Search Results (1,449)

This research aims to investigate the Italian architectural heritage in Brazil through the analysis of the ‘Santa Casa de Misericórdia’ hospital in São Carlos, in the state of São Paulo. As part of the KNOW.IT national project, the work aims to recover and digitally enhance Italian heritage abroad from the 19th and 20th centuries. The buildings analysed were either designed or built by Italian architects who emigrated to South America or constructed using materials and techniques typical of Italian architecture of those years. The hospital, designed by the Italian architect Samuele Malfatti in 1891, was chosen for its historical value and its role in the urban context of the city of São Carlos, which, moreover, continues to perform its function even today. The study aims to create a digital archive with 3D models and two-dimensional graphical drawings. The methodology includes historical analysis, photogrammetric survey, and digital modelling using Agisoft Metashape and 3DF Zephyr software. A total of 636 images were processed, with the maximum resolution achieved in the models being 3526 × 2097 pixels. The results highlight the influence of Italian architecture on late 19th-century São Carlos and promote its virtual accessibility and wide-ranging knowledge. Full article

The use of high-resolution sea level anomaly (SLA) data in climate change research and ocean forecasting has become increasingly important. However, existing datasets often lack the fine spatial resolution required for capturing mesoscale ocean processes accurately. This has led to the development of conventional deep learning models for SLA spatial downscaling, but these models often overlook spatial disparities between land and ocean regions and do not adequately address the spatial structures of SLA data. As a result, their accuracy and structural consistency are suboptimal. To address these issues, we propose a Deep Separable Distillation Network (DSDN) that integrates Depthwise Separable Distillation Blocks (DSDB) and a Landmask Contextual Attention Mechanism (M_CAMB) to achieve efficient and accurate spatial downscaling. The M_CAMB employs geographically-informed land masks to enhance the attention mechanism, prioritizing ocean regions. Additionally, we introduce a novel Pixel-Structure Loss (PSLoss) to enforce spatial structure constraints, significantly improving the structural fidelity of the SLA downscaling results. Experimental results demonstrate that DSDN achieves a root mean square error (RMSE) of 0.062 cm, a peak signal-to-noise ratio (PSNR) of 42.22 dB, and a structural similarity index (SSIM) of 0.976 in SLA downscaling. These results surpass those of baseline models and highlight the superior precision and structural consistency of DSDN. Full article

To promote the automation and intelligence of rail freight, the accurate identification and localization of bulk cargo unloading hoppers have become a key technical challenge. Under the technological wave driven by the deep integration of Industry 4.0 and artificial intelligence, the bulk cargo unloading process is undergoing a significant transformation from manual operation to intelligent control. In response to this demand, this paper proposes a vision-based 3D localization system for unloading hoppers, which adopts a single visual sensor architecture and integrates three core modules: object detection, corner extraction, and 3D localization. Firstly, a lightweight hybrid attention mechanism is incorporated into the YOLOv5 network to enable edge deployment and enhance the detection accuracy of unloading hoppers in complex industrial scenarios. Secondly, an image processing approach combining depth consistency constraint (DCC) and geometric structure constraints is designed to achieve sub-pixel level extraction of key corner points. Finally, a real-time 3D localization method is realized by integrating corner-based initialization with an RGB-D SLAM tracking mechanism. Experimental results demonstrate that the proposed system achieves an average localization accuracy of 97.07% under challenging working conditions. This system effectively meets the comprehensive requirements of automation, intelligence, and high precision in railway bulk cargo unloading processes, and exhibits strong engineering practicality and application potential. Full article

