Search Results (177)

The imaging modalities of correlation-assisted techniques utilize the inherent information present in the spatial correlation of random intensity patterns for the successful reconstruction of object information. However, most correlation approaches focus only on the reconstruction of amplitude information, as it is a direct byproduct of the correlation, disregarding the phase information. Complex-field reconstruction requires additional experimental or computational schemes, alongside conventional correlation geometry. The resurgence of holography in recent times, with advanced digital techniques and the adoption of the full-field imaging potential of holography in correlation with imaging techniques, has paved the way for the development of various state-of-the-art approaches to correlation optics. This review article provides an in-depth discussion of the recent developments in speckle-correlation-assisted techniques by focusing on various quantitative imaging scenarios. Furthermore, the recent progress and application of correlation-assisted holographic imaging techniques are reviewed, along with its potential challenges. Full article

In order to solve the problem of small and dense targets in drone aerial images, a small target detection model based on dual branches and adaptive feature fusion is proposed. The model first constructs a small target detection framework with dual branches to improve the detection accuracy while reducing the number of parameters. Secondly, the model introduces semantic and detail injection (SDI) in the neck network and embeds bidirectional adaptive feature fusion in the detection head to innovate and optimize the feature fusion mechanism, achieve the full interaction of deep and shallow information, enhance the feature representation of small targets, and overcome the problem of scale inconsistency. Finally, in order to focus on the target area more accurately, we introduce the large separable kernel attention mechanism into the convolutional layer to provide it with a richer and more comprehensive feature representation, which significantly improves the detection accuracy of targets of different scales. The experimental results show that the model algorithm performs well in the VisDrone2019 dataset. Compared with the original model, the mAP50 of this model increases by 20.9%, the mAP50–95 increases by 23.7%, and the total number of parameters decreases by 61.3%, making it more suitable for drones. Full article

Winter wheat yield prediction is critical for optimizing field management plans and guiding agricultural production. To address the limitations of conventional manual yield estimation methods, including low efficiency and poor interpretability, this study innovatively proposes an intelligent yield estimation method based on a convolutional neural network (CNN). A comprehensive two-factor (fertilization × irrigation) controlled field experiment was designed to thoroughly validate the applicability and effectiveness of this method. The experimental design comprised two irrigation treatments, sufficient irrigation (C) at 750 m³ ha⁻¹ and deficit irrigation (M) at 450 m³ ha⁻¹, along with five fertilization treatments (at a rate of 180 kg N ha⁻¹): (1) organic fertilizer alone, (2) organic–inorganic fertilizer blend at a 7:3 ratio, (3) organic–inorganic fertilizer blend at a 3:7 ratio, (4) inorganic fertilizer alone, and (5) no fertilizer control. The experimental protocol employed a DJI M300 RTK unmanned aerial vehicle (UAV) equipped with a multispectral sensor to systematically acquire high-resolution growth imagery of winter wheat across critical phenological stages, from heading to maturity. The acquired multispectral imagery was meticulously annotated using the Labelme professional annotation tool to construct a comprehensive experimental dataset comprising over 2000 labeled images. These annotated data were subsequently employed to train an enhanced CNN model based on ResNet50 architecture, which achieved automated generation of panicle density maps and precise panicle counting, thereby realizing yield prediction. Field experimental results demonstrated significant yield variations among fertilization treatments under sufficient irrigation, with the 3:7 organic–inorganic blend achieving the highest actual yield (9363.38 ± 468.17 kg ha⁻¹) significantly outperforming other treatments (p < 0.05), confirming the synergistic effects of optimized nitrogen and water management. The enhanced CNN model exhibited superior performance, with an average accuracy of 89.0–92.1%, representing a 3.0% improvement over YOLOv8. Notably, model accuracy showed significant correlation with yield levels (p < 0.05), suggesting more distinct panicle morphological features in high-yield plots that facilitated model identification. The CNN’s yield predictions demonstrated strong agreement with the measured values, maintaining mean relative errors below 10%. Particularly outstanding performance was observed for the organic fertilizer with full irrigation (5.5% error) and the 7:3 organic-inorganic blend with sufficient irrigation (8.0% error), indicating that the CNN network is more suitable for these management regimes. These findings provide a robust technical foundation for precision farming applications in winter wheat production. Future research will focus on integrating this technology into smart agricultural management systems to enable real-time, data-driven decision making at the farm scale. Full article

