Inspection Of Images Research Articles

APbI3-A2AgBiI6 Double-Layer Perovskite Film for a Self-Powered and High-Stability X-ray Detector.

The development of lead halide perovskite X-ray detectors has promising applications in medical imaging and security inspection but is hindered by poor long-term stability and drift of the dark current and photocurrent. Herein, we design a (Cs0.05MA0.65FA0.3)PbI3-(Cs0.1MA1.3FA0.6)AgBiI6 double-layer perovskite film to assemble a self-powered flat-panel X-ray detector. The demonstrated X-ray detector achieves an outstanding self-powered sensitivity of 80 μC Gyair-1 cm-2 under a 0 V bias. More importantly, owing to the inhibition of the phase transition process and ion migration of (Cs0.05MA0.65FA0.3)PbI3 by the (Cs0.1MA1.3FA0.6)AgBiI6 layer, the device exhibits excellent continuous operating stability with a retention rate of 99% dark current and photocurrent over X-ray pulses of up to 4000 s and excellent long-term stability without a loss of the original response current after 150 days in an air environment. The strategy of double-layer perovskites improves the stability and sensitivity of devices, which paves a path for the industrial application of lead halide perovskite X-ray detectors.

AI-Dentify: deep learning for proximal caries detection on bitewing x-ray - HUNT4 Oral Health Study.

Dental caries diagnosis requires the manual inspection of diagnostic bitewing images of the patient, followed by a visual inspection and probing of the identified dental pieces with potential lesions. Yet the use of artificial intelligence, and in particular deep-learning, has the potential to aid in the diagnosis by providing a quick and informative analysis of the bitewing images. A dataset of 13,887 bitewings from the HUNT4 Oral Health Study were annotated individually by six different experts, and used to train three different object detection deep-learning architectures: RetinaNet (ResNet50), YOLOv5 (M size), and EfficientDet (D0 and D1 sizes). A consensus dataset of 197 images, annotated jointly by the same six dental clinicians, was used for evaluation. A five-fold cross validation scheme was used to evaluate the performance of the AI models. The trained models show an increase in average precision and F1-score, and decrease of false negative rate, with respect to the dental clinicians. When compared against the dental clinicians, the YOLOv5 model shows the largest improvement, reporting 0.647 mean average precision, 0.548 mean F1-score, and 0.149 mean false negative rate. Whereas the best annotators on each of these metrics reported 0.299, 0.495, and 0.164 respectively. Deep-learning models have shown the potential to assist dental professionals in the diagnosis of caries. Yet, the task remains challenging due to the artifacts natural to the bitewing images.

Image-based novel fault detection with deep learning classifiers using hierarchical labels

One important characteristic of modern fault classification systems is the ability to flag the system when faced with previously unseen fault types. This work considers the unknown fault detection capabilities of deep neural network-based fault classifiers. Specifically, we propose a methodology on how, when available, labels regarding the fault taxonomy can be used to increase unknown fault detection performance without sacrificing model performance. To achieve this, we propose to utilize soft label techniques to improve the state-of-the-art deep novel fault detection techniques during the training process and novel hierarchically consistent detection statistics for online novel fault detection. Finally, we demonstrated increased detection performance on novel fault detection in inspection images from the hot steel rolling process, with results well replicated across multiple scenarios and baseline detection methods.

Sustained release gel based on CT image inspection for treatment of diabetes fundus macular lesions

Currently, slow-release gel therapy is considered to be an effective treatment for fundus macular disease, but the lack of effective evaluation methods limits its clinical application. Therefore, the purpose of this study was to investigate the application and clinical effect of slow-release gel based on CT image examination in the treatment of diabetic fundus macular disease. CT images of fundus macular lesions were collected in a group of diabetic patients. Then the professional image processing software is used to process and analyze the image and extract the key parameters. A slow-release gel was designed and prepared, and applied to the treatment of diabetic fundus macular disease. CT images before and after treatment were compared and analyzed, and the effect of slow-release gel was evaluated. In a certain period of time after treatment, the lesion size and lesion degree of diabetic fundus macular disease were significantly improved by using slow-release gel therapy with CT image examination. No significant adverse reactions or complications were observed during the treatment. This indicates that the slow-release gel based on CT image examination is a safe, effective and feasible treatment method for diabetic fundus macular disease. This method can help improve the vision and quality of life of patients, and provide a new idea and plan for clinical treatment.

