Stress measurement and evaluation of ferromagnetic components using electromagnetic nondestructive testing (ENDT) are affected by the microstructure properties of the material, such as inhomogeneity and anisotropy. This leads to magnetic induction sensitivity in the variety of both the stress and microstructure. To address this challenge, a hybrid magnetic sensing structure for measuring eddy current, alternating electromagnetic (EM) field, Barkhausen noise, and incremental permeability signals is designed and instrumented. Additionally, a stress detection verification test for oriented silicon steel samples is carried out, and a comparison with commercial instrument in terms of sensitivity and repeatability is performed. The results show that the proposed sensor has advantages in detection repeatability, good sensitivity as well as high signal-to-noise (SNR) for MBN features, which brings potential in stress characterization and assessment.

Magnetic domain motion has been widely studied in the fields of spintronics, nanowires, and thin films. However, there is a lack of such studies on industrial steels, especially for domain motion under the action of varying stress. Understanding domain motion under stress is helpful for the improvement of evaluation accuracy and the establishment of theoretical models of passive, nondestructive testing technology. This paper presents the influence of varying tensile stresses on the magnetic domain motion of silicon steel sheets. Magnetic domain rotation and domain wall displacement were characterized using magnetic domain images, and their motion mechanisms under elastic and plastic stresses are presented. The results show that the domain rotation under stress involves reversible and irreversible changes. The effect of material rearrangement on domain rotation and domain wall displacement after plastic deformation is discussed. Based on the motion mechanism, a threshold stress value (TSV) required for the complete disappearance of the supplementary domains in the elastic range is proposed, enabling the classification of the elastic stress ranges in which the reversible and irreversible domain rotations occur. In addition, the effect of microstructure on TSV is also discussed, and the results show that the regions far away from the grain boundary need larger stresses to complete an irreversible domain rotation. Additionally, the domain width and orientation also affect the TSV. These findings regarding the domain motion mechanism and TSV can help to explain the sequence of domain rotation under stress and modify the stress assessment under dynamic loads in electromagnetic nondestructive evaluation, especially in the magnetic memory method.

Deep insights into microstructures and domain wall behaviors in the evaluation of different material statuses under elastic and plastic stress ranges have essential implications for magnetic sensing and nondestructive testing and evaluation (NDT&E). This paper investigates the repeatability and stability of residual magnetic field (RMF) signals using a magneto-optical Kerr effect microscope for the stress characterization of silicon steel sheets beyond their elastic limit. Real-time domain motion is used for RMF characterization, while both the repeatability under plastic ranges after the cyclic stress rounds and stability during relaxation time are studied in detail. The distinction between elastic and plastic materials is discussed in terms of their spatio-temporal properties for further residual stress measurement since both ranges are mixed. During the relaxation time, the RMF of the plastic material shows a two-stage change with apparent recovery, which is contrasted with the one-stage change in the elastic material. Results show that the grain boundary affects the temporal recovery of the RMF. These findings concerning the spatio-temporal properties of different RMFs in plastic and elastic materials can be applied to the design and development of magnetic NDT&E for (residual) stress measurement and material status estimation.

Infrared-visible image fusion quality is critical for nighttime perception in autonomous driving and surveillance but suffers severe degradation under extreme low-light conditions, including irreversible texture loss in visible images, thermal boundary diffusion artifacts, and overexposure under dynamic non-uniform illumination. To address these challenges, a Decomposition–Disentanglement–Dynamic Compensation framework, D3Fusion, is proposed. Firstly, a Retinex-inspired Decomposition Illumination Net (DIN) decomposes inputs into enhanced images and degradative illumination maps for joint low-light recovery. Secondly, an illumination-guided encoder and a multi-scale differential compensation decoder dynamically balance cross-modal features. Finally, a progressive three-stage training paradigm from illumination correction through feature disentanglement to adaptive fusion resolves optimization conflicts. Compared to State-of-the-Art methods, on the LLVIP, TNO, MSRS, and RoadScene datasets, D3Fusion achieves an average improvement of 1.59% in standard deviation (SD), 6.9% in spatial frequency (SF), 2.59% in edge intensity (EI), and 1.99% in visual information fidelity (VIF), demonstrating superior performance in extreme low-light scenarios. The framework effectively suppresses thermal diffusion artifacts while mitigating exposure imbalance, adaptively brightening scenes while preserving texture details in shadowed regions. This significantly improves fusion quality for nighttime images by enhancing salient information, establishing a robust solution for multimodal perception under illumination-critical conditions.

The Magnetic Barkhausen Noise (MBN) measurement method is widely used to evaluate stress, microstructure, plasticity and other material states of ferromagnetic materials. However, stress assessment is generally affected by material inhomogeneity or anisotropy caused by microstructure properties such as grain size, leading to low signal-to-noise ratio and poor repeatability when characterizing stress using MBN signals. In this paper, a high spatial resolution MBN sensor based on a recording magnetic head is designed. A statistical method based on grain statistics is proposed to determine the minimum number of grains required for stable MBN signals. The method is experimentally validated on fine-grained Q235 carbon steel and 27QG110 oriented silicon steel with large grains. Different feature extraction methods, including RMS, amplitude histogram and pulse height distribution of MBN signals are applied and compared for robustness of material inhomogeneity. The results demonstrate that the average effect achieved by a sufficiently large measurement area reduces the impact of material heterogeneity on MBN signals, which is particularly relevant for large-grain materials. Additionally, the normalized threshold pulse count (ThPC) feature of amplitude histogram, defined as the number of upward threshold crossings, enhances robustness to material inhomogeneity and provides good repeatability in stress characterization across materials with different microstructurual scales. Therefore, this study provides practical guidance for quantitative stress evaluation using MBN measurement.