Magnetic Defects Research Articles

Adaptive Segmentation Algorithm for Subtle Defect Images on the Surface of Magnetic Ring Using 2D-Gabor Filter Bank

In order to realize the unsupervised segmentation of subtle defect images on the surface of small magnetic rings and improve the segmentation accuracy and computational efficiency, here, an adaptive threshold segmentation method is proposed based on the improved multi-scale and multi-directional 2D-Gabor filter bank. Firstly, the improved multi-scale and multi-directional 2D-Gabor filter bank was used to filter and reduce the noise on the defect image, suppress the noise pollution inside the target area and the background area, and enhance the difference between the magnetic ring defect and the background. Secondly, this study analyzed the grayscale statistical characteristics of the processed image; the segmentation threshold was constructed according to the gray statistical law of the image; and the adaptive segmentation of subtle defect images on the surface of small magnetic rings was realized. Finally, a classifier based on a BP neural network is designed to classify the scar images and crack images determined by different threshold segmentation methods. The classification accuracies of the iterative method, the OTSU method, the maximum entropy method, and the adaptive threshold segmentation method are, respectively, 85%, 87.5%, 95%, and 97.5%. The adaptive threshold segmentation method proposed in this paper has the highest classification accuracy. Through verification and comparison, the proposed algorithm can segment defects quickly and accurately and suppress noise interference effectively. It is better than other traditional image threshold segmentation methods, validated by both segmentation accuracy and computational efficiency. At the same time, the real-time performance of our algorithm was performed on the advanced SEED-DVS8168 platform.

Robustness defect detection: Improving the performance of surface defect detection in interference environment

In industrial production environments, the performance of surface defect detection methods for products is affected by disturbances in the equipment, manufacturing processes, and the environment, resulting in poor robustness of the training model. In this paper, an improved defect detection method based on feature extraction fusion and dynamic label assignment is proposed. Three improved network structures, i.e., the adaptive feature recognition convolution (AFRC), path aggregation balanced feature pyramid, and dynamic label assignment, are proposed for the feature extraction, feature fusion, and detection phases of the existing Faster RCNN model, respectively. The AFRC is employed to improve the feature extraction ability of the model for irregular defects, the path balance feature pyramid is utilized to enhance the recognition ability of the model for different levels of feature maps, and the dynamic label assignment is employed to improve the quality of the candidate frames. Under disturbed environments, the results of the proposed improvements outperform the benchmark faster RCNN model in terms of average precision (AP) by 4.1%, 3.7%, and 4.2% on an aluminum surface defect dataset and by 1.8%, 2.2%, and 3.7% AP on a magnetic tile surface defect dataset. The experimental results demonstrate that the proposed scheme effectively improves the robustness of the baseline model in environments with various disturbances.

Optimising digital signal processor‐based defect detection in smart manufacturing with lightweight convolutional neural networks

AbstractIndustrial defect detection is an important part of intelligent manufacturing, and Internet of things (IoT)‐based defect detection is receiving more and more attention. Although deep learning (DL) can help defect detection reduce the cost and improve the accuracy of traditional manual quality inspection, DL requires huge computational resources and is difficult to be simply deployed on IoT devices with limited computational power and memory resources. Digital signal processor (DSP) is an important IoT device with small size, high performance and low energy consumption, which has been widely used in intelligent manufacturing. In order to perform accurate defect detection on DSP, the authors proposed various optimisation strategies and then used a parallel scheme to scale the model to execute on multiple cores. The authors’ method evaluated it on Northeastern University Surface Defect Dataset, Magnetic Tile Defect Dataset, Rail Surface Defect Dataset and Silk Cylinder Defect Dataset, and the experimental results showed that the authors’ method obtains faster speeds without accuracy loss compared to running the same Convolutional Neural Networks model on a mainstream desktop CPU. This means that the authors’ method can realise efficient and accurate defect detection on IoT devices with limited computational power and memory resources, which opens up new possibilities for future development in the field of smart manufacturing.

Probing geometry-induced magnetic defects in cylindrical modulated nanowires with optically detected spin resonance in nitrogen-vacancy center in diamond.

Cylindrical magnetic nanowires (NWs) have gained significant interest as building-blocks of spintronics devices and magnetic sensors thanks to their geometry-tunable magnetic properties and anisotropy. While the synthesis and compositional control of NWs have seen major improvements in recent years, considerable challenges remain for the characterization of local magnetic features at the nanoscale. Here, we demonstrate non-perturbative field distribution mapping and minimally invasive magnetic imaging with scanning nitrogen-vacancy magnetometry. This enables a sensitivity down to 3 μT Hz-1/2 used to localize ultra-scaled magnetic defects with lateral dimensions below 50 nm. The imaging reveals the presence of magnetic inhomogeneities in correspondence of periodical geometrical modulations/anti-notches in axial magnetized nanowires that are largely undetectable with standard metrology. The features induce local fluctuations of the NWs' magnetization orientation that are sensed by SNVM and compared with magnetic force microscopy. Finally, the strong magnetic field confinement in the nanowires is leveraged to study the interaction between the stray magnetic field and the fluorescence generated by two nitrogen-vacancies contained in the probe sensor, thus clarifying the contrast formation mechanisms.

