Nondestructive Defect Detection Research Articles

Quasi-visualizable detection of deep sub-wavelength defects in patterned wafers by breaking the optical form birefringence

Abstract In integrated circuit (IC) manufacturing, fast, nondestructive, and precise detection of defects in patterned wafers, realized by bright-field microscopy, is one of the critical factors for ensuring the final performance and yields of chips. With the critical dimensions of IC nanostructures continuing to shrink, directly imaging or classifying deep-subwavelength defects by bright-field microscopy is challenging due to the well-known diffraction barrier, the weak scattering effect, and the faint correlation between the scattering cross-section and the defect morphology. Herein, we propose an optical far-field inspection method based on the form-birefringence scattering imaging of the defective nanostructure, which can identify and classify various defects without requiring optical super-resolution. The technique is built upon the principle of breaking the optical form birefringence of the original periodic nanostructures by the defect perturbation under the anisotropic illumination modes, such as the orthogonally polarized plane waves, then combined with the high-order difference of far-field images. We validated the feasibility and effectiveness of the proposed method in detecting deep subwavelength defects through rigid vector imaging modeling and optical detection experiments of various defective nanostructures based on polarization microscopy. On this basis, an intelligent classification algorithm for typical patterned defects based on a dual-channel AlexNet neural network has been proposed, stabilizing the classification accuracy of λ/16-sized defects with highly similar features at more than 90%. The strong classification capability of the two-channel network on typical patterned defects can be attributed to the high-order difference image and its transverse gradient being used as the network's input, which highlights the polarization modulation difference between different patterned defects more significantly than conventional bright-field microscopy results. This work will provide a new but easy-to-operate method for detecting and classifying deep-subwavelength defects in patterned wafers or photomasks, which thus endows current online inspection equipment with more missions in advanced IC manufacturing.

Nondestructive Defect Detection in 3D X-ray Microscopy Data of Ball Grid Array Solder for Void Detection in Solder Joints using Deep Learning

Abstract A deep learning-based nondestructive approach for void segmentation in BGA solder balls using 3D x-ray microscopy is presented.

Multi-perspective feature collaborative perception learning network for non-destructive detection of pavement defects

Pavement defects detection has made significant progress with the development of convolutional neural networks. Due to the topological complexity of pavement defects regions, most existing detection methods are limited to a limited number of types of defects, which results in unsatisfactory detection results for multi-classified defects. Several studies have used enhanced data to address this issue, however, few have considered the inefficiency caused by imbalanced differences between classes from the perspective of the network model itself. In light of this issue, our goal is to fill this gap by better exploring the fusion of different feature information to improve the performance of the network. As a result, we propose a multi-perspective feature collaborative perception learning network (MFCPLNet), which uses the model's contextual information to guide high-quality multi-classified defects detection and that can be adapted to various pavement conditions. Specifically, for adaptive feature extraction, an irregular perceptual extraction network (IPENet) with enhanced edge information is first designed to draw upon hybrid deformable convolution for efficient network block hybridization. Additionally, a multi-balanced collaborative learning mechanism (MCLM) consisting of auxiliary axial information and 3D feature guidance is constructed to enhance deep semantic features, improve the localization of defects in defective regions, and compensate for weak category supervision. Lastly, in the detection head, an iterative soft aggregation module has been designed to enhance scalable spatial interactions. The results of extensive testing were based on publicly available pavement defects datasets. Based on the proposed network, the mAP index reaches 29.6% and 33.8%, respectively, an increase of 8.3% and 4.6% over the baseline, which is qualitatively and quantitatively superior to mainstream methods.

FFF print defect characterization through in-situ electrical resistance monitoring

Fused filament fabrication is a popular fabrication technique. Currently there is a need for in-situ monitoring modalities to gather real-time information on prints, both for quality control and closed-loop control. Despite current advancements, effective and affordable in-situ monitoring techniques for non-destructive defect detection of voids and bonding quality are still limited. This work demonstrates in-situ monitoring of fused filament fabrication through electrical resistance measurements as an alternative to thermal and optical methods. A new, easy-to-implement setup is demonstrated which measures the electrical resistance of a conductively doped filament between the nozzle and single or multi-electrodes on the bed. Defects can be located in an unprecedented way with the use of encoded axes in combination with the observed resistance variations throughout the part. A model of the anisotropic electrical conduction is used to interpret the measurements, which matches well with the data. Warping, inter-layer adhesion, under-extrusion and overhang sagging print defects can be observed in the measurements of parts with a complex geometry, which would be difficult to measure otherwise. Altogether in-situ electrical resistance monitoring offers a tool for optimising prints by online studying the influence of the print parameters for quality assessment and it opens up possibilities for closed-loop control.

