X-ray Inspection Research Articles

Image enhancement method for low-light pipeline weld X-ray radiographs based on weakly supervised deep learning

X-ray inspection is the most intuitive approach for the non-destructive testing (NDT) of pipeline weld defects to avoid pipeline safety accidents. However, identifying pipeline weld defects in dark X-ray images is difficult due to low greyscale values. This paper proposed a weakly supervised network for denoising and enhancing low-light pipeline weld X-ray images. First, a semi-supervised network based on an improved Retinex-Net which implemented by self-paced learning was proposed to enhance illumination, yielding more natural X-ray images without artifacts, distortion, and overexposure. A new denoising network constrained by the X-ray images themselves was designed to achieve denoising while preserving the image detail. Qualitative comparison and quantitative analysis indicated that the proposed method outperformed other industrial image enhancement methods used for pipeline weld detection in terms of both subjective visual effects and objective metric values.

On revealing hidden entrainment damage during in situ tensile testing of cast aluminum alloy components

Tensile tests have been conducted on die-cast coupons of an Al-Si-Cu alloy, which have been found to contain no pores via X-ray inspection. Due to the digital image correlation used during tensile testing, either single or multiple strain concentrations have been detected and subsequently characterized. In all cases, oxide films have been found on fracture surfaces at the site of the strain concentrations. The analysis of a crack away from the fracture surface has also shown it to be an oxide bifilm. These results pour doubt on the effectiveness of quality assurance practices employed in industry.

Airtight welding technology for aircraft composite mold panel

In this paper, the airtightness problem caused by welding defects in the manufacturing process of composite mold is analyzed. The mechanism analysis of the weld defects of composite mold is conducted. According to the mechanism of weld failure, the welding process monitoring, weld metallographic analysis, weld X-ray inspection, panel airtightness detection, and other methods are used to evaluate the welding quality of different welding forms in the welding process of composite mold panels. The measures to optimize the welding quality of composite mold panels are put forward. Finally, through the actual verification, it is proved that the optimized air-tightness welding method is stable and reliable.

Methylammonium-Based Quasi-Two-Dimensional Perovskite Single Crystals for Highly Sensitive X-ray Detection and Imaging.

The strategy of introducing large organic cations into three-dimensional perovskites could reduce the dimensionality of perovskites to form quasi-two-dimensional (quasi-2D) perovskites, resulting in increased stability and reduced detection limits due to less ion migration. Herein, a quasi-2D perovskite single crystal (BDA)(MA)2Pb3Br10 (BDA = NH3C4H8NH3, MA = CH3NH3) with a layered structure was grown by the temperature-cooling solution method. The X-ray detector based on the (BDA)(MA)2Pb3Br10 single crystal has a sensitivity as high as 1984 μC Gy-1 cm-2 at 55.6 V/mm, and it could detect X-rays as low as 28.12 nGy s-1 at 22.2 V/mm. In addition, the X-ray imaging system based on the single-crystal device easily distinguishes between metals and plastics and exhibits a spatial resolution estimated as 250 μm, indicating the feasibility of (BDA)(MA)2Pb3Br10 crystals for X-ray imaging. This research offers a method for the design of quasi-2D layered perovskites and enhances photoelectronic applications in X-ray inspection and imaging.

Comparing the results of testing a longitudinal tube weld by nondestructive methods

Defects of longitudinal tube welds identified by X-ray inspection are studied in order to determine possible causes of their occurrence and to assess the reliability of nondestructive testing results. Metallographic analysis shows that the base metal and weld metal meet the requirements of regulatory documents. The strength properties and the chemical composition conform to the standards. The contamination with non-metallic inclusions in the segregation streamers of the sheet extending to the edge and falling into the fusion zone was detected in X-ray photographs and incorrectly interpreted as a defect. This assumption is based on the correlation of X-ray patterns with the position of the grooves in the zones of the studied welded joints, but it ignores the results of studying by other nondestructive testing methods, and most importantly, it is not confirmed by metallographic studies. Neither were these defects found in the tubes during magnetic flaw detection by means of a UMD-101M device in the line of production of longitudinally electric-welded tubes. Erroneous defect identification during X-ray inspection happens in tube making. The results of the study show the expediency of the integrated application of different testing methods for a reliable assessment of the quality of tubes or pipes of both small and large diameters.

