基于变异系数法的工业产品表面缺陷快速检测应用研究

为了提高工业产品质量控制中表面缺陷检测的准确性和快速性,提出了一种基于机器视觉的工业产品表面缺陷快速检测方法。方法引用变异系数的概念,通过待检测图像和模板图像进行差运算得到差影图像,并通过计算其变异系数以确定阈值,利用分割定位缺陷,从而实现缺陷的快速检测。壁纸表面缺陷检测实验结果表明,本方法提高了图像检测系统的准确性和鲁棒性,误检率大大降低。     

基于图像数据的塑料餐盒质量诊断系统

对塑料餐盒表面缺陷检测系统进行设计,并以图像相似度匹配作为图像处理的核心算法,通过理论分析和实验,研究基于欧式距离、归一化相关系数以及余弦相似度的图像相似度。结果表明:余弦相似度算法在塑料餐盒图像相似度判断中的准确性高达90%以上,不易误判,且对不同类型塑料餐盒表面缺陷检测的适用性高。     

基于大数据的胶料表面缺陷视觉识别检测研究

胶料表面缺陷影响胶料使用,实际生产过程中要强化表面缺陷检测。对人工检测胶料表面缺陷存在的效率低、精度低、成本高等问题,提出了基于大数据的胶料表面缺陷检测方法。采用线扫描工业相机获取胶料表面缺陷图像,对获取的图像灰度化处理。将图像白色树脂转换为白雾,并去除白雾。在此基础上,采用灰度变换法对图像进行增强。采用Darknet-19卷积神经网络提取图像特征,并通过YOLOv4算法进行表面缺陷检测。胶料表面缺陷检测结果表明：对裂缝缺陷检出率最高,误检率最低;对毛团缺陷检出率最低,误检率最高;对三种表面缺陷的检测均能够满足实际胶料生产的精度和实时性要求。     

碳纤维拉挤板材表面缺陷检测

碳纤维拉挤成型工艺是一种可以制造出高强度、高刚度、轻量化复合材料的方法。该工艺通过在高温高压下将通过模具的碳纤维和树脂复合材料拉挤成型，由于碳纤维方向的一致性，从而提高了材料的力学性能。因其需要将不同材料复合，出现的缺陷会影响材料力学性能。传统人工肉眼初筛缺陷的方法效率低、易遗漏。采用多目摄像头采集图像与模糊聚类算法处理取代人工缺陷检测。机器视觉可在生产过程中对碳纤维拉挤板材表面进行实时缺陷检测。采集图像时，应用光度立体学法采集复合板材表面图像，并通过图像灰度计算形成图像梯度矩阵。在缺陷检测过程中，先将梯度矩阵经过模糊聚类法区分正常区域与缺陷区域。再通过缺陷隶属度模型划分不同缺陷种类，最后通过模糊信息熵模型标记缺陷，完成碳纤维拉挤板材表面缺陷检测。实验结果表明，表面缺陷检测准确率达98%以上，缺陷检测准确率高，误差范围小，实现了不同缺陷种类的分类划分。在碳纤维拉挤板材生产过程中具备优异的缺陷检测性能与实用价值。     

基于机器视觉的氮化硅陶瓷轴承表面缺陷检测技术

为提高缺陷检测效率和精度,针对人工检测氮化硅陶瓷轴承球表面缺陷成本高、效率低的缺陷,分析机器视觉技术特点.通过自主搭建氮化硅陶瓷轴承球检测平台,采用CCD相机采集表面缺陷图像,分析氮化硅陶瓷轴承球表面缺陷特征;采用灰度处理、中值滤波、缺陷提取及边缘检测等操作,有效识别氮化硅陶瓷球表面缺陷.该方法对氮化硅陶瓷球表面缺陷的...     