Controlling the spin current around the rectangular cavities in two-dimensional topological insulators.

Controlling spin current in topological insulators (TIs) is a crucial requirement for applications in quantum computing and spintronics. Using the non-equilibrium Keldysh Green's function formalism, we demonstrate that such control can be established around rectangular cavities of two-dimensional TIs by breaking their time reversal symmetry via exchange magnetic fields and magnetic defects. In the presence of magnetic defects with xy symmetry or Ising symmetry, the density of states is localized, and the spin current forms a current loop around the rectangular cavity in TIs interfacing with two ferromagnetic stripes. We also observe that the spin direction of the traveling electrons is inverted under the reversal of bias and gate voltages. The change in the spin-polarized current around the cavities is predicted by varying the strength of Rashba spin-orbit coupling. This result allows for the creation and control of nearly fully spin-polarized currents with various spatial patterns around the cavities in TIs, and the design of tunable spin diodes for highly integrated spintronics.

Controlling spin polarization of gapless states in defected trilayer graphene with a gate voltage

Trilayer graphene exhibits valley-protected gapless states when the stacking order changes from ABC to CBA and a gate voltage is applied to outer layers. Some of these states survive strong distortions of the trilayer. For example, they persist when the outer layers are partially devoid yielding a system of two trilayers of different stacking order connected by a strip of a single graphene layer. Here we investigate how these states respond to another perturbation, i.e., the presence of magnetic defects, which we model as π-vacancies. We show that the gap states hybridize with the defect states and strongly spin-split. More importantly, it is demonstrated that by changing the gate voltage value one can change the spin density of the gap states and the corresponding currents at the Fermi level.

Surface defect segmentation of magnetic tiles based on cross self-attention module

The detection of magnetic tile quality is an essential link before the assembly of permanent magnet motor. In order to meet the high standard of magnetic tile surface defect detection and realize the rapid and automatic segmentation of magnetic tile defects, a magnetic tile surface defect segmentation algorithm based on cross self-attention model (CSAM) is proposed. It adopts high-low level semantic feature fusion method to build the dependency relationship between the deep and shallow features. Multiple auxiliary loss functions are used to constrain the network and reduce the noise in the deep features. In addition, an image enhancement method is also designed to solve the problem of insufficient annotated data. The experimental results show that the network can achieve 79.6% mIoU and 98.5% PA, which can meet the high standard requirements of magnetic tile manufacturing.

A Fast Magnetic Flux Leakage Small Defect Detection Network

To solve the problem of the difficult and slow detection of small defects in magnetic flux leakage (MFL), we propose a fast magnetic flux leakage small defect detection network (FSDDNet). First, we introduce COMSOL data augmentation method that utilizes COMSOL simulation software to obtain high-resolution images of defects, which allows the network to capture complete defect features. Furthermore, this method addresses the issue of the relatively monotonic nature of the MFL defect data set used in the experiment. Deep learning networks usually use stride = 2 or max pooling to downsample the feature map, but this method will make the feature map lose some information, and small targets will lose more fine-grained information. Therefore, we introduce an SPD-Conv method to downsample the feature map, which can effectively avoid the loss of information. Meanwhile, an improved C3 network is introduced in the backbone network of FSDDNet. It greatly decreases the computational effort of the network and improves the detection speed. Finally, we add a small target detection head, which effectively improves the small target accuracy. FSDDNet is improved on the basis of YOLOv5, and after the above improvements, FSDDNet obtains very good results in the problem of MFL small defect detection. Experiments show that the accuracy of this algorithm is 95.2% when IOU = 0.5, and the latency is 7.9 ms.

Topological magnetic line defects in Fe(Te, Se) high-temperature superconductors

The realization of Majorana zero modes in condensed matter have been attracting enormous interests from fundamental science such as topological quantum computation. Recently iron based superconductors were identified as a high-temperature platform for realizing topological superconductivity and Majorana modes. As unconventional superconductors, one of the most important characteristics of them is that they are in the vicinity of magnetic states due to the strong Hund’s coupling in iron atoms. Here we propose that the line defects with missing Te/Se anions in Fe(Se, Te) superconductors provide the realization of intrinsic antiferromagnetic (AFM) chains with Rashba spin-orbit coupling. Against conventional wisdom, Majorana zero modes (MZMs) can be robustly generated at these AFM chain ends. These results can consistently explain the recent experimental observation of zero-energy end states in line defects of monolayer Fe(Te, Se)/SrTiO3 by scanning tunneling microscopy (STM) measurements. Our research not only demonstrates an unprecedented interplay among native line defect, emergent magnetism and topological superconductivity but also explores a high-temperature platform for Majorana fermions.

