Search Results (179)

Although GPUs are widely used in Computer-Generated Holography (CGH), their specific application to concrete problems such as occlusion or speckle filtering through temporal multiplexing is not yet standardized and has not been fully explored. This work aims to optimize the software architecture by taking the GPU architecture into account in a novel way for these particular tasks. We present an optimized algorithm for CGH computation that provides a joint solution to the problems of speckle noise and occlusion. The workflow includes the generation and illumination of a 3D scene, the calculation of the CGH including color, occlusion, and temporal speckle-noise filtering, followed by scene reconstruction through both simulation and experimental methods. The research focuses on implementing a temporal multiplexing technique that simultaneously performs speckle denoising and occlusion culling for point clouds, evaluating two types of occlusion that differ in whether the occlusion effect dominates over the depth effect in a scene stored in a CGH, while leveraging the parallel processing capabilities of GPUs to achieve a more immersive and high-quality visual experience. To this end, the total computational cost associated with generating color and occlusion CGHs is evaluated, quantifying the relative contribution of each factor. The results indicate that, under strict occlusion conditions, temporal multiplexing filtering does not significantly impact the overall computational cost of CGH calculation. Full article

Aiming at the challenges faced by the detection of small-scale ship targets in Synthetic Aperture Radar (SAR) images, this paper proposes a novel deep learning network named FAWT-Net based on attention-matrix despeckling and Haar wavelet reconstruction. This network collaboratively optimizes the detection performance through three core modules. First, during the feature transfer stage from backbone to the neck, a filtering module based on attention matrix is designed, which can suppress the speckle noise. Then, during feature upsampling stage, a wavelet transform feature upsampling method for reconstructing image details is designed to enhance the distinguishability of target boundaries and textures. At the same time, the network also combines sub-image feature stitching downsampling to avoid losing key details in small targets, and adopts a scale-sensitive detection head. By adaptively adjusting the shape constraints of prediction boxes, it effectively solves the regression deviation problem of ship targets with inconsistent aspect ratios. Verified by experiments on SSDD and LS-SSDD, the proposed method improves AP50 by 1.3% and APS by 0.8% on the SSDD. Meanwhile, it is verified that the proposed method has higher precision and recall rates on the LS-SSDD, and the recall rate has been increased by 2.2%. Full article

A common drawback of denoising methods of images is that all pixels are filtered regardless of the amount of noise affecting them individually. Since the essence of denoising is lowpass filtering, subjecting clean pixels to denoising results in blurring. In this paper, a filtering framework is introduced where a fitness function is incorporated in a Kalman filter (KF) to assess the suitability of accepting the value recommended by KF or retaining the existing value of a pixel. Furthermore, a limit on the number of iterations is imposed to avoid over filtering that leads to shrinkage of pixel value ranges of the channels and loss of spectral signatures. In post processing, the means of the filtered channels are shifted to their original values prior to filtering, to spread the pixel value ranges and regain important spectral signatures. The experiments involve the implementation of KF, extended Kalman filter (EKF), Kalman smoother (KS), extended Kalman smoother (EKS) and moving average filter (MAF) in filtering noisy channels of oil palm hyperspectral data under the same framework. Their performances are compared in terms of execution time, SNR gain, NIQE and SSIM metrics. In the second set of experiments, the performance of the improved KF with a fitness function and mean restoration is compared to those of KF and MAF. The results show that the improved KF outperforms the other two filters in the spectral signature characteristics and pixel value ranges of the denoised channels. Full article

Ship detection in synthetic aperture radar (SAR) remote sensing imagery is of great significance in military and civilian applications. However, two factors limit detection performance: (1) a high prevalence of small-scale ship targets with limited information content and (2) interference affecting ship detection from speckle noise and land–sea clutter. To address these challenges, we propose a novel end-to-end (E2E) transformer-based SAR ship detection framework, called Flow-Aligned Nested Transformer for SAR Small Ship Detection (FANT-Det). Specifically, in the feature extraction stage, we introduce a Nested Swin Transformer Block (NSTB). The NSTB employs a two-level local self-attention mechanism to enhance fine-grained target representation, thereby enriching features of small ships. For multi-scale feature fusion, we design a Flow-Aligned Depthwise Efficient Channel Attention Network (FADEN). FADEN achieves precise alignment of features across different resolutions via semantic flow and filters background clutter through lightweight channel attention, further enhancing small-target feature quality. Moreover, we propose an Adaptive Multi-scale Contrastive Denoising (AM-CDN) training paradigm. AM-CDN constructs adaptive perturbation thresholds jointly determined by a target scale factor and a clutter factor, generating contrastive denoising samples that better match the physical characteristics of SAR ships. Finally, extensive experiments on three widely used open SAR ship datasets demonstrate that the proposed method achieves superior detection performance, outperforming current state-of-the-art (SOTA) benchmarks. Full article