Quantifying the effect of X-ray scattering for data generation in real-time defect detection.

X-ray imaging is widely used for the non-destructive detection of defects in industrial products on a conveyor belt. In-line detection requires highly accurate, robust, and fast algorithms. Deep Convolutional Neural Networks (DCNNs) satisfy these requirements when a large amount of labeled data is available. To overcome the challenge of collecting these data, different methods of X-ray image generation are considered. Depending on the desired degree of similarity to real data, different physical effects should either be simulated or can be ignored. X-ray scattering is known to be computationally expensive to simulate, and this effect can greatly affect the accuracy of a generated X-ray image. We aim to quantitatively evaluate the effect of scattering on defect detection. Monte-Carlo simulation is used to generate X-ray scattering distribution. DCNNs are trained on the data with and without scattering and applied to the same test datasets. Probability of Detection (POD) curves are computed to compare their performance, characterized by the size of the smallest detectable defect. We apply the methodology to a model problem of defect detection in cylinders. When trained on data without scattering, DCNNs reliably detect defects larger than 1.3mm, and using data with scattering improves performance by less than 5%. If the analysis is performed on the cases with large scattering-to-primary ratio (1<SPR<5), the difference in performance could reach 15% (approx. 0.4mm). Excluding the scattering signal from the training data has the largest effect on the smallest detectable defects, and the difference decreases for larger defects. The scattering-to-primary ratio has a significant effect on detection performance and the required accuracy of data generation.

Nondestructive detection of CFRP subsurface defects using transient lock-in thermography

Nondestructive detection of CFRP subsurface defects using transient lock-in thermography

Non-destructive detection and analysis of weld defects in dissimilar pulsed GMAW and FSW joints of aluminium castings and plates through 3D X-ray computed tomography

AbstractThis work focuses on porosity formation during the welding of dissimilar aluminium alloys (cast and sheet) by pulsed gas metal arc welding (GMAW) with different travel speeds (12–14 mm/s) and by friction stir welding (FSW). The case study concerns the assembling of a battery-pack enclosure prototype. The welded specimens were scanned by 3D X-ray computed tomography. The cast base material (BM) shows a porosity percentage of 1.45%, and it is characterized by pores with a strong hyperbolic relationship between equivalent diameter and sphericity. Considering the GMAW beads, porosity rises with the travel speed (from 1.80 to 5.12%), due to the reduction of the opening window in which pores can escape. Pores with volume higher than 0.10 mm3 rise with the travel speed, representing from 9.75 to 32.98% of the total porosity. These pores are responsible for the weaker hyperbolic connection for sphericity found for the GMAW beads. On the other hand, FSW mixes and homogenizes the pores in the cast BM. The novelty of the paper lays in proving the strong potentialities of FSW for weld porosity reduction. A re-designing of the battery-pack enclosures is necessary to limit arc welding in marginal areas, which are not crucial for sealing but necessary to create a stable platform to be subsequently sealed with FSW.

Preliminary study on terahertz non-destructive testing for defect detection in hot melt joints of polyethylene pipes

With the rapid development of the economy and society, non-metallic pipelines are more and more widely used in oil and gas fields, water supply projects, urban gas systems, and so on. Among the conventional nondestructive testing (NDT) methods, only ultrasonic, radiographic, and penetration testing tools are applicable to non-metallic pipe examination from a theoretical perspective, while their actual efficiency is quite low. In this work, terahertz (THz) non-destructive detection of typical welding defects in polyethylene-pipe hot melt joints is carried out. The characteristic THz wave signal is obtained, and the relationship between the THz wave and the defect type is established. Finally, a THz signal processing model for defect detection in polyethylene-pipe hot melt joints is established. Thus, this research lays a foundation for the application of the THz non-destructive testing technology to polyethylene pipeline engineering.

Effect of beam and gate offset size on x-ray focal spot resolution of a cold cathode electron beam

In x-ray imaging, high resolution is essential, particularly in sectors such as medical and industries where the need for nondestructive defect detection is required. Previous research has shown that altering beam design and the number of gates offset holes has an impact on focal spot size (FSS). However, the specific effects of beam size and offset size were not thoroughly assessed. In the present study, the influence of beam size and gate offset size was evaluated by utilizing a cold cathode electron beam. Various beam sizes were employed to achieve a small FSS, and subsequently, the smallest beam was utilized to examine the impact of gate offset size. In doing so, the smallest FSS of 0.25 mm vertical and 0.33 mm horizontal was attained without the utilization of any additional focusing lens. This illustrates that by adjusting the beam size and gate offset size, it is possible to attain a small FSS, facilitating the development of an economically viable x-ray imaging beam.

