边缘提取。实验结果表明,所提方法能准确分割显著边缘与细节边缘,精准提取景观边缘,且边缘线连贯、平滑,有效避免了伪轮廓问题,对实现城市与景观的合理规划具有较高的实际应用价值。

基于变分水平集的多尺度纹理图像分割方法

为了更有效地对纹理图像进行多尺度分割,我们提出了一种混合模型。该模型使用全变分模型来控制纹理图像的多尺度表示,并且引入局部自相似性以及局部高斯分布拟合函数作为边界提取项,最终得到一个混合的多尺度纹理图像分割模型。实验结果表明,所提模型能够在不同尺度下对纹理图像进行较好地分割,使分割过程更加符合人类视觉的感知机制,并且当图像受到不同程度的高斯噪声污染时,该模型也能够对目标边界进行精确地提取。