Background: Speckled lentiginous nevus (SLN), also known as nevus spilus (NS), is a variant of congenital melanocytic nevus. Although historically considered to have low malignant potential, recent studies have reported melanoma arising within SLN. This study presents a systematic review of multiple melanomas occurring in association with SLN and includes a representative clinical case. Methods: We conducted a systematic review in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. A comprehensive search of PubMed, Scopus, Web of Science, and Google Scholar was performed from 1957 to 2025 using the terms “melanoma” and “nevus spilus” or “speckled lentiginous nevus.” Filters were applied for original reports, case series, and case reports. Studies were selected based on predefined criteria, with data independently extracted by two reviewers. A case of a 66-year-old male with three melanomas (two within and one outside SLN) over nine years is also presented. Because the evidence base consisted almost exclusively of case reports and small series, meta-analysis and formal risk-of-bias assessment were not feasible; findings were therefore synthesized qualitatively. Results: We first describe an illustrative case of a 66-year-old male who developed three melanomas (two within and one outside SLN) over a nine-year period, underscoring the challenges of diagnosis and long-term monitoring. In the systematic review, we identified 41 eligible publications describing 51 patients, and in our illustrative case, we identified a total of 52 with melanoma on SLN; 9/52 (17.3%) developed multiple melanomas (24 total), and in our illustrative case, we identified a total of 52. Most were male (seven of nine), with the first melanoma diagnosed at a mean age of 52.4 years. The majority (21/24) occurred within SLNs ≥5 cm and were of the superficial spreading type (16/17 where specified). Of 24 tumors, 19 (79.2%) were synchronous, and among the 16 invasive melanomas, the mean Breslow thickness was 1.17 mm (median 0.95 mm, IQR 0.56–1.40 mm). Conclusions: Large or segmental SLNs may carry a clinically relevant risk for developing multiple melanomas. Regular full-body skin examinations and dermoscopic monitoring are recommended for early detection and management. As the synthesis is based mainly on case reports and small series, these conclusions are necessarily descriptive and exploratory, providing a qualitative mapping of the available evidence rather than definitive risk estimates. Full article

Digital holography enables the reconstruction of both 2D and 3D object information from interference patterns captured by digital cameras. A major challenge in this field is speckle noise, which significantly degrades the quality of the reconstructed images. We propose a novel speckle noise reduction method based on 3D adaptive filtering. Our technique processes a stack of holograms, each with an uncorrelated speckle pattern, using an adapted 3D Frost filter. Unlike conventional filtering techniques, our approach exploits statistical adaptivity to enhance noise suppression while preserving fine image details in the reconstructed holograms. Both numerical simulations and optical experiments confirm that our 3D filtering technique significantly enhances reconstruction quality. Specifically, it reduces the normalized standard deviation by up to 40% and improves the structural similarity index by up to 60% compared to classical 2D, 3D median, BM3D, and BM4D filters. Optical experiments validate the method’s effectiveness in practical digital holography scenarios by local and global image quality estimation metrics. These results highlight adaptive 3D filtering as a promising approach for mitigating speckle noise while maintaining structural integrity in digital holography reconstructions. Full article

Neural networks have shown significant promise in the field of image processing, particularly for tasks such as denoising and restoration, due to their capacity to model complex nonlinear relationships between inputs and outputs. In this study, we explored the application of a complex-valued neural network—a multilayer neural network with multi-valued neurons (MLMVN)—for filtering two types of noise in digital images: additive Gaussian noise and multiplicative speckle noise. The proposed approach involves processing images as a set of overlapping patches in the frequency domain using MLMVN. Training was performed using a batch learning algorithm, which proved to be more efficient for big learning sets: it results in fewer learning epochs and a better generalization capability. Experimental results demonstrated that MLMVN achieves noise filtering quality comparable to well-established methods, such as the BM3D, Lee, and Frost filters. These findings suggest that MLMVN offers a viable framework for image denoising, particularly in scenarios where frequency domain processing is advantageous. Also, complex-valued logistic and hyperbolic tangent activation functions were used for multi-valued neurons for the first time and have shown their efficiency. Full article

