Detection Of Internal Defects Research Articles

Toward robust and high-throughput detection of seed defects in X-ray images via deep learning

BackgroundThe detection of internal defects in seeds via non-destructive imaging techniques is a topic of high interest to optimize the quality of seed lots. In this context, X-ray imaging is especially suited. Recent studies have shown the feasibility of defect detection via deep learning models in 3D tomography images. We demonstrate the possibility of performing such deep learning-based analysis on 2D X-ray radiography for a faster yet robust method via the X-Robustifier pipeline proposed in this article.Results2D X-ray images of both defective and defect-free seeds were acquired. A deep learning model based on state-of-the-art object detection neural networks is proposed. Specific data augmentation techniques are introduced to compensate for the low ratio of defects and increase the robustness to variation of the physical parameters of the X-ray imaging systems. The seed defects were accurately detected (F1-score >90%), surpassing human performance in computation time and error rates. The robustness of these models against the principal distortions commonly found in actual agro-industrial conditions is demonstrated, in particular, the robustness to physical noise, dimensionality reduction and the presence of seed coating.ConclusionThis work provides a full pipeline to automatically detect common defects in seeds via 2D X-ray imaging. The method is illustrated on sugar beet and faba bean and could be efficiently extended to other species via the proposed generic X-ray data processing approach (X-Robustifier). Beyond a simple proof of feasibility, this constitutes important results toward the effective use in the routine of deep learning-based automatic detection of seed defects.

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 <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 < 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.

Monitoring of Älvkarleby test embankment dam using 3D electrical resistivity tomography for detection of internal defects

AbstractElectrical resistivity tomography (ERT) is a potential-based method for detecting internal erosion in the core of embankment dams using the electrodes installed outside. This study aims at evaluating the practical capability of ERT monitoring for detecting internal defects in embankment dams. A test embankment dam with in-built well-defined defects was built in Älvkarleby, Sweden, to assess different monitoring systems including ERT and the defect locations were unknown to the monitoring teams. Between 7500 and 14,000 ERT data points were acquired daily, which were used to create the distribution of electrical resistivity models of the dam using 3D time-lapse inversion. The inversion models revealed a layered resistivity structure in the core that might be related to variations in water content or unintentional variations in material properties. Several anomalous zones that were not associated with the defects were detected, which might be caused by unintentional variations in material properties, temperature, water content, or other installations. The results located two out of five defects in the core, horizontal and vertical crushed rock zones, with a slight location shift for the horizontal zone. The concrete block defect in the core was indicated, although not as distinctly and with a lateral shift. The two remaining defects in the core, a crushed rock zone at the abutment and a wooden block and a crushed rock zone in the filter, were not discovered. The results cannot be used to fully evaluate the capability of ERT in detecting internal erosion under typical Swedish conditions due to limited seepage associated with the defects. Furthermore, scale effects need to be considered for larger dams.

Reliability of Noninvasive Sonic Tomography for Detection and Quantification of Internal Defects in Yoshino Cherry

Abstract Preventive maintenance through monitoring and early detection is important for trees that are structurally weak due to internal decay. Once internal decay has occurred, recovery of lost strength is irreversible. The decaying tree becomes vulnerable to external impacts, which may lead to breakage or windthrow. The Yoshino cherry (Cerasus×yedoensis (Matsum.) “Somei-yoshino”) accounts for a high proportion of landscape trees in Asia. Testing the reliability of noninvasive sonic tomography (SoT) to estimate internal defects could facilitate structural stability evaluation and proactive preventive management. We performed noninvasive SoT and invasive resistance microdrilling to detect and quantify internal defects in old and large Yoshino cherry trees. Generally, SoT reflected the areas and location of severe structural defects, showing a correlation with the resistance microdrilling results (R² = 0.542, p < .001). However, when cracks were present in disks, the SoT overestimated defects and errors mainly observed in the transition zone (green). Based on these findings, when evaluating the SoT, careful decisions are needed to distinguish between actual defects and broad sonic shadows. Furthermore, additional evidence from drilling resistance is required to differentiate between incipient decay and cracks. The results of this study provide valuable insights for enhancing the interpretation of sonic tomograms. Study Implications: We suggest defect boundaries in sonic tomograms should be redefined to improve the accuracy of interpretation for any new species being studied. Careful drilling resistance measurements should be taken, especially in transition zones (green areas in tomogram) and incipient decay zone (light brown areas). Noninvasive sonic tomograms only reflect acoustic properties of the tested cross-section of a tree and cannot be considered an actual representation of the internal conditions until the readings are calibrated to destructive sampling results.