Accurate analysis of medical videos remains a major challenge in deep learning (DL) due to the need for effective spatiotemporal feature mapping that captures both spatial detail and temporal dynamics. Despite advances in DL, most existing models in medical AI focus on static images, overlooking critical temporal cues present in video data. To bridge this gap, a novel DL-based framework is proposed for spatiotemporal feature extraction from medical video sequences. As a feasibility use case, this study focuses on gastrointestinal (GI) endoscopic video classification. A 3D convolutional neural network (CNN) is developed to classify upper and lower GI endoscopic videos using the hyperKvasir dataset, which contains 314 lower and 60 upper GI videos. To address data imbalance, 60 matched pairs of videos are randomly selected across 20 experimental runs. Videos are resized to 224 × 224, and the 3D CNN captures spatiotemporal information. A 3D version of the parallel spatial and channel squeeze-and-excitation (P-scSE) is implemented, and a new block called the residual with parallel attention (RPA) block is proposed by combining P-scSE3D with a residual block. To reduce computational complexity, a (2 + 1)D convolution is used in place of full 3D convolution. The model achieves an average accuracy of 0.933, precision of 0.932, recall of 0.944, F1-score of 0.935, and AUC of 0.933. It is also observed that the integration of P-scSE3D increased the F1-score by 7%. This preliminary work opens avenues for exploring various GI endoscopic video-based prospective studies. Full article

Objectives: This study aimed to implement a high-resolution Saudi voxel-based anthropomorphic phantom in the OpenMC Monte Carlo (MC) simulation framework. The objective was to evaluate its applicability in radiological simulations, including radiographic imaging and effective dose calculations, tailored to the Saudi population. Methods: A voxel phantom comprising 30 segmented organs/tissues and over 32 million voxels were constructed from full-body computed tomography data and integrated into OpenMC. The implementation involved detailed voxel mapping, material definition using ICRP/ICRU-116 recommendations, and lattice geometry construction. The simulations included X-ray radiography projections using mesh tallies and anterior–posterior effective dose calculations across 20 photon energies (10 keV–1 MeV). The absorbed dose was calculated using OpenMC’s heating tally and converted to an effective dose using tissue weighting factors. Results: The phantom was successfully modeled and visualized in OpenMC, demonstrating accurate anatomical representation. Radiographic projections showed optimal contrast at 70 keV. The effective dose values for 29 organs were calculated and compared with MCNPX, the ICRP-116 reference phantom, and XGBoost-based machine learning (ML) predictions. OpenMC results showed good agreement, with maximum deviations of −35.5% against ICRP-116 at 10 keV. Root mean square error (RMSE) comparisons confirmed reasonable alignment, with OpenMC displaying higher RMSEs relative to other methods due to expanded organ modeling and material definitions. Conclusions: The integration of the Saudi voxel phantom into OpenMC demonstrates its utility for high-resolution dosimetry and radiographic simulations. OpenMC’s Python (version 3.10.14) interface and open-source nature make it a promising tool for radiological research. Future work will focus on combining MC and ML approaches for enhanced predictive dosimetry. Full article

A comprehensive framework for ultrasound imaging simulations is presented. Solutions to an inhomogeneous wave equation are provided, yielding a linear model for characterizing ultrasound propagation and scattering in soft tissue. This simulation approach, which is based upon the fast nearfield method, is implemented in the Fast Object-oriented C++ Ultrasound Simulator (FOCUS) and is extended to a range of imaging modalities, including synthetic aperture, B-mode, plane wave, and color Doppler imaging. The generation of radiofrequency (RF) data and the receive beamforming techniques employed for each imaging modality, along with background on color Doppler imaging, are described. Simulation results demonstrate rapid convergence and lower error rates compared to conventional spatial impulse response methods and Field II, resulting in substantial reductions in computation time. Notably, the framework effectively simulates hundreds of thousands of scatterers without the need for a full three-dimensional (3D) grid, and the inherent randomness in the scatterer distributions produces realistic speckle patterns. A plane wave imaging example, for instance, achieves high fidelity using 100,000 scatterers with five steering angles, and the simulation is completed on a personal computer in a few minutes. Furthermore, by modeling scatterers as moving particles, the simulation framework captures dynamic flow conditions in vascular phantoms for color Doppler imaging. These advances establish FOCUS as a robust, versatile tool for the rapid prototyping, validation, and optimization of both established and novel ultrasound imaging techniques. Full article

