表面缺陷检测的机器视觉技术研究进展

物体表面缺陷检测技术是工业质检领域的一项重大课题,对工业生产有着重要的意义。针对近些年基于机器视觉的表面缺陷检测技术进行梳理总结。首先,列举了几种缺陷检测在工业领域的应用场景;其次从特征提取和分类算法的角度简要阐述了传统的机器视觉方法;重点探讨了缺陷检测中常用的经典神经网络结构和缺陷检测算法的最新发展,并介绍了两种常用的缺陷检测算法优化方式;最后,分析了缺陷检测领域面临的三大挑战：实时性问题、小样本问题和小目标问题,目的是为工业表面缺陷检测的研究提供有益的参考和脉络梳理。     

铁轨表面缺陷的视觉检测与识别算法

提出一种铁轨表面缺陷的视觉检测与识别算法。设计铁轨表面缺陷视觉检测与识别系统的总体结构,基于水平投影法提取铁轨表面区域,采用逻辑操作组合检测结果,使用 BP 神经网络进行缺陷分类。实验结果表明,该算法能准确地检测与识别铁轨表面的疤痕和波纹擦伤这2种缺陷,分类正确率分别达到99%和95%。     

基于YOLOv5的工具表面缺陷检测系统

在工业生产过程中，产品质量极易受到现有生产技术等客观条件的影响，因此需要对产品进行质量检验，其中，表面缺陷是产品质量合格的重要指标之一。现如今在表面缺陷检测方面的常见技术有渗透探伤、超声波检测、机器视觉等。利用基于深度学习的YOLOv5算法通过机器视觉识别工具表面不同缺陷种类，为工具产品的质量检验提供便利。     

基于光度立体和图像显著性的皮革缺陷检测

为了提高皮革缺陷检测效率，提出一种基于光度立体视觉和图像显著性的皮革缺陷检测算法。搭建光度立体视觉平台，完成不同角度的皮革样本采集，利用光度立体视觉技术计算皮革样本的合成图和表面法向量图；对表面法向量图进行曲率滤波操作，用近似表面粗糙度特征自适应选择合成图或滤波图；利用显著性目标检测算法完成皮革缺陷检测与定位。实验结果表明，与现有皮革缺陷检测方法相比，该算法能很好地检测不同材质皮革的多种缺陷，且准确率高，速度快。     

带钢表面质量在线检测系统研究与设计

为了提高现有带钢表面质量检测技术在缺陷检测精度与识别率上存在的问题,设计了基于机器视觉技术的带钢表面质量自动检测系统,从系统整体构成、视觉传感系统、软件开发及检测与分类算法等方面进行了深入地研究,实现了包括图像的采集、传输、缺陷的实时检测和定位,缺陷的分类以及缺陷的存储与报警等功能;实验结果表明,该系统可以对带钢表面常见的边裂、氧化、结疤等几十种不同类型的缺陷进行精确地检测;与现有的缺陷检测技术相比,该系统中设计的算法在检测精度和实时吞吐量上都具有很大优势.     

钢板表面质量机器视觉检测系统设计

针对国内钢厂采用人工方法检查钢板表面缺陷存在可靠性差的问题,开发设计了基于机器视觉技术的带钢表面缺陷自动检测系统.系统通过摄像头采集带钢表面的图像,然后采用图像处理及模式识别算法对图像进行实时处理和分析,从而检测出钢扳表面缺陷,并对缺陷进行自动分类识别.实验结果表明,系统能够对带钢表面进行实时在线监测,并能正确识别常见的带钢表面缺陷.     

机器视觉表面缺陷检测综述

目的 工业产品的表面缺陷对产品的美观度、舒适度和使用性能等带来不良影响,所以生产企业对产品的表面缺陷进行检测以便及时发现并加以控制。机器视觉的检测方法可以很大程度上克服人工检测方法的抽检率低、准确性不高、实时性差、效率低、劳动强度大等弊端,在现代工业中得到越来越广泛的研究和应用。方法 以机器视觉表面缺陷检测为研究对象,在广泛调研相关文献和发展成果的基础上,对基于机器视觉在表面缺陷检测领域的应用进行了综述。分析了典型机器视觉表面缺陷检测系统的工作原理和基本结构,阐述了表面缺陷视觉检测的研究现状、现有视觉软件和硬件平台,综述了机器视觉检测所涉及到的图像预处理算法、图像分割算法、图像特征提取及其选择算法、图像识别等相关理论和算法研究,并对每种主要方法的基本思想、特点和存在的局限性进行了总结,对未来可能的发展方向进行展望。结果 机器视觉表面缺陷检测系统中,图像处理和分析算法是重要内容,算法各有优缺点和其适应范围。如何提高算法的准确性、实时性和鲁棒性,一直是研究者们努力的方向。结论 机器视觉是对人类视觉的模拟,机器视觉表面检测涉及众多学科和理论,如何使检测进一步向自动化和智能化方向发展,还需要更深入的研究。     