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.

A simple and precise alignment calibration method for cone-beam computed tomography with the verifications

Cone-beam computed tomography (CBCT) is widely used in dental imaging, small animal imaging, radiotherapy, and non-destructive industrial inspection. The quality of CBCT images depends on the precise knowledge of the CBCT system’s alignment. We introduce a distinct procedure, ‘precision alignment loop (PAL)’, to calibrate any CBCT system with a circular trajectory. We describe the calibration procedure by using a line-beads phantom, and how PAL determines the misalignments from a CBCT system. PAL also yields the uncertainties in the simulated calibration to give an estimate of the errors in the misalignments. From the analytical simulations, PAL can precisely obtain the source-to-rotation axis distance, and the geometric center G, ‘the point in z-axis meets the detector’, where the z-axis is coincident with the line from the x-ray source that intersects the axis of the rotation orthogonally. The uncertainties of three misalignment angles of the detector are within ±0.05°, which is close to ±0.04° for the results of Yang et al (2017 Med. Phys. 44 1692), but our method is easy and simple to implement. Our distinct procedure, on the other hand, yields the calibration of a micro-CT system and an example of reconstructed images, showing our calibration method for the CBCT system to be simple, precise, and accurate.

Application of gel based on medical image inspection in total closure resection of oral cancer patients

Oral cancer is a common malignant tumor, and total closed resection is a common treatment. However, it has always been a challenge to determine the exact extent of excision during surgery. The application of medical image examination in surgery can provide important reference information, but the current methods still have some limitations. This study explored the application of gels based on medical image examination in the total closed resection of oral cancer patients to improve the accuracy of resection range and surgical treatment effect. The study collected medical image data of patients with oral cancer for image enhancement and determination of resection boundaries. By comparing the results of the experimental group and the control group, the application effect of gel in operation was evaluated. Through the application of medical image inspection technology, the determination of surgical resection boundary is more accurate, and the positive incisal margin of patients is effectively avoided. Gel technology improves the success rate and efficacy of surgery, and this method helps to improve the accuracy of surgery and the certainty of the scope of resection, which is of great significance for improving the surgical treatment effect and the survival rate of patients.

Simulation of defect detection for the buried petroleum pipe by the X-ray backscatter imaging

Ensuring the integrity, safety, and compliance of oil pipelines is crucial for operators. Two significant factors affecting pipeline integrity are metal loss due to corrosion or air gouging and the presence of cracks. This paper introduces an X-ray backscatter nondestructive testing model on a smart pig(an intelligent robot) for internal inspection of petroleum pipelines. The paper investigates the X-ray source and detection system, essential components of the backscatter imaging technology, to provide technical guidance for model design. The imaging system’s layout is modeled using GEANT4, and its performance in detecting defects of various formats on pipe walls is studied. The research demonstrates the feasibility of using X-ray backscatter imaging for pipeline inspection and provides insights into different defect detection. The findings contribute to designing and optimizing the robot’s scanning head for effective pipeline inspection.