浅析无线物联网技术在陶瓷制品表面检测中的应用

随着无线物联网技术的深入应用,工业产品的检测方式步入无线化、智能化时代。针对该背景,笔者主要提出一种基于无线物联网的卫生陶瓷产品表面缺陷检测技术。检测系统采用Cortex-A8作为核心芯片构建嵌入式主控平台,并搭载基于OV2710的CMOS检测相机以及MF210型3G模块,在客户端基于Qt开发适用于工控的UI界面,采集到实时陶瓷产品表面缺陷图像后,采取灰色模型算法完成缺陷识别。笔者设计的表面缺陷识别系统能够高效处理陶瓷产品图像,工程实现难度低,组网灵活,适应性强,工程应用意义突出。     

轮胎断面测量系统的快速图像拼接方法

针对轮胎断面测量系统中扫描图像分辨率高、视场小,需拼接图像数量多的特点,在传统拼接算法基础上,提出一种基于小波金字塔的快速图像拼接方法。通过计算最小区域差异度,结合小波金字塔的搜索策略,提高了图像拼接的准确性和效率。     

基于机器视觉的玻璃表面缺陷检测系统的研究

介绍了玻璃表面缺陷检测方法与原理,研究开发了基于机器视觉的玻璃表面检测系统.针对实际的目标图像,通过对其进行图像获取、图像预处理、图像分割、特征提取和区域参数计算研究, 从而获得玻璃表面缺陷图像的识别信息.实验表明,该算法简单、速度快、系统抗干扰能力强,能满足实际使用的要求.     

钢化玻璃碎片状态快速检测装置研究

简述了钢化玻璃碎片状态检测方法的研究发展历程,针对现有检测方法的不足设计了相应的快速检测装置,具体包括玻璃试样快速固定装置、图像采集装置和图像处理软件.该装置使工作效率提高了10倍、降低了人为因素的差错、提高了检测结果的准确性和公正性.     

基于机器视觉的汽车零件涂装缺陷检测方法

设计了一种基于机器视觉的涂装缺陷检测系统来检测汽车零件表面的孔洞边缘缺陷,并在Halcon软件中实现了相关检测算法.首先搭建了检测平台获取图像信息,运用迭代加权拟合的方法实现了对孔洞感兴趣区域(ROI)的定位与提取,从而缩短检测时间;再使用一种带阻滤波器对图像进行预处理,减少了表面反光;最后运用阈值分割等方法对涂装缺陷特征进行识别和提取.该法能快速、准确地识别提取涂装缺陷特征,平均单次识别时间为320 ms,识别准确率为97％,满足工业涂装缺陷检测的要求.     

A nondestructive testing method for detecting surface defects of Si3N4-Bearing cylindrical rollers based on an optimized convolutional neural network

Aiming at solving the problems of low detection efficiency, poor accuracy rate, and low applicability of the traditional detection methods for surface defects on the silicon nitride (Si3N4) bearing cylindrical rollers. In this study, to detect the surface defects on Si3N4-bearing cylindrical rollers, a nondestructive testing (NDT) method based on an optimized convolutional neural network is proposed, which uses machine vision system to detect the surface defects. The optimized convolutional neural network is a two-stage network. It combines a semantic segmentation sub-network and a decision sub-network. The semantic segmentation sub-network performs an end-to-end learning based on the features of the original image. It completes defect feature extraction to segment the surface defect area from the normal area. The decision sub-network classifies the defects in the segmented surface defect area. The experimental results show that the detection time of surface defects on Si3N4-bearing cylindrical rollers by the proposed network was 72 ms. In addition, its Accuracy (Acc), Percision (Pre), Sensitivity (Sen), and Specificity (Spe) were 97.5%, 99.9%, 99.0%, 98.6%, respectively, which significantly improved the Acc and Pre of the detection of surface defects on Si3N4-bearing cylindrical rollers compared with the traditional detection method. On the other hand, its mIoU, mPA were 84.4%, 92.9%, respectively, meaning brilliant ability of image segmentation for this network. To sum up, the ability of the proposed network to detect and classify surface defects on Si3N4-bearing cylindrical rollers is excellent.     

Infrared radiation properties of the carbon-carbon composite and their application to nondestructive detection of its defects

Infrared radiation properties and surface characteristics of C/C composites and graphites were examined at temperatures in the range of 293-373 K from the viewpoint of the nondestructive detection of defects in these materials. The radiation temperature of specimen surface and its variation were quantitatively evaluated on the basis of true specimen temperature, ambient temperature, emissivity and radiosity coefficient to obtain data applicable to the thermographic detection of defects. It was found that the larger the pore size and roughness of specimen, the larger the variation of data points. Graphite specimens with different artificial flaws 1-10 mm in diameter and 1-8 mm in depth were examined by thermography, and the minimum difference in radiation temperature at a defect to be detected was obtained with regard to the flaw size.     