周期纹理背景中的表面缺陷检测技术

基于机器视觉原理的自动光学表面缺陷检测技术是当今工业生产中在线检测表面缺陷的一种新的技术方法,是精密制造与组装工业过程中保证零部件表面质量的重要检测手段.以液晶面板TFT阵列表面缺陷自动光学检测为例,介绍了表面缺陷自动光学检测的基本组成原理,阐述了周期纹理背景表面上的表面缺陷检测方法、缺陷信息处理的基本过程与实用算法.针对表面缺陷检测图像处理技术难题,详细论述了表面缺陷扫描图像中的周期纹理背景傅里叶变换频域滤波方法、缺陷分割双阈值统计控制法,并用实验结果给出了例证.     

基于神经网络的零件缺陷机器视觉识别系统

零件缺陷检测是保证零件使用安全的重要手段。传统的零件缺陷检测法需要有操作人员参与其中,易受主观因素影响,检测的效率及精度得不到良好的保证。而采用机器视觉技术的检测法可实现实时在线的自动检测,无需人工参与,这就极大的提高了生产效率。本文以小轴承表面为研究对象,针对微小轴承的表面结构、尺寸、检测精度和缺陷特征,设计了基于BP神经网络的零件缺陷机器视觉在线自动检测系统,其采用机器视觉技术,构建了BP神经网络检测识别模型,采用进行图像特征提取的间接识别方法,对微小轴承缺陷进行实时检测。实验结果证明了人工神经网络模型的检测能力的可靠性。     

基于GPU的金属工件表面缺陷检测算法的研究

在金属工件的生产过程中,不可避免地会生产出一些不良品,必须进行快速识别。缺陷检测系统需使用图像采集设备采集金属工件的表面图像,完成混合噪声滤除等预处理后,进行图像配准并使用差影法分割图像,然后标记缺陷和提取缺陷纹理特征,最后进行工件缺陷的分类和识别。为了提高金属工件表面检测系统的检测速度,以满足高速生产流水线对检测系统的高实时性要求,依托GPU平台设计了一套合理的并行算法来完成不合格工件的自动检出工作。实验结果表明,在满足检测精度的前提下,基于GPU的并行图像处理算法相对于串行算法能取得较好的加速效果(实验环境为3.2~10.3倍加速比),为工件表面缺陷的快速检测提供了一种新的途径。     

基于超像素分割和随机森林的橡胶柱塞缺陷检测方法

针对直径为3 mm的小尺寸橡胶柱塞件端面,其受光斑、灰尘及纹理干扰不易分割提取缺陷轮廓的问题,提出一种结合SLIC（简单线性迭代聚类）和RF（随机森林）算法的缺陷检测系统。首先利用霍夫变换和各向异性扩散滤波对图像预处理,然后采用基于超像素分割的SLIC算法分割和提取缺陷区域,最后把获得的缺陷区域的五维形状特征作为RF分类器特征向量进行缺陷分类预测。结果表明,SLIC算法较传统的自适应阈值分割算法快了0.128 s,并且分割效果远好于传统算法,能够准确分割出小至0.5 mm的缺陷,整体检测流程平均耗时小于1.5 s,同时RF分类结果准确率达到97.3%。因此,本文的缺陷检测系统满足在线检测准确性和实时性的要求,可在实际工作中使用。     

工件表面缺陷检测投影分析与算法研究

以iphone4S工件为例,针对工件表面竖纹缺陷和注塑缺陷检测过程中受高频噪声影响的不足,提出了基于投影分析的缺陷检测算法。通过分析缺陷的特征,设计了峰谷值判定的竖纹缺陷检测算法、基于分块投影的工件边缘线检测和注塑缺陷积分值判定算法,以实现计算机对工件表面竖纹和注塑缺陷的自动检测。通过大量实验表明：以新算法为核心技术的检测算法,能够对竖纹缺陷和注塑缺陷检测准确率达到95%以上。     

Effect of bright and shade,and luminance difference of defect on defect detection in appearance inspection utilizing peripheral vision

This study considers the effect of bright and shade, and luminance difference of defects on defect detection in appearance inspection utilizing peripheral vision experimentally. Specifically, bright and shade of defect, luminance difference of defect, surface luminance of defect (evaluation index of the difficulty of defect detection, which was proposed in the previous study), and location of defect are designed as experimental factors, and their effects on defect detection rate are evaluated. As a result, it is clarified that the defect detection rate of the shade defects is lower than that of the bright defects, even if the surface luminance is at an identical level. It is also clarified that the defect detection rate of the luminance difference of 10 cd/m² becomes lower even if the surface luminance is at an identical level. Furthermore, these two trends are particularly remarkable for defects detected in the peripheral vision. Based on the results, in actual appearance inspection utilizing peripheral vision, it is necessary to take measures to detect the shade defects and/or the luminance difference of approximately 10 cd/m².     