Research Status and Progress on Non-Destructive Testing Methods for Defect Inspection of Micro-Electronic Packaging

Abstract In a highly competitive and demanding micro-electronics market, non-destructive testing (NDT) technology has been widely applied to defect detection and evaluation of micro-electronic packaging. However, the trend of micro-electronic packaging toward miniaturization, high-density, ultrathin, ultralight, and with small chip footprint, poses an urgent demand for novel NDT methods with high-resolution and large penetration depth, which is utilized for internal defect detection and identification of advanced complicated packages. The conventional NDT methods for micro-electronic packaging mainly include optical visual inspection, X-ray inspection, active infrared thermography, scanning acoustic microscopy (SAM), atomic force microscopy (AFM), laser Doppler vibration measuring technique, scanning SQUID microscopy (SSM), electrical impedance spectroscopy, scanning electron microscope (SEM), and so on. This paper aims to provide a review of addressing their basic principles, advantages, limitations, and application researches in the field of defect inspection of micro-electronic packaging. Moreover, in order to overcome the shortcomings of the existing NDT methods, this paper emphasizes a novel NDT approach, called hybrid ultrasonic-laser digital holographic microscopy (DHM) imaging inspection method, and discusses its basic principle, merits, key techniques, system construction, and experimental results in detail. When some key technical problems can be solved in further research, this method will become a potentially promising technique for defect detection and evaluation of advanced complicated packages.

Accuracy improvement of inner defects of cylindrical components using ultrasonic detection with modified ALOK method

The accuracy of defect localisation and its size quantification is poor in the detection of internal defects of cylindrical components using the ultrasonic Amplituden und Laufzeit Orts-Kurven (ALOK) method. The influence of acoustic beam spread is not taken into consideration in the ultrasonic ALOK method, resulting in difficulties with the precise characterisation of the defect state. To address this, the relationship between the acoustic distance, amplitude, ultrasonic frequency, size and depth of hole defects was studied. The acoustic distance curve and the amplitude curve were fitted and then the localisation model of the defect was obtained. The acoustic beam spreading angle and echo sound pressure were introduced and then the size quantification model for defects was acquired based on principal component analysis (PCA). Both the simulated and experimental results show that the modified ALOK algorithm improved the detection accuracy of the defect location and its size and the relative error of defect sizing decreased by more than 35% compared with the original algorithm.

Non-Destructive Testing of Laser Welded Hydrogen Storage Tank Liner Using Optical Coherence Tomography

The Type 4 hydrogen storage tank has an internal liner made of polymer material. Instead of the joining of two cylinder-shaped liners through butt-joint thermal fusion bonding, a faster joining can be achieved through lap-joint laser welding. Non-destructive testing is required to inspect the quality of laser-welded areas and to detect pore defects. In this study, the interface of a laser-welded region is investigated using optical coherence tomography to detect internal pore defects, where the maximum thickness of the liner is found to be approximately 3 mm. Optical coherence tomography images with defects are then used for supervised YOLOv2 deep learning. Real-time detection of internal defects is successfully performed on laser-welded liner samples.

Research on electric field characteristics under different length interface air gap defects in cable terminals of high-speed train

It is vital that accurately and effectively detecting the insulation status of cable terminal to ensure the safe and stable operation of high-speed trains. Accordingly, a novel monitoring system for the internal defect detection of cable terminal was proposed based on electric field strength. Firstly, partial discharge and electric field strength along the surface of the air gap defects cable terminal umbrella skirt were measured and analyzed. The development of internal defects in cable terminals can be divided into three stages: S1, S2, and S3. Then, the interquartile range, variance, average value and dispersion coefficient of electric field strength have been proposed to characterize the development state of air gap defects in cable terminal. The results shows that the internal defects of cable terminals develop from 20%l to 100%l, the slope of the interquartile range of the distribution of magnitudes of the electric field intensity increases from 0.126 to 0.671, and the absolute value of variance and offset increase by about 81.3% and 80.66%, respectively. Therefore, it is feasible and effective to detect the development state of air gap defects inside the cable terminal by measuring and analyzing the electric field strength characteristics along the surface of the cable terminal umbrella skirt.

