基于自适应阈值与圆形相似度的禁令标志检测

交通标志检测是影响交通标志识别速度和正确率的重要因素。采用一种 HSI自适应阈值粗分割方法获取场景图像的红色二值图,克服固定阈值分割算法受光照等外界因素影响较大的缺点。然后进行连通区域标记,排除了部分非禁令标志红色区域,并对多个相连禁令标志作有效分割,提取出禁令标志的候选区域。最后,通过分析候选区域外边缘的圆形相似度参数,确定了禁令标志区域。复杂道路环境下的场景实验表明：本方法具有较高的鲁棒性。     

基于二邻域局部阈值的矾花图像分割

本文提出一种采用二邻域类间方差局部阈值化算法对矾花图像进行分割，实践证明该方法非常有效，能基本解决常规局部阈值分割中出现的问题。     

基于多分辨率阈值的非均匀光照微视觉图像实时分割技术研究

为了满足微纳和生物医学领域对微视觉图像目标信息高精度实时提取的需求,提出一种基于多分辨率阈值的非均匀光照微视觉图像实时分割技术.针对传统方法所存在的在非均匀光照微视觉图像分割过程中难以精确估计阈值和实时性差等问题,首先根据微视觉图像的梯度概率密度稀疏分布特性,建立灰度强度分布估计目标函数,并利用迭代加权最小二乘法实现灰...     

利用超声波图像检测提取钻孔参数

利用超声波对钻成孔进行检测,可以直观地给出钻孔的垂直度、孔径等参数。但是,目前的超声波钻孔检测设备只能给出钻孔的观测图像,而要得到孔径和垂直度等参数,需要从回放的数据中读取（比如中科院东海研究站研制的UDM100Q钻孔检测仪）,或从打印输出的图纸上测量（比如进口的超声波钻孔检测仪）。通过对超声波成孔检测记录图用数字图像处理方法进行处理,提取出钻孔的孔壁线,并使其坐标映射到实际测量的坐标范围,从而能通过提取壁线的坐标计算出孔径、垂直度等参数。     

一种边界-区域相结合的图像分割算法

提出了一种模仿人类视觉机制的区域-细节的图像分割算法.首先提取图像边缘,然后将边缘分段切割,得到端点集合,然后从端点集合生成Delaunay三角形网络,以Delatmay三角形为顶点,相邻三角形的属性差异作为边的权重,构造图;9以基于图的分割算法生成最小生成树,划分区域.最后用Snake模型精确确定区域边界,生成准确的区域边缘.实验证明,这种区域分割和边缘检测相结合的方法能准确地分割非纹理图像,较好地克服了块现象和非连续边界,相比单一区域分割或者边缘检测方法有更好的分割结果,并且计算速度比较快.     

超声手术刀的优化设计

为适应微创手术的要求,设计了频率为55.5 kHz的细长结构超声手术刀。采用有限元方法对其进行模态分析和谐振分析,获取振幅大小、应力分布等参数。通过对手术刀刀身结构进行二次优化,使所设计的手术刀在50~60 kHz频段内以纵振模式为主,刀头和刀身应力分布均匀,最大应力由577 MPa降到375 MPa,远小于医用钛合金材质的超声手术刀的最大抗拉强度。为验证有限元分析结果,分别试制结构优化前后的手术刀模型并进行实验。结果表明:结构优化之后的手术刀工作时,应力分布均匀,径向无明显扰度,轴向输出振幅可以满足生物组织切割的要求。     

K-均值聚类中心分析法实现红外人体目标分割

针对由于不同红外成像设备参数差异以及目标周围环境影响而造成的红外目标分割阈值自动选取算法的鲁棒性差这一问题,本文从红外成像的机理出发,提出了一个新的解决方案并加以实现.首先对图像的直方图采用K-均值聚类,然后通过对聚类中心分布特点的分析,确定图像分割的阈值.该方法不需要事先对图像进行均衡和对背景分布进行假设.实验结果表明,算法具有良好的鲁棒性.     