The manufacturing methodologies for carbon fiber triaxial woven fabric composites demonstrate significant variability, resulting in the failure mechanisms under tensile loading conditions, and the fundamental role of interweaving points remains unclear. Moreover, the mechanisms of destruction under tensile loads have not been sufficiently studied. In this study, the resin transfer molding and resin film infusion were selected to fabricate carbon fiber triaxial woven fabric composites, with a specific focus on their effects on the tensile properties of carbon fiber triaxial woven composites. Compared with ordinary materials, the tensile load of carbon fiber triaxial woven fabric composites after yarn spreading has increased by more than 30%. The strength can reach 1133 MPa after yarn spreading of 3k carbon fiber, which was 39% higher than the original. Furthermore, acoustic emission monitoring shows that the counts of acoustic signals in the first half dropped from 10,000 to around 3000, mostly due to the reduction of resin and fiber/matrix debonding. The digital image correlation provided full-field strain analysis, which proved that the strain of the fibers at the interweaving points decreased significantly during the stretching process after yarn spreading. Full article

A conventional microchannel plate framing camera is typically utilized for inertial confinement fusion diagnosis. However, as a vacuum electronic device, it has inherent limitations, such as a complex structure and the inability to achieve single-line-of-sight imaging. To address these challenges, a CMOS image sensor that can be seamlessly integrated with an electronic pulse broadening system can provide a viable alternative to the microchannel plate detector. This paper introduces the design of an 8 × 8 pixel-array ultrashort shutter-time single-framing CMOS image sensor, which leverages silicon epitaxial processing and a 0.18 μm standard CMOS process. The focus of this study is on the photodiode and the readout pixel-array circuit. The photodiode, designed using the silicon epitaxial process, achieves a quantum efficiency exceeding 30% in the visible light band at a bias voltage of 1.8 V, with a temporal resolution greater than 200 ps for visible light. The readout pixel-array circuit, which is based on the 0.18 μm standard CMOS process, incorporates 5T structure pixel units, voltage-controlled delayers, clock trees, and row-column decoding and scanning circuits. Simulations of the pixel circuit demonstrate an optimal temporal resolution of 60 ps. Under the shutter condition with the best temporal resolution, the maximum output swing of the pixel circuit is 448 mV, and the output noise is 77.47 μV, resulting in a dynamic range of 75.2 dB for the pixel circuit; the small-signal responsivity is 1.93 × 10⁻⁷ V/e⁻, and the full-well capacity is 2.3 Me⁻. The maximum power consumption of the 8 × 8 pixel-array and its control circuits is 0.35 mW. Considering both the photodiode and the pixel circuit, the proposed CMOS image sensor achieves a temporal resolution better than 209 ps. Full article

Background and Clinical Significance: Non-prosthetic peri-implant fractures (NPPIFs) are rare injuries occurring around internal fixation devices, and are distinct from periprosthetic fractures. While most studies focus on the femur, humeral NPPIFs remain poorly documented. This case illustrates a complex humeral NPPIF and highlights key surgical considerations. Case Presentation: A 62-year-old woman presented with a spiral humeral shaft fracture (AO 12B2) after a fall. Following closed reduction and antegrade intramedullary nailing, an intraoperative peri-implant fracture occurred at the distal interlocking screw. CT imaging revealed a complex fracture extending from the lateral condyle to the proximal humerus. Treatment included implant removal and open reduction with dual plate fixation—lateral distal and helically contoured proximal plates—plus cerclage bands and antibiotic-loaded beads. Recovery was uneventful, with a full range of motion achieved at six months. At one year, the DASH score and MEPS were 86 and 75, respectively. Conclusions: Humeral NPPIFs are challenging and require individualized, biomechanically sound strategies. This case reinforces the importance of intraoperative assessment and careful implant selection in humeral fracture management. Full article