A Case of 3Cs: Coronary Artery Calcification Masquerading as a Stent on Chest X-ray

Coronary artery calcification in patients with coronary artery disease is associated with increased major adverse cardiovascular events. They are usually diagnosed on a computed tomography scan or on an invasive coronary angiography. A 60-year-old male was diagnosed with acute coronary syndrome after presentation with one month of worsening angina. Coronary angiography showed triple vessel disease (mid left anterior descending artery and mid right coronary artery having total occlusion and a 70% stenosis in the left circumflex artery). Apart from this, there was a heavily calcified right coronary artery with a C configuration evident on fluoroscopy. A computed tomography coronary angiography was done to visualize the distal vessels. The calcium score was 5988. Retrospectively, on inspection of the chest X-ray, there was calcification on the right heart border. The patient was medically managed and referred to a cardiothoracic vascular surgeon for coronary artery bypass grafting. Coronary artery calcification on X-ray is rare and hence being presented along with the review of the literature.

A Novel 2D/3D X-ray Microscopy Alignment and Inspection Solution for Thermocompression Bonding (TCB) in a Highly Integrated Flip Chip Fan-Out Wafer Level Package (FO-WLP)

As packaging drives to extend the capabilities of Moore’s Law via heterogenous integration, numerous bonding technologies have emerged as potential pathways, allowing for integration of both a variety of devices in the lateral space as well as vertical integration of chips via stacking. As vertical integration on interposer becomes more complex due to shrinking CDs and increasing die layers, alignment and shape control in the bonding process becomes critical. Thermocompression bonding (TCB) is particularly well-suited for larger die, where localized reflow allows a better control of chip gap height and tilt throughout the bonding process. TCB systems require both high accuracy and repeatability in all dimensions to provide a reliable system in package. While a 2D system can provide a plan view assessment of a single bond layer between two chips, it falls short in characterizing the critical z-dimension in a highly integrated chip stacking scenario. Therefore a robust solution is needed for both system alignment validation in the x,y plane as well as a capability to explore gap height and bump quality/buried issues in high aspect ratio TSV. This paper will explore an x-ray application of a stacked 15-layer die-to-interposer structure to qualitatively validate overall health of the TSV (bond, pad, fill) via a nondestructive volumetric method as well as quantitatively evaluate stacked die for via/ball/pad alignment to determine overall alignment shift during the bonding process. In the test vehicle for a flip chip FO-WLP process flow first presented by IMEC, a silicon (Si) bridge is utilized to connect the logic and through-package via dies with a bump pitch of 20µm for the logic die interconnect. This is done by carefully aligning the logic dies on the carrier and then placing the through-package dies aligned to the carrier and logic dies. In the final step, a high accuracy placement and stacking TCB tool attaches the Si bridge. To perform an accurate assessment for bump alignment and bond quality, an initial high-resolution 3D x-ray scan is performed on a 1.5mm3 region of interest with over 2,000 individual projections at a 1.4µm voxel resolution with a 4x objective, taking approximately two hours. The dataset is put through a deep-learning algorithm which generates an AI model. This model is then applied to a faster scan of 15 minutes with a 1.4µm voxel resolution for the same field of view (FOV) on a nearby array and the data is collected and processed. To verify alignment of the entire sample, data is collected on the tightest pitch (in this case 20µm) oriented in both the vertical (x) and horizontal (y) plane in the through-silicon via (TSV) region. The z-plane is analyzed and processed by computer vision algorithms, and individual interconnects are then quantified and sorted for solder height (z), width (x,y) and TSV misalignment (deviation from expected center, x1y1 → x2y2). To validate the model, a high-resolution uncorrected 20h scan and a 15min scan based on deep learning are collected for the same region of interest (ROI). The data is comparable. Furthermore, the global misalignment data between both datasets demonstrates similar misalignment propagating through the 15-layer stack.

Implementation of dual-energy material decomposition technique in stationary CT baggage scanner withπ-angle sparsity for enhancing threat detection

