Search Results (28)

To address the challenges of image inpainting in scenarios with extensive or irregular missing regions—particularly detail oversmoothing, structural ambiguity, and textural incoherence—this paper proposes an Image Structure-Guided (ISG) framework that hierarchically integrates structural priors with semantic-aware texture synthesis. The proposed methodology advances a two-stage restoration paradigm: (1) Structural Prior Extraction, where adaptive edge detection algorithms identify residual contours in corrupted regions, and a transformer-enhanced network reconstructs globally consistent structural maps through contextual feature propagation; (2) Structure-Constrained Texture Synthesis, wherein a multi-scale generator with hybrid dilated convolutions and channel attention mechanisms iteratively refines high-fidelity textures under explicit structural guidance. The framework introduces three innovations: (1) a hierarchical feature fusion architecture that synergizes multi-scale receptive fields with spatial-channel attention to preserve long-range dependencies and local details simultaneously; (2) spectral-normalized Markovian discriminator with gradient-penalty regularization, enabling adversarial training stability while enforcing patch-level structural consistency; and (3) dual-branch loss formulation combining perceptual similarity metrics with edge-aware constraints to align synthesized content with both semantic coherence and geometric fidelity. Our experiments on the two benchmark datasets (Places2 and CelebA) have demonstrated that our framework achieves more unified textures and structures, bringing the restored images closer to their original semantic content. Full article

Mine ventilation systems are critical for ensuring operational safety, yet air leakage remains a pervasive challenge, leading to energy inefficiency and heightened safety risks. Traditional tracer gas methods, while effective in simple networks, exhibit significant errors in complex multi-entry systems due to static empirical parameters and environmental interference. This study proposes an integrated methodology that combines multi-path airflow analysis with dynamic longitudinal dispersion coefficient correction to enhance the accuracy of air leakage detection. Utilizing sulfur hexafluoride (SF6) as the tracer gas, a phased release protocol with temporal isolation was implemented across five strategic points in a coal mine ventilation network. High-precision detectors (Bruel & Kiaer 1302) and the MIVENA system enabled synchronized data acquisition and 3D network modeling. Theoretical models were dynamically calibrated using field-measured airflow velocities and dispersion coefficients. The results revealed three deviation patterns between simulated and measured tracer peaks: Class A deviation showed 98.5% alignment in single-path scenarios, Class B deviation highlighted localized velocity anomalies from Venturi effects, and Class C deviation identified recirculation vortices due to abrupt cross-sectional changes. Simulation accuracy improved from 70% to over 95% after introducing wind speed and dispersion adjustment coefficients, resolving concealed leakage pathways between critical nodes and key nodes. The study demonstrates that the dynamic correction of dispersion coefficients and multi-path decomposition effectively mitigates errors caused by turbulence and geometric irregularities. This approach provides a robust framework for optimizing ventilation systems, reducing invalid airflow losses, and advancing intelligent ventilation management through real-time monitoring integration. Full article

The circadian clock orchestrates nearly every aspect of physiology, aligning metabolic processes with environmental cues, such as light and food intake. While the central pacemaker in the suprachiasmatic nucleus synchronizes peripheral clocks across key metabolic tissue, feeding behavior emerges as the dominant cue for peripheral clock alignment. This interaction reveals a crucial link between circadian biology and metabolism. Disruption of these processes, whether from shift work, irregular eating patterns or lifestyle misalignment, has been strongly associated with metabolic disorders, including obesity, insulin resistance and cardiometabolic diseases. Within the field of chrononutrition, strategies, such as time-restricted feeding (TRF), have gained attention for their potential to restore circadian alignment and improve metabolic health. However, translational gaps persist, as most mechanistic insights are derived from nocturnal murine models, limiting their applicability to diurnal human physiology. Moreover, human studies are confounded by interindividual variability in chronotype, behavioral patterns, and dietary compliance. This review explores the molecular underpinnings of zeitgeber signals and critically assesses the translational barriers to implementing chrononutrition across species. By integrating insights from both preclinical and clinical research, we aim to refine the potential of circadian-based dietary interventions for metabolic disease prevention and personalized nutrition. Full article