新的Otsu阈值改进方法的红外小目标检测

针对红外小目标区域灰度对比度很差的现象,以及红外小目标实时性检测的要求,在对Otsu阈值方法进行研究的基础上,结合红外小目标图像的自身特点,提出了一种新的Otsu阈值改进的红外小目标检测方法。该方法不仅继承了Otsu阈值方法比较简单,计算速度较快的优点,更重要的是很好地解决了照度不均匀的图像分割时候多个红外小目标粘连的问题,使红外小目标能够被清晰地检测出来。实验结果表明,新的Otsu阈值改进方法检测出红外小目标的准确率比Otsu阈值检测方法提高了20%。     

Anatomical region segmentation method from dermoscopic images of pigmented skin lesions

Melanoma tumor can cause a serious life threatening problem in humans, if left untreated for a long time without early diagnosis. For early diagnosis of melanoma, it is more significant to develop novel methods based on biophysics analyses, molecular targets recognitions, and new image analysis criteria. In this article, anatomical region segmentation and diameter identification is proposed to detect melanoma from dermoscopic images. Four main steps of the proposed system are as follows: In the first step, the preprocessing is performed to smooth the melanoma extraction process. The region segmentation is done in the second step using watershed segmentation and Sobel operator. In the third step, the postprocessing procedures like as morphological open, canny edge detection also applied to improve the region of interest. Finally, the melanoma region is identified using color symmetry features. The proposed method is tested with two data sets to prove the performance proposed method. The proposed method achieved an accuracy of 95.31% and specificity of 98.3%, which is better than other methods. Experimental results show that the effectiveness of the proposed method and illustrate viability of real-time clinical applications.     

红外图像的动态阈值分割

在深入探讨目前图像分割算法的基础上，针对实际红外图像的分割提出了一种新的方法－－动态阈值分割算法。它利用局部空间阈值适度调整全局统计最优阈值，因而其分割阈值是动态的。实验表明该算法能够突出图像中感兴趣的细节，单动态阈值的分割效果相当于常规固定双域值的分割。     

Automatic annotation of liver computed tomography images based on a vessel-skeletonization method

Anatomical analysis of liver region is an essential and key step for liver-related disease diagnosis and treatment. One of the challenging issues is to annotate the functional regions of liver automatically or semi- automatically by analyzing Computed Tomography (CT) images. The present study developed a complete liver annotation system with an improved vessel-skeletonization method is proposed for CT images. In the first step, an automatic level set method and a customized region-growing method are applied to extract the liver region including vessels and tumors. Next, a modified iterative thinning method is developed to obtain the geometric structure of liver vessels and mark a vessel skeleton. The three-dimensional information is transformed into a tree data structure for storage. Based on the branch distribution of portal vein skeleton, a model-based method with a modified nearest neighbor segment approximation (NNSA) algorithm is adopted for the functional liver anatomy. Three experiments involving five 64-row liver CT datasets are performed. The accuracies of segmentation and annotation results were validated by an experienced doctor. Compared with different methods, our proposed vessel skeletonization method can simultaneously preserve the connectivity of the vasculature topology and generate the skeleton in a shorter time. Furthermore, our proposed annotation system can provide both visual and measurable information of livers. These experimental results demonstrate the usefulness and effectiveness of our proposed method. Our liver annotation system is helpful to evaluate the function of liver system and support diagnosis of liver disease.     

眼底OCT图像中糖尿病性黄斑水肿的分割

糖尿病性黄斑水肿(DME)是导致成年人失明的主要原因之一,它的病理特征主要表现在视网膜内黄斑处液体的积聚。由于光学相干层析成像(OCT)具有无创性、成像安全快速等特点,在临床上被广泛用于眼科疾病的诊断。本文提出一种分割OCT图像中糖尿病性黄斑水肿的方法,首先通过对图像的预处理,排除散斑噪声和血管对最终分割结果的影响,再用改进的水平集方法有效地解决分割的问题,并计算水肿区域的面积,为后期的临床诊断治疗提供定性和定量的工具,最后将本文方法在15个患有DME成人的OCT视网膜图像上进行验证。结果表明：囊状水肿分割的精确度、灵敏度和Dice相似性系数分别为81.12%、86.90%、80.05%。     