Visual geolocalization for aerial vehicles based on an analysis of Earth observation imagery is an effective method in GNSS-denied environments. However, existing methods for geographic location estimation have limitations: one relies on high-precision geodetic elevation data, which is costly, and the other assumes a flat ground surface, ignoring elevation differences. This paper presents a novel aerial vehicle geolocalization method. It integrates 2D information and relative depth information, which are both from Earth observation images. Firstly, the aerial and reference remote sensing satellite images are fed into a feature-matching network to extract pixel-level feature-matching pairs. Then, a depth estimation network is used to estimate the relative depth of the satellite remote sensing image, thereby obtaining the relative depth information of the ground area within the field of view of the aerial image. Finally, high-confidence matching pairs with similar depth and uniform distribution are selected to estimate the geographic location of the aerial vehicle. Experimental results demonstrate that the proposed method outperforms existing ones in terms of geolocalization accuracy and stability. It eliminates reliance on elevation data or planar assumptions, thus providing a more adaptable and robust solution for aerial vehicle geolocalization in GNSS-denied environments. Full article

In remote sensing, object classification often suffers from severe degradation caused by atmospheric turbulence and low-signal conditions. Traditional image reconstruction approaches are computationally expensive and fragile under such conditions. In this work, we propose a novel image-free classification framework using single-pixel imaging (SPI), which directly classifies targets from 1D measurements without reconstructing the image. A learnable sampling matrix is introduced for structured light modulation, and a hybrid CNN-Transformer network (Hybrid-CTNet) is employed for robust feature extraction. To enhance resilience against turbulence and enable efficient deployment, we design a $(N+1)\times L$ hybrid strategy that integrates convolutional and Transformer blocks in every stage. Extensive simulations and optical experiments validate the effectiveness of our approach under various turbulence intensities and sampling rates as low as 1%. Compared with existing image-based and image-free methods, our model achieves superior performance in classification accuracy, computational efficiency, and robustness, which is important for potential low-resource real-time remote sensing applications. Full article

To meet the growing demand for secure and reliable image protection in digital communication, this paper proposes a novel image encryption framework that addresses the challenges of high plaintext sensitivity, resistance to statistical attacks, and key security. The method combines a two-dimensional dynamically coupled map lattice (2D DCML) with elementary cellular automata (ECA) to construct a heterogeneous chaotic system with strong spatiotemporal complexity. To further enhance nonlinearity and diffusion, a multi-source perturbation mechanism and adaptive DNA encoding strategy are introduced. These components work together to obscure the image structure, pixel correlations, and histogram characteristics. By embedding spatial and temporal symmetry into the coupled lattice evolution and perturbation processes, the proposed method ensures a more uniform and balanced transformation of image data. Meanwhile, the method enhances the confusion and diffusion effects by utilizing the principle of symmetric perturbation, thereby improving the overall security of the system. Experimental evaluations on standard images demonstrate that the proposed scheme achieves high encryption quality in terms of histogram uniformity, information entropy, NPCR, UACI, and key sensitivity tests. It also shows strong resistance to chosen plaintext attacks, confirming its robustness for secure image transmission. Full article

The article presents a three-stage methodology for calibrating 3D printing using mSLA technology, aimed at improving dimensional accuracy and print repeatability. The proposed approach is based on procedures that enable the collection and analysis of numerical data, thereby minimizing the influence of the operator’s subjective judgment, which is commonly relied upon in traditional calibration methods. In the first stage, compensation for the uneven illumination of the LCD matrix was performed by establishing a regression model that describes the relationship between UV radiation intensity and pixel brightness. Based on this model, a grayscale correction mask was developed. The second stage focused on determining the optimal exposure time, based on its effect on dimensional accuracy, detail reproduction, and model strength. The optimal exposure time is defined as the duration that provides the highest possible mechanical strength without significant loss of detail due to the light bleed phenomenon (i.e., diffusion of UV radiation beyond the mask edge). In the third stage, scale correction was applied to compensate for shrinkage and geometric distortions, further reducing the impact of light bleed on the dimensional fidelity of printed components. The proposed methodology was validated using an Anycubic Photon M3 Premium printer with Anycubic ABS-Like Resin Pro 2.0. Compensating for light intensity variation reduced the original standard deviation from 0.26 to 0.17 mW/cm², corresponding to a decrease of more than one third. The methodology reduced surface displacement due to shrinkage from 0.044% to 0.003%, and the residual internal dimensional error from 0.159 mm to 0.017 mm (a 72% reduction). Full article