Punctuation restoration plays an essential role in the postprocessing procedure of automatic speech recognition, but model efficiency is a key requirement for this task. To that end, we present EfficientPunct, an ensemble method with a multimodal time-delay neural network that outperforms the current best model by 1.0 F1 point while using less than a tenth of its network parameters for inference. This work further streamlines a speech recognizer and a BERT implementation to efficiently output hidden layer acoustic embeddings and text embeddings in the context of punctuation restoration. Here, forced alignment and temporal convolutions are used to eliminate the need for attention-based fusion, greatly increasing computational efficiency and improving performance. EfficientPunct sets a new state of the art with an ensemble that weighs BERT’s purely language-based predictions slightly more than the multimodal network’s predictions. Although EfficientPunct shows great promise, from a different perspective, to date, another important challenge in the field has been the fact that punctuation restoration models have been evaluated almost solely on well-structured, scripted corpora. However, real-world ASR systems and postprocessing pipelines typically apply to spontaneous speech with significant irregularities, stutters, and deviations from perfect grammar. To address this important discrepancy, we also introduce SponSpeech, a punctuation restoration dataset derived from informal speech sources, which includes punctuation and casing information. In addition to publicly releasing the dataset, the authors have contributed by providing a filtering pipeline that can be used to generate more data. This filtering pipeline examines the quality of both the speech audio and the transcription text. A challenging test set is also carefully constructed, aimed at evaluating the models’ ability to leverage audio information to predict, otherwise grammatically ambiguous, punctuation. SponSpeech has been made available to the public, along with all code for dataset building and model runs. Full article

Deep learning outperforms traditional interpolation methods in 3D geological modeling due to its ability to model nonlinear relationships and its flexibility in incorporating diverse geological data. However, acquiring geological data for practical applications is challenging, and the quality of the data can vary significantly, which limits the effectiveness of purely data-driven deep learning models in 3D geological modeling. To address this challenge, this paper introduces GeoSAE, a geoconstraint-driven 3D geological modeling method. GeoSAE improves potential field prediction by employing a stacked autoencoder network (SAE) and incorporating geological constraints as a loss function during model training. This approach generates a geologically consistent, smooth, and continuous 3D stratigraphic model. To validate the method, this study applies it to a 60-square-kilometer region in Jiangdong new district, Haikou city, China. Stratigraphic interface points were utilized to predict the 3D potential field, with PyVista (version 0.44.2) enabling the accurate extraction of stratigraphic interfaces. Model quality was evaluated through comprehensive assessments of loss function analysis, data fitting, and the verification of stratigraphic smoothness constraints. Results indicate that the stratigraphic model generated by GeoSAE closely aligns with the actual data, accurately capturing stratigraphic geometry. Additionally, incorporating smoothness constraints enhances model smoothness, minimizes irregular stratigraphic fluctuations, and produces a more natural and continuous stratigraphic morphology. Full article

Addressing the challenge of weak interface strength in 3D-printed mortars, this study introduces a novel technique using sinuous printing trajectories. The self-locking interface is formed by different meandering print trajectories, and the changes in the strength of the test interface are investigated by adjusting the trajectories to form different amplitudes. This ensures alignment of peaks and troughs between layers, aiming for enhanced interfacial cohesion. Experimental tests measured mechanical properties of printed mortar specimens with varying amplitudes. Using Digital Image Correlation technology, strain fields and fracture surfaces were analyzed. Initial results revealed a 28% decrease in shear resistance for side-by-side printed interfaces compared to traditional layered interfaces. As amplitude increased, shear load-bearing capacity improved. Specifically, a 15 mm amplitude saw a 40% rise in interlayer shear strength. However, a 20 mm amplitude led to reduced shear capacity, with even slight forces causing potential fractures. Tensile strength also increased with amplitude. Specimens up to 15 mm amplitude primarily followed the printing interface in fractures, while a 20 mm amplitude cut through mortar strips. Post-fracture analysis showed the highest surface irregularity at a 15 mm amplitude, aligning with tensile load-bearing capacity. Full article

To address the challenges of accurately segmenting irregular building boundaries in complex urban environments faced by existing remote sensing change detection methods, this paper proposes a building change detection network based on multilevel geometric representation optimization using frame fields called BuildingCDNet. The proposed method employs a multi-scale feature aggregation encoder–decoder architecture, leveraging contextual information to capture the characteristics of buildings of varying sizes in the imagery. Cross-attention mechanisms are incorporated to enhance the feature correlations between the change pairs. Additionally, the frame field is introduced into the network to model the complex geometric structure of the building target. By learning the local orientation information of the building structure, the frame field can effectively capture the geometric features of complex building features. During the training process, a multi-task learning strategy is used to align the predicted frame field with the real building outline, while learning the overall segmentation, edge outline, and corner point features of the building. This improves the accuracy of the building polygon representation. Furthermore, a discriminative loss function is constructed through multi-task learning to optimize the polygonal structured information of the building targets. The proposed method achieves state-of-the-art results on two commonly used datasets. Full article

