基于迭代拟合的集装箱角件边缘检测算法

噪声通常是影响集装箱角件图像中低层次语义信息提取精度的重要因素,传统的边缘检测算法通常通过改进滤波器和阈值来消除图像中的物理噪声和环境噪声,但是却无法去除边缘检测后的噪声,为解决这一问题,提出了一种基于迭代拟合的边缘检测算法。首先,对角件图像进行一系列预处理操作获取边缘点集,其次,使用拟合算法处理点集并且得到函数表达式,然后定义偏差值度量并计算,用于衡量目标点集到拟合或者检测结果的偏差,最后,去除定义下距离拟合结果最远的指定数量的点,如此迭代拟合直至评价函数收敛。实验结果与分析表明,该算法可以有效地去除边缘点集中的非真实边缘点,相比于传统的边缘检测算法更能去除特殊噪声,算法具有收敛速度快、准确率较高、灵活性好等特点。     

Magnetic Resonance Imaging Segmentation via Weighted Level Set Model Based on Local Kernel Metric and Spatial Constraint

Magnetic resonance imaging (MRI) segmentation is a fundamental and significant task since it can guide subsequent clinic diagnosis and treatment. However, images are often corrupted by defects such as low-contrast, noise, intensity inhomogeneity, and so on. Therefore, a weighted level set model (WLSM) is proposed in this study to segment inhomogeneous intensity MRI destroyed by noise and weak boundaries. First, in order to segment the intertwined regions of brain tissue accurately, a weighted neighborhood information measure scheme based on local multi information and kernel function is designed. Then, the membership function of fuzzy c-means clustering is used as the spatial constraint of level set model to overcome the sensitivity of level set to initialization, and the evolution of level set function can be adaptively changed according to different tissue information. Finally, the distance regularization term in level set function is replaced by a double potential function to ensure the stability of the energy function in the evolution process. Both real and synthetic MRI images can show the effectiveness and performance of WLSM. In addition, compared with several state-of-the-art models, segmentation accuracy and Jaccard similarity coefficient obtained by WLSM are increased by 0.0586, 0.0362 and 0.1087, 0.0703, respectively.     

一种基于双分支改良编解码器的图像去噪算法

针对传统图像去噪算法多噪声去除难,深层卷积神经网络去噪模型网络复杂、训练时间长等问题,提出一种基于自编码器结构的双分支改良编解码网络,实现高效图像去噪。双分支结构之一采用降-升采样实现点噪声消除,另一分支专注于宏观的图像修复和伪像去除,后端利用残差结构进行整合,实现数字图像混合噪声去噪。实验结果显示:对于含有标准差为15,均值为0的高斯噪声、噪声密度为5%的椒盐噪声和散粒噪声的混合噪声图像测试集,实验去噪效果相较于输入混合噪声图像峰值信噪比,平均提升了5.3%。与12层全卷积神经网络相比,去噪效果相当,训练速度提升了25.4%,体现了其“轻量级”的优点。实验表明:该方法相较于深层卷积神经网络,训练速度快,网络简单;相较于传统图像去噪算法,噪声去除效果也较为明显。该算法可应用于轻量级视觉平台后端去噪。     

基于加权块稀疏联合非凸低秩约束的高光谱图像去条带方法

针对高光谱图像(hyperspectral images,HSI)去条带易引起影像结构细节丢失问题,提出一种基于加权块稀疏(weighted block sparsity,WBS)正则化联合最小最大非凸惩罚(minimax concave penalty,MCP)约束的HSI去条带方法.本算法采用加权?2,1范数和MC...     

基于C-V模型的改进快速水平集图像分割法

针对水平集方法计算复杂度高,无法满足实时系统要求的缺陷,提出一种改进的快速水平集算法。该算法对快速水平集算法进行简化,采用单链表表示轮廓曲线。利用C-V模型的二值拟合项来设计曲线演化的速度函数,保留了C-V模型的全局优化特性。还给出了一个基于单链表中轮廓点个数变化的水平集演化终止准则。该算法不仅明显提高了分割速度,且对噪声图像也能实现高效的分割。     

