Search Results (3,195)

This study examines the pore structure and distribution in 3D printed and cast foamed concrete using protein-based and synthetic foaming agents alongside various stabilizing additives. In 3D printed samples, pores are irregular and flattened due to mechanical forces during printing, whereas cast samples display uniform, spherical pores from homogeneous foam distribution. Samples containing the CA stabilizer show higher apparent densities (up to 2.05 g/cm³ for printed samples), correlating with lower water absorption. Protein-based foaming agents (PS) produce smaller, more evenly distributed pores, while synthetic agents (AS) result in larger, less uniform pores. Stabilizers significantly influence pore characteristics: commercial stabilizers yield smaller, more uniform pores, while recycled industrial oil (UO) leads to larger, more variable pores. Protein-based agents improve structural stability and reduce water absorption through uniform pore distributions, while synthetic agents lower density and increase water absorption. The highest sorption values were observed in samples with AS without stabilizer (1.7 kg/m²h¹/²) and AS and UO (1.6 kg/m²h¹/²) in a vertical orientation, with the horizontal orientation of sample AS and UO achieving a peak value of 2.0 kg/m²h¹/². Moreover, stabilization using UO resulted in higher sorption coefficients than stabilization with CA. High porosity in M1 resulted in low strength (0.2 MPa bending, 0.1 MPa perpendicular compression), while M5 showed superior performance (11.5 MPa perpendicular compression). PS-foamed samples (M4, M6) with uniform pores had the highest strengths, with M6 achieving 3.8 MPa bending and 10.3 MPa perpendicular compression. Perpendicular compression (M5: 11.5 MPa) was up to three times stronger than parallel compression due to weak interlayer bonds in 3D printing. Full article

In real-world scenarios, posture variation and low-quality image resolution are two well-known factors that compromise the accuracy and reliability of face recognition system. These challenges can be overcome using various methods, including Generative Adversarial Networks (GANs). Despite this, concerns over the accuracy and reliability of GAN methods are increasing as the facial recognition market expands rapidly. The existing framework such as Two-Pathway GAN (TP-GAN) method has demonstrated that it is superior to numerous GAN methods that provide better face-texture details due to its unique deep neural network structure that allows it to perceive local details and global structure in a supervised manner. TP-GAN overcomes some of the obstacle associated with face frontalization tasks through the use of landmark detection and synthesis functions, but it remains challenging to achieve the desired performance across a wide range of datasets. To address the inherent limitations of TP-GAN, we propose a novel face frontalization method (NFF) combining landmark detection, decision forests, and data augmentation. NFF provides 2D landmark detection to integrate global structure with local details of the generator model so that more accurate facial feature representations and robust feature extractions can be achieved. NFF enhances the stability of the discriminator model over time by integrating decision forest capabilities into the TP-GAN discriminator core architecture that allows us to perform a wide range of facial pose tasks. Moreover, NFF uses data augmentation techniques to maximize training data by generating completely new synthetic data from existing data. Our evaluations are based on the Multi-PIE, FEI, and CAS-PEAL datasets. NFF results indicate that TP-GAN performance can be significantly enhanced by resolving the challenges described above, leading to high quality visualizations and rank-1 face identification. Full article

Due to its miniature size and single-pass nature, Unmanned Aerial Vehicle (UAV)-borne array synthetic aperture radar (SAR) is capable of obtaining three-dimensional (3D) electromagnetic scattering information with a low cost and high efficiency, making it widely applicable in various fields. However, the limited payload capacity of the UAV platform results in a limited number of array antennas and affects 3D resolution. This paper proposes a 3D imaging method for UAV-borne SAR based on nested difference co-arrays and azimuth multi-snapshots. We first designed an antenna arrangement based on nested arrays, generating a virtual antenna twice as long as the original one. Then, we used a difference co-array method for 3D imaging. The required multi-snapshot data were obtained through azimuth down-sampling, rather than traditional spatial averaging methods. Due to the slow flight of the UAV, this method could generate multiple SAR images without affecting the two-dimensional resolution. Based on simulations and real data verification, the proposed algorithm overcomes the problem of two-dimensional resolution decline caused by traditional spatial averaging methods and improves three-dimensional resolution ability, theoretically achieving half the Rayleigh resolution. Full article