Accurate estimation of the leaf area index (LAI), a key indicator of canopy development and light interception, is essential for improving productivity in greenhouse tomato cultivation. This study presents a non-destructive LAI estimation method using side-view images captured by a vertical scanning system. The system recorded the full vertical profile of tomato plants grown under two deleafing strategies: modifying leaf height (LH) and altering leaf density (LD). Vegetative and leaf areas were extracted using color-based masking and semantic segmentation with the Segment Anything Model (SAM), a general-purpose deep learning tool. Regression models based on leaf or all vegetative pixel counts showed strong correlations with destructively measured LAI, particularly under LH conditions (R² > 0.85; mean absolute percentage error ≈ 16%). Under LD conditions, accuracy was slightly lower due to occlusion and leaf orientation. Compared with prior 3D-based methods, the proposed 2D approach achieved comparable accuracy while maintaining low cost and a labor-efficient design. However, the system has not been tested in real production, and its generalizability across cultivars, environments, and growth stages remains unverified. This proof-of-concept study highlights the potential of side-view imaging for LAI monitoring and calls for further validation and integration of leaf count estimation. Full article

This paper proposes a dual-mode CMOS analog front-end (AFE) circuit for short-wave infrared (SWIR) image sensors, which integrates a hybrid readout circuit (ROIC) and a 12-bit two-step single-slope analog-to-digital converter (TS-SS ADC). The ROIC dynamically switches between buffered-direct-injection (BDI) and direct-injection (DI) modes, thus balancing injection efficiency against power consumption. While the DI structure offers simplicity and low power, it suffers from unstable biasing and reduced injection efficiency under high background currents. Conversely, the BDI structure enhances injection efficiency and bias stability via an input buffer but incurs higher power consumption. To address this trade-off, a dual-mode injection architecture with mode-switching transistors is implemented. Mode selection is executed in-pixel via a low-leakage transmission gate and coordinated by the column timing controller, enabling low-current pixels to operate in low-noise BDI mode, whereas high-current pixels revert to the low-power DI mode. The TS-SS ADC employs a four-terminal comparator and dynamic reference voltage compensation to mitigate charge leakage and offset, which improves signal-to-noise ratio (SNR) and linearity. The prototype occupies 2.1 mm × 2.88 mm in a 0.18 µm CMOS process and serves a 64 × 64 array. The AFE achieves a dynamic range of 75.58 dB, noise of 249.42 μV, and 81.04 mW power consumption. Full article

Medical images demand robust privacy protection, driving research into advanced image encryption (IE) schemes. However, current IE schemes still encounter certain challenges in both security and efficiency. Fractional-order Hopfield neural networks (HNNs) demonstrate unique advantages in IE. The introduction of fractional-order calculus operators enables them to possess more complex dynamical behaviors, creating more random and unpredictable keystreams. To enhance privacy protection, this paper introduces a high-performance medical IE scheme that integrates a novel 4D fractional-order HNN with a differentiated encryption strategy (MIES-FHNN-DE). Specifically, MIES-FHNN-DE leverages this 4D fractional-order HNN alongside a 2D hyperchaotic map to generate keystreams collaboratively. This design not only capitalizes on the 4D fractional-order HNN’s intricate dynamics but also sidesteps the efficiency constraints of recent IE schemes. Moreover, MIES-FHNN-DE boosts encryption efficiency through pixel bit splitting and weighted accumulation, ensuring robust security. Rigorous evaluations confirm that MIES-FHNN-DE delivers cutting-edge security performance. It features a large key space ($2^{383}$), exceptional key sensitivity, extremely low ciphertext pixel correlations (<0.002), excellent ciphertext entropy values (>7.999 bits), uniform ciphertext pixel distributions, outstanding resistance to differential attacks (with average NPCR and UACI values of $99.6096\%$ and $33.4638\%$, respectively), and remarkable robustness against data loss. Most importantly, MIES-FHNN-DE achieves an average encryption rate as high as 102.5623 Mbps. Compared with recent leading counterparts, MIES-FHNN-DE better meets the privacy protection demands for medical images in emerging fields like medical intelligent analysis and medical cloud services. Full article

