Frontiers in Optics + Laser Science 2022 (FIO, LS), Journal Year: 2024, Volume and Issue: unknown, P. JD4A.76 - JD4A.76
Published: Jan. 1, 2024
This paper presents a motion detection framework that is based on optical fiber reflections integrated with machine learning. The presented achieves accurate and cost-effective sensing by reliably identifying reflectance spectra enabling versatile applications
Language: Английский
Smart Cities, Journal Year: 2024, Volume and Issue: 7(4), P. 1888 - 1906
Published: July 22, 2024
The timely and accurate recognition of multi-type structural surface damage (e.g., cracks, spalling, corrosion, etc.) is vital for ensuring the safety service performance civil infrastructure accomplishing intelligent maintenance smart cities. Deep learning computer vision have made profound impacts on automatic using nondestructive test techniques, especially non-contact vision-based algorithms. However, accuracy highly depends training data volume completeness in conventional supervised pipeline, which significantly limits model under actual application scenarios; stability categories are still challenging. To address above issues, this study proposes a dual-stage optimization-based few-shot segmentation method only few images with information recognition. A optimization paradigm established encompassing an internal network based meta-task external meta-learning machine meta-batch. underlying image features pertinent to various types learned as prior knowledge expedite adaptability across diverse via samples. Furthermore, mathematical framework formulated intuitively express perception mechanism. Comparative experiments conducted verify effectiveness necessity proposed small-scale set. results show that could achieve higher accuracies than directly original network. In addition, generalization ability unseen category also validated. provides effective solution image-based high robustness bridges buildings, assists unmanned inspection drones robotics
Language: Английский
Citations
3
Advanced Optical Materials, Journal Year: 2025, Volume and Issue: unknown
Published: March 17, 2025
Abstract Mechanoluminescence (ML) sensor‐derived distributing measurement urgently needs to overcome the trade‐off between luminous intensity and afterglow duration. In this article, a strontium aluminate (SrAl 2 O 4 ) based ML sensing candidate is controllably synthesized by solid‐solution reaction of powdered precursors SrCO 3 Al under hybrid doping rare earth cations (Eu 2+ , Dy 3+ Nd at 1400 °C. Compared with traditional SrAl : Eu (SAOEDN) has demonstrated highly enhanced (over two orders increase), robust behavior (300 cycles), tunable performance (50 325 s) after synergistic regulation trap depth (from 0.2 0.88 eV). After in situ compounding SAOEDN epoxy resin matrix, flexible film created for distributed detection engineering strain distribution. The effect triggered mechanical deformation presented an approximately linear dependence higher spatiotemporal resolution. As result, field reconstructed via deep learning‐derived image‐to‐image mapping process eliminating disturbance afterglow. Moreover, capable accurately detecting capturing fracture propagation materials. It suggested promising potential non‐contact stress fields applications.
Language: Английский
Citations
0
Intelligent Buildings International, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 17
Published: May 7, 2025
Language: Английский
Citations
0
Eng—Advances in Engineering, Journal Year: 2025, Volume and Issue: 6(5), P. 97 - 97
Published: May 13, 2025
The gradual deterioration of underground water infrastructure requires constant condition monitoring to prevent catastrophic failures, reduce leaks, and avoid costly unexpected repairs. However, given the large scale tight budgets utilities, it is essential implement strategies for optimal selection deployment technologies. This article introduces a unified framework methods optimally selecting technologies while locating their at most vulnerable pipe segments. approach underpinned by an R-E-R-A-V (Redundant, Established, Reliable, Accurate, Viable) principle asset management concepts. proposed method supported thorough review assessment technologies, as well common sensor placement approaches. selects technology using combination readiness levels SFAHP (Spherical Fuzzy Analytic Hierarchy Process). Optimal achieved with k-Nearest Neighbors (kNN) model tuned minimal topological physical pipeline system features. Feature engineering performed OPTICS (Ordering Points Identify Clustering Structure) evaluating segment vulnerability failure-prone areas. Both are integrated through algorithm. demonstrated modified benchmark network (Net3). results reveal accurate robust performance harmonic mean precision recall approximately 65%. effectively identifies segments requiring failures over period 11 years. benefits areas future exploratory research explained encourage improvements additional applications.
Language: Английский
Citations
0
Deleted Journal, Journal Year: 2025, Volume and Issue: unknown
Published: May 16, 2025
Language: Английский
Citations
0
The international archives of the photogrammetry, remote sensing and spatial information sciences/International archives of the photogrammetry, remote sensing and spatial information sciences, Journal Year: 2024, Volume and Issue: XLVIII-4/W10-2024, P. 93 - 100
Published: May 31, 2024
Abstract. Quantum sensing technologies provide future cities with unimaginable techniques for solving their complex problems. sensors, through the utilization of quantum effects such as superposition, entanglement, and tunneling, can an unmatched level sensitivity, precision, durability against traditional technologies. This study explores potential applications in four critical urban infrastructure domains: water, energy, transport, construction. Throughout this study, we determine most promising each domain. Besides, discuss technical progress these sensors advantages they have comparison classical devices, well organizational issues face when implementing sensors. Our results indicate that will be a enabler smart cities, generating advanced monitoring, control, decision-making capabilities across various sectors. Nevertheless, taking advantage demand close partnership industry, academia, policymakers to guide complicated adoption process.
Language: Английский
Citations
1
Electronics, Journal Year: 2024, Volume and Issue: 13(18), P. 3715 - 3715
Published: Sept. 19, 2024
In this study, we compared the performance of quantum temperature sensors (QTSs) with conventional (CSs), highlighting differences in measurement accuracy and stability. Quantum (QSs), known for their ability to provide ultra-precise measurements (UPMs), were tested across a range −10 40 °C. The results indicate that QSs offer superior accuracy, lower average error smaller standard deviation CSs, indicating better For comparison, utilized Python scripts conduct simulations statistical analyses, leading precise reproducible results. sensor was simulated controlled environment, obtained data experimental This comparison reveals are more reliable applications requiring high precision, such as those Internet Things (IoT) domain. These findings underscore potential advantage critical systems where is paramount.
Language: Английский
Citations
1
Frontiers in Optics + Laser Science 2022 (FIO, LS), Journal Year: 2024, Volume and Issue: unknown, P. JD4A.76 - JD4A.76
Published: Jan. 1, 2024
This paper presents a motion detection framework that is based on optical fiber reflections integrated with machine learning. The presented achieves accurate and cost-effective sensing by reliably identifying reflectance spectra enabling versatile applications
Language: Английский
Citations
0