中阶梯光栅光谱仪谱图背景去除算法

中阶梯光栅光谱仪凭借交叉色散特性实现全谱瞬态直读,面阵探测器接收的二维光谱图像需要还原成一维谱图以提取有效波长。由于二维谱图含有庞大的数据,且有效信息仅占极小比例,因此在谱图还原前进行背景去除能够减小数据量、提高运算速度。详细分析了中阶梯光栅光谱仪二维图像的特点,并针对其特点提出了背景去除算法。将图像边缘检测方法应用于弥散光斑的检测中,选择合适的边缘检测算子与原始图像卷积得到边缘图像,设置边缘图像的全局阈值对其进行二值分割,最终利用二值边缘图像映射原始图像得到去除背景的二维谱图。依据不同元素灯在不同积分时间下所拍摄的谱图,对比不同边缘检测算法的背景去除效果,分析了各算子对算法速度、精度的影响。实验结果表明本文提出的算法运算简单、边缘图像阈值易于计算、目标提取精度高,处理后的图像可以与谱图还原算法有效对接,谱图处理速度显著提升。     

Block-based disparity estimation by partial finite ridgelet distortion search (PFRDS)

In stereo vision applications, computing the disparity map is an important issue. Performance of different approaches totally depends on the employed similarity measurements. In this paper finite ridgelet transform is used to define an edge sensitive block distortion similarity measure. Simulation results emphasize to outperform in the conventional criteria and is less sensitive to noise, especially at the edge set of images. To speed computations, a new partial search algorithm based on energy conservation property of FRIT is proposed.     

Developing a Hong Kong based speed correction factor by CPX method

A systematic procedure for developing speed correction factor to adjust measured tyre/road noise is proposed in this paper. The draft ISO/CD 11819-2 on the Close-Proximity method to measure tyre/road noise stipulates that speed correction is needed to convert the tyre/road noise level at the sampling speeds to that at the reference speeds. However, the ISO standard does not mention the exact methodology for developing the speed correction factor development, although it recommends local authorities to develop their own. In this study, the tyre/road noise and speed data over 59 runs on a high speed porous asphalt surface is recorded by our CPX trailer in Hong Kong. The instantaneous noise and speed data are then analysed using this proposed procedure and the set of speed correction factor developed is found to be more stable and repeatable.     

Application of radial basis functions to evolution equations arising in image segmentation

In this paper, radial basis functions are used to obtain the solution of evolution equations which appear in variational level set method based image segmentation. In this method, radial basis functions are used to interpolate the implicit level set function of the evolution equation with a high level of accuracy and smoothness. Then, the original initial value problem is discretized into an interpolation problem. Accordingly, the evolution equation is converted into a set of coupled ordinary differential equations, and a smooth evolution can be retained. Compared with finite difference scheme based level set approaches, the complex and costly re-initialization procedure is unnecessary. Numerical examples are also given to show the efficiency of the method.     

基于水平集的闪光照相图像分割算法

针对Chan-Vese(CV)模型局部控制能力差的缺点,将基于区域的CV模型和分割曲线的局部信息结合起来,提出了一种新的水平集图像分割算法。该算法以CV法的分割曲线为初始曲线,以获得全局收敛性,在后继分割中引入分割曲线的局部信息,以提高模型对图像中微弱信号的分割能力。对闪光照相图像的数值实验表明,该算法噪声抵抗能力强,对初始轮廓位置不敏感,能实现对含细长拓扑结构和微小孔洞的弱边界闪光图像的自动分割。     

The adaptive sinogram restoration algorithm based on anisotropic diffusion by energy minimization for low-dose X-ray CT

Apparent streak artifacts will present in reconstructed images due to excessive quantum noise in low-dose X-ray imaging process. Estimating a noise-free sinogram to satisfy the filtered back-projection (FBP) reconstruction is an effective way to solve this problem. In this paper, we propose a novel sinogram noise reduction method by energy minimization. An adaptive smoothness parameter based on a modified anisotropic diffusion coefficient is applied for an optimal estimation. The smoothness parameter can make the method effectively adjust the degree of smoothness according to the noise level and the region feature in the sinogram. Visual effect together with quantitative analysis of the experimental result shows the developed approach has the excellent performance in protection of the edge and removal of streak artifacts in the reconstructed image.     