Nondestructive detection of fruit fly (Bactrocera dorsalis) infestation in Alphonso and Totapuri varieties of mango fruits by thermal imaging technique

AbstractThermal imaging is a noncontact and nondestructive sensing technique that provides thermal signatures of objects by detecting the infrared radiations emitted by them above 1 K. The nondestructive detection of internal defects and infestations has vast applications in the Food Industry. In the present experiment, a C3 thermal camera (FLIR systems) having a thermal sensitivity of &lt;0.10°C was used for the detection of fruit fly infestation in mango fruits. The healthy mango fruits of both varieties had a uniform surface temperature with a variation of only ±0.1°C. The maximum temperature difference between the healthy and fruit fly infested regions in fruit fly infested Totapuri mango was found to be 4.56°C in the heated and dark conditions. The maximum difference in the surface temperature of 4.3°C was observed between the healthy and infested regions of the infested Alphonso mango under heated and dark conditions. The difference in temperature between the healthy and infested regions of mangoes can be attributed to the changes in the internal structure, thermal properties, and biochemical properties of the healthy and infested regions of the infested mango fruits. The statistical analysis showed a significant difference (p &lt; 0.05) in surface temperature between the infested and healthy regions of fruit fly infested Alphonso and Totapuri mango fruits. This envisages that nondestructive fruit fly infestation in Alphonso and Totapuri varieties of mango fruits can be detected under active thermography.Practical ApplicationsThe quality of mango fruit can be described in terms of external and internal quality. The external quality parameters such as color, size, and weight are usually considered during the sorting of mangoes. The internal defects such as diseases, physiological disorders, and damage due to pests are also significant in mango processing. The visual inspection methods for detecting pest infestation in fruits are nonreliable, time‐consuming, and subjective. Therefore, there is a need for efficient, reliable, and accurate methods to detect fruit fly infestation in mango fruits. A nondestructive method for the detection of fruit fly infestation has the potential to increase fruit quality and acceptability, thereby benefiting processors, traders, and consumers. Thermal imaging has a distinct advantage over other nondestructive quality assessment techniques as it is noncontact, practically feasible, and economically viable.

Improved automatic defect detection from X-ray scans for aluminum conductor composite core wire based on modified Skip-GANomaly

The Aluminium Conductor Composite Core (ACCC) is a critical component of power transmission that needs automated nondestructive defect detection to prevent wire breaking accidents and ensure regular city operations. Unfortunately, the current supervised target detection scheme is limited in its generalization ability due to the diversity of defect morphology and the high cost of acquiring samples. To overcome these challenges, we propose an automatic defect detection method based on a modified Skip-GANomaly model using X-ray images of ACCC wires as input. We experiment with semi-supervised deep learning reconstruction methods to mitigate the lack of abnormal samples and improve the generalization ability of unknown defects. We also explore hierarchical reconstruction and Skip-Connection adjustment to enhance the reconstruction effect and defect identification based on Resnet50. The results show that our method can achieve high accuracy, recall, and precision levels of over 99 %, significantly highlighting the difference in defect morphology. Compared to existing supervised schemes, our Defect Detection Scheme with Hierarchical Reconstruction and Anomaly-Subtracted Image Discrimination (DHRAD) eliminates dependence on the number of abnormal samples and the diversity of defect types in the training set, and has the ability to detect defects with unknown morphology. Therefore, DHRAD provides an effective detection scheme for automatic nondestructive defect detection of ACCC in operation, based on computer vision and semi-supervised learning.

Особенности применения установок Плазменный фокус в радиационном и космическом материаловедении

The features of the use of Plasma Focus (PF) devices in the field of radiation and space materials science related to the specific generation of several types of radiation in the working chambers of the PF are considered. It is noted that the main role in the damage and modification of the structural state of the surface layers of materials irradiated in the cathode zone of the PF working chamber is played by powerful pulsed beams of working gas ions and flows of high-temperature plasma in the nano- and microsecond range of pulse duration. It is shown that in addition to the common combined beam-plasma effect on a target sample, PF installations allow the possibility of separate irradiation of the target material with ion and plasma flows. Three methods for separating pulsed flows of hot plasma and fast ions in space and time in a PF installation, described in the literature, are presented. It is noted that PF installations are a promising tool for solving fundamental and applied problems in radiation and space materials science, including for simulation experiments focused on problems of thermonuclear energy and aerospace engineering. It is also indicated that hard X-ray pulses with a duration in the nanosecond range, arising from pulsed discharges in the PF chamber, can be used for diagnosing fast processes, in particular for non-destructive flaw detection of machines and mechanisms during their operation.

