基于Zernike矩的亚像素边缘定位算法

亚像素软件处理技术可以在一定程度上补偿图像测量系统由硬件限制引起的边缘定位精度,提出了一种基于Zernike矩的亚像素边缘定位算法。该算法利用Zernike多项式的正交性与完备性以及复数矩幅度的旋转不变性计算出边缘的亚像素位置。计算出了像素为正方形与矩形(长宽比为4∶3)时的矩模板,使其应用范围进一步扩大。分析了由于矩模板产生的原理偏差,并提出了修正公式,使定位精度得到提高。进一步推导出了实际边缘模型的定位公式,导出了其存在的偏差,并进行了补偿。最终的实验结果表明,补偿后的算法对直线的定位精度达0.05像素,对曲线的定位精度可达0.07像素。     

改进的Zernike矩工业CT图像边缘检测

为提高工业计算机断层扫描(CT)图像亚像素边缘检测的精度和速度,研究了一种改进的Zernike矩边缘检测方法。该方法采用Sobel边缘算子快速检测出图像所有可能的边缘,通过Zernike矩算子对所有可能的边缘进行重新检测,最后,检测出图像的亚像素边缘并计算其精确位置。由于采用Sobel算子检测出可能的边缘使后续Zernike矩算子检测范围缩小,从而减小了运算量,提高了运算速度。对实际CT图像进行的实验结果表明:改进的Zernike矩工业CT图像边缘检测精度绝对误差<0.24 pixel,改进算法的运算速度提高了约70%。     

基于灰度矩的CCD图像亚像素边缘检测算法研究

根据灰度矩算子在目标成像前后的矩不变特性,利用Tabatabai等人提出的前三阶灰度矩建立了亚像素边缘检测算法的完备数学模型。通过实验对该算法的有效性和检测精度进行了研究,给出了工件的实测尺寸对比结果。实验表明:基于灰度矩算子的亚像素边缘检测算法比传统算子具有更高的检测定位精度,可满足图像目标高精度实时在线测量的要求。     

一种改进的Zernike正交矩亚像素边缘检测算法

在分析Ghosal提出的基于Zernike正交矩亚像素边缘检测算法的基础上,对Ghosal算法进行了改进。指出了Ghosal算法的两点不足,即没有考虑模板效应和没有考虑边缘梯度方向上的一阶导数模型。针对这种不足提出了改进方法。改进后的算法在计算边缘亚像素坐标时更准确,同时还具有细化边缘的能力,使边缘定位更精确。     

基于Zernike矩的保持架直径测量方法

为了解决轴承保持架人工抽检费时费力等问题,同时提高工业生产自动化水平,简述了一种基于Zernike矩的保持架直径测量方法;以型号32007E的圆锥滚动轴承筐形保持架为例,提出了基于视觉的直径测量方法,分析CCD相机采集到的轴承保持架大小端面图像,进行图像预处理后,对Sobel算子边界点阈值进行重新设定,快速检测出保持架两端圆面可能存在的边缘点集,增加了有效圆检测算法,剔除部分偏离有效圆的点,再利用 Zernike 矩算子对有效的边缘点进行重新定位,检测出保持架两端圆面的亚像素边缘并计算其精确位置,最后对所得到的亚像素边缘点集进行最小二乘法拟合,获取保持架两端直径具体尺寸;实验表明,该方法测量结果与人工测量精度接近,甚至更高,具有良好的效果和实用价值。     

基于Zernike正交矩的亚像素图像线宽测量算法

针对图像内的细窄线宽,提出了一种基于Zernike正交矩的亚像素图像线宽检测算法。该算法具有明确的几何模型,通过计算图像的2阶和4阶Zernike正交矩,推导出了亚像素线宽表达式。根据数字图像的离散性,给出了计算正交矩所需的模板系数,并分析了由离散性造成的原理误差。将所提出的亚像素线宽检测技术应用于安瓿内异物粒径的标定实验,结果表明:该算法可有效地测量弱小目标在图像中的亚像素线宽值,从而得到了异物粒径大小与亚像素线宽之间的标定曲线。     

基于改进Zernike矩的远场激光高精度中心测量

针对远场激光光斑分布不均匀、形状不规则的特性，提出一种基于改进的Zernike矩的远场激光高精度中心测量方法。在传统Zernike矩亚像素边缘检测基础上，使用新型的logistic边缘检测模型和阶跃阈值自适应提取方法，在减少人工对阶跃阈值误判的同时，提高对实际边缘的识别精度，最后使用最小二乘法椭圆拟合得到高精度激光光斑中心。该方法在远场激光中心检测中，单帧误差在0.5 pixel左右，连续多帧中心偏差波动在1 pixel以内，拥有较高的精度和可靠的稳定性。     