Two-dimensional X-ray inspection systems are widely used in aviation security applications; however, they have inherent limitations in recognizing the three-dimensional (3D) shapes of hidden objects. Therefore, there is a growing demand for the implementation of advanced 3D X-ray inspection systems at airports for more accurate detection of threats in luggage and personal belongings. In this study, we designed a new stationary computed tomography (CT) baggage scanner with π-angle sparsity (i.e., 20 pairs of X-ray sources and line detectors were placed within a scan angle of 180°) and compressed sensing (CS)-based reconstruction, and implemented a dual-energy material decomposition (DEMD) technique in the proposed system to separate soft and dense materials of an examined object to enhance threat detection. To validate the efficacy of the proposed approach (CS/180°/P20), we conducted a feasibility study using numerical simulation before its practical implementation. Polychromatic projections were emulated at X-ray tube voltages of 60 and 140 kVp, and DEMD was applied to the projections prior to CT reconstruction. Conventional and dual-energy CT images were reconstructed using both standard filtered-backprojection (FBP) and state-of-the-art CS-based algorithms to compare the image quality. According to our simulation results, the CS-reconstructed images were almost unaffected by the clearly evident streak artifacts on the FBP-reconstructed images because of the use of 20 extreme sparse-view projections, and the image quality of the dual-energy CT images obtained using the proposed CT configuration was similar to that obtained using the conventional CT configuration with 720 dense projections, indicating the efficacy of the proposed approach. Consequently, high-quality dual-energy CT images of soft and dense materials were successfully obtained using the proposed stationary CT configuration.

Enhanced Identification of Internal Casting Defects in Vehicle Wheels Using YOLO Object Detection and X-Ray Inspection

Enhanced Identification of Internal Casting Defects in Vehicle Wheels Using YOLO Object Detection and X-Ray Inspection

Особенности формирования сварных соединений алюминиевого сплава АК9, полученного селективным лазерным плавлением

The problem of obtaining high-quality plate welds made of AL9 alloy and powder by selective laser melting is viewed. Welded butt joints of plates of AK9CH alloy with a thickness of 3.0 mm were made by melting welding (automatic argon arc and electron beam welding) and friction stir welding. After welding the base metal of the AK9 alloy plates and their welded joints were subjected to X-ray transmission, computed tomography, metallographic analysis and mechanical tests for static tension and static three-point bending. X-ray inspection of welded joints made by automatic argon arc and electron beam welding revealed multiple porosity inside welds. The nuclei of pores in the weld metal are spherical micropores, formed in the parent metal during selective laser melting. The diameter of these micropores is 150...200 microns. In the welding bath during fusion welding, micropores develop up to 420...1070 microns in diameter in case of argon arc welding and 215...420 microns in case of electron beam welding. Metallographic analysis of cross-sections of welded joints performed by fusion welding revealed a characteristic pore distribution in the weld metal. So in case of argon arc welding, the largest pores were located near the front surface of the weld. In the fusion zone at the border of the weld and the base metal, pores with a diameter of 80.220 microns were located in the form of chains along the entire thickness of the plates being welded. When testing welded joints for static tension, the destruction of joints occurred precisely in this zone. The strength coefficient of the AK9 alloy joints obtained by fusion welding is as follows: for automatic argon arc welding 0.46, for electron beam welding - 0.66. It is established that the problem of porosity of welded joints is eliminated when using solid -phase welding for jointing (friction stir welding). The strength coefficient of the jointing of the AK9CH alloy plates made by friction stir welding is at the level of 0.81.0.86 of the time resistance of the base metal. The destruction of welded joints occurs along the mixing zone. In the weld zone during friction stir welding, as a result of dynamic recrystallization, a fine-grained equiaxed structure with an average grain size of 4.5...6.2 microns was formed with virtually no pores. The parent metal is represented by a cellular structure in which there are pores up to 168 microns in size.

Dense metal corrosion depth estimation

Introduction: Metal corrosion detection is important for protecting lives and property. X-ray inspection systems are widely used because of their good penetrability and visual presentation capability. They can visually display both external and internal corrosion defects. However, existing X-ray-based defect detection methods cannot present and estimate the dense corrosion depths. To solve this problem, we propose a dense metal corrosion depth estimation method based on image segmentation and inpainting.Methods: The proposed method employs an image segmentation module to segment metal corrosion defects and an image inpainting module to remove these segmented defects. It then calculates the pixel-level dense corrosion depths using the X-ray images before and after inpainting. Moreover, to address the difficulty of acquiring training images with ground-truth dense corrosion depth annotations, we propose a virtual data generation method for creating virtual corroded metal X-ray images and their corresponding ground-truth annotations.Results: Experiments on both virtual and real datasets show that the proposed method successfully achieves accurate dense metal corrosion depth estimation.Discussion: In conclusion, the proposed virtual data generation method can provide effective and sufficient training samples, and the proposed dense metal corrosion depth estimation framework can produce accurate dense corrosion depths.