AI applications are increasingly prevalent in radiotherapy, including commercial software solutions for automatic segmentation of anatomical structures for 3D CT. However, their use in intraoperative electron radiotherapy (IOERT) remains limited. In particular, no AI solution is available for contouring cone beam CT (CBCT) images acquired with a mobile CBCT device. The U-Net convolutional neural network architecture has gained huge success for medical image segmentation but still has difficulties capturing the global context. To increase the accuracy in CBCT segmentation for IOERT, three different AI architectures were trained and evaluated. The features of the natural language processing models Transformer and xLSTM were added to the popular U-Net architecture and compared with the standard U-Net and manual segmentation performance. These networks were trained and tested using 55 CBCT scans obtained from breast cancer patients undergoing IOERT in the department of radiotherapy and radiation oncology in Salzburg, and each architecture’s segmentation performance was assessed using the dice coefficient (DSC) as a similarity measure. The average DSC values were 0.83 for the standard U-Net, 0.88 for the U-Net with transformer features, and 0.66 for the U-Net with xLSTM. The hybrid U-Net architecture, including Transformer features, achieved the best segmentation accuracy, demonstrating an improvement of 5% on average over the standard U-Net, while the U-Net with xLSTM showed inferior performance compared to the standard U-Net. With the help of automatic contouring, synthetic CT images can be generated, and IOERT challenges related to the time-consuming nature of 3D image-based treatment planning can be addressed. Full article

This study examined the impact of the long-term application (25 years) of tea waste (TW), compost (COM), and neem oil cake (NOC) compared to conventional synthetic fertilizers (CONV) on soil thermal and physical properties of a tea-cultivated Ultisol. Soil samples were collected from 0–15 cm and 15–30 cm depths of an experimental site of the Tea Research Institute in Sri Lanka. These samples were analyzed for soil thermal conductivity (k), volumetric heat capacity (C), thermal diffusivity (D), bulk density (BD), aggregate stability, soil organic carbon (SOC), and volumetric water contents at 0 kPa (θ₀) and 10 kPa (θ₁₀). TW and COM significantly (p < 0.05) increased surface SOC, leading to better aggregation, lower BD, and, consequently, a substantial reduction in k and D compared to CONV plots. Further, TW and COM amendments slightly increased C compared to CONV plots due to elevated SOC and water content. However, NOC had no impact on soil thermal and physical properties compared to CONV. The reduced thermal conductivity and thermal diffusivity indicated an improved soil capacity to buffer extreme temperature fluctuations. Moreover, soils treated with TW and COM exhibited greater water retention and improved soil resistance to erosion. The findings suggest that the long-term application of tea waste and compost could be a sustainable soil management strategy for improving soil health and enhancing resilience to climate change in tea-cultivated Ultisols. Full article

With the rapid development of the Internet of Vehicles (IoV) and autonomous driving technologies, robust and accurate visual pose perception has become critical for enabling smart connected vehicles. Traditional deep learning-based localization methods face persistent challenges in real-world vehicular environments, including occlusion, lighting variations, and the prohibitive cost of collecting diverse real-world datasets. To address these limitations, this study introduces a novel approach by combining Vision-LSTM (ViL) with synthetic image data generated from high-fidelity 3D models. Unlike traditional methods reliant on costly and labor-intensive real-world data, synthetic datasets enable controlled, scalable, and efficient training under diverse environmental conditions. Vision-LSTM enhances feature extraction and classification performance through its matrix-based mLSTM modules and advanced feature aggregation strategy, effectively capturing both global and local information. Experimental evaluations in independent target scenes with distinct features and structured indoor environments demonstrate significant performance gains, achieving matching accuracies of 91.25% and 95.87%, respectively, and outperforming state-of-the-art models. These findings underscore the innovative advantages of integrating Vision-LSTM with synthetic data, highlighting its potential to overcome real-world limitations, reduce costs, and enhance accuracy and reliability for connected vehicle applications such as autonomous navigation and environmental perception. Full article

