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,本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,本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。     

A Study on Railway Surface Defects Detection Based on Machine Vision

The detection of rail surface defects is an important tool to ensure the safe operation of rail transit. Due to the complex diversity of track surface defect features and the small size of the defect area, it is difficult to obtain satisfying detection results by traditional machine vision methods. The existing deep learning-based methods have the problems of large model sizes, excessive parameters, low accuracy and slow speed. Therefore, this paper proposes a new method based on an improved YOLOv4 (You Only Look Once, YOLO) for railway surface defect detection. In this method, MobileNetv3 is used as the backbone network of YOLOv4 to extract image features, and at the same time, deep separable convolution is applied on the PANet layer in YOLOv4, which realizes the lightweight network and real-time detection of the railway surface. The test results show that, compared with YOLOv4, the study can reduce the amount of the parameters by 78.04%, speed up the detection by 10.36 frames per second and decrease the model volume by 78%. Compared with other methods, the proposed method can achieve a higher detection accuracy, making it suitable for the fast and accurate detection of railway surface defects.     

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。     

基于激光多普勒频移的钢轨缺陷监测

针对现阶段我国铁路上应用的探伤设备只能在天窗时间进行人工巡检,无法在线监测的问题,提出一种基于超声导波的激光多普勒频移法钢轨内部缺陷监测方法。首先,引入环境温度作为变量改进了半解析有限元方法,并应用该方法获得了我国无缝线路CHN60钢轨在特定温度下的频散曲线。通过分析振型并结合激励响应算法确定了适于检测缺陷的模态及其激励方式,从而激励该超声导波模态使其在钢轨中传播。然后,应用半反半透玻璃镜将激光分为参考光和测量光,测量光通过Bragg Cell进行频偏照射钢轨表面,通过反射光产生的多普勒频移与参考光干涉得到光强度变化曲线,经过信号处理及标定测得钢轨内部缺陷的回波速度信号,再经过数字化处理和计算得到缺陷的位置。最后,在北京环形铁路试验基地进行了现场实验,以钢轨接地孔模拟钢轨内部核伤,得到缺陷定位误差均小于0. 5 m,验证了该方法的可行性。使用激光多普勒频移方法检测导波信号从而定位缺陷的方法可以有效避免由于换能器接触性测量而产生的误差。该方法在不影响列车的正常运营的同时,实现了全天候无间断的在线监测,提高了检测效率。     

圆结构光视觉三维点管道缺陷检测及重构

管道内表面的缺陷检测对于保证介质运输安全,避免泄漏和爆炸事故非常重要。在分析三维点分布几何特征的基础上,对基于视觉检测获得的管道内表面三维点,通过判断同一圆周上相邻点法线夹角变化检测管道内表面的凹凸缺陷,采用空间点相邻三角平面法线加权平均获取空间点的法线。依据三维点呈圆周分布的特点,采用相邻圆周上点顺次连接进行快速三角剖分。基于上述方法对实际测量的管道内表面三维点和仿真三维点分别进行了凹凸缺陷检测和三维重构。该方法能实现径向变化小于0.1 mm的管道凹凸缺陷的检测和识别,三维测量精度为0.081 mm。     

基于改进梯度幅值的包装缺陷检测算法研究及应用

针对包装质量检测精度易受外界光照影响的问题，在已有基于梯度幅值相似性的缺陷检测算法基础上，将局部二值模式算子引入到该算法中，提出了一种基于改进梯度幅值相似性的缺陷检测算法。该算法利用局部二值模式算子的旋转不变性和灰度不变性的特点，并将其与图像的梯度幅值特征进行融合后用于包装的缺陷检测中，提升了缺陷检测算法对光照的鲁棒性。实验结果表明，相比传统梯度幅值缺陷检测算法，该算法具有更好的抗光照影响能力，并且对于不同光照情况下的包装缺陷，该算法的检测准确率可达96.57%。因而，该算法能够被广泛地用于包装缺陷检测中，提高缺陷检测的精度。     

板状结构Lamb波频散补偿与聚焦接收成像检测

对板状结构中Lamb波的频散和多模现象进行了理论分析,采用有限元仿真方法获取了Lamb波与缺陷作用后的回波信号,对信号进行聚焦接收处理,即在一定距离范围内利用导波频散特性对各个模式信号进行频散补偿,提取补偿后信号的幅值,结果表明当所得信号幅值最大时,所对应的补偿距离等于缺陷与换能器之间的实际距离.利用这一结论,提出了以...     