Thermal testing of corundum ceramic plates

A review of the methods of production of corundum ceramics with specified characteristics is carried out and possible causes of hidden defects are indicated. Ceramic samples for research were obtained by the method of slip casting and their physical properties were determined. The possibility of diagnosing the integrity of corundum ceramic plates was investigated by the method of active thermal non-destructive testing. Various testing schemes and sources of thermal stimulation of the studied samples were applied. As a reference method, X-ray inspection was used, which confirmed the characteristics of the defect detected by the thermal method in the control sample.

YOLO-CID: Improved YOLOv7 for X-ray Contraband Image Detection

Currently, X-ray inspection systems may produce false detections due to factors such as the varying sizes of contraband images, complex backgrounds, and blurred edges. To address this issue, we propose the YOLO-CID method for contraband image detection. Firstly, we designed the MP-OD module in the backbone network to enhance the model’s ability to extract key information from complex background images. Secondly, at the neck of the network, we designed a simplified version of BiFPN to add cross-scale connection lines in the feature fusion structure, to preserve deeper semantic information and enhance the network’s ability to represent objects in low-contrast or occlusion situations. Finally, we added a new object detection layer to improve the model’s accuracy in detecting small objects in dense environments. Experimental results on the PIDray public dataset show that the average accuracy rate of the YOLO-CID algorithm is 82.7% and the recall rate is 81.2%, which are 4.9% and 3.2% higher than the YOLOv7 algorithm, respectively. At the same time, the mAP on the CLCXray dataset reached 80.2%. Additionally, it can achieve a real-time detection speed of 40 frames per second and 43 frames per second in real scenes. These results demonstrate the effectiveness of the YOLO-CID algorithm in X-ray contraband detection.

Reduction of rejects by combining data from the casting process and automatic X-ray inspection

Automatic inspection of castings with X-rays (radiographic and computed tomography) is widespread for parts that are relevant for safety or have high quality requirements. Examples in the automotive sector are aluminum wheels, chassis parts and new parts within the electric power train. Those parts are automatically inspected, which means that both the image acquisition and the evaluation of the images is done fully automatically. Today, in most industrial implementations, the generated data with a size up to several gigabytes per part is summarized to a simple good or bad decision, according to specification. All other data is dismissed, although this information can be valuable to optimize production processes and thus minimize rejects. This contribution gives an overview about the results of the project Cast Control, which is a collaboration of Fraunhofer Development Center for X-ray Technology EZRT, Fraunhofer Center for Applied Research on Supply Chain Services SCS and industry partner RONAL GROUP. RONAL GROUP is a major aluminum wheel manufacturer, mainly for the OEM market. Within the project we combined serial production data from the low pressure die casting process from a foundry of the RONAL GROUP with the data generated in the automatic X-ray inspection. After collecting a large base of sample data, we were able train a neural network for the prediction of error metrics obtained by X-ray inspection. We apply a combination of layer-wise relevance propagation and dimensionality reduction to find correlations between data of the casting machines (process and sensor) and the characteristics of anomalies detected during X-ray inspection. With this information, it is possible to adjust the casting process in an early stage – even before rejects are produced. This enables the foundry to reduce their rejects rate, which saves costs and energy and results in a better competitivenessin a better competitiveness.

Improving Patient Safety in the X-ray Inspection Process with EfficientNet-Based Medical Assistance System.

Patient safety is a paramount concern in the medical field, and advancements in deep learning and Artificial Intelligence (AI) have opened up new possibilities for improving healthcare practices. While AI has shown promise in assisting doctors with early symptom detection from medical images, there is a critical need to prioritize patient safety by enhancing existing processes. To enhance patient safety, this study focuses on improving the medical operation process during X-ray examinations. In this study, we utilize EfficientNet for classifying the 49 categories of pre-X-ray images. To enhance the accuracy even further, we introduce two novel Neural Network architectures. The classification results are then compared with the doctor's order to ensure consistency and minimize discrepancies. To evaluate the effectiveness of the proposed models, a comprehensive dataset comprising 49 different categories and over 12,000 training and testing sheets was collected from Taichung Veterans General Hospital. The research demonstrates a significant improvement in accuracy, surpassing a 4% enhancement compared to previous studies.

Study on Porosity Defect Detection in Narrow Gap Laser Welding Based on Spectral Diagnosis.

