基于PCA和边缘不变矩的车标识别新方法

车标识别技术是进行车辆类型识别的关键内容。在车牌定位准确的基础上,利用车头图像中车牌、车标位置关系定位到车标大致区域;接着对车标进行精确定位、归一化处理,并提出似真度函数对车标进行确认;然后利用边缘不变矩的最小距离进行车标识别。该方法不受原始车标图像大小、变形等影响,能够自适应地进行目标定位确认和识别,具有较强的鲁棒性。通过实测图像测试,结果表明该方法是有效和可行的。

基于泽尼克矩与边缘梯度共生矩的商标检索算法

为提高商标的检索精度,定义基于泽尼克矩与边缘梯度共生矩的商标检索方案。引入中值滤波,对商标实施去噪处理;分别提取商标的泽尼克矩(Zernike moment,ZM)和边缘梯度共生矩(edge gradient co-occurrence matrix,EGCM);根据得到的ZM,通过Euclidean距离度量方法,计算查询商标与商标库的全局相异性及平均全局相异性;将平均全局相异性作为阈值,与全局相异性实施比较,剔除不相关商标;根据EGCM计算局部相异性,联合全局相异性,通过对二者实施加权得到总相异性,以输出检索结果。实验结果表明,与已有商标检索方案相比,所提方案具有更高的检索精度与鲁棒性。

边缘不变矩匹配算法的列车闸瓦图像识别试验

研究一种基于边缘不变矩匹配算法的列车闸瓦图像识别新方法。在列车运动过程中,应用视觉传感器摄取闸瓦图像,根据闸瓦轮廓经图像旋转、缩放和平移变换后边缘特征点保持不变的统计特性,以其作为主要特征参数来自动识别具有复杂背景的闸瓦目标。针对闸瓦表面图像特征与轮对相近、图像灰度分布不均匀和存在遮挡等特点,研究改进的边缘不变矩匹配算法对闸瓦轮廓进行分析和特征提取,实现列车运动过程中闸瓦图像的识别。在传统不变矩算法计算的基础上,用原点矩代替中心矩,并且选用4个低阶矩作为特征分量,减少计算误差。试验结果表明,新方法可以解决闸瓦灰度与轮对相近和存在遮挡等问题,能够有效地识别闸瓦,为解决长期存在的闸瓦动态识别的技术难题提供一条新的途径。

灰度矩边缘精确定位法的测量应用探讨

针对运用图像处理法进行二维尺寸精密测量的一个关键问题—图像目标边缘精确定位,提出了运用一维灰度矩亚像素边缘定位原理,分别计算若干条扫描线边缘,通过统计计算确定直线边缘的方法。在分析图像目标边缘像素灰度分布的基础上,确定了计算用像素点的选取原则。试验证明这种方法可以进行精确边缘定位,定位精度可以达到小于0.01个像素的水平。

预应力混凝土空心桩换算截面受拉边缘弹性抵抗矩公式推导及分析

根据预应力混凝土空心桩配筋均匀布置的特点,推导出预应力混凝土管桩和空心方桩的换算截面受拉边缘弹性抵抗矩公式,并做简单分析。以期在标准设计中采用。

基于粒子群算法与改进BP神经网络的水电机组轴心轨迹识别

在水电机组状态检修系统中,轴心轨迹是判断机组状态的一个重要特征。该文提出边缘检测和矩特征提取相结合的方法,利用粒子群寻优算法来获取与待识别样本最接近的已知样本,应用改进的BP神经网络进行识别,将轴心轴迹的不变性矩作为神经网络的特征参数,对几种典型的轴心轨迹进行了辨识。某水电站机组试验表明该方法识别速度快、精度高,具有较高的实用价值。

高速的9×9尺寸模板Zernike矩边缘算子

根据Zernike矩基本原理,提出了基于9×9模板尺寸的Zernike矩边缘算子。首先推导Zernike矩模板计算过程,然后计算出2个9×9模板,并选取二值图像和经过5次平滑的模糊图像为对象对新算子进行了测试。结果表明:对于理想的二值图像,新算子具有较强的边缘细化能力,且边缘检测精度也较高;而对于模糊图像,虽然新算子的边缘检测效果也较好,但应注意边缘图像表征与亚像素坐标数据间存在不一致的现象。最后,采用先粗后精定位思想,使新算子的运算速度大幅提高,耗时仅为0.19s。