基于机器视觉的圆锥滚子外观缺陷检测系统研究

回顾了近年来国内外工业生产中圆锥滚子外观缺陷分割领域的研究进展,为了实现更加简单高效的滚子外观缺陷检测,设计了一套合适的光学成像系统用于图像采集,提出了最大类间方差法和局部阈值算法的缺陷分割,对分割后图像进行特征提取并进行SVM分类。实验结果表明,该算法有较高的准确率和较好的适应性,优于传统的滚子外观缺陷分割算法,可以满足工业生产中的实际需求。     

Recognition algorithm of internal defect images of thermal battery北大核心CSCD

针对目前热电池内部装配缺陷检测效率低、准确度不高的问题,研究了一种可精准分割内部电池堆图像并能够准确识别缺陷种类的方法。首先采用水平、垂直积分投影法对目标电池堆边缘特征进行提取,利用局部自适应对比度增强算法对局部不清晰部分进行细节纹理增强;然后研究了缺陷结构的灰度特性,计算提取出缺陷特征参数;最后使用BP（back propagation）神经网络和CART（classification and regression tree）决策树对特征参数分类识别,并根据分类准确度进行权重分配,将加权融合后的结果作为检测的最终判据。实验结果表明:该方法对2 000个样本的检测准确度达98.9%,为热电池的X射线缺陷检测提供了有效的途径。     

基于数字图像处理技术的树脂眼镜镜片缺陷检测研究

通过设计的镜片自动检测光学硬件系统,给出了树脂镜片精细杂质图像处理措施,将图像差分法应用于镜片图像边缘的去除;同时解决了镜片检测灰尘影响的问题;另外运用改进型的归一法解决杂质计数问题,为自动化分级分拣系统的研发提供重要的数据支持,为镜片缺陷检测系统的后续分级分拣的数字控制模块提供了可操作性。该检测系统检测精度为0.03 mm、速度为1片/s,检测误差控制在5%。     

基于CiteSpace的内表面缺陷检测研究进展与趋势

为分析内表面缺陷检测的发展历程、趋势和研究动态，通过对WoS和CNKI数据库中该领域相关文献的检索，共搜集相关文献英文4 708篇，中文818篇，利用可视化分析软件CiteSpace对文献数据开展共现分析、聚类分析等知识图谱研究，分析内表面缺陷检测领域在国家、机构及研究人员层面的分布现状及合作情况，梳理研究热点和前沿趋势。研究发现内表面缺陷检测研究具有明显的多学科交叉属性，主要涉及分析化学、材料科学、光谱学、仪器仪表、机械工程和计算机等学科。近几年WoS数据库相关主题收录文献年增长率超过10%, CNKI年增长率超过20%,中美两国为本领域研究最为活跃的国家，两国发文量约占总发文量的40%,中国学者在无损检测、图像处理等领域的研究明显落后于国外学者，但在机器视觉和深度学习领域实现赶超。按照研究路线可将相关研究分为基于声光电热磁的检测和基于视觉成像的检测两类，其中前者包括采用不同技术手段获取光谱、超声和电磁图像并借助图像处理技术实现缺陷检测，而后者主要基于视觉图像进行缺陷识别和分类，目前已成为该领域主要的研究热点。内表面缺陷检测发展历程分为缺陷识别、缺陷分类、缺陷分析三个阶段，2000年...     

基于机器视觉的电机换向器质量在线检测系统开发

电机换向器质量检测是换向器生产线中的一个重要工序,但其仍采用人工方式,存在检测效率低、漏检率高等问题。针对此问题,本文运用图像处理和机器视觉技术,开发换向器质量在线视觉检测系统。该系统针对生产过程工艺多变,造成形状检测标准不一致问题,提出自适应学习模板方法;在轴孔孔径检测,提出基于Freeman链码改进的孔径快速检测算法;在端面缺陷中,提出基于改进视觉注意力模型的端面缺陷检测方法。实验结果表明,系统检测精度达到99.80%,漏检率为0%,F-measure值为99.89%。该系统能够快速有效检测换向器存在的外观质量问题,可满足换向器在线质量检测需求。     

A new synthetic detection technique for trackside acoustic identification of railroad roller bearing defects

The detection of railroad roller bearing defects is of great importance for railway traffic. The acoustic defective bearing detector (ADBD) via a trackside acoustic detection system (TADS) is a popular and reliable tool for trackside identification of railroad bearing incipient defects. However, the defective trackside acoustic signal of bearing is confronted with three challenges: demodulation of acoustic signal, removal of Doppler effect and enhancement of defect frequency. This paper proposes a synthetic detection technique to overcome these difficulties successively. In the technique, the signal envelope is first extracted by a new variable-resolution ridge demodulation (VRRD) method. Then the Doppler effect is removed in the envelope via a dynamic signal resampling method. Finally, the rectified defect frequency is enhanced through the means of logarithmic transformation and piecewise linear detrending. The VRRD technique extracts the envelope from the signal time–frequency distribution (TFD) along the Doppler shift curve of resonance frequency traced by an improved ridge extraction algorithm. The resampling method is carried out on the envelope according to the Doppler shift curve of defect frequency obtained by the same ridge extraction algorithm. The sidebands around defect frequency can be suppressed by the logarithmic transformation and the low-frequency non-linear trend can be removed by the detrending method. The signal amplitude is unified at the same time. Two practical bearing signals with inner-race and outer-race defects separately verify the effectiveness of proposed technique.     