Applying deep neural networks (DNNs) to broadband seismic wave impedance inversion is challenging, especially in generalizing from synthetic to field data, which limits the exploitation of their nonlinear mapping capabilities. While many research studies are about advanced and enhanced architectures of DNNs, this article explores how variations in input data affect DNNs and consequently enhance their generalizability and inversion performance. This study introduces a novel data pre-processing strategy based on histogram equalization and an iterative testing strategy. By employing a U-Net architecture within a fully convolutional neural network (FCN) exclusively trained on synthetic and monochrome data, including post-stack profile, and 1D linear background impedance profiles, we successfully achieve broadband impedance inversion for both new synthetic data and marine seismic data by integrating imaging profiles with background impedance profiles. Notably, the proposed method is applied to reverse time migration (RTM) data from the Ceduna sub-basin, located in offshore southern Australia, significantly expanding the wavenumber bandwidth of the available data. This demonstrates its generalizability and improved inversion performance. Our findings offer new insights into the challenges of seismic data fusion and promote the utilization of deep neural networks for practical seismic inversion and outcomes improvement. Full article

While traditional self-supervised learning methods improve performance and robustness across various medical tasks, they rely on single-vector embeddings that may not capture fine-grained concepts such as anatomical structures or organs. The ability to identify such concepts and their characteristics without supervision has the potential to improve pre-training methods, and enable novel applications such as fine-grained image retrieval and concept-based outlier detection. In this paper, we introduce ConceptVAE, a novel pre-training framework that detects and disentangles fine-grained concepts from their style characteristics in a self-supervised manner. We present a suite of loss terms and model architecture primitives designed to discretise input data into a preset number of concepts along with their local style. We validate ConceptVAE both qualitatively and quantitatively, demonstrating its ability to detect fine-grained anatomical structures such as blood pools and septum walls from 2D cardiac echocardiographies. Quantitatively, ConceptVAE outperforms traditional self-supervised methods in tasks such as region-based instance retrieval, semantic segmentation, out-of-distribution detection, and object detection. Additionally, we explore the generation of in-distribution synthetic data that maintains the same concepts as the training data but with distinct styles, highlighting its potential for more calibrated data generation. Overall, our study introduces and validates a promising new pre-training technique based on concept-style disentanglement, opening multiple avenues for developing models for medical image analysis that are more interpretable and explainable than black-box approaches. Full article

Electrochemical enzyme biosensors are extensively utilized in clinical analysis and environmental monitoring, yet achieving effective enzyme immobilization while maintaining high activity remains a challenge. In this work, we developed a flow-through enzyme biosensor system using a 3D-printed flow-through electrochemical cell fabricated from commercially available poly (lactic acid). After modification with thiacalixarene-functionalized oligo (lactic acids) (OLAs), the material enabled efficient immobilization of uricase on the inner surface of a replaceable reactor of the cell. Swelling and hydrolytic stability of OLAs in cone, partial cone, and 1,3-alternate conformations were studied, with 1,3-alernate conformation demonstrating superior stability and enzyme immobilization performance. The use of OLAs enhanced immobilization efficiency by over 30% and protected the reactor from swelling, hydrolytic degradation, and enzyme loss. The biosensor was validated for amperometric uric acid determination, with a screen-printed carbon electrode modified with carbon black and Prussian Blue. This modification reduced the cathodic potential for uric acid detection to –0.05 V. The biosensor exhibited a linear detection range of 10 nM to 30 μM with a detection limit of 7 nM, and it performed effectively in artificial urine and synthetic blood plasma. The novel cell design, featuring easy assembly and low-cost replaceable parts, makes this biosensor a promising candidate for routine clinical analysis and other practical applications. Full article