Color constancy using color edge moments and regularized regression in anchored neighborhood

To improve the accuracy of illumination estimation while maintaining a relative fast execution speed, a novel learning-based color constancy using color edge moments and regularized regression in an anchored neighborhood is proposed. First, scene images are represented by the color edge moments of various orders. Then, an iterative regression with a squared Frobenius norm（F-norm） regularizer is introduced to learn the mapping between the edge moments and illuminants in the neighborhood of the anchored sample.Illumination estimation for the test image finally becomes the nearest anchored point search followed by a matrix multiplication using the associated mapping matrix which can be precalculated and stored. Experiments on two standard image datasets show that the proposed approach significantly outperforms the state-of-the-art algorithms with a performance increase of at least 10. 35% and 7. 44% with regard to median angular error.

圆桩换算截面受拉边缘的弹性抵抗矩公式推导与分析

根据圆桩均匀布筋的特点,推导出圆桩换算截面受拉边缘弹性抵抗矩的理论公式。结合常见圆桩算例,与《建筑桩基技术规范》中的简化公式进行对比、分析。经过对比,规范中的公式精度能够满足工程精度要求。

Visual positioning of rectangular lead components based on Harris corners and Zernike moments

With the increasing necessities for reliable printed circuit board(PCB) product, there has been a considerable demand for high speed and high precision vision positioning system. To locate a rectangular lead component with high accuracy and reliability, a new visual positioning method was introduced. Considering the limitations of Ghosal sub-pixel edge detection algorithm, an improved algorithm was proposed, in which Harris corner features were used to coarsely detect the edge points and Zernike moments were adopted to accurately detect the edge points. Besides, two formulas were developed to determine the edge intersections whose sub-pixel coordinates were calculated with bilinear interpolation and conjugate gradient method. The last experimental results show that the proposed method can detect the deflection and offset, and the detection errors are less than 0.04° and 0.02 pixels.

Sub-pixel edge detection method for miniature parts in microscopic field of view

In order to improve the edge detection precision of miniature parts in microscopic field of viewa sub-pixel edge detectionalgorithm combining non-orthogonal quadratic B-spline wavelet transform algorithm and Zernike moment algorithm is proposed.Non-orthogonal quadratic B-spline wavelet transform algorithm is adopted to get the pixel edge of miniature parts?andthe moment invariant of Zernike moment algorithm is used for refining the pixel edge to get sub-pixel edges.A real-time detectionsystem based on the proposed algorithm for miniature parts is established.The general system structure and operational principle are given,the real-time image acquisition and detection are completed,the results of edge detection are analyzed and the detection precision is evaluated.The results show that parts size can be0.01-10mm and the detection precision reaches0.01%-0.1%.Therefore,the edge of the miniature parts can be accurately identified and the detection precision can be improved to sub-pixel level,which meets the requirements of miniature parts precision detection.

基于分区域自适应阈值Zernike矩的骨签轮廓尺寸测量方法

汉长安城遗址出土骨签数量庞大,基于数字图像技术的骨签尺寸自动测量方法能够提高工作效率。针对骨签纹理繁多、边缘对比度低导致测量精度不高的问题,提出一种基于分区域自适应阈值Zernike矩的骨签图像尺寸测量方法。首先,利用Canny算子进行像素级定位;然后,对称设定边缘有效分区域,以欧氏距离为比例系数计算各个分区域像素点与中心点的加权灰度值,作为各区域Zernike矩提取边缘的判定阈值,提取骨签亚像素级边缘;其次,引入边缘轮廓判别条件剔除无效纹理,获取骨签图像精确轮廓;最后,使用最小外接矩算法计算骨签不规则轮廓,结合相机标定完成骨签尺寸测量。实验结果表明,与其他方法相比,该方法测量的骨签长宽的均方根误差分别降低1.3 mm和1.2 mm,平均相对误差分别降低3%和6.3%,平均绝对误差分别降低1.23 mm和1.08 mm,能够有效提高复杂边缘轮廓的骨签尺寸测量精度。