Laser Powder Bed Fusion Printing of CoCrFeMnNi High Entropy Alloy: Processing, Microstructure, and Mechanical Properties

Equiatomic CoCrFeMnNi high entropy alloy (HEA) powder was processed by laser powder bed fusion (LPBF) additive manufacturing (AM). The properties of the spherical pre-alloyed CoCrFeMnNi powder were characterized and its processability using LPBF AM was systematically investigated through the volumetric energy density (VED) based on the surface roughness, defects (micro-cracks and porosity) and densification. After optimization, LPBF processing at a VED of 104 J/mm3 achieved highly dense and crack-free vertical and horizontal test specimens with a porosity fraction lower than 0.01% and micro-pores having a mean size of, respectively, 25.9 μm and 13.4 μm, as determined from X-ray micro-computed tomography (μCT) inspection. Scanning electron microscope (SEM) analysis of the as-built (AB) CoCrFeMnNi processed at a VED of 104 J/mm3 showed a heterogeneous solidification microstructure, consisting of columnar grains with a cellular subgrain structure, and electron backscattered diffraction (EBSD) revealed a crystallographic texture mainly along the < 100 > direction. Post treatment with hot isostatic pressing (HIP) was effective in closing the remnant micro-pores in the bulk volume of the AB CoCrFeMnNi. Also, the cellular sub-grain structure in the AB CoCrFeMnNi completely disappeared after HIP and the resulting microstructure consisted of recrystallized equiaxed grains with annealing twins. The room temperature tensile response was anisotropic for AB CoCrFeMnNi with horizontally built specimens exhibiting higher strength and fracture strains (global and local) compared to vertically built ones; HIP reduced the anisotropy in the tensile properties and led to similar tensile strength with elongation values that were ~ 50% higher than in the AB condition. The HIPed CoCrFeMnNi also displayed higher Charpy impact toughness and absorbed energy at both room and liquid nitrogen temperatures compared to the AB material. Examination of the fracture surfaces after tensile and Charpy impact testing revealed ductile features with characteristic dimpled appearance and pointed to the important role of the remnant micro-pores on failure in the AB CoCrFeMnNi. Tribological assessments pointed to the superior low-stress abrasion resistance of AB and HIPed CoCrFeMnNi compared to 316L stainless steel (SS), which was included in this study to reinforce the analysis. SEM observations revealed that scratching and micro-fracture are the dominant wear mechanisms for the CoCrFeMnNi HEA, whereas ploughing and cutting parallel to the abrasive flow direction are the dominant mechanisms for 316L SS. To the authors’ knowledge, this study is the first to evaluate and report the low-stress abrasion resistance of any high entropy alloy. To understand the corrosion behavior, polarization curves of AB and HIPed CoCrFeMnNi were measured in 3.5 wt% NaCl and 1N H2SO4 solutions, and the results were compared to those of 316L SS. The findings indicate that AB and HIPed CoCrFeMnNi outperform 316L SS in a chloride-containing environment, but not in an acid-containing environment. Additionally, observations of hydrogen permeability revealed that AB CoCrFeMnNi permeates a lower volume of hydrogen atoms (by ~ 5 times) compared to 316L SS, despite its higher (by nearly 3 times) diffusion coefficient. Electrochemical hydrogen permeation data showed that the concentration of atomic hydrogen in the sub-surface of AB and HIPed CoCrFeMnNi was, respectively, about 32 and 26 times lower than in 316L SS. This study provides important material–structure–property data and indicates a promising outlook for LPBF of the CoCrFeMnNi HEA with high-performance.

Automated Bridge Inspection Image Retrieval Based on Deep Similarity Learning and GPS

Automated Bridge Inspection Image Retrieval Based on Deep Similarity Learning and GPS

Adaptive defogging method for transmission line inspection images based on multilayer perceptual fusion

Existing image defogging methods generally have problems such as incomplete defogging and color distortion. To address this problem, this paper proposes an adaptive defogging method for transmission line images based on multilayer perceptual fusion, which uses dynamic convolution, dense residuals, and attention mechanism to design an adaptive feature enhancement network containing six Dy-namic Residual Components (DRC) and two Dy-namic Skip-Connected Feature Fusion Component (DSCFF) composed of adaptive feature enhancement network, which prevents the problem of features being forgotten in the early stage of the network, and enhances the expressive ability of the model. For the decoding network, the de-fogging effect of the model is further strengthened by introducing a decoder module based on the SOS enhancement model, and finally, by comparing the experiments with the current de-fogging methods with more advanced performance, the results show that the method has good de-fogging effect and can retain the image details better with high color retention.