Bacterium-triggered carious lesions implicate dental hard tissue destruction and the simultaneous initiation of regenerative events comprising dental stem cell activation. Streptococcus mutans (S. mutans) is a prominent pathogen of the oral cavity and the principal cause of caries. S. mutans generates complex products involved in interbacterial interactions, including Mutanobactin-D (Mub-D), which belongs to a group of non-ribosomal cyclic lipopeptides. In the present study, we aimed to analyse the potential role of the synthetic Mub-D peptide in cell populations involved in tissue regenerative processes. To this end, we assessed the in vitro effects of Mub-D in human dental pulp stem cells (hDPSCs) and human bone marrow stem cells (hBMSCs). Our data demonstrated a concentration-dependent effect of Mub-D on their viability and a significant increase in their proliferation and osteogenic/odontogenic differentiation. These events were associated with specific changes in gene expression, where CCDN-1, RUNX-2, OSX, OCN, DMP-1, DSPP, and BMP-2 genes were upregulated. The ability of Mub-D to modulate the osteogenic/odontogenic differentiation of both hDPSCs and hBMSCs and considerably enhance mineralisation in a controlled and concentration-dependent manner opens new perspectives for stem cell-based regenerative approaches in the clinics. Full article

Continuous fibers with outstanding mechanical performance due to the continuous enhancement effect, show wide application in aerospace, automobile, and construction. There has been great success in developing continuous synthetic fiber-reinforced composites, such as carbon fibers or glass fibers; however, most of which are nonrenewable, have a high processing cost, and energy consumption. Bio-sourced materials with high reinforced effects are attractive alternatives to achieve a low-carbon footprint. In this study, continuous bamboo fiber-reinforced polyethylene (CBF/PE) composites were prepared via a facile two-step method featuring alkali treatment followed by 3D printing. Alkali treatment as a key processing step increases surface area and surface wetting, which promote the formation of mechanical riveting among bamboo fibers and matrix. The obtained treated CBF (T-CBF) also shows improved mechanical properties, which enables a superior reinforcement effect. 3D printing, as a fast and local heating method, could melt the outer layer PE tube and impregnate molten plastics into fibers under pressure and heating. The resulting T-CBF/PE composite fibers can achieve a tensile strength of up to 15.6 MPa, while the matrix PE itself has a tensile strength of around 7.7 MPa. Additionally, the fracture morphology of printed bulks from composite fibers shows the alkali-treated fibers–PE interface is denser and could transfer more load. The printed bulks using T-CBF/PE shows increased tensile strength and Young’s modulus, with 77%- and 1.76-times improvement compared to pure PE. Finally, the effect of printing paraments on mechanical properties were analyzed. Therefore, this research presents a potential avenue for fabricating continuous natural fiber-reinforced composites. Full article