Image super-resolution (SR) aims to reconstruct high-resolution (HR) images from low-resolution (LR) inputs. It plays a crucial role in applications such as medical imaging, surveillance, and remote sensing. However, due to the ill-posed nature of the task and the inherent limitations of imaging sensors, obtaining accurate HR images remains challenging. While numerous methods have been proposed, the traditional approaches suffer from oversmoothing and limited generalization; CNN-based models lack the ability to capture long-range dependencies; and Transformer-based solutions, although effective in modeling global context, are computationally intensive and prone to texture loss. To address these issues, we propose a hybrid CNN–Transformer architecture that cascades a pixel-wise self-attention non-local means module (PSNLM) and an adaptive dual-path multi-scale fusion block (ADMFB). The PSNLM is inspired by the non-local means (NLM) algorithm. We use weighted patches to estimate the similarity between pixels centered at each patch while limiting the search region and constructing a communication mechanism across ranges. The ADMFB enhances texture reconstruction by adaptively aggregating multi-scale features through dual attention paths. The experimental results demonstrate that our method achieves superior performance on multiple benchmarks. For instance, in challenging ×4 super-resolution, our method outperforms the second-best method by 0.0201 regarding the Structural Similarity Index (SSIM) on the BSD100 dataset. On the texture-rich Urban100 dataset, our method achieves a 26.56 dB Peak Signal-to-Noise Ratio (PSNR) and 0.8133 SSIM. Full article

This study presents the design and development of electromagnetic dipole magnets for use as beam dumps and spectrometers in the MIR and THz free-electron laser (FEL) beamlines at the PBP-CMU Electron Linac Laboratory (PCELL). The magnets were optimized to achieve a 60-degree bending angle for electron beams with energies up to 30 MeV, without requiring water cooling. Using CST EM Studio for 3D magnetic field simulations and ASTRA for particle tracking, the THz dipole (with 414 turns) and MIR dipole (with 600 turns) generated magnetic fields of 0.1739 T and 0.2588 T, respectively, while both operating at currents below 10 A. Performance analysis confirmed effective beam deflection, with the THz dipole showing that it was capable of handling beam energies up to 20 MeV and the MIR dipole could handle up to 30 MeV. The energy measurement at the spectrometer screen position was simulated, taking into account transverse beam size, fringe fields, and space charge effects, using ASTRA. The energy resolution, defined as the ratio of energy uncertainty to the mean energy, was evaluated for selected cases. For beam energies of 16 MeV and 25 MeV, resolutions of 0.2% and 0.5% were achieved with transverse beam sizes of 1 mm and 4 mm, respectively. All evaluated cases maintained energy resolutions below 1%, confirming the spectrometer’s suitability for high-precision beam diagnostics. Furthermore, the relationship between the initial and measured energy spread errors, taking into account a camera resolution of 0.1 mm/pixel, was evaluated. Simulations across various beam energies (10–16 MeV for the THz dipole and 20–25 MeV for the MIR dipole) confirmed that the measurement error in energy spread decreases with smaller RMS transverse beam sizes. This trend was consistent across all tested energies and magnet configurations. To ensure accurate energy spread measurements, a small initial beam size is recommended. Specifically, for beams with a narrow initial energy spread, a transverse beam size below 1 mm is essential. Full article