Magnetic flux leakage defect size estimation method based on physics-informed neural network.

Magnetic flux leakage (MFL) is a magnetic method of non-destructive testing for in-pipe defect detection and sizing. Despite the fact that recent developments in machine learning have revolutionized disciplines like MFL defect size estimation, the most current methods for quantifying pipeline defects are primarily data-driven, which may violate the underlying physical knowledge. This paper proposes a physics-informed neural network-based method for MFL defect size estimation. The training process of neural network is guided by the MFL data and the physical constraints that is mathematically represented by the magnetic dipole model. We use synthetic MFL data produced by a virtual MFL testing of pipeline defects to validate the proposed method through a comparison to purely data-driven neural networks and support vector machines. The findings imply that the physics-informed strategy can both improve predictive accuracy and mitigate physical violations in MFL testing, providing us with a better knowledge of how neural networks perform in defect size estimation. This article is part of the theme issue 'Physics-informed machine learning and its structural integrity applications (Part 2)'.

Current-driven motion of magnetic topological defects in ferromagnetic superconductors

Current-driven motion of magnetic topological defects in ferromagnetic superconductors

Improved Leakage Detection and Recognition Algorithm for Residual Neural Networks Based on Transfer Learning

Due to the lack of other component information in traditional magnetic leakage signal defects and the low accuracy of prediction methods, this paper proposes an improved residual network for magnetic leakage detection defect recognition method that predicts defect size and different detection speeds. A new defect diagnosis method based on ResNet18 on the Convolutional Neural Network (CNN) is proposed in this study. This method transfers the pre-trained ResNet18 network and replaces the activation function in the transferred network structure. It extracts features from transformed two-dimensional images obtained by converting the original experimental signals and signals with added noise, removing the influence of manual features. The results demonstrated that the improved ResNet18 network model, after transfer learning, achieved 100% prediction accuracy for all 10,000 grayscale images generated with defect lengths of 50 mm; width of 2 mm; and depths of 2 mm, 4 mm, 6 mm, and 8 mm. Moreover, the prediction accuracies for the quasi-static, slow, compensated fast, and fast scanning speeds were 99.20%, 98.50%, 93.30%, and 94.00%, respectively, for defect depths of 2 mm, 4 mm, 6 mm, and 8 mm. These accuracies surpass those of other models, demonstrating the significant improvement in prediction accuracy achieved by this method.

Recent progress in the JARVIS infrastructure for next-generation data-driven materials design

The joint automated repository for various integrated simulations (JARVIS) infrastructure at the National Institute of Standards and Technology is a large-scale collection of curated datasets and tools with more than 80 000 materials and millions of properties. JARVIS uses a combination of electronic structure, artificial intelligence, advanced computation, and experimental methods to accelerate materials design. Here, we report some of the new features that were recently included in the infrastructure, such as (1) doubling the number of materials in the database since its first release, (2) including more accurate electronic structure methods such as quantum Monte Carlo, (3) including graph neural network-based materials design, (4) development of unified force-field, (5) development of a universal tight-binding model, (6) addition of computer-vision tools for advanced microscopy applications, (7) development of a natural language processing tool for text-generation and analysis, (8) debuting a large-scale benchmarking endeavor, (9) including quantum computing algorithms for solids, (10) integrating several experimental datasets, and (11) staging several community engagement and outreach events. New classes of materials, properties, and workflows added to the database include superconductors, two-dimensional (2D) magnets, magnetic topological materials, metal-organic frameworks, defects, and interface systems. The rich and reliable datasets, tools, documentation, and tutorials make JARVIS a unique platform for modern materials design. JARVIS ensures the openness of data and tools to enhance reproducibility and transparency and to promote a healthy and collaborative scientific environment.

Inter-layer propagation of spiral waves: Effects of time-varying defect blocks and magnetic flows

The variation of magnetic flux with distance and time is relatively common in nature. This paper investigates the inter-layer propagation of spiral waves in bi-layer Hindmarsh-Rose networks under gradient magnetic field and defect blocks motion, with the following conclusions: (1) Magnetic flux slows down the propagation of waves. (2) For defect blocks moving linearly or away from each other, spiral waves appear in the network, while the magnetic flux suppresses the spiral wave formation. (3) For defect blocks close to or merging with each other, the magnetic flow can only convert spiral waves in the network into “eye-like waves”. These findings imply that the propagation of spiral waves is sensitive to the action site of magnetic flux, which may be important for scar tissue (e.g., cardiac tissue with poor postoperative recovery). Moreover, three statistical parameters defined to detect waveform evolution in excitable media show good performance.