Characterization and depth detection of internal delamination defects in CFRP based on line laser scanning infrared thermography

In this article, the shape characterization and depth detection of delamination defects in carbon fiber reinforced polymer (CFRP) sandwich are achieved by detecting delamination defects through line laser infrared thermography. In terms of defect shape characterization, the temperature matrix difference imaging method is proposed to characterize the shape of internal defects, and the defect shape distortion is corrected by the principal component analysis algorithm to optimize the effect of defect shape characterization. In terms of defect depth detection, this study constructs a dataset of cooling curves for defects of different depths and builds an encoder–decoder convolutional neural network structure based on the attention mechanism, which is centered on a streamlined encoder–decoder structure as well as a highly integrated attention gate, which is capable of accelerating the extraction of features and reducing the number of iterations of the algorithm. The results demonstrate that we achieved an average classification accuracy of 88.9% for defects with a depth gradient difference of 0.2 mm, attaining the highest classification accuracy of 95% for a single defect. Furthermore, our approach outperformed several classical classifiers by achieving superior accuracy with fewer iterations. The experimental results demonstrate that the line laser scanning inspection method is effective in the task of detecting internal defects in CFRP materials.

Novel Coil Transducer Induced Thermoacoustic Detection of Rail Internal Defects Towards Intelligent Processing

A novel resonant tri-coil transducer is proposed to induce thermoacoustic (TA) signals for detecting rail internal defects. It consists of a ferrite plate-backed tri-coil and its associated circuits, a shielding layer, and a three-dimensional printed enclosure for supporting and securing. General TA detection theory is illustrated to understand how to detect rail internal defects. For achieving electromagnetic (EM) field shaping and focusing, the physical structure of the proposed transducer is optimized. It reduces backward EM fields and interference. Then, an equivalent circuit model between the proposed transducer and the rail is constructed to analyze optimal power transfer efficiency as a function of distance and frequency. To validate the design concept, the proposed transducer is fabricated for inducing TA signals, and experiments are performed. Several engineered rail specimens (UIC-60E) with different types of defects are prepared. The induced TA waves are found to be compressional. With selected features, a support vector machine algorithm-based multinomial classifier model is trained and employed for intelligent processing. Then, the proposed transducer is integrated into the self-designed automatic rail monitoring system. Experimental results show that the self-designed system can achieve an accuracy rate of 96.4%, successfully identifying defects in defect classification. The proposed transducer has low cost in fabrication and maintenance, focused magnetic fields with a large lift-off distance and enhanced acoustic wave intensity. Compared to earlier version of the system, the spacing between the transmitter and the rail has been further increased.

Automated Identification and Localization of Rail Internal Defects Based on Object Detection Networks

The timely identification of rail internal defects and the application of corresponding preventive measures would greatly reduce catastrophic failures, such as rail breakage. Ultrasonic rail defect detection is the current mainstream rail defect detection method thanks to its advantages of strong penetration, high accuracy, and ease to deploy. The 2D B-scan image output by ultrasonic detectors contains rich features of defects; however, rail engineers manually identify and localize the defect image, which can be time-consuming, and the image may be subject to missing identification or mistakes. This paper adopted state-of-the-art deep learning algorithms for novel B-scan images for the automatic identification and localization of rail internal tracks. First, through image pre-processing of classification, denoising, and augmentation, four categories of defect image datasets were established, namely crescent-shaped fatigue cracks, fishbolt hole cracks, rail web cracks, and rail base transverse cracks; then, four representatives of deep learning object detection networks, YOLOv8, YOLOv5, DETR, and Faster R-CNN, were trained with the defects dataset and further applied to the testing dataset for defect identification; finally, the performances of the three detection networks were compared and evaluated at the data level, the network structure level, and the interference adaptability level, respectively. The results show that the YOLOv8 network can effectively classify and localize four categories of internal rail defects in B-scan images with a 93.3% mean average precision at three images per second, and the detection time is 58.9%, 376.8%, and 123.2% faster than YOLO v5, DETR, and Faster R-CNN, respectively. The proposed approach could ensure the real-time, accurate, and efficient detection and analysis of internal defects to a rail.

Electromagnetic Thermoacoustic Technique for Online Dynamic Detection of High-Speed Rail Internal Defect at Early Stage: Theory, Implementation, and Field Trial

Electromagnetic Thermoacoustic Technique for Online Dynamic Detection of High-Speed Rail Internal Defect at Early Stage: Theory, Implementation, and Field Trial

Research on the detection of surface and internal defects in cold rolled steel

Research on the detection of surface and internal defects in cold rolled steel

Modeling of Ultrasonic Flaw Detection Processes in the Task of Searching and Visualizing Internal Defects in Assemblies and Structures