Autonomous vehicles are a promising solution to traffic congestion, air pollution, accidents, wasted time, and resources. However, remote driver intervention may be necessary in extreme situations to ensure safe roadside parking or complete remote takeover. In these cases, high-quality real-time video streaming is crucial for remote driving. In a preliminary study, we presented a region of interest (ROI) High-Efficiency Video Coding (HEVC) method where the image was segmented into two categories: ROI and background. This involved allocating more bandwidth to the ROI, which yielded an improvement in the visibility of classes essential for driving while transmitting the background at a lower quality. However, migrating the bandwidth to the large ROI portion of the image did not substantially improve the quality of traffic signs and lights. This study proposes a method that categorizes ROIs into three tiers: background, weak ROI, and strong ROI. To evaluate this approach, we utilized a photo-realistic driving scenario database created with the Cognata self-driving car simulation platform. We used semantic segmentation to categorize the compression quality of a Coding Tree Unit (CTU) according to its pixel classes. A background CTU contains only sky, trees, vegetation, or building classes. Essentials for remote driving include classes such as pedestrians, road marks, and cars. Difficult-to-recognize classes, such as traffic signs (especially textual ones) and traffic lights, are categorized as a strong ROI. We applied thresholds to determine whether the number of pixels in a CTU of a particular category was sufficient to classify it as a strong or weak ROI and then allocated bandwidth accordingly. Our results demonstrate that this multi-category ROI compression method significantly enhances the perceptual quality of traffic signs (especially textual ones) and traffic lights by up to 5.5 dB compared to a simpler two-category (background/foreground) partition. This improvement in critical areas is achieved by reducing the fidelity of less critical background elements, while the visual quality of other essential driving-related classes (weak ROI) is at least maintained. Full article

Laser marking on wafers can introduce various defects such as inconsistent mark quality; under- or over-etching, and misalignment. Excessive laser power and inadequate cooling can cause burning or warping. These defects were inspected using machine vision, confocal microscopy, optical and scanning electron microscopy, acoustic/ultrasonic methods, and inline monitoring and coaxial vision. Machine learning has been successfully applied to improve the classification accuracy, and we propose a random forest algorithm with a training database to not only detect the defect but also trace its cause. Four causes have been identified as follows: unstable laser power, a dirty laser head, platform shaking, and voltage fluctuation of the electrical power. The object-matching technique ensures that a visible image can be utilized without a precise location. All inspected images were compared to the standard (qualified) product image pixel-by-pixel, and then the 2D matrix pattern for each type of defect was gathered. There were 10 photos for each type of defect included in the training to build the model with various labels, and the synthetic testing images altered by the defect cause model for laser marking defect inspection had accuracies of 97.0% and 91.6% in sorting the error cause, respectively Full article

LiDAR and camera sensors are widely utilized in autonomous vehicles (AVs) and robotics due to their complementary sensing capabilities—LiDAR provides precise depth information, while cameras capture rich visual context. However, effective multi-sensor fusion remains challenging due to discrepancies in resolution, data format, and viewpoint. In this paper, we propose a robust pattern matching algorithm that leverages singular value decomposition (SVD) and gradient descent (GD) to align geometric features—such as object contours and convex hulls—across LiDAR and camera modalities. Unlike traditional calibration methods that require manual targets, our approach is targetless, extracting matched patterns from projected LiDAR point clouds and 2D image segments. The algorithm computes the optimal transformation matrix between sensors, correcting misalignments in rotation, translation, and scale. Experimental results on a vehicle-mounted sensing platform demonstrate an alignment accuracy improvement of up to 85%, with the final projection error reduced to less than 1 pixel. This pattern-based SVD-GD framework offers a practical solution for maintaining reliable cross-sensor alignment under calibration drift, enabling real-time perception systems to operate robustly without recalibration. This method provides a practical solution for maintaining reliable sensor fusion in autonomous driving applications subject to long-term calibration drift. Full article