基于红外成像测温的储罐液位传热反问题识别方法

传统的利用红外热像仪进行储罐液位的检测主要是基于对红外图像进行数字处理,因此测量精度不高。通过建立储罐的传热模型,利用有限体积法,对筒体外表面的温度分布进行了数值模拟,并提出了基于筒体外表面温度分布对筒内液位进行定量识别的传热反问题方法,同时分析了初始假设、测量误差(σ)和最大温差对液位识别精度的影响。结果表明:在工程允许的测量误差范围内(σ=2℃),利用红外热像仪检测储罐外表面温度反演估计储罐液位和罐内流体温度有较高的识别精度,误差在2%以内。该方法为储罐液位和内部温度识别提供了一种新的思路和途径。     

An enhancement algorithm based on adaptive updating template with Gaussian model for Si3N4 ceramic bearing roller surface defects detection

The method based on machine vision image processing is used to detect the surface defects of Si₃N₄ bearing roller. Owing to the variety of defects, small area and low contrast, it is easy to miss or error detection. In this paper, an adaptive update template defect enhancement algorithm based on Gaussian model is proposed. First, a large number of surface images of Si₃N₄ bearing roller are collected to obtain the non-defect background statistical feature, and the background characteristic curve is fitted by Gaussian model. Further, the initial background template is gained according to the Gaussian curve. Then, combined with the gray distribute of defect images and initial background template, unique adaptive update template can be established. Finally, subtraction operation and nonlinear enhancement are used to improve the comparison of defect information and background. Through inverse sorting, adaptive threshold segmentation and Canny operation, the precise positioning of defects is realized. The enhancement algorithm can effectively enhance the contrast and eliminate the influence of noise. The average detection time is 0.84s, and the detection accuracy is 96.2%.     

气体密封性能检测仪设计

针对压差法气体密封性能检测设计的缺点,提出容腔压降方案,可以避开温度扰动障碍,满足精度要求。     

基于机器视觉的管纱污渍检测算法研究

针对管纱表面污渍的质量问题,提出基于机器视觉的管纱表面污渍检测算法。先对图像进行预处理以及Sobel算法处理,消除光照不均以及背景纹理的影响,再用双高斯算法增强目标与背景的差异,然后用灰度矩形法削弱背景的噪声,最后采用动态阈值的方法提取目标。实验结果表明,该算法对管纱表面污渍的检测有一定的准确性与实时性。     

Accurate measurement of the kinetic coefficient of friction between a surface and a granular mass

A device has been developed for correct measurement of the kinematic coefficient of friction between a cohesionless granular material and a surface. Particle size may range from 0.5 up to about 9 mm, depending somewhat on the desired accuracy. Sliding velocity of the granules with respect to the surface was varied up to 33 m/s. The accuracy of the magnitude of the friction coefficient is within ±1.5%, whereas the variation of this coefficient, as caused by the sliding velocity, can be obtained with an accuracy of ±0.5%.     

注塑制品表面缺陷自动识别软件的构建

以注塑制品的四种常见表面缺陷(油污、点缺陷、压痕、划痕)为研究对象,设计了基于机器视觉技术和图像处理技术的自动识别系统,对注塑制品表面缺陷进行检测。介绍了软件系统结构及人机界面设计,并对图像处理的实验结果进行了讨论,制品缺陷的平均识别率达84.44%。系统在解决传统检测方法难题的同时,提高了注塑制品表面缺陷识别的准确性和自动化程度。     

SSAW焊管超声检测试验研究

分析了脉冲反射法和TOFD法的检测原理,对缺陷进行了定位、定量的方法研究。然后制作了模拟SSAW焊管缺陷试块,用两种方法对这些缺陷进行深度和长度测量,对检测结果进行分析得知,对于坡口未熔合等方向性缺陷,TOFD检测精度远大于常规超声波检测精度。最后,用手动检测模拟SSAW自动检测,发现超声检测存在漏检现象,对焊缝中部的漏检缺陷可以采用TOFD法进行补充检测。TOFD法定量精度高,但不适合测定表面缺陷,两种方法结合对SSAW焊管进行缺陷检测,则获得了理想的检测效果。     

基于机器视觉的塑胶制品表面缺陷自动检测技术

详细阐述了机器视觉的照明系统、CCD摄像机、图像增强、图像分割、图像识别等关键技术的发展现状,并介绍了机器视觉技术在塑料电子元件、带材、PVC建材、薄膜、皮革、塑料瓶子等塑胶制品表面缺陷自动检测中的应用现状.