Introduction. Inverse problems are a specific type of tasks where the consequences of phenomena are studied to identify their causes. They are widely used in scientific studies, specifically, those dealing with large amounts of experimental data. In the presented paper, inverse problems in mechanical engineering and structural diagnostics are considered. These areas require precise methods to identify internal defects in various materials, which can be critical to ensure the safety and efficiency of technical structures. Despite the many flaw detection methods available, there is a need for innovative developments that can provide higher accuracy and efficiency. This study integrates different scientific methods and technologies. It opens up new perspectives in nondestructive testing for the detection of internal defects in various materials and structures. Its objective is to develop and implement nondestructive testing methods based on a neural network device to improve the accuracy of defect identification, as well as to build a neural network model and evaluate its effectiveness for the refinement of ultrasonic visualization of internal defects in solid materials. In this regard, the task to be solved is to create a reliable tool for accurate visualization of sizes, shapes, location and orientation of internal defects in various materials.Materials and Methods. The technique of determining the geometric parameters of defects in materials through nondestructive testing is used. The approach combining modeling of ultrasonic wave propagation in acoustic medium and artificial neural network technologies is applied. This approach identifies nonlinear relationships between the geometry of defects and the amplitude-frequency and amplitude-time data obtained during signal analysis. Artificial neural networks are a model that can be trained on examples, which provides for an effective solution to problems that are difficult to express in traditional forms. The study uses the finite difference method in the time domain. It is applied to identify and visualize internal defects in materials using ultrasonic nondestructive testing and convolutional generative neural networks.Results. A convolutional neural network has been developed to visualize internal defects using ultrasonic nondestructive testing techniques. This neural network successfully determines the size of defects, their location, shape and orientation with high accuracy and reliability.Discussion and Conclusion. The authors highlight the key influence of defect size on the accuracy of ultrasonic imaging in various scenarios. The validation of the model for three different cases of defects with different mechanical parameters has shown that for successful visualization of defects, the wavelength of the ultrasonic pulse must be ten times smaller than the size of the defect. When analyzing the impact of defect size on the accuracy of the neural network, it is found that the visualization error increases for defects of smaller size. It has also been found that the relative speed of sound in materials has a greater effect on the accuracy of the method than the relative density of the material. Based on the results obtained by the authors, it can be argued that the developed methods and technical solutions are of great importance for future research in the field of flaw detection. They have significant potential for scientific and practical applications.

Nondestructive testing of internal defects by ring-laser-excited ultrasonic

ABSTRACT To improve the excitation efficiency, detection utilisation rate, and signal-to-noise ratio of the laser ultrasonic non-destructive testing of internal defects, the spatial distribution of the excitation light source was changed to a ring-shaped laser source to excite ultrasound. Finite element analysis method was used to simulate and analyse the ultrasonic characteristics excited by the ring laser. The superposition of shear waves on the central axis of a ring laser exhibits considerable advantages for internal defect detection. Consequently, a method was proposed for detecting internal defects using a ring laser to excite and receive ultrasonic waves at the concentric and opposite positions of the ring laser. A ring-source laser ultrasonic detection system was designed to scan and detect circular hole defects with different diameters as well as crack defects with different depths, lengths, and inclination angles. The experimental results showed that the ring-source laser ultrasonic technique could effectively detect defect positions and depths and describe the defect length and external contour under certain conditions. Moreover, it could achieve more complex detection tasks. This study provides an effective reference for the application of laser ultrasonic non-destructive testing of internal defects.

Rail Internal Defect Detection Method Based on Enhanced Network Structure and Module Design Using Ultrasonic Images

Improving the detection accuracy of rail internal defects and the generalization ability of detection models are not only the main problems in the field of defect detection but also the key to ensuring the safe operation of high-speed trains. For this reason, a rail internal defect detection method based on an enhanced network structure and module design using ultrasonic images is proposed in this paper. First, a data augmentation method was used to extend the existing image dataset to obtain appropriate image samples. Second, an enhanced network structure was designed to make full use of the high-level and low-level feature information in the image, which improved the accuracy of defect detection. Subsequently, to optimize the detection performance of the proposed model, the Mish activation function was used to design the block module of the feature extraction network. Finally, the proposed rail defect detection model was trained. The experimental results showed that the precision rate and F1\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${F}_{1}$$\\end{document} score of the proposed method were as high as 98%, while the model’s recall rate reached 99%. Specifically, good detection results were achieved for different types of defects, which provides a reference for the engineering application of internal defect detection. Experimental results verified the effectiveness of the proposed method.