纸塑复合袋表面缺陷图像分割算法的设计与实现

目的 纸塑复合袋表面缺陷图像受到噪声、光照不均以及自身缺陷等因素的影响,在对图像缺陷区域进行分割时会造成过分割或欠分割.针对此现象提出一种将边缘检测和自适应区域生长法相结合的纸塑复合袋表面缺陷图像的分割算法.方法 首先利用Sobel算子和形态学运算对双边滤波后的缺陷图像进行第1次分割;然后对缺陷区域进行最小外接矩形标记并计算其形状特征,通过判定形状特征大小来决定是否继续分割;最后将符合继续分割的图像缺陷区域质心作为初始种子点,在原始图像上进行自适应区域生长,形成第2次分割结果,完成缺陷图像分割.结果 与其他算法相比,该算法对各类常见缺陷均能取得较好的分割效果,Dice系数均在0.93以上.结论 该算法分割精度较高,有较强的鲁棒性,可以满足工业上的生产需求.     

基于局部Otsu分割与Hough变换的海天线检测

海面波浪、船只与光照等因素的影响,使得可见光海面图像中的海天线难以被准确检测。为提高海天线检测的准确性与鲁棒性,提出了基于局部Otsu分割与Hough变换的海天线检测方法。首先,通过纵向中值滤波快速地抑制灰度图像中的光斑等高频噪声。然后,根据图像特点进行纵向分块处理来补偿光照的不均匀性并将船只的干扰范围限定在部分图像块中,再进行局部Otsu分割得到二值图像并提取其中的边缘像素,抑制了波浪边缘的干扰。最后,采用Hough变换拟合边缘像素以得到海天线。实验结果表明所提方法具有较高的准确性、鲁棒性与实时性,其检测准确率达93.0%,显著高于三种代表性的海天线检测方法。     

A practical determination strategy of optimal threshold parameter for matrix compression in wavelet BEM

A practical strategy is developed to determine the optimal threshold parameter for wavelet‐based boundary element (BE) analysis. The optimal parameter is determined so that the amount of storage (and computational work) is minimized without reducing the accuracy of the BE solution. In the present study, the Beylkin‐type truncation scheme is used in the matrix assembly. To avoid unnecessary integration concerning the truncated entries of a coefficient matrix, a priori estimation of the matrix entries is introduced and thus the truncated entries are determined twice: before and after matrix assembly. The optimal threshold parameter is set based on the equilibrium of the truncation and discretization errors. These errors are estimated in the residual sense. For Laplace problems the discretization error is, in particular, indicated with the potential's contribution ∥ c ∥ to the residual norm ∥ R ∥ used in error estimation for mesh adaptation. Since the normalized residual norm ∥ c ∥/∥ u ∥ ( u : the potential components of BE solution) cannot be computed without main BE analysis, the discretization error is estimated by the approximate expression constructed through subsidiary BE calculation with smaller degree of freedom (DOF). The matrix compression using the proposed optimal threshold parameter enables us to generate a sparse matrix with O(N^1+γ) (0≤γ<1) non‐zero entries. Although the quasi‐optimal memory requirements and complexity are not attained, the compression rate of a few per cent can be achieved for N～1000. Copyright © 2003 John Wiley & Sons, Ltd.     

基于局域最大亮度的靶状激光图像边缘检测

针对工程应用中靶状激光图像带有较强噪声干扰和要求图像处理速度快的特点,本文提出基于局域最大亮度的靶状激光图像边缘检测方法.该方法根据靶状激光图像由多个明暗相间的圆环组成的特征,采用一个大小适当的圆形检测模板,让模板沿着亮环按一定规则在一定范围内移动,计算模板内各像素点的亮度和,使亮度和为最大值的模板的中心点为这个区域的边缘点.用最小二乘法将由此方法获得的边缘点拟合定中.用摄像机(640像素×480像素,光靶分辨率为0.0504mm/像素)距光源10m处拍取50幅图像,处理后的定中标准差为0.0237mm.实验和应用结果表明,此方法速度快,精度高且抗噪声性能好.     