As an advanced connection technology for large thick-walled components, narrow gap laser welding has the advantages of small heat input and high efficiency and quality. However, porosity defects are prone to occur inside the weld due to the complex welding environment. In this study, the influence of the process parameters and pollutants such as water and oil on the porosity defect were explored. The action mechanism of water on the electron temperature and spectral intensity of the laser-induced plasma was analyzed. The results showed that the spectral intensity during narrow gap laser welding was weaker than that of flat plate butt welding. Under the optimal welding process conditions, the electron temperature during narrow gap laser self-fusion welding was calculated as 7413.3 K by the Boltzmann plot method. The electron density was 5.6714 × 1015 cm-3, conforming to the thermodynamic equilibrium state. With six groups of self-fusion welding parameters, only sporadic porosity defects were observed according to the X-ray detection. When there was water on the base metal surface, a large number of dense pores were observed on the weld surface and in the weld through X-ray inspection. Compared with the spectral data obtained under the normal process, the relative light intensity of the spectrometer in the whole band was reduced. The electron temperature decreased to the range of 6900 to 7200 K, while the electron density increased. The spectrum variation during narrow gap laser wire filling welding was basically the same as that of laser self-fusion welding. The porosity defects caused by water and oil pollutants in the laser welding could be effectively identified based on the intensity of the Fe I spectral lines.

Multivariate polynomial regression based selection and analysis of process parameters for X-ray inspection of pipeline containing medium

Pressure pipelines use pressure to transport gas, liquefied gas, and steam in the process of use. If there are defects in the pipeline, it is easy to leak or burst, and even cause secondary accidents. To ensure safety, it is necessary to inspect the pressure pipelines in use regularly. However, it is difficult for the pressure pipeline in use to empty the internal medium. Therefore, it is of great significance to inspect the defects of the industrial pressure pipeline under operating conditions. In this paper, X-ray testing experiments are carried out for weld defects of pipelines with the medium. Using the experimental data, multivariate polynomial regression equations of negative blackness and tube voltage, and other variables are established, which provides an effective tool for the formulation of the optimal process route for testing pressure pipelines containing medium.

X-ray Inspection Model Validation with Physical Dosimetry

X-ray Inspection Model Validation with Physical Dosimetry

Combination of deep learning with representation learning in X-ray prohibited item detection

During X-ray inspection detection, a detector converts the collected X-rays from objects into electrical signals, which are then transmitted to a computer for image processing and analysis. From the aspect of digital image processing, detection tasks mainly focus on data processing and transformation to identify valuable features, which make the algorithms more effective. The consistent requirement for speed and accuracy in X-ray prohibited item detection is still not fully satisfied, especially in pictures obtained under special imaging conditions. For noisy X-ray images with heavy occlusion, a direct and suitable approach of representation learning is the optimal solution. According to our study, we realized that heterogeneous information fusion from different extraction approaches can be applied effectively to overcome this issue. We proposed two innovative algorithms to extract effective features of X-ray objects to significantly improve the efficiency of X-ray prohibited item detection. The brief model we proposed fuses the representations learned from the noisy X-ray images and outperforms the best model (DOAM-O) so far on OPIXray. Furthermore, the attention module we designed to select information on deep learning and representation strengthens the model; considering this, the model utilizes lesser time for both training and inference, which makes it easier to be trained on a lightweight computing device.

Cs3Cu2I5 Nanocrystals with Near-Unity Photoluminescence Quantum Yield for Stable and High-Spatial-Resolution X-ray Imaging

With almost negligible self-absorption and high photoluminescence quantum yield (PLQY), zero-electronic-dimensional Cs3Cu2I5 emerges as an excellent scintillator. Ball-milling is a desired method to prepare Cs3Cu2I5 scintillators at a large scale; however, the defects generated during ball-milling and large particle size hinder its scintillation properties. Theoretical analysis suggests that utilizing a molecule with a small surface energy and a passivation group during ball-milling could be a two-birds-with-one-stone strategy, which can simultaneously improve the PLQY of particles by defect passivation and decrease the particle size by minimizing the surface energy of particles and inhibiting particle rewelding. Accordingly, we successfully prepared Cs3Cu2I5 nanoparticles with a near-unity PLQY and smaller size using a passivator (nonaethylene glycol monododecyl ether, C12E9)-assisted ball-milling method. The interaction of C12E9 with Cs3Cu2I5 has been verified and could enhance the luminescence performance of Cs3Cu2I5. Further, the Cs3Cu2I5 nanoparticle coating delivered a spatial resolution of 17.5 lp mm–1@MTF = 0.2 in X-ray imaging and excellent stability under prolonged X-ray irradiation with total dose of 10.9 Gyair, equal to the dose of 108,000 medical chest X-ray inspections. These results suggest that this in situ passivator-assisted ball-milling method is effective for large-scale production of high-quality Cs3Cu2I5 nanoparticles with excellent scintillation performance.