Automated detection and segmentation of internal defects in reinforced concrete using deep learning on ultrasonic images

Periodic inspection of concrete structures for detecting internal defects through nondestructive imaging is recommended for ensuring structural safety and integrity. Ultrasonic imaging is one of the emerging and widely accepted techniques in nondestructive diagnostics of concrete structures. However, manually screening the ultrasonic images to detect and localize anomalies over large areas especially those from public infrastructure is tedious and prone to misjudgments. This work proposes a Region-based CNN to automatically detect, localize, and segment those signatures corresponding to defects from the noisy ultrasonic images with multiple features. This network replaces the conventional RoI pooling adopted in RCNNs with RoIAlign. To optimize the model for real-life complexity and noise levels, the network is trained on real experimental data. The ultrasonic image dataset has been generated by applying the imaging technique based on the Synthetic Aperture Focusing Technique on reflected waveform data obtained on various specimens with installed defects. The proposed network has been trained on the labelled image dataset containing signatures corresponding to multiple flaws like debonded rebars, delaminations, and planar crack-like defects in concrete members. The predictions of the trained model over the test datasets detect and masks the pixels corresponding to defects in ultrasonic images with a best mAP of 0.98. Segmentation results of the developed model upon testing are found superior to other state-of-the-art defect detection networks.

Efficiency enhancement and application of laser ultrasonic longitudinal wave based on glass constraint

Due to the weak longitudinal signals generated by laser ultrasound in the thermoelastic mechanism, the characteristic echoes are weak when evaluating the interior of solids, thus limiting its application to internal defect detection. A glass confinement layer is introduced to enhance the longitudinal excitation effectiveness. Specially, a thermoelastic model of laser ultrasound with a glass confinement is established to explain the mechanism of the enhancement of the longitudinal wave effectiveness, and the effect of the glass width on the longitudinal wave generated by the base ultrasound is investigated. The effect of the glass confinement layer on the enhancement of the effectiveness of the internal defects detection is studied. The simulation and experimental results show that the longitudinal waves with high signal-to-noise ratio induced from thermoelastic effect are excited similar to the ablation mechanism, which greatly improves the excitation efficiency of the longitudinal waves. The deep detection defects and the accurate localization of depth information are realized with an error of no more than 1.2%.

Comparison of In-situ Nondestructive Testing and Ex-situ Methods in Additive Manufactured Specimens for Internal Feature Detection

ABSTRACT The effectiveness and repeatability of Additive Manufacturing (AM) technologies has been often measured through rigorous testing, both in-situ and ex-situ. In-situ nondestructive testing (NDT) has been used to understand the AM process and detect anomalies during the build. Ex-situ testing has been used for material characterization and internal defect detection. In this work, in-situ NDT and ex-situ testing methods are compared as applied to a Fused Filament Fabrication (FFF) process to detect defects that have formed internally in the part. Intentionally embedded features and defects are monitored and analyzed in-situ and ex-situ. Results from an automated image-based real-time defect detection system, an ultrasound ex-situ system, and a tensile testing system are compared to detect defects and their effects on mechanical properties. Real-time image-based defect detection is found to effectively spot defects as they form and allow the potential for implementing defect correction methods, while ex-situ tensile testing can be used as a form of validation to in-situ results.

Artificial intelligence-aided detection of rail defects based on ultrasonic imaging data

Railroads are one of society’s fundamental infrastructures, facilitating the transportation of passengers and goods over vast distances. Rail status data is immensely important for ensuring the safe and efficient operation of railroad networks. However, analyzing ultrasonic inspection data is a labor-intensive process and relies heavily on the expertise of experienced inspectors. To detect internal defects of the rail accurately and automatically, this paper proposes a customized image recognition method based on a convolutional neural network with limited B-scan rail image data collected within the industry. The proposed method uses EfficientNet-b7 as the backbone network to fully extract the B-scan rail image data features. With the help of transfer learning and data augmentation techniques, the backbone network is substantially enhanced so that it can understand high-level features of the object without being trained with large-scale B-scan image data. We establish a real-world internal rail defect dataset with 280 B-scan images and test our proposed method. The results reveal that the highest accuracy of the other mainstream CNN-based methods is 76.25% and the accuracy of the traditional method based on a support vector machine classifier trained with Tamura texture and LBP features is 60.00%. Our proposed EfficientNet-b7 model classifies rail defect B-scan images with an accuracy of 85.00%, precision of 84.71%, and recall of 85.00%. Compared to other rail internal defect detection methods, this method is more accurate. With the help of transfer learning and data augmentation, our proposed method achieves better performance and requires less data.