Comparison of clinical geneticist and computer visual attention in assessing genetic conditions.

Artificial intelligence (AI) for facial diagnostics is increasingly used in the genetics clinic to evaluate patients with potential genetic conditions. Current approaches focus on one type of AI called Deep Learning (DL). While DL- based facial diagnostic platforms have a high accuracy rate for many conditions, less is understood about how this technology assesses and classifies (categorizes) images, and how this compares to humans. To compare human and computer attention, we performed eye-tracking analyses of geneticist clinicians (n = 22) and non-clinicians (n = 22) who viewed images of people with 10 different genetic conditions, as well as images of unaffected individuals. We calculated the Intersection-over-Union (IoU) and Kullback-Leibler divergence (KL) to compare the visual attentions of the two participant groups, and then the clinician group against the saliency maps of our deep learning classifier. We found that human visual attention differs greatly from DL model's saliency results. Averaging over all the test images, IoU and KL metric for the successful (accurate) clinician visual attentions versus the saliency maps were 0.15 and 11.15, respectively. Individuals also tend to have a specific pattern of image inspection, and clinicians demonstrate different visual attention patterns than non-clinicians (IoU and KL of clinicians versus non-clinicians were 0.47 and 2.73, respectively). This study shows that humans (at different levels of expertise) and a computer vision model examine images differently. Understanding these differences can improve the design and use of AI tools, and lead to more meaningful interactions between clinicians and AI technologies.

Design of a new detachable pedicle screw based on medical optical imaging inspection to improve osteoporosis and enhance vertebral body revision effect

Osteoporosis is a common bone disease that often leads to difficulty in vertebrae revision. Traditional pedicle screws are often complicated to operate and have poor visibility during implantation. A new detachable pedicle screw is needed to improve the revision effect. The aim of this study was to design a new detachable pedicle screw based on medical optical imaging to improve the outcome of vertebral revision in osteoporosis, and to improve operational feasibility and visibility. In this study, the parameters related to the degree of osteoporosis were obtained by optical imaging detection of the osteoporotic vertebral body. Then a new detachable pedicle screw was designed according to the test results to improve the effect of vertebral body revision. By preparing and optimizing the material and structure of the screw, it is ensured that it has sufficient mechanical strength and stability. Finally, the visibility and operability of the improved screw during implantation were verified by medical optical imaging. Compared with traditional screws, the new detachable pedicle screw can improve the vertebral body revision in the case of osteoporosis. The optical imaging test results show that the new screw has good visibility and maneuverability, providing more accurate guidance and positioning for the vertebral body revision operation.

Efficient leukocytes detection and classification in microscopic blood images using convolutional neural network coupled with a dual attention network

Leukocytes, also called White Blood Cells (WBCs) or leucocytes, are the cells that play a pivotal role in human health and are vital indicators of diseases such as malaria, leukemia, AIDS, and other viral infections. WBCs detection and classification in blood smears offers insights to pathologists, aiding diagnosis across medical conditions. Traditional techniques, including manual counting, detection, classification, and visual inspection of microscopic images by medical professionals, pose challenges due to their labor-intensive nature. However, traditional methods are time consuming and sometimes susceptible to errors. Here, we propose a high-performance convolutional neural network (CNN) coupled with a dual-attention network that efficiently detects and classifies WBCs in microscopic thick smear images. The main aim of this study was to enhance clinical hematology systems and expedite medical diagnostic processes. In the proposed technique, we utilized a deep convolutional generative adversarial network (DCGAN) to overcome the limitations imposed by limited training data and employed a dual attention mechanism to improve accuracy, efficiency, and generalization. The proposed technique achieved overall accuracy rates of 99.83%, 99.35%, and 99.60% for the peripheral blood cell (PBC), leukocyte images for segmentation and classification (LISC), and Raabin-WBC benchmark datasets, respectively. Our proposed approach outperforms state-of-the-art methods in terms of accuracy, highlighting the effectiveness of the strategies employed and their potential to enhance diagnostic capabilities and advance real-world healthcare practices and diagnostic systems.