Coal mining-induced ground subsidence is a severe hazard that can damage property, infrastructure, and the environment in the vicinity when the deformation is not negligible. The boundary of a mining-induced subsidence-affected zone refers to the area beyond which the ground subsidence is less concerned. Accurately measuring mining-induced ground deformation is essential for delineating the irregular boundary of the impacted area. This study employs multitemporal interferometric synthetic aperture radar (MT-InSAR) techniques, including differential InSAR (DInSAR), InSAR stacking, and interferometric point target analysis (IPTA), to analyze coal mine subsidence and delineate the boundaries of the mining-impacted zones. DInSAR accurately reconstructs, locates, and detects the trend in mining-induced subsidence and correlates well with documented mining operations. The InSAR stacking method maps the spatial variation of the ground’s average line-of-sight (LOS) velocity over the mining area, delineating the boundary of the impacted zone. IPTA analysis combining multilook and single-pixel phases achieves millimeter-level surface measurement above tunnel alignments and measures unevenly distributed deformation fields. This study considers an average of 4 cm per year of surface deformation in the LOS direction as the subsidence threshold value for delineating the boundary of the mining-induced subsidence-affected (MISA) zone during the active coal mining stage. Interestingly, there are twin transportation tunnels near the mining area. The twin tunnels completed before the coal mining activities started were functioning well, but damage was observed after the mining began. Our study reveals the tunnels are located within the InSAR-derived MISA zone, although the tunnels approach the MISA boundary. As direct signs of subsidence, ground fissures have been identified near the tunnels via field investigations and UAV photogrammetry. Furthermore, the derived distribution of ground fissures validates and verifies InSAR measurements. The integrated approach of MT-InSAR, UVA photogrammetry, and field investigation developed in this study can be applied to delineate the irregular boundary of the MISA zone and study the accumulating effects of mining-induced subsidence on the performance of infrastructure in areas proximate to coal mining activities. Full article

Stereo matching technology, enabling the acquisition of three-dimensional data, holds profound implications for marine engineering. In underwater images, irregular object surfaces and the absence of texture information make it difficult for stereo matching algorithms that rely on discrete disparity values to accurately capture the 3D details of underwater targets. This paper proposes a stereo method based on an energy function of Markov random field (MRF) with 3D labels to fit the inclined plane of underwater objects. Through the integration of a cross-based patch alignment approach with two label optimization stages, the proposed method demonstrates features akin to segment-based stereo matching methods, enabling it to handle images with sparse textures effectively. Through experiments conducted on both simulated UW-Middlebury datasets and real deteriorated underwater images, our method demonstrates superiority compared to classical or state-of-the-art methods by analyzing the acquired disparity maps and observing the three-dimensional reconstruction of the underwater target. Full article

The unusual daytime F-region Field-Aligned Irregularities (FAIs) were observed by the HCOPAR and the satellites at low latitudes on 13 June 2022. These irregularities survived from night-time to the following afternoon at 15:00 LT. During daytime, they appeared as fossil structures with low Doppler velocities and narrow spectral widths. These characteristics indicated that they drifted along the magnetic field lines without apparent zonal velocity to low latitudes. Combining the observations of the ICON satellite and the Hainan Digisonde, we derived the movement trails of these daytime irregularities. We attributed their generation to the rapid ascent of the F-layer due to the fluctuation of IMF Bz during the quiet geomagnetic conditions. Subsequently, the influence of the substorm on the low-latitude ionosphere was investigated and simulated. The substorm caused the intense Joule heating that enhanced the southward neutral winds, carrying the neutral compositional disturbances to low latitudes and resulting in a negative storm effect in Southeast Asia. The negative storm formed a low-density circumstance and slowed the dissipation of the daytime FAIs. These results may provide new insights into the generation of post-midnight irregularities and their relationship with daytime fossil structures. Full article

In this study, we demonstrate a single-track magnetic code tape-based absolute position sensor system. Unlike traditional dual-track systems, our method simplifies manufacturing and avoids crosstalk between tracks, offering higher tolerance to alignment errors. The sensing system employs an array of magnetic field sensing elements that recognize the bit sequence encoded on the tape. This approach allows for accurate position determination even when the number of sensing elements is fewer than the number of bits covered, and without the need for specific spacing between sensing elements and bit length. We demonstrate the system’s ability to learn and adapt to various magnetic code patterns, including those that are irregular or have been altered. Our method can identify and localize the sensed magnetic field pattern directly within a self-learned magnetic field map, providing robust performance in diverse conditions. This self-adaptive capability enhances operational safety and reliability, as the system can continue functioning even when the magnetic tape is misaligned or has undergone changes. Full article