Introduction: Paediatric hip disorders present unique challenges for artificial intelligence (AI)-aided assessments of medical imaging due to disease-related and age-dependent changes in hip morphology. This scoping review aimed to describe current and emerging applications of AI in medical imaging for paediatric hip disorders. Methods: A descriptive synthesis of articles identified through PubMed, Embase, Cochrane Library, Web of Science, Emcare, and Academic Search Premier databases was performed including articles published up until June 2024. Original research articles’ titles and abstracts were screened, followed by full-text screening. Two reviewers independently conducted article screening and data extraction (i.e., data on the article and the model and its performance). Results: Out of 871 unique articles, 40 were included. The first article was dated from 2017, with annual publication rates increasing thereafter. Research contributions were primarily from China (17 [43%]) and Canada (10 [25%]). Articles mainly focused on developing novel AI models (19 [47.5%]), applied to ultrasound images or radiographs of developmental dysplasia of the hip (DDH; 37 [93%]). The three remaining articles addressed Legg–Calvé–Perthes disease, neuromuscular hip dysplasia in cerebral palsy, or hip arthritis/osteomyelitis. External validation was performed in eight articles (20%). Models were mainly applied to the diagnosis/grading of the disorder (22 [55%]), or on screening/detection (17 [42.5%]). AI models were 17 to 124 times faster (median 30) in performing a specific task than experienced human assessors, with an accuracy of 86–100%. Conclusions: Research interest in AI applied to medical imaging of paediatric hip disorders has expanded significantly since 2017, though the scope remains restricted to developing novel models for DDH imaging. Future studies should focus on (1) the external validation of existing models, (2) implementation into clinical practice, addressing the current lack of implementation efforts, and (3) paediatric hip disorders other than DDH. Full article

Phase wrapping is a common phenomenon in optical full-field imaging or measurement systems. It arises from large phase retardations and results in wrapped-phase maps that contain essential information about surface roughness and topology. However, these maps are often degraded by noise, such as speckle and Gaussian, which reduces the measurement accuracy and complicates phase reconstruction. Denoising such data is a fundamental problem in computer vision and plays a critical role in biomedical imaging modalities like Full-Field Optical Interferometry. In this paper, we propose WPD-Net (Wrapped-Phase Denoising Network), a lightweight deep learning-based neural network specifically designed to restore phase images corrupted by high noise levels. The network architecture integrates a shallow feature extraction module, a series of Residual Dense Attention Blocks (RDABs), and a dense feature fusion module. The RDABs incorporate attention mechanisms that help the network focus on critical features and suppress irrelevant noise, especially in high-frequency or complex regions. Additionally, WPD-Net employs a growth-rate-based feature expansion strategy to enhance multi-scale feature representation and improve phase continuity. We evaluate the model’s performance on both synthetic and experimentally acquired datasets and compare it with other state-of-the-art deep learning-based denoising methods. The results demonstrate that WPD-Net achieves superior noise suppression while preserving fine structural details even with mixed speckle and Gaussian noises. The proposed method is expected to enable fast image processing, allowing unwrapped biomedical images to be retrieved in real time. Full article

Mental images such as foci of attention can significantly enhance the quality of movements, providing a positive effect on human performance, e.g., in dancers or athletes. Thirteen participants (height = 161 ± 13 cm, mass = 46.4 ± 17.3 kg, and age = 21 ± 8.4 years) with varying levels of experience in classical ballet were divided into three groups (amateur, professional, and children). Each participant performed three sauté en suite jumps, followed by an instruction to imagine “taking the floor with them” during the jump. The study aimed to assess the effect of this external focus on jumping performance using biomechanical modeling. Results showed a statistically significant increase in jump height and an expanded range of motion in the hip and knee joints after the intervention, suggesting a positive influence on movement quality. However, results varied among groups, with no significant change in leg stiffness across participants, though tendencies appeared within each group. These findings indicate that an external focus of attention could be a useful tool in dance pedagogy, enhancing performance quality across experience levels and supporting individual progress. The study recommends further research to explore the full impact of psychologically effective instructions on various aspects of physical performance. Full article

