Search

Single-resolution homography estimation of infrared and visible images is a significant and challenging research area within the field of computing, which has attracted a great deal of attention. However, due to the large modal differences between infrared and visible images, existing methods are difficult to stably and accurately extract and match features between the two image types at a single resolution, which results in poor performance on the homography estimation task. To address this issue, this paper proposes an end-to-end unsupervised single-resolution infrared and visible image homography estimation method based on graph neural network (GNN), homoViG. Firstly, the method employs a triple attention shallow feature extractor to capture cross-dimensional feature dependencies and enhance feature representation effectively. Secondly, Vision GNN (ViG) is utilized as the backbone network to transform the feature point matching problem into a graph node matching problem. Finally, this paper proposes a new homography estimator, residual fusion vision graph neural network (RFViG), to reduce the feature redundancy caused by the frequent residual operations of ViG. Meanwhile, RFViG replaces the residual connections with an attention feature fusion module, highlighting the important features in the low-level feature graph. Furthermore, this model introduces detail feature loss and feature identity loss in the optimization phase, facilitating network optimization. Through extensive experimentation, we demonstrate the efficacy of all proposed components. The experimental results demonstrate that homoViG outperforms existing methods on synthetic benchmark datasets in both qualitative and quantitative comparisons. [ABSTRACT FROM AUTHOR]

Rotating compressor blades experience complex alternating loads during service, altering their stress–strain distributions and peak stress positions over time. Accurate measurement of these strains is crucial for identifying the areas of stress concentration. This paper presents a structural health monitoring system using fiber Bragg grating (FBG) sensors to record dynamic strains on laboratory-scale rotating blades, and a tailored full-field strain reconstruction methodology, which successfully identifies the magnitude of the strains and the areas of stress concentration of the blades at different rotational speeds. First, dynamic strain at selected blade points was monitored using FBG sensors, with raw signal data enhanced by the empirical wavelet transform method to reduce noise and clarify signals. An analytical framework was developed to relate blade rotational velocity to signal period, enabling precise speed calculation and accurate strain analysis. The improved-Kriging interpolation technique was then used to reconstruct comprehensive strain profiles. A comparative analysis showed an average strain relative error of 7.4% between predicted and actual values, demonstrating the methodology's robustness and precision. [ABSTRACT FROM AUTHOR]

A rigorous solution for the kinematic response of the disconnected piles under vertically incident P-waves is presented. A new soil–pile column model is proposed with two fictitious soil columns as the extensions of the shaft along the pile top and tip. The discontinuity of the pile at the two ends is modeled using the Heaviside function to simulate the continuity of the soil–pile column. The total displacement of the soil is divided into the free-field and scattered components. The displacement of the soil–pile system is calculated in terms of the method of separation of variables and the soil–pile interaction. A parametric study indicates that for the disconnected pile, the displacement at the pile top increases significantly with the increase of the cushion thickness, and its effect on the scattered field is weak. The axial force of the disconnected pile decreases gradually with the increase of the cushion thickness. The displacement and axial force of the disconnected pile are significantly influenced by the slenderness ratio of the pile. [ABSTRACT FROM AUTHOR]

Multi-layer complex structures are widely used in large-scale engineering structures because of their diverse combinations of properties and excellent overall performance. However, multi-layer complex structures are prone to interlaminar debonding damage during use. Therefore, it is necessary to monitor debonding damage in engineering applications to determine structural integrity. In this paper, a damage information extraction method with ladder feature mining for Lamb waves is proposed. The method is able to optimize and screen effective damage information through ladder-type damage extraction. It is suitable for evaluating the severity of debonding damage in aluminum-foamed silicone rubber, a novel multi-layer complex structure. The proposed method contains ladder feature mining stages of damage information selection and damage feature fusion, realizing a multi-level damage information extraction process from coarse to fine. The results show that the accuracy of damage severity assessment by the damage information extraction method with ladder feature mining is improved by more than 5% compared to other methods. The effectiveness and accuracy of the method in assessing the damage severity of multi-layer complex structures are demonstrated, providing a new perspective and solution for damage monitoring of multi-layer complex structures. [ABSTRACT FROM AUTHOR]

Driven by the rapid advancement in manufacturing technologies, engineering structures with complex geometries are increasingly applied in various industries, posing challenges to the applicability and adaptability of existing structural health monitoring methods based on guided ultrasonic waves. To fulfill the characterization of defects in complex structures, a novel approach featuring a conjunction of zero-group-velocity (ZGV) Lamb waves and polarized poly(vinylidenefluoride-co-trifluoroethylene) (PVDF-TrFE) transducers is proposed. In this approach, the PVDF-TrFE solvent is deposited and in situ polarized on the structure surface to form thin and flexible coatings, with which the ZGV waves can be excited efficiently and measured reliably. On this basis, the defect can be characterized by investigating the defect-induced alteration in ZGV wave features. In experimental validations, disbond defects in multilayer bonded structures are evaluated using the ZGV waves measured with fabricated PVDF-TrFE transducers. For the first time, the ZGV waves are measured in a contact and in situ manner. Compared with conventional noncontact measurement of ZGV waves, the proposed approach features a remarkably improved reliability, convenience for narrowband excitation, immunity to measurement uncertainty and capability of in situ monitoring. The proposed approach can advance the ZGV wave-based methods toward the in situ health monitoring and enable the defect evaluation in emerging complex structures. [ABSTRACT FROM AUTHOR]