Automatic recognition of defects behind railway tunnel linings in GPR images using transfer learning

Ground-penetrating radar (GPR) is widely used for non-destructive detection of hidden defects behind railway tunnel linings, including non-compactness, air- and water-filled voids, fractures, and insufficient lining thickness. However, manual interpreting a substantial volume of GPR B-scan images is time-consuming, and the results heavily depend on the professional knowledge of the practitioner and the available prior information. This paper presents an automatic recognition algorithm for detecting hidden defects, as well as estimating lining thickness in GPR B-scan images of railway tunnels using a You Only Look Once for Panoptic driving perception network (YOLOP) model. The process mainly contains three steps. Firstly, a real GPR dataset of underground objects is employed for transfer learning to enhance the robustness of the prediction model. Secondly, a dataset containing 2,245 real GPR B-scan images of railway tunnel linings is created. Thirdly, the established dataset is used to retrain the YOLOP model and assess the prediction accuracy for identifying hidden defects and lining thickness in GPR images. Field experiment results demonstrate that the trained hidden defects recognition algorithm achieves an average accuracy of 80.6%, with a detection time of only 0.02 s for a GPR image sizing of 1024 × 512. Therefore, it is concluded that the proposed algorithm is efficient and reliable for automatic recognition of hidden defects behind railway tunnel linings.

Nondestructive Defect Detection in Battery Pouch Cells: A Comparative Study of Scanning Acoustic Microscopy and X‐Ray Computed Tomography

The identification and location of critical defects inside battery cells before the performance decreases or safety issues arise remain a challenge. This study compares two nondestructive testing methods for the 3D visualization of defects at different depths inside a pouch battery cell: scanning acoustic microscopy (SAM) and X‐ray computed tomography (CT). A manufactured pouch cell with eight electrode sheets is used for this investigation. SAM using a 15 MHz transducer in reflection mode can detect defects at depths of up to four electrode sheets with a lateral resolution of 150 μm in 2 min. CT can locate defects on all eight stacked electrode sheets inside the pouch cell. The CT measurements take about 12.5 h. Both methods can complement each other in detecting defects inside thin pouch cells as an end‐of‐line test after the production or for qualifying individual battery cells for second‐life applications. As an in‐line quality check, SAM has proven to be a cost‐effective and efficient method for detecting defects such as misalignment on stacked electrodes. Both methods have the potential to expand the portfolio of nondestructive quality assurance tests in the production of lithium‐ion battery cells. This contributes to increasing the safety and productivity of battery technology.

Defect-depth-field algorithm for simulating magnetic flux leakage signals based on discrete magnetic dipole model

Magnetic flux leakage (MFL) testing is widely used for non-destructive detection of defects in ferromagnetic materials, in which fast and accurate processing of MFL signals is critical for reconstructing and evaluating the defects. Of the detection technologies, the existing magnetic dipole model (MDM) can simulate the MFL signals with regular or limited types of defects, but not with complex defects, such as naturally occurring corrosion material losses. With the help of finite element method (FEM), MFL signals of complex defects can be analysed but demanding time-consuming calculation. In this paper, an improved MDM method, i.e., the discrete MDM (DMDM) is proposed, and an algorithm for automatic MFL signals processing is further developed based on the DMDM and the defect-depth-field concept. By comparison, the difference of the simulated MFL signals between the DMDM and the FEM is less than 3%, and the maximum discrepancy of the MFL signals between the simulated and the experimental is about 6%. In addition, the simulation with the DMDM can be thousands of times more efficient than the FEM, indicating significant advantage over the existing method in both improving the accuracy and enabling the fast signal processing of reconstruction for any types of defects in MFL testing.

Infrared in-line monitoring of flaws in steel welded joints: a preliminary approach with SMAW and GMAW processes

The non-destructive full-field non-contact thermographic technique is applied for non-destructive flaw detection of the welded joints, in real-time and offline configuration. In this paper, a thermographic procedure for real-time flaw detection in manual arc welding process is presented. Surface temperature acquisitions by means of an IR camera were performed during arc welding process of 8 specimen both for calibration and validation of the numerical model. The investigated variables are the technique (manual stick arc (SMAW) and gas arc (GMAW) welding) and the joint shape (butt and T joint) for steel joints, in sound conditions and with artificial flaws. Numerical simulation of welding thermal transients was run to obtain the expected surface temperature fields and thermal behavior for different welding parameter configurations. Hardness measurement and micro-graphic analysis were performed to validate numerical simulation results. The real-time thermographic study of the weld pool gives direct indications of anomalies; local studies of the thermal transient and thermal profiles can detect some kind of flaws; microstructural analysis of Heat-Affected Zone (HAZ) and surrounding areas higlights the presence of austenite and martensite distribution which justifies the thermal transients and thermal profiles for different welding configurations. Comparing real-time IR acquisition of the welding process with simulated thermal contours of sound processes provides information of presence of some kind of flaws. Since most of the flaws are generated in the weld pool, it is possible to recognize anomalies directly from the thermal acquisitions or with post-processing the acquired data.