Image fusion methodology for efficient interpretation of multiband images in 3D high‐resolution ultrasonic transmission tomography

With the critical innovations of using submillimeter transducers and multiband analysis of the first arrival pulse, a high‐resolution ultrasonic transmission tomography (HUTT) system has been built and tested to produce multiband images of biological organs at submillimeter resolution. Since the resulting multiband images consist of frequency‐dependent attenuation coefficients (relative to water reference) of transmitted ultrasound pulses, their contrast and sharpness depend on the specific frequency band(s) used for image formation. Even though this multiband representation provides a powerful tool for soft‐tissue differentiation, it hinders visual inspection and limits the visual interpretation of image contents in a short time. To facilitate the visual interpretation of HUTT multiband images, this article presents an efficient image fusion methodology called local principal component analysis with structure tensor (LPCA‐ST). The LPCA has been known as a feasible tool for the fusion of spectral data, since it utilizes the principal components of spectral data as a fusion‐weighting vector of local area. Nonetheless, the LPCA‐fused image often suffers from oversmoothness because of the redundancy of the spectral data. To prevent this problem, we propose a structure tensor as the metric used to select the most informative bands for subsequent LPCA fusion. Our preliminary studies have shown that the contrast of the LPCA‐fused image improves dramatically only when multiband images whose values of the respective structure tensor are the highest are used in the LPCA fusion process. This is achieved in 3D without increasing the computational complexity of the fusion process. © 2009 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 19, 277–282, 2009     

圆柱形三自由度超声电机定子的结构动力学优化设计

在已有研究的基础上,利用结构动力学方法,对圆柱形三自由度超声电机的定子进行了优化设计。首先根据该种电机的设计要求和先前的经验,确定了其定子的结构形式。然后应用精度较高的体积单元,建立了定子的结构动力学有限元模型。在灵敏度分析、确定设计变量的基础上,建立起定子的优化设计数学模型。最后,采用约束变尺度优化方法,在M ATLAB环境下编制了计算程序,得到了定子的优化设计方案,并制作了定子。实验表明:定子的工作模态满足预期的设计要求;实验结果与理论设计结果相符。同时,运用优化设计方法大大提高了设计效率。     

反相叠加减小超声检测盲区的方法

对于收发合置应用的换能器,在发射脉冲信号的激励下,会产生一定的拖尾,在大尺度声学测量应用中这种拖尾可以被忽略,但是在短距离声学测量应用中,回波信号会淹没在激励信号的拖尾之中,导致测量盲区的加大.为了提高超声波的短距离测量能力,从换能器等效电路入手,通过分析换能器的频率特性,设计了换能器的等效高斯带通滤波器模型,提出了一...     

激光点云中散乱电子元器件分割方法

目的 针对散乱电子元器件计数过程中电子元器件分割困难的问题,提出一种基于点云簇平均法线夹角、平均点云密度边缘提取和区域生长阈值自适应的散乱电子元器件分割方法。方法 通过体素化处理、RANSAC算法和统计离群滤波算法对原始点云数据进行预处理,去除大量无关点云;使用欧式聚类算法对预处理结果粗分割得到电子元器件点云簇,以点云簇为阈值设置单元,避免阈值设置不合理的情况;通常边缘点较非边缘点法线夹角更大、邻域点更少,提出通过点云簇平均法线夹角和平均点云密度自适应约束来去除点云簇中边缘点的方法;对去边缘点后的点云簇细分割,根据细分割后点云簇的平均法线夹角进行区域生长阈值的自适应选择,通过改进的区域生长算法将每个电子元器件从点云簇中分割出来。结果 实验结果证明,文中方法分割正确率达97%以上,每10个目标分割耗时约345 ms。结论 提出的方法具有良好的准确性和实用性,分割效果优于传统分割算法,能够准确地将每个电子元器件从复杂场景中分割出来。