An automated radiographic NDT system for weld inspection: Part II-Flaw detection

An automated radiographic NDT system for weld inspection has been developed. The entire system consists of two major components: weld extraction and flaw detection. Part I of the system implemented a weld extraction methodology, which has been presented in a previous paper. This paper presents Part II of the system dealing with the detection of welding flaws. The flaw detection methodology is developed based on the fitted line profiles of a weld image and consists of four modules: preprocessing, curve fitting, profile-anomaly detection, and postprocessing. The successful detection rate and false alarm rate of the methodology are reported based on the test results of 24 representative welds.     

Flaw detection in radiographic weld images using morphological approach

It is necessary to detect suspected defect regions in the radiographic weld images to find the type of flaw and its causative factors. This requires processing of radiographic images by a suitable approach. This paper presents an approach to process these radiographic weld images of the weld specimens considering morphological aspects of the image. The proposed approach first determines the flaw boundaries by applying the Canny operator after choosing an appropriate threshold value. The boundaries are then fixed using a morphological image processing approach i.e. dilating few similar boundaries and eroding some irrelevant boundaries decided on the basis of pixel characteristics. The flaws detected by this approach are categorized according to their properties.     

基于改进卷积神经网络的管焊缝X射线图像缺陷识别方法

针对卷积神经网络(CNN)应用于焊缝探伤图像识别时,目标区域占比小,局部信息冗余,激活函数小于零时出现硬饱和区导致模型对输入变化较敏感、网络参数难以训练的问题,采用超像素分割算法(SLIC)和改进的ELU激活函数构建CNN模型进行焊缝探伤图像缺陷识别. 首先,在CNN模型中使用ELU激活函数,在缓解梯度消失时对输入噪声产生更好的鲁棒性,同时,利用SLIC算法对图像像素进行像素块处理的特点,增大焊缝探伤图像中感兴趣区域的占比,降低局部冗余信息,提高模型在训练过程中的特征提取能力. 通过对焊缝探伤图像感兴趣区域提取并与所述CNN模型进行对比试验. 结果表明,该方法在焊缝探伤图像特征提取、训练耗时及识别准确率方面较传统卷积神经网络有更好的表现.     

An automatic system of classification of weld defects in radiographic images

In this paper, we describe an automatic detection system to recognise welding defects in radiographic images. In a first stage, image processing techniques, including noise reduction, contrast enhancement, thresholding and labelling, were implemented to help in the recognition of weld regions and the detection of weld defects. In a second stage, a set of geometrical features which characterise the defect shape and orientation was proposed and extracted between defect candidates. In a third stage, an artificial neural network (ANN) for weld defect classification was used. With the aim of obtaining the best performance of ANN three different methods for improving network generalisation was used. The results was compared with a method without generalisation. For the input layer, the principal component analysis technique was used in order to reduce the number of feature variables; and, for the hidden layer, a different number of neurons was used.     

Automatic weld defect identification from radiographic images

This paper presents a new approach for weld defect identification from radiographic images. This approach is based on the generation of a database of defect features using Mel-Frequency Cepstral Coefficients (MFCCs) and polynomial coefficients extracted from the Power Density Spectra (PDSs) of the weld segmented areas after performing pre-processing and segmentation stages. Artificial Neural Networks (ANNs) are used for the feature matching process in order to automatically identify defects in radiographic images. The performance of the proposed approach is evaluated using 150 radiographic images in the presence of various types of noise and blurring. The experimental results show that the proposed approach can be used in a reliable way for automatic weld defect identification from radiographic images in noisy environments, and can achieve high recognition rates.     

Improving the accuracy of computer-aided radiographic weld inspection by feature selection

This paper presents new results of our continuous effort to develop a computer-aided radiographic weld inspection system. The focus of this study is on improving accuracy by feature selection. To this end, we propose two versions of ant colony optimization (ACO)-based algorithms for feature selection and show their effectiveness to improve the accuracy in detecting weld flaws and the accuracy in classifying weld flaw types. The performances of ACO-based methods are compared with that of no feature selection and that of sequential forward floating selection, which is a known good feature selection method. Four different classifiers, including nearest mean, k-nearest neighbor, fuzzy k-nearest neighbor, and center-based nearest neighbor, are employed to carry out the tasks of weld flaw identification and weld flaw type classification.     

基于超声信号和图像融合的焊缝缺陷识别

超声无损检测已被广泛用来检测材料内部的缺陷,然而对缺陷性质的识别始终是检测的难点,为此研究了一种基于超声信号和图像融合的焊缝缺陷识别新方法.该方法充分利用检测数据,通过对缺陷回波信号特征与缺陷形态特征的数据融合,实现了焊缝缺陷的有效识别.利用自主研制的超声成像手动检测系统对含有气孔、夹渣、裂纹、未焊透和未熔合五类典型焊接缺陷的焊件进行了检测,分别提取缺陷的超声回波信号特征和缺陷图像的形态特征,构建神经网络实现超声信号和图像特征的数据融合.结果表明,该方法实现了多类缺陷的识别,提高了缺陷识别率,有助于焊缝质量评定.     