To date, there is little publicly available scientific data on unidentified aerial phenomena (UAP) whose properties and kinematics purportedly reside outside the performance envelope of known phenomena. To address this deficiency, the Galileo Project is designing, building, and commissioning a multi-modal, multi-spectral ground-based observatory to continuously monitor the sky and collect data for UAP studies via a rigorous long-term aerial census of all aerial phenomena, including natural and human-made. One of the key instruments is an all-sky infrared camera array using eight uncooled long-wave-infrared FLIR Boson 640 cameras. In addition to performing intrinsic and thermal calibrations, we implement a novel extrinsic calibration method using airplane positions from Automatic Dependent Surveillance–Broadcast (ADS-B) data that we collect synchronously on site. Using a You Only Look Once (YOLO) machine learning model for object detection and the Simple Online and Realtime Tracking (SORT) algorithm for trajectory reconstruction, we establish a first baseline for the performance of the system over five months of field operation. Using an automatically generated real-world dataset derived from ADS-B data, a dataset of synthetic 3D trajectories, and a hand-labeled real-world dataset, we find an acceptance rate (fraction of in-range airplanes passing through the effective field of view of at least one camera that are recorded) of 41% for ADS-B-equipped aircraft, and a mean frame-by-frame aircraft detection efficiency (fraction of recorded airplanes in individual frames which are successfully detected) of 36%. The detection efficiency is heavily dependent on weather conditions, range, and aircraft size. Approximately 500,000 trajectories of various aerial objects are reconstructed from this five-month commissioning period. These trajectories are analyzed with a toy outlier search focused on the large sinuosity of apparent 2D reconstructed object trajectories. About 16% of the trajectories are flagged as outliers and manually examined in the IR images. From these ∼80,000 outliers and 144 trajectories remain ambiguous, which are likely mundane objects but cannot be further elucidated at this stage of development without information about distance and kinematics or other sensor modalities. We demonstrate the application of a likelihood-based statistical test to evaluate the significance of this toy outlier analysis. Our observed count of ambiguous outliers combined with systematic uncertainties yields an upper limit of 18,271 outliers for the five-month interval at a 95% confidence level. This test is applicable to all of our future outlier searches. Full article

Isoxazole derivatives (isoxazoles, isoxazolines, and isoxazolidines) are present in the structure of several natural products and/or pharmaceutically interesting compounds. In this work, a synthetic study for the preparation of fused isoxazoline/isoquinolinone hybrids is presented. The initial approach involving the sequential 1,3-dipolar cycloaddition of nitrile oxides to indenone (to obtain the isoxazoline ring) and a Beckmann rearrangement (to construct the isoquinolinone lactam system) was complicated by the formation of fragmentation products during the latter. Therefore, the desired hybrids were successfully reached by applying DDQ-mediated oxidation of the respective isoxazolidines. Based on the results, key observations were made regarding the mechanism of the Beckmann reaction. Moreover, selected isoxazole benzamides and fused isoxazoline/isoxazolidine isoquinolinones were in vitro evaluated against a series of fungi strains (including a 2D checkerboard assay with ketoconazole), revealing that some of these compounds exhibit promising antifungal activity. Full article

Simultaneous localization and mapping (SLAM) is the basis for intelligent robots to explore the world. As a promising method for 3D reconstruction, 3D Gaussian splatting (3DGS) integrated with SLAM systems has shown significant potential. However, due to environmental uncertainties, errors in the tracking process with 3D Gaussians can negatively impact SLAM systems. This paper introduces a novel dense RGB-D SLAM system based on 3DGS that refines Gaussians through sub-Gaussians in the camera coordinate system. Additionally, we propose an algorithm to select keyframes closely related to the current frame, optimizing the scene map and pose of the current keyframe. This approach effectively enhances both the tracking and mapping performance. Experiments on high-quality synthetic scenes (Replica dataset) and low-quality real-world scenes (TUM-RGBD and ScanNet datasets) demonstrate that our system achieves competitive performance in tracking and mapping. Full article

Point clouds are a crucial element in the process of scanning and reconstructing 3D environments, such as buildings or heritage sites. They allow for the creation of 3D models that can be used in a wide range of applications. In some cases, however, only the 3D model of an environment is available, and it is necessary to obtain point clouds with the same characteristics as those captured by a laser scanner. For instance, point clouds may be required for surveys, performance optimization, site scan planning, or validation of point cloud processing algorithms. This paper presents a new terrestrial laser scanner (TLS) simulator, designed as a Blender add-on, that produces synthetic point clouds from 3D models in real time. The simulator allows users to adjust a set of parameters to replicate real-world scanning conditions, such as noise generation, ensuring the synthetic point clouds closely mirror those produced by actual laser scanners. The target meshes may be derived from either a real-world scan or 3D designs created using design software. By replicating the spatial distributions and attributes of real laser scanner outputs and supporting real-time generation, the simulator serves as a valuable tool for scan planning and the development of synthetic point cloud repositories, advancing research and practical applications in 3D computer vision. Full article