Lightweight Detection Method for X-ray Security Inspection with Occlusion

Identifying the classes and locations of prohibited items is the target of security inspection. However, X-ray security inspection images with insufficient feature extraction, imbalance between easy and hard samples, and occlusion lead to poor detection accuracy. To address the above problems, an object-detection method based on YOLOv8 is proposed. Firstly, an ASFF (adaptive spatial feature fusion) and a weighted feature concatenation algorithm are introduced to fully extract the scale features from input images. In this way, the model can learn further details in training. Secondly, CoordAtt (coordinate attention module), which belongs to the hybrid attention mechanism, is embedded to enhance the learning of features of interest. Then, the slide loss function is introduced to balance the simple samples and the difficult samples. Finally, Soft-NMS (non-maximum suppression) is introduced to resist the conditions containing occlusion. The experimental result shows that mAP (mean average precision) achieves 90.2%, 90.5%, 79.1%, and 91.4% on the Easy, Hard, and Hidden sets of the PIDray and SIXray public test set, respectively. Contrasted with original model, the mAP of our proposed YOLOv8n model increased by 2.7%, 3.1%, 9.3%, and 2.4%, respectively. Furthermore, the parameter count of the modified YOLOv8n model is roughly only 3 million.

Enlightening the Universe behind the Milky Way bulge

Context. The location of the Solar System complicates the detection extragalactic sources beyond the Milky Way plane. The optical observations are hampered in the so-called zone of avoidance (ZOA), where stellar crowding and Galactic absorption are severe. Observations at longer wavelengths are needed in order to discover new background galaxies and complete the census in the ZOA. Aims. The goal of this work is to identify galaxies behind the Milky Way bulge using near-infrared (NIR) data from the VISTA Variables in Vía Láctea (VVV) survey. Methods. To this end, we made use of different VISTA Science Archive (VSA) tools in order to extract relevant information from more than 32 billion catalogued sources in the VVV bulge region. We find that initial photometric restriction on sources from the VSA vvvSource table combined with restrictions on star–galaxy separation parameters obtained from Source Extractor is a successful strategy for achieving acceptable levels of contamination (60%) and high completeness (75%) in the construction of a galaxy target sample. To remove contaminating Galactic sources from the initial target sample, our methodology also incorporates a visual inspection of false-colour RGB images, a crucial quality control carried out following a specifically defined procedure. Results. Under this methodology, we find 14 480 galaxy candidates in the VVV bulge region, making this sample the largest catalogue to date in the ZOA. Moreover, these new sources provide a fresh picture of the Universe hidden behind the curtain of stars, dust, and gas in the unexplored Milky Way bulge region. Conclusions. The results from this work further demonstrate the potential of the VVV/VVVX survey to find and study a large number of galaxies and extragalactic structures obscured by the Milky Way, expanding our knowledge of the Universe in this challenging and impressive region of the sky.