Automated detection of welding defects in pipelines from radiographic images DWDI

This paper presents a method for the automatic detection and classification of defects in radiographic images of welded joints obtained by exposure technique of double wall double image (DWDI). The proposed method locates the weld bead on the DWDI radiographic images, segments discontinuities (potential defects) in the detected weld bead and extracts features of these discontinuities. These features are used in a feed-forward multilayer perceptron (MLP) with backpropagation learning algorithm to classify descontinuities in “defect and no-defect”. The classifier reached an accuracy of 88.6% and a F-score of 87.5% for the test data. A comparison of the results with the earlier studies using SWSI and DWSI radiographic images indicates that the proposed method is promising. This work contributes towards the improvement of the automatic detection of welding defects in DWDI radiographic image which results can be used by weld inspectors as a support in the preparation of technical reports.     

炭素制品X射线图像的缺陷特征分析与选择

针对炭素制品X射线检测图像的特点，对缺陷及其特征提取与选择技术进行了研究。分析了炭素制品生产中易产生的缺陷类型及缺陷的成像特征，在此基础上，从缺陷样本中提取了19个特征值。以特征组合分类能力数学模型为适应度函数，设计了基于遗传算法的特征选择策略，实现了对缺陷原始特征量的优化选择。利用BP神经网络分类器及选择的特征值对缺陷进行了模式分类。研究结果表明，提出的选择方法是比较有效的，可以用于缺陷的识别与分类。     

An adaptive-network-based fuzzy inference system for classification of welding defects

In this paper, we describe an adaptive-network-based fuzzy inference system to recognise welding defects in radiographic images. In a first stage, image processing techniques, including noise reduction, contrast enhancement, thresholding and labelling, were implemented to help in the recognition of weld regions and the detection of weld defects. In a second stage, a set of 12 geometrical features which characterise the defect shape and orientation was proposed and extracted between defect candidates. In a third stage, an adaptive-network-based fuzzy inference system (ANFIS) for weld defect classification was used. With the aim of obtaining the best performance to automate the process of the classification of defects, of all possible combinations without repetition of the 12 features chosen, four were used as input for the ANFIS. The results were compared with the aim to know the features that allow the best classification. The correlation coefficients were determined obtaining a minimum value of 0.84. The accuracy or the proportion of the total number of predictions that were correct was determined obtaining a value of 82.6%.     

Edge detection and its application to recognition of arc weld image

Abstract Image sensor has been one of the key technologies in intellectualized robotics welding. Edge detection plays an important role when the vision technology is applied in intellectualized welding robotics technologies. There are all kinds of noises in welding environment. The algorithms in common use cannot be applied to the recognition of welding environment directly. The edge of images can be fell into four types. The weld images are classified by the characteristic of welding environment in this paper. This paper analyzes some algorithms of edge detection according to the character of welding image, some relative advantages and disadvantages are pointed out when these algorithms are used in this field, and some suggestions are given. The feature extraction of weld seam and weld pool are two typical problems in the realization of intellectualized welding. Their edge features are extracted and the results show the applicability of different edge detectors. The trndeoff between precision and calculated time is also considered for different application.     

带管端探伤的全管体自动超声波探伤系统

设计了一种带管端探伤的全管体自动超声波探伤系统。系统由超声波探头、超声波探伤仪器、图像跟踪系统、PLC电气控制系统和全管体探伤机械装置组成,把焊缝探伤和母材探伤合为一体,实现螺旋管全管体的自动化超声波探伤。在全管体探伤机械装置中的焊缝管端检测装置后增加焊缝跟踪装置,在母材管端检测装置后依次增加管尾检测装置、母材管端探头体和管头检测装置,实现了管端自动探伤,解决了管端盲区问题,提高了探伤速度,保证了螺旋管探伤的可靠性和准确性,降低了人力资源成本。     

Welding defect detection from radiography images with a cepstral approach

This paper presents a new approach for feature extraction from radiography images acquired with gamma rays in order to detect weld defects. In this approach, images are lexicographically ordered into 1D signals. Then, Mel-Frequency Cepstral Coefficients (MFCCs) and polynomial coefficients are extracted from these signals, one of their transforms, or both of them. Discrete Wavelet Transform (DWT), Discrete Cosine Transform (DCT), and Discrete Sine Transform (DST) are tested and compared for efficient feature extraction. Neural networks are used for feature matching in the proposed approach. Sixteen radiography images containing seventy three weld defects are used to evaluate the performance of the proposed approach. For performance evaluation, the tested images are degraded by Gaussian, impulsive, speckle, or Poisson noises with and without blurring. The experimental results show that the proposed approach can be used in a reliable way for automatic defect detection from radiography images in the presence of noise and blurring.     

Geometric characterization of a circumferential seam by automatic segmentation of digitized radioscopic images

The present study deals with the nondestructive control of a circumferential seam by digital radioscopy. A series of images for one complete revolution of the welded component is available. We first resort to a joint approach by simulation and experimentation. This approach allows the detection of the molten zone limits for an initial image. We then develop a segmentation method that permits automatic extraction of the geometric characteristics of the set of images representative of the weld. These measures supply fast and automatic control of the weld quality. Results are shown for real components.     