Metasurfaces, composed of engineered nanoantennas, enable unprecedented control over electromagnetic waves by leveraging multipolar resonances to tailor light–matter interactions. This review explores key physical mechanisms that govern their optical properties, including the role of multipolar resonances in shaping metasurface responses, the emergence of bound states in the continuum (BICs) that support high-quality factor modes, and the Purcell effect, which enhances spontaneous emission rates at the nanoscale. These effects collectively underpin the design of advanced photonic devices with tailored spectral, angular, and polarization-dependent properties. This review discusses recent advances in metasurfaces and applications based on them, highlighting research that employs full-wave numerical simulations, analytical and semi-analytic techniques, multipolar decomposition, nanofabrication, and experimental characterization to explore the interplay of multipolar resonances, bound and quasi-bound states, and enhanced light–matter interactions. A particular focus is given to metasurface-enhanced photodetectors, where structured nanoantennas improve light absorption, spectral selectivity, and quantum efficiency. By integrating metasurfaces with conventional photodetector architectures, it is possible to enhance responsivity, engineer photocarrier generation rates, and even enable functionalities such as polarization-sensitive detection. The interplay between multipolar resonances, BICs, and emission control mechanisms provides a unified framework for designing next-generation optoelectronic devices. This review consolidates recent progress in these areas, emphasizing the potential of metasurface-based approaches for high-performance sensing, imaging, and energy-harvesting applications. Full article

For the Norman kings of Sicily and the ecclesiastical authorities who ruled their dioceses, Byzantine art served as both a symbol of luxury and a model of prestige. Similarly to the mosaics of Palermo, Monreale, and Cefalú, as well as textiles and goldsmithing, the manuscripts preserved in the National Library of Madrid stand as prime examples of the fascination that the dignitaries of the Kingdom of Sicily had for Byzantine esthetics. Among these manuscripts, the Sacramentary of Messina (Madrid, BNE Ms. 52) is perhaps the most striking. This Latin sacramentary, comprising 303 folios, features illuminated initials, a calendar with depictions of classical topics, such as the Spinario and a compelling depiction of August inspired by the Byzantine Koimesis, the months and zodiac, and two full-page illustrations depicting the Virgin Glykophilousa, the Crucifixion, and the Deesis. This study has a dual focus. First, it aims to analyze the iconographic peculiarities of the monthly images in this Latin calendar. Second, it seeks to provide new insights into the manuscript’s patronage and its place of origin. In this context, one of the most striking and significant aspects of the sacramentary’s iconography is the prominent role of the Virgin, a theme that will also be examined in this study. Archbishop Richard Palmer emerges as the leading candidate to have been the driving force in the patronage of the manuscript to the Royal scriptoria of Palermo. Full article

Sheet music is one of the most efficient methods for storing music. Meanwhile, a large amount of sheet music-image data is stored in paper form, but not in a computer-readable format. Therefore, digitizing sheet music is an essential task, such that the encoded music object could be effectively utilized for tasks such as editing or playback. Although there have been a few studies focused on recognizing sheet music images with simpler structures—such as monophonic scores or more modern scores with relatively simple structures, only containing clefs, time signatures, key signatures, and notes—in this paper we focus on the issue of classical sheet music containing dynamics symbols and articulation signs, more than only clefs, time signatures, key signatures, and notes. Therefore, this study augments the data from the GrandStaff dataset by concatenating single-line scores into multi-line scores and adding various classical music dynamics symbols not included in the original GrandStaff dataset. Given a full-scale piano score in pages, our approach first applies three YOLOv8 models to perform the three tasks: 1. Converting a full page of sheet music into multiple single-line scores; 2. Recognizing the classes and absolute positions of dynamics symbols in the score; and 3. Finding the relative positions of dynamics symbols in the score. Then, the identified dynamics symbols are removed from the original score, and the remaining score serves as the input into a Convolutional Recurrent Neural Network (CRNN) for the following steps. The CRNN outputs KERN notation (KERN, a core pitch/duration representation for common practice music notation) without dynamics symbols. By combining the CRNN output with the relative and absolute position information of the dynamics symbols, the final output is obtained. The results show that with the assistance of YOLOv8, there is a significant improvement in accuracy. Full article