Measurements of visual double stars with PISCO2 at the Nice 76‐cm refractor in 2016–2017

AbstractWe present relative astrometric and photometric measurements of visual double stars made in 2016–2017, with PISCO2 installed at the 76‐cm refractor of Côte d'Azur Observatory in Nice (France). Our observing list contains orbital couples as well as double stars whose motion is still uncertain. Three different techniques were used for obtaining measurements: Lucky Imaging, Speckle Interferometry and the direct vector autocorrelation (DVA) method. From our observations of 1,510 multiple stars, we obtained 2,918 new measurements with angular separations in the range 0″.1–20″ and an average accuracy of . The mean error on the position angles is 1°.1. Most of the position angles were determined without the usual 180° ambiguity with the application of the DVA technique and/or by inspection of the Lucky images or the long integration files. We managed to routinely monitor faint systems () with large magnitude differences (up to ). We have thus been able to measure six systems containing red dwarf stars that had been poorly monitored since their discovery, from which we estimated the stellar masses thanks to Gaia measurements. We also measured the magnitude difference of the two components of 593 double stars with a mean error of 0.1 mag. Thanks to good seeing images and with the use of high‐contrast numerical filters, we have also been able to obtain 46 measurements with an angular separation smaller than the diffraction limit of our instrumentation. Except for a few objects that have been discussed, our measurements are in good agreement with the ephemerides computed with published orbital elements, even for the double stars whose separation is smaller than the diffraction limit. We also report the measurements of 205 new double stars that we found in the files obtained during the observations.

Attention-Enhanced Co-Interactive Fusion Network (AECIF-Net) for automated structural condition assessment in visual inspection

Efficiently monitoring the condition of civil infrastructure requires automating the structural condition assessment in visual inspection. This paper proposes an Attention-Enhanced Co-Interactive Fusion Network (AECIF-Net) for automatic structural condition assessment in visual bridge inspection. AECIF-Net can simultaneously parse structural elements and segment surface defects on the elements in inspection images. It integrates two task-specific relearning subnets to extract task-specific features from an overall feature embedding. A co-interactive feature fusion module further captures the spatial correlation and facilitates information sharing between tasks. Experimental results demonstrate that the proposed AECIF-Net outperforms the current state-of-the-art approaches, achieving promising performance with 92.11% mIoU for element segmentation and 87.16% mIoU for corrosion segmentation on the test set of the new benchmark dataset Steel Bridge Condition Inspection Visual (SBCIV). An ablation study verifies the merits of the designs for AECIF-Net, and a case study demonstrates its capability to automate structural condition assessment.

Research on image classification of power inspection using less sample learning technique

Abstract This study proposes a few-sample self-supervised power inspection image classification method based on Vision Transformer and masked autoencoders (MAE). Initially, an MAE is employed to pretrain an extensive collection of unlabeled power inspection images, leveraging the masking technique to encourage the model to acquire an understanding of the overarching features within these images. Then, the pretrained encoder part is combined with the visual transformer to further extract image features. For the subsequent classification task, the encoder’s pretrained weights are frozen and subsequently incorporated into a novel classification network. By fine-tuning this network with a limited amount of labeled data, experimental findings indicate a substantial enhancement in the classification accuracy of power inspection images, particularly in scenarios with limited sample sizes. By introducing visual transformers and MAE, this article provides an efficient and reliable solution for intelligent classification of power inspection images, which has important practical application value and broad market prospects.

NeXtResUNet: A neural network for industrial CT image denoising

Image denoising is a critical issue in industrial computed tomography (CT) inspection. Most existing noise reduction algorithms are based on synthetic data, resulting in the loss of fine details, local region distortions, and other problems that arise during application in practical scenarios. As such, a fusion network, combining ConvNeXt, ResNet, and UNet is proposed in this study to address these issues. This algorithm was applied to a self-constructed industrial CT image denoising dataset, achieving a peak signal-to-noise ratio (PSNR) of 47.20 dB, which is 6% (44.48 dB) higher than that of the benchmark DnCnn. This also represents a 0.5% improvement over the current state-of-the-art (SOTA) SCUNet (46.94 dB). More importantly, the proposed network is superior to other existing models in terms of the overall visual effects and degree of image detail preservation. The network structure is also concise yet simple to expand. As such, it can be applied not only to image denoising, but also to CT image segmentation, super resolution, and related tasks.