Antiferromagnetic bimeron dynamics controlled by magnetic defects

We investigate dynamics of an isolated bimeron in the antiferromagnetic system, in the presence of magnetic defects. The bimeron is driven by torques due to a spin-polarized current, which is incorporated into the Landau–Lifshitz–Gilbert (LLG) equation. We numerically solve the LLG equation, and find a breathing dynamics of the bimeron as well as its transverse motion owing to the defect potential. We also analyze the potential landscapes due to the interaction between the bimeron and a defect, and give a qualitative picture of the numerically obtained dynamics.

Persistent dynamic magnetic state in artificial honeycomb spin ice

Topological magnetic charges, arising due to the non-vanishing magnetic flux on spin ice vertices, serve as the origin of magnetic monopoles that traverse the underlying lattice effortlessly. Unlike spin ice materials of atomic origin, the dynamic state in artificial honeycomb spin ice is conventionally described in terms of finite size domain wall kinetics that require magnetic field or current application. Contrary to this common understanding, here we show that a thermally tunable artificial permalloy honeycomb lattice exhibits a perpetual dynamic state due to self-propelled magnetic charge defect relaxation in the absence of any external tuning agent. Quantitative investigation of magnetic charge defect dynamics using neutron spin echo spectroscopy reveals sub-ns relaxation times that are comparable to the relaxation of monopoles in bulk spin ices. Most importantly, the kinetic process remains unabated at low temperature where thermal fluctuation is negligible. This suggests that dynamic phenomena in honeycomb spin ice are mediated by quasi-particle type entities, also confirmed by dynamic Monte-Carlo simulations that replicate the kinetic behavior. Our research unveils a macroscopic magnetic particle that shares many known traits of quantum particles, namely magnetic monopole and magnon.

Dissipationless Circulating Currents and Fringe Magnetic Fields Near a Single Spin Embedded in a Two-Dimensional Electron Gas.

Theoretical calculations predict the anisotropic dissipationless circulating current induced by a spin defect in a two-dimensional electron gas. The shape and spatial extent of these dissipationless circulating currents depend dramatically on the relative strengths of spin-orbit fields with differing spatial symmetry, offering the potential to use an electric gate to manipulate nanoscale magnetic fields and couple magnetic defects. The spatial structure of the magnetic field produced by this current is calculated and provides a direct way to measure the spin-orbit fields of the host, as well as the defect spin orientation, e.g., through scanning nanoscale magnetometry.

Magnetic defect characterization by polarized neutron diffraction

Magnetic defect characterization by polarized neutron diffraction

Coercivity modulation of FeCoCrMoTi films by artificial magnetic phase defects engineering based on multilayer structure

FeCoCr-based films have been considered as promising candidates for next-generation magnetic code disk materials used in high-accuracy magnetic rotary encoders. However, realizing high coercivity with sufficient remanent magnetization in FeCoCr-based films with in-plane anisotropy to ensure suitable disturbance-rejection performance and detectability of magnetic pole pairs is a critical issue, which hinders their practical applications. In this study, we demonstrated an in-plane anisotropic [Ta(20 nm)/FeCoCrMoTi(100 nm)]2/Ta(10 nm) multilayer film (M20 sample) with a high coercivity of 945 Oe and a sufficient remanent magnetization of 2890 Oe, which is 92.1% and 63.1% higher than FeCoCrMoTi(200 nm)/Ta(10 nm) film (SL sample), respectively. Detailed microstructure characterization using atomic resolution scanning transmission electron microscopy (STEM) indicates that the high coercivity originates from the domain-wall pinning effect induced by three factors including grain refinement and heterogeneous phases generation and composition fluctuation of modulated phases. X-ray absorption fine structure (XAFS) and X-ray photoelectron spectroscopy (XPS) analyses reveal that the high remanent magnetization is ascribed to the higher relative content of Fe2Ta phases with strong ferromagnetism than that of Co2Ta phases with weak ferromagnetism.

Electrically-Excited Motion of Topological Defects in Multiferroic Materials

Topological magnetic defects in multiferroic materials acquire an electric charge or dipole moment due to the inverse Dzyaloshinskii-Moriya mechanism. This magnetoelectric coupling makes possible to excite large-amplitude collective motion of topological magnetic textures with an oscillating electric field. Here, I discuss electric excitation of a polar optical mode in a vortex-antivortex crystal and the electrically-induced spin precession in a magnetic skyrmion, which gives rise to rotation of skyrmions around each other and a translational motion of skyrmion-antiskyrmion pairs. The electric manipulation of magnetic topological defects in Mott insulators can find applications in magnetoelectric memory and logical devices.