Numerical analysis of dipole sound source around high speed trains

As the maximum speed of high speed trains increases, the effect of aeroacoustic noise on the sound level on the ground becomes increasingly important. In this paper, the distribution of dipole sound sources at the bogie section of high speed trains is predicted numerically. The three-dimensional unsteady flow around a train is solved by the large eddy simulation technique. The time history of vortices shows that unstable shear layer separation at the leading edge of the bogie section sheds vortices periodically. These vortices travel downstream while growing to finally impinge upon the trailing edge of the section. The wavelength of sound produced by these vortices is large compared to the representative length of the bogie section, so that the source region can be regarded as acoustically compact. Thus a compact Green's function adapted to the shape can be used to determine the sound. By coupling the instantaneous flow properties with the compact Green's function, the distribution of dipole sources is obtained. The results reveal a strong dipole source at the trailing edge of the bogie section where the shape changes greatly and the variation of flow with time is also great. On the other hand, the bottom of the bogie section where the shape does not change, or the leading edge and boundary layer where the variation of flow with time is small, cannot generate a strong dipole source.     

A robust level set method based on local statistical information for noisy image segmentation

The paper presents an improved local region-based active contour model for image segmentation, which is robust to noise. A data fitting energy functional is defined in terms of curves and the energy terms which are based on the differences between the local average intensity and the global intensity means. Then the energy is incorporated into a level set variational formulation, from which a curve evolution equation is derived for energy minimization. And then the level set function is regularized by Gaussian filter to keep smooth and eliminate the re-initialization. By using the local statistical information, the proposed model can handle with noisy images. The proposed model is first presented as a two-phase level set formulation and then extended to a multi-phase one. Experimental results show desirable performances of the proposed model for both noisy synthetic and real images, especially with high level noise.     

A kernel machine-based fMRI physiological noise removal method

Functional magnetic resonance imaging (fMRI) technique with blood oxygenation level dependent (BOLD) contrast is a powerful tool for noninvasive mapping of brain function under task and resting states. The removal of cardiac- and respiration-induced physiological noise in fMRI data has been a significant challenge as fMRI studies seek to achieve higher spatial resolutions and characterize more subtle neuronal changes. The low temporal sampling rate of most multi-slice fMRI experiments often causes aliasing of physiological noise into the frequency range of BOLD activation signal. In addition, changes of heartbeat and respiration patterns also generate physiological fluctuations that have similar frequencies with BOLD activation. Most existing physiological noise-removal methods either place restrictive limitations on image acquisition or utilize filtering or regression based post-processing algorithms, which cannot distinguish the frequency-overlapping BOLD activation and the physiological noise. In this work, we address the challenge of physiological noise removal via the kernel machine technique, where a nonlinear kernel machine technique, kernel principal component analysis, is used with a specifically identified kernel function to differentiate BOLD signal from the physiological noise of the frequency. The proposed method was evaluated in human fMRI data acquired from multiple task-related and resting state fMRI experiments. A comparison study was also performed with an existing adaptive filtering method. The results indicate that the proposed method can effectively identify and reduce the physiological noise in fMRI data. The comparison study shows that the proposed method can provide comparable or better noise removal performance than the adaptive filtering approach.     

Vortex shedding noise of low tip speed,axial flow fans

Noise and performance tests were conducted on three low tip speed, half-stage, axial flow fans to determine the nature of the vortex shedding noise mechanism. Each fan was 356 mm in diameter and had eight equally spaced, variable pitch blades. The noise measurements were made in a free field environment and the fan back pressure and speed were varied during the tests. An acenaphthene coating on the blades was used to determine the regions of laminar and turbulent flow.Vortex shedding can be a significant source of noise when the fan is operated in a lightly loaded condition. Essentially it is due to instabilities in the laminar boundary layer on the suction side of the blade where these instabilities are in the form of Tollmien-Schlichting (T-S) waves. These instabilities interact with the trailing edge of the blade and generate acoustic waves which radiate from the trailing edge and form a feedback loop with the source of the instabilities. Vortex shedding noise can contribute as much as 5 dB in overall noise level and up to 22 dB at higher frequencies (8–14 kHz).Serrations located at the leading edge, at the mid-chord, or near the trailing edge on the suction side were found to reduce the vortex shedding noise significantly. The mid-chord location was found to be the most satisfactory because, as well as eliminating the noise, the serrations provided a 3% improvement in peak efficiency. This improvement occurred because separation of the laminar boundary layer was prevented on the suction side. On the other hand, serrations placed at the other two locations tended to degrade fan performance.     