像增强器视场缺陷检测方法研究

在数字式像增强器性能综合检测系统的基础上，研究了像增强器视场缺陷的检测方法。通过数字视频摄像机采集缺陷图像，利用数字图像处理方法确定缺陷性质，缺陷处理算法基于图像灰度特征而不是边缘特征提取。该方法适用于对检测对象有具体量化和定位要求，而不是简单的只有数量和形状特征要求的场合。实验表明，该检测方法能够较准确地检测像增强器的缺陷。     

多尺度传感网络失效节点检测系统设计

对多尺度传感网络中的失效节点进行准确检测与定位,实现故障节点的高效检测,保障传感网络的可靠运行。提出一种基于多传感器量化融合跟踪滤波检测的失效节点检测算法,并进行检测系统优化设计。构建多尺度传感网络的节点分布实体对象模型,进行失效节点检测系统总体设计和技术指标分析。设计基于多传感器量化融合跟踪滤波检测的失效节点检测算法。进行系统的硬件设计,包括A/D模块设计、时钟电路设计、程序加载电路设计、传感器通信模块设计和系统电源模块设计。在ARM Cortex?-M0平台上进行检测系统软件开发。系统仿真结果表明,该系统进行多尺度传感网络失效节点检测的准确度较高,提高了传感器网络的寿命周期。     

一种平板玻璃缺陷在线检测系统的研究

根据光干涉的原理和视觉检测方法,研究开发了基于机器视觉的玻璃缺陷检测系统。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,从而获得玻璃表面锡点、气泡、玻筋等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。系统处理周期最大不超过400 m s,能检测出最小0.3 mm的缺陷。     

基于空耦超声Lamb波的金属板状结构内部缺陷检测方法

针对接触式超声检测方法在金属板结构内部缺陷实际工程检测中存在的环境要求高、效率低、操作难度高等问题,提出了空耦超声Lamb波检测方法,该方法能更好地适应现场应用环境,提高检测效率,减少传感器数量。通过有限元仿真和实验分析比较了空耦超声检测与接触式超声检测两种方法接收到的信号和成像效果。结果表明:有限元仿真和实验中,空耦超声检测方法对缺陷位置的定位误差分别为2 mm和3.6 mm,接触式检测方法对缺陷位置的定位误差分别为2 mm和11.3 mm,空耦检测具有较高的定位精度;单侧激励条件下,适合采用A0模态Lamb波对板内缺陷进行检测;空耦超声检测可以通过调整信号接收角度接收单一模态Lamb波,避免伪像产生。该方法为后续金属板状结构内部缺陷的空耦超声检测提供参考。     

Leak Detection in Water Pipes Based on Maximum Entropy Version of Least Square Twin K-Class Support Vector Machine

Numerous novel improved support vector machine (SVM) methods are used in leak detection of water pipelines at present. The least square twin K-class support vector machine (LST-KSVC) is a novel simple and fast multi-classification method. However, LST-KSVC has a non-negligible drawback that it assigns the same classification weights to leak samples, including outliers that affect classification, these outliers are often situated away from the main leak samples. To overcome this shortcoming, the maximum entropy (MaxEnt) version of the LST-KSVC is proposed in this paper, called the MLT-KSVC algorithm. In this classification approach, classification weights of leak samples are calculated based on the MaxEnt model. Different sample points are assigned different weights: large weights are assigned to primary leak samples and outliers are assigned small weights, hence the outliers can be ignored in the classification process. Leak recognition experiments prove that the proposed MLT-KSVC algorithm can reduce the impact of outliers on the classification process and avoid the misclassification color block drawback in linear LST-KSVC. MLT-KSVC is more accurate compared with LST-KSVC, TwinSVC, TwinKSVC, and classic Multi-SVM.     

Computing communities in complex networks using the Dirichlet processing Gaussian mixture model with spectral clustering

Community detection becomes a significant tool for the complex network analysis. The study of the community detection algorithms has received an enormous amount of attention. It is still an open question whether a highly accurate and efficient algorithm is found in most data sets. We propose the Dirichlet Processing Gaussian Mixture Model with Spectral Clustering algorithm for detecting the community structures. The combination of traditional spectral algorithm and new non-parametric Bayesian model provides high accuracy and quality. We compare the proposed algorithm with other existing community detecting algorithms using different real-world data sets and computer-generated synthetic data sets. We show that the proposed algorithm results in high modularity, and better accuracy in a wide range of networks. We find that the proposed algorithm works best for the large size of the data sets.     

Independently tunable multichannel terahertz filters

We numerically demonstrate terahertz multichannel filters with independently tunable defect modes based on fractal photonic crystals. Single defect and multiple defects models are proposed to fabricate the multichannel terahertz filters. The facts that the wave functions of the defect states do not overlap and their bases are orthogonal lead to the independency among the defect modes. The simulated results theoretically provide the principle for fabricating independently tunable multichannel terahertz filters by utilizing one-dimensional photonic crystals with defects.