A new strategy for defect inspection by the virtual inspection in semiconductor wafer fabrication

The defect of process equipments is a major factor that impairs the yields in the mass production of semiconductor wafer fabrication and it is a main supervision means to use high-resolution defect inspection tools to detect and monitor the defect damage. Due to the high investment costs of these inspection tools and the resulting decrease in the throughput, how to improve the sampling rate is an important issue for the associated inspection strategy. This paper proposes a new concept and implementation of virtual inspection (VI) to enhance the detection and monitoring of defect in semiconductor production process. The underlying theory of the VI concept is that the state variables identifications (SVIDs) of process equipments can reflect the process quality effectively and loyally. The approach of VI is to combine the application of the fault detection and classification (FDC), and the defect library and the re-engineering of inspection procedure to reach the full-scope of strategic objective. VI enables the defect monitoring to enter a new era by promoting the monitoring level of defect inspection from the previous lot-sampling basis to the wafer-sampling level, and hence upgrades the sampling strategy from random-sampling to full and right-sampling. In this study, various typical defect cases are utilized to illustrate how to create VI models and verify the reliability of the proposed approach. Furthermore, a feasible architecture of the VI implementation for mass production in semiconductor factory is presented in the paper.     

Fabric defect detection based on saliency histogram features

In order to increase the automatic quality control level in the textile industry, depending on the big data collected by the Internet of things of the textile factories, this paper proposes a novel visual saliency–based defect detection algorithm, which has the capability of automatically detecting defect in both nonpatterned and patterned fabrics. The algorithm employs the histogram features extracted from the saliency maps to detect the fabric defects. The algorithm involves three main steps: (1) saliency map generation to highlight the defective regions and suppress the defect‐free regions, (2) saliency histogram features extraction and selection to obtain the feature vectors that can effectively discriminate between the defective and defect‐free fabric images, and (3) fabric defect detection using a two‐class support vector machine classifier that has been trained using sets of feature vectors extracted from defective and defect‐free fabric samples. Experimental results show that our method yields accurate detections, outperforming other state‐of‐the‐art algorithms.     

基于机器视觉方法的焊缝缺陷检测及分类算法

对于在工业生产中如何有效地识别薄壁金属罐焊缝的缺陷及其类型判别的问题,提出了一种基于机器视觉技术的自动化焊缝缺陷检测及分类算法。利用混合高斯模型,提出了一种改进的背景差分法,主要用来提取焊缝缺陷的特征区域。在此基础上,以不同缺陷类型的缺陷面积、亮度及波形特征等差别作为依据,对焊缝缺陷进行了分类。实验检测结果表明,算法可以对主流的薄壁金属制罐焊缝缺陷类型进行准确的识别和归类,达到了96%以上的精确度。同时,算法的运算时间也能够满足在实际生产中的高实时性需求。     

Deep-learning-based anomaly detection for lace defect inspection employing videos in production line

Defect inspection plays an essential role in ensuring quality of industrial products. The most widely used human visual inspection method has some drawbacks such as high cost and low efficiency, which bring an eager demand for the application of automatic defect inspection algorithm in actual production. However, few industrial production lines use automatic detection devices due to the gap between data collected in the actual production environment and ready-made datasets. Lace is one of the industrial products which completely depends on manual defect inspection. The complex and fine texture of lace makes it difficult to extract regular patterns using the existing image-based defect inspection methods. In this paper, we propose to collect lace videos in the weaving stage and design a deep-learning-based anomaly detection framework to detect lace defects. The framework contains three stages, namely video pre-processing stage, pixel reconstruction stage and pixel classification stage. In the offline phase, only defect-free lace videos are needed to train the pixel reconstruction model and calculate the detection threshold by our adaptive thresholding method. In the online phase, the proposed framework reconstructs lace videos and performs defect inspection using reconstruction error and the pre-set threshold. As far as we know, this paper the first to detect fabric defects by videos. Experimental results on artificial defect videos demonstrate the effectiveness of the proposed framework.     

Further Experiences with Scenarios and Checklists

Software inspection is one of the best methods of verifying software documents. Software inspection is a complex process, with many possible variations, most of which have received little or no evaluation. This paper reports on the evaluation of one component of the inspection process, detection aids, specifically using Scenario or Checklist approaches. The evaluation is by subject-based experimentation, and is currently one of three independent experiments on the same hypothesis. The paper describes the experimental process, the resulting analysis of the experimental data, and attempts to compare the results in this experiment with the other experiments. This replication is broadly supportive of the results from the original experiment, namely, that the Scenario approach is superior to the Checklist approach; and that the meeting component of a software inspection is not an effective defect detection mechanism. This experiment also tentatively proposes additional relationships between general academic performance and individual inspection performance; and between meeting loss and group inspection performance.