Chan-Vese模型两相自适应窄带的阴影区检测方法

为得到快速高精度的声呐图像阴影区检测效果,提出Chan-Vese模型两相自适应窄带检测方法。利用各向异性二阶邻域马尔可夫模型估计声呐图像的纹理特征参数,实现原始图像平滑去噪;由块方式的k-均值聚类算法确定图像的初始两类分割,初步确定阴影区大致位置,并根据此大致位置,自适应初始化零水平集函数,来减少人为干预,提高检测速度;在此基础上,提出建立Chan-Vese模型两相窄带水平集进行声呐图像检测,完成局部寻优,排除全局图像中孤立区对检测的影响,使阴影区检测结果更加精确。通过对真实声呐图像的检测实验结果分析,验证提出的检测方法能够去除原始图像的部分噪声,提高检测精度和速度,有一定的自动性和适应性。      

基于改进倾斜刃边法的光电成像系统MTF高准确度测试

提出一种能高准确度测试光电成像系统调制传递函数的改进的倾斜刃边法.利用一种自动确定高低阈值的改进Canny算子提取边缘位置点,对提取到的边缘位置点进行线性拟合求得刃边倾角,在利用费米函数拟合边缘扩散函数的过程中,提出一种邻域平均值和设定误差容限的多重去噪改进方法.通过仿真与物理实验证明了本文方法的有效性,并分析了刃边倾角、噪声变化对本文方法与传统方法MTF测试准确度的影响.实验结果表明:本文方法的调制传递函数测试准确度达到0.004 9,重复准确度达到0.003 1,皆优于传统方法;说明其具有较高的测试准确度和良好的测试重复性,并且具有较强的抗噪性,基本不受刃边倾角变化的影响.     

基于分数阶微分的消除局部强反射的孔径测量方法

针对视觉测量硬盘圆孔直径易受到圆孔周围高反光面强反射的影响而导致测量精度不高的问题,提出了一种基于分数阶微分的图像去噪声的处理方法。通过分数阶微分算法对相机采集的带有强反射、高反光的硬盘圆孔孔径图像进行处理,消除圆孔周围强反射表面等不相关信息对圆孔边缘提取带来的影响,通过实验分别与Prewitt、Soble、Laplacian算子进行比较,证明分数阶微分可有效减少所要分析的图像信息量,更好地增强圆孔边缘轮廓信息,可达到更好的视觉效果。对分数阶微分算法处理后的圆孔图像进行Canny边缘检测,提取出有效的圆孔边缘,利用最小二乘法对孔径边缘进行直径测量。实验结果表明:该文算法与其他算法相比,在保证精确提取孔径边缘信息的基础上,对圆孔周围的强反射面进行了抑制,误差控制在0.05 mm左右,保证了测量精度。     

振动噪声等高图的电驱动桥噪声预测及优化*

在项目开发前期通过优化电驱动桥扭矩特性的设计,可以规避噪声大问题。根据电驱动桥台架在对应扭矩下的振动噪声特性,提出了一套稳定高效的测试流程和分析方法。首先设计了一套完整的试验流程,制定了精准的数据分析方法。然后绘制出能够全面反映电驱动桥振动噪声特性的等高图。最后利用电驱动桥台架的振动噪声等高图,准确评估电驱动桥加速工况下振动噪声风险,为主机厂和电驱动桥零部件企业提供电驱动桥扭矩特性设计前期指导。利用该方法成功识别到某电驱动桥匀速及加速工况下电机及齿轮的啸叫问题。通过优化电驱动桥扭矩特性设计,电机48阶噪声峰值降低了8.5dB(A),确认该方法准确可靠,具备推广应用价值。     

Tensor diffusion level set method for infrared targets contours extraction

A tensor diffusion level set method is presented to extract infrared (IR) targets contour under a sky-mountain-water complex background. The proposed model combines tensor diffusion operator and the eigenvalues of tensor-image into a common energy minimization level set framework. By incorporating the information of image tensor diffusion operator into the external energy term, the level set function can move in a specific way. And eigenvalues of tensor-image are used for the regularization of zero level curves in order to diminish the influence of image ‘clutter’ and noise. An additional benefit of the proposed method is robust to initial conditions. Experimental results show very good performance of the tensor diffusion level set method for IR targets contours extraction.