基于FGT-FBP重构算法的焊接缺陷检测

针对焊接过程产生的缺陷,提出一种磁光成像传感的模糊灰度变换和滤波反投影(FGT-FBP)重构检测方法. 研究焊接缺陷的几何特征,通过分析裂纹和未熔合两种不同焊接缺陷在交变磁场励磁下的磁光成像特征,设计模糊规则,对磁光图像进行模糊灰度变换. 增强磁光图像对比度,使焊接缺陷形态趋势可视化,实现描述磁光成像焊接缺陷细节的无参考型图像评估方法. 对FGT处理的焊接缺陷磁光图进行旋转投影,并经过快速傅里叶变换和改进的滤波器进行滤波去噪,消除伪影后进行反投影变换实现焊接缺陷图像的重构. 利用滤波反投影重构算法进行去噪,可有效突出焊接缺陷特征. 最后结合阈值分割和边缘检测实现焊接缺陷检测. 结果表明,该方法能较准确检测裂纹和未熔合两种焊接缺陷.     

Enhancement of ultrasonic images for sizing of defects by time-of-flight diffraction

A signal processing technique is presented for significantly sharpening the resolution of ultrasonic images, similar to those acquired in the nondestructive evaluation of girth welds in oil/gas pipelines. This enhancement allows a much improved estimate of the exact size of any detected anomaly in the weld, such that fracture mechanics can be used to gauge the probability of weld failure. The algorithm is based on the synthetic aperture focusing technique (SAFT), combined with a variation of Wiener filtering and autoregressive spectral extrapolation. An analytical model of the transducer is used to construct an appropriate reference spectrum for the deconvolution operation, and accounts for the dependence of a beam's frequency spectrum on the position of a flaw relative to the transmitter. Experimental results are used to provide an estimate of the improvement in flaw sizing accuracy.     

基于B样条曲线的X射线图像焊缝缺陷分割与提取

针对大部分X射线数字图像低对比度、背景起伏大以及纹理复杂的问题,在X射线数字焊缝图像预处理之后,采用B样条曲线对列灰度曲线进行拟合,获得光滑而且顺畅的曲线.在此基础上进一步提取曲线的极值点,并通过定义的波动阈值以及边界阈值进行两次极值点集合的修正.最后利用数学形态学及中值滤波对缺陷的形状和大小进行了修正.结果表明,该技术有效地解决了X射线图像由于焊缝纹理复杂导致缺陷提取困难的问题,有利于实现X射线图像焊接缺陷的自动提取.     

一种焊缝X射线数字图像的缺陷提取算法

准确提取焊缝X射线图像中的缺陷是图像自动识别的主要问题。针对这一难题,提出了一种以邻域灰度平方差变换为基础,结合灰度级形态学重构和边缘检测,辅之边缘保持滤波和已标注连接分量图像融合的缺陷提取新算法。该方法能可靠地提取出低对比度X射线数字图像中的焊缝缺陷部分。     

基于改正二维主成分分析的焊缝表面缺陷检测

机器人焊接因零件形状不规则和焊接工艺复杂不可避免带来各种焊缝缺陷. 针对二维主成分分析应用于焊缝表面缺陷检测时面临计算复杂度高、分类准确率低及无法进行增量学习等问题,提出了一种基于均值更新的增量二维主成分分析(mean updated incremental two-dimensional principal component analysis,MUI2DPCA)算法,并将MUI2DPCA和前馈神经网络( feedforward neural network,FNN)相结合进行焊缝表面缺陷在线检测. 首先,对相机捕获的视频帧图像进行预处理得到焊缝局部块图像. 然后,利用MUI2DPCA在线提取局部块图像的模式特征. MUI2DPCA对图像的特征主成分进行增量迭代估计,降低计算复杂度,并且能够增量更新当前的样本均值,减少无关特征变化对主成分收敛性的影响. 最后,利用FNN建立提取的模式特征与焊缝类别之间的联系,实时返回焊缝表面缺陷的检测信息. 试验结果表明,该检测方法平均分类准确率为95.40%,平均处理速度可达29帧/s,能够满足焊缝在线检测的实时性要求.     

微间隙焊缝磁光成像分形维数识别方法

针对紧密对接、无坡口、肉眼难以分辨的微间隙焊缝,采用磁光成像方法获取焊缝位置信息,并运用分形维数方法解决焊缝磁光图像存在较多干扰的问题.根据图像大视野相关信息处理图像,避免受到图像微小细节干扰对焊缝位置识别精度的影响,较传统图像处理法具有显著的抗干扰性.对焊缝磁光图像进行滤波去噪,将图像细分成块并计算出每个图像块的分形维数,再选取合适阈值对图像进行分割,准确地提取出焊缝中心位置.试验结果表明,运用分形维数提取磁光图像焊缝边缘区域特征,能够获取较准确的焊缝位置信息,为焊缝跟踪控制提供重要基础.