Space plasma turbulence plays a relevant role in several plasma environments, such as solar wind and the Earth’s magnetosphere–ionosphere system, and is essential for describing their complex coupling. This interaction gives rise to various phenomena, including ionospheric irregularities and the amplification of magnetospheric and ionospheric currents. The structure and dynamics of these currents have relevant implications, for example, in studying ionospheric heating and the nature of electric and magnetic field fluctuations in the auroral and polar environments. In this study, we investigate the nature of small-scale fluctuations characterizing the ionospheric magnetic field in response to different geomagnetic conditions. We use high-resolution (50 Hz) magnetic data from the ESA’s Swarm mission, collected during a series of high-latitude crossings, to probe the scaling features of magnetic field fluctuations in auroral and polar cap regions at spatial scales still poorly explored. Our findings reveal that magnetic field fluctuations in field-aligned currents (FACs) and polar cap regions across both hemispheres are characterized by different scaling properties, suggesting a distinct driver of turbulence. Furthermore, we find that geomagnetic activity significantly influences the nature of energy dissipation in FAC regions, leading to more localized filamentary structures toward smaller scales. Full article

The claim that incoherent scatter radar data show electron density enhancements of 50–80% during some ionospheric heating experiments in a recent paper is questioned. The backscatter from the monostatic radar can indeed be enhanced during these experiments, but the conclusion that a large electron density increase is the cause is almost certainly wrong. Some natural plasma line data are presented in support of our claim. Previously published studies of similar events and a possible explanation for the observed increases in backscattered power are pointed out. Full article

On the basis of interferometry measurement made with the Chung-Li VHF radar, we investigated the effects of upward propagating gravity waves on the spatial structures and dynamic behavior of the 3 m field-aligned irregularities (FAIs) of the sporadic E (Es) layer. The results demonstrate that the quasi-periodic gravity waves oscillating at a dominant wave period of about 46.3 min propagating from east-southeast to west-northwest not only modulated the Es layer but also significantly disturbed the Es layer. Interferometry analysis indicates that the plasma structures associated with gravity wave propagation were in clumpy or plume-like structures, while those not disturbed by the gravity waves were in a thin layer structure that descended over time at a rate of about 2.17 km/h. Observation reveals that the height of a thin Es layer with a thickness of about 2–4 km can be severely modulated by the gravity wave with a height as large as 10 km or more. Moreover, sharply inclined plume-like plasma irregularities with a tilted angle of about 55° or more with respect to the zonal direction were observed. In addition, concave and convex shapes of the Es layer caused by the gravity wave modulations were also found. Some of the wave-generated electric fields were so intense that the corresponding E × B drift velocities of the 3 m Es FAIs approximated 90 m s⁻¹. Most interestingly, sharp Doppler velocity shear as large as 68 m/s/km of the Es FAIs at a height of around 108 km, which bore a strong association with the result of the gravity wave propagation, was provided. The plausible mechanisms responsible for this tremendously large Doppler velocity shear are discussed. Full article

Ducts (field-aligned plasma density enhancements) provide a link into the magnetosphere and can guide whistler waves. Inside ducts, wave-particle interactions occur efficiently; therefore, their presence contributes to the removal of energetic particles from the magnetosphere. We present experimental results concerning the characteristics, behavior, and excitation thresholds of ducts induced by extraordinary (X-mode) polarized high-power HF radio waves emitted towards the magnetic zenith (MZ) into the upper ionosphere at EISCAT (European Incoherent SCATter). The features and behavior of ducts were diagnosed by the EISCAT incoherent scatter radar (ISR) at Tromsø and the CUTLASS (SuperDARN) Finland radar at Hankasalmi. The state of the ionosphere was monitored by the Dynasonde in Tromsø. It was found that the electron density N_e enhancements inside ducts were of 50–80% above the background N_e values and their transverse size (normal to the magnetic flux tube) corresponded to about 3–4° in the north–south direction. They were generated during magnetically quiet periods and extended from ~300 to 320 km up to the upper altitude limit of the EISCAT radar measurements (600–700 km), when heater frequencies were both below and above the critical frequency of the F2 layer (f_H ≤ f_oF2 and f_H > f_oF2), regardless of whether HF-induced plasma and ion lines were generated or not. Comparing the O-/X-mode effects from the EISCAT radar observations, it was shown that the creation of the strong N_e ducts is a typical characteristic of the X-mode pulses. As a rule, electron density enhancements were not observed during O-mode pulses. A plausible mechanism for the creation of X-mode artificial ducts is discussed. Full article