Deep Learning Based Online Nondestructive Defect Detection for Self-Piercing Riveted Joints in Automotive Body Manufacturing

Self Piercing Riveting (SPR) is widely used for joining lightweight and dissimilar materials in automotive body manufacturing, the quality of which directly affects the safety of vehicles. However, there is still no reliable method that can be used for SPR quality control without destructive test and manual intervention. This paper presents an online non-destructive SPR defect detection method based on deep learning. By learning the temporal dependencies of punch force varying with rivet displacement under different joint combinations, the proposed method can provide real-time defect alarms and avoid the enormous cost of joint dissection. We develop an SPR parameter selection mechanism to rule out the irrelevant parameters, which enhances the learning performance. For the problem of model overfitting caused by the savage imbalance of SPR data, we design a conditional generative adversarial network based data generation model. In order to accommodate the difference in defect patterns between factory and laboratory, we devise a transfer learning based model migration method, which substantially reduces the amount of labeled factory data for model training. The evaluations on real SPR data collected from two car assembly lines of Audi and NIO verify that the proposed method achieves a high detection accuracy and a low missing rate in SPR defect detection.

Fruit Freshness Evaluation using a Real-Time Industrial Framework for Deep Learning Ensemble Approaches

Abstract: Consumers give a high value on fruits' freshness, and manual visual grading presents challenges due to labor effort and inconsistent results. This research suggests an effective machine vision system for automating a visual assessment of fruit freshness and attractiveness based on cutting-edge deep learning algorithms and ensemble methodologies. The suggested architecture enables the non-destructive and economical detection of fruit defects by utilizing convolutional neural networks (CNNs). To attain high classification accuracy, which acts as the performance metric, the system utilizes ensemble deep learning models. Fruit photographs are used to train the algorithm, enabling precise fruit quality assessment. This framework revolutionizes the inspection process by using computer vision in real-time for industrial applications in fruit freshness detection

Detection of grinding-induced defects in titanium matrix composites by polarized laser scattering

In high-speed grinding of TiC-reinforced titanium matrix composites (TiC-TMCs), grinding-induced defects, e.g., chinks and cracks, respectively caused by titanium matrix smearing and TiC fracturing, can reduce the fatigue life of a ground part. Since TiC-TMCs have heterogeneous and discontinuous material properties, it is difficult to detect the defects of a ground part by the traditional non-destructive methods. In this study, the polarized laser scattering (PLS) detection method was applied for the first time to detecting the grinding-induced defects of the additively manufactured (AMed) TiC-TMCs. Different surface morphologies were obtained at various grinding speeds. A correlation between the polarized laser scattering signal and the percentage of defects was revealed. With an increase in grinding speed, the surface and subsurface defects were reduced, and the PLS signal intensity was decreased correspondingly. This study provides a new non-destructive defect detection method for metal matrix composite parts.

Research on internal defect detection of epoxy glass fiber composites based on terahertz imaging technology

Composite materials are characterized by light weight and high specific strength, which have been widely used in the field of transportation. The rapid and non-destructive detection of defects in the process of composite material production can be regarded as the key and guarantee for the safety. In this paper, the terahertz imaging technology was used to explore the detection of epoxy glass fiber composite samples with varied defects. Terahertz time-domain spectroscopy (THz-TDS) of different depths of defects in epoxy glass fiber samples were corrected by multiple pretreatment methods. Different variable selection methods were used to screen the spectral data. The Partial Least Squares (PLS) and Least Squares Support Vector Machines (LS-SVM) defect depth quantitative models were established, respectively. The visualization of the defect area is achieved by combining with different types of binary threshold segmentation methods. It was found that the ALS-SPA-LS-SVM model presents the best effects, and Rp and RMSEP of prediction set are 0.9998 and 0.0152 respectively. The grayscale histogram bimodal method has the best effect on detecting the defect area. It is feasible to detect the defects in epoxy glass fiber by THz-TDS combined with LS-SVM and image processing technology, which can achieve rapid and nondestructive detection of the location, depth and area of the defects. It provides the technical basis for defect detection in the process of composite material production. In this way, this study provides new methods and means for defect detection and quality supervision of other types of composite materials.