椭圆目标的亚像素边缘定位方法研究

分析和比较了三种常见的基于矩的亚像素边缘定位方法,得出三种方法获取的边缘角度以及OFMM法与ZOM法获取的亚像素边缘相同的结论。提出了一种针对椭圆目标的亚像素边缘定位方法。在ROI提取和SOBEL算子初步检测边缘的基础上,利用初步获取的椭圆目标的几何信息,将参与计算的模板个数减少至一个,从而大大减少了运算时间,提高了算法的运算速度。同时实验结果表明,与其他基于矩的亚像素边缘定位方法相比,新方法不仅有更高的运算速度,而且精度和抗噪性能都有很大地提高。     

基于局部区域灰度矩图像边缘定位方法的研究

在采用机器视觉测量零件尺寸的过程中,提出了一种基于局部区域灰度矩图像边缘定位方法。该方法先利用中值滤波对采集到的图像进行处理,然后通过Prewitt算子对边缘进行初步定位,在此基础上,对图像边缘附近的9个像素的灰度值进行一维灰度矩定位,最后采用误差分析和处理的方法,获得更精确的边缘位置。实验证明,该方法能有效提高边缘检测精度,在检测分辨率为1000DPI情况下,尺寸检测误差小于2um。     

快速多项式拟合亚像素边缘检测算法的研究

传统的亚像素边缘检测都是先粗定位,再进行亚像素细分,因而运行时间较长.采用了一种基于多项式拟合的快速亚像素边缘检测算法.该算法根据图像的灰度分布,运用三次多项式拟合图像的边缘实现亚像素定位,从而达到较高的定位精度.采用在边缘附近取点,进行亚像素边缘检测,减少了运行时间.仿真实验给出的实验结果证明了该算法的有效性.     

The high precision positioning algorithm of circular landmark center in visual measurement

For vision measurement in the center circle landmark location problem, an improved center of the ellipse based on Zernike moments sub-pixel positioning algorithm was proposed. First, using Sobel operator edge of the image pixel level positioning and then use the constructed Zernike moments to solve the model, combined with Zernike orthogonal polynomials and completeness and plural moment magnitude rotational invariance calculated edge sub-pixel position; followed by analysis of the principle deviation generated by the moments template and ideals model, a correction formula was proposed to compensate and improved edge criterion used to image sub-pixel edge positioning; finally, using the least-squares ellipse fitting algorithm fitting circle center, reverse edge point, filtered residuals larger point, and then precise positioning of the ellipse center. The experimental results show that this method has good stability and high positioning accuracy can be efficiently used in many applications.     

A method of moments for calculating dynamic responses beyond linear response theory

A method of moments for calculating the dynamic response of periodically driven overdamped nonlinear stochastic systems in the general response sense is proposed, which is a modification of the method of moments confined within linear response theory. The calculating experience suggests that the proposed technique is simple and efficient in implementation, and the comparison with stochastic simulation shows that the first three orders of susceptibilities calculated by the proposed technique have high accuracy. The dependence of the spectral amplification parameters at the first three harmonics on the noise intensity is also investigated, and another observed phenomenon of stochastic resonance in the systems induced by the location of a single periodic orbit is disclosed and explained.     

A Legendre orthogonal moment based 3D edge operator

This paper presents a new 3D edge operator based on Legendre orthogonal moments. This operator can be used to extract the edge of 3D object in any window size, with more accurate surface orientation and more precise surface location. It also has full geometry meaning. Process of calculation is considered in the moment based method. We can greatly speed up the computation by calculating out the masks in advance. We integrate this operator into our rendering of medical image data based on ray casting algorithm. Experimental results show that it is an effective 3D edge operator that is more accurate in position and orientation.     

双目视觉测量系统特征点提取与匹配技术研究

自主研发了一种光笔式双目立体视觉大工件尺寸测量系统,对测量系统中特征点的提取及其匹配技术进行了研究。测量系统采用 Canny 算子和 Zernike 矩相结合的算法实现椭圆光斑的亚像素边缘提取,根据得到的亚像素边缘点,采用基于最小二乘的曲线拟合法得到椭圆光斑的中心坐标。针对特征点的匹配问题,提出了一种基于位置约束的快速匹配方法。实验结果显示:所提方法能提取到椭圆光斑亚像素边缘,可精确计算出椭圆光斑中心坐标,匹配率达到95%以上。     

Further improvement of edge location accuracy of double fiber spherical coupling sensor using orthogonal Jacobi–Fourier moments

In order to further improve the edge location accuracy of double fiber spherical coupling sensor, an approach based on OJFMs is proposed. In this paper, OJFMs with high order spatial moment and simple calculation is built. Then, its edge location method and fast discrete algorithm is proposed. Finally, the proposed method is tested by locating the edge of standard pattern and the image acquired by double fiber spherical coupling sensor with lot of noise. Experiment results show that the location accuracy is up to 0.05 pixel and 0.08 pixel for standard lines and standard circle, respectively. The accuracy of double fiber spherical coupling sensor is 0.026 pixel in 2σ band and a actual accuracy of 36 nm is obtained by OJFMs after a moving average filter (N = 7).