Crop height is a key indicator of plant development and growth dynamics, offering valuable insights for temporal crop monitoring. However, modeling its variation across phenological stages remains challenging due to canopy structural changes. This study aimed to predict wheat height throughout the growth cycle by integrating radar remote sensing data with a phenology-informed clustering approach. The research was conducted in three wheat fields in Umbria, Italy, from 30 January to 10 June 2024, using in-field height measurements, phenological observations, and Sentinel-1 acquisitions. Backscatter variables (VH, VV, and CR) were processed using two speckle filters (Lee 7 × 7 and Refined Lee), alongside additional radar-derived parameters (entropy, anisotropy, alpha, and RVI). Fuzzy C-means clustering enabled the classification of observations into two phenological groups, supporting the development of stage-specific linear regression models. Results demonstrated high accuracy during early growth stages (tillering to stem elongation), with R² values of 0.76 (RMSE = 6.88) for Lee 7 × 7 and 0.79 (RMSE = 6.35) for Refined Lee. In later stages (booting to maturity), model performance declined, with Lee 7 × 7 outperforming Refined Lee (R² = 0.51 vs. 0.33). These findings underscore the potential of phenology-based modeling approaches to enhance crop height estimation and improve radar-driven crop monitoring. Full article

The increasing demand for accurate and accessible medical imaging has driven efforts to develop technologies that overcome limitations associated with conventional imaging techniques, such as MRI and CT scans. This study presents the design and implementation of an electronic interface for acquiring signals from a piezoelectric ultrasound sensor with the aim of improving image reconstruction quality by addressing electromagnetic interference and speckle noise, two major factors that degrade image fidelity. The proposed interface is installed between the ultrasound transducer and acquisition system, allowing real-time signal capture without altering the medical equipment’s operation. Using a printed circuit board with 110-pin connectors, signals from individual piezoelectric elements were analyzed using an oscilloscope. Results show that noise amplitudes occasionally exceed those of the acoustic echoes, potentially compromising image quality. By enabling direct observation of these signals, the interface facilitates the future development of analog filtering solutions to mitigate high-frequency noise before digital processing. This approach reduces reliance on computationally expensive digital filtering, offering a low-cost, real-time alternative. The findings underscore the potential of the interface to enhance diagnostic accuracy and support further innovation in medical imaging technologies. Full article

In high-resolution microscopic imaging, using shorter-wavelength ultraviolet (UV) lasers as illumination sources is a common approach. However, the high spatial coherence of such lasers, combined with the surface roughness of the sample, often introduces disturbances in the received optical field, resulting in strong speckle noise. This paper presents a novel speckle noise suppression method specifically designed for coherent laser-based microscopic imaging. The proposed approach integrates statistical physical modeling and image gradient discrepancy into the training of a Cycle Generative Adversarial Network (CycleGAN), capturing the perturbation mechanism of speckle noise in the optical field. By incorporating these physical constraints, the method effectively enhances the model’s ability to suppress speckle noise without requiring annotated clean data. Experimental results under high-resolution laser microscopy settings demonstrate that the introduced constraints successfully guide network training and significantly outperform traditional filtering methods and unsupervised CNNs in both denoising performance and training efficiency. While this work focuses on microscopic imaging, the underlying framework offers potential extensibility to other laser-based imaging modalities with coherent noise characteristics. Full article

Accurate and timely flood mapping is critical for informing emergency response and risk mitigation during extreme weather events. This study presents a synthetic aperture radar (SAR)-based approach for rapid flood extent mapping using Sentinel-1 imagery, demonstrated for Tropical Storm Imelda (17–21 September 2019) in southeastern Texas. Dual-polarization Sentinel-1 SAR data (VH and VV) were processed by computing the VH/VV backscatter ratio, and the resulting ratio image was classified using a supervised Random Forest classifier to delineate water and land. All Sentinel-1 images underwent radiometric calibration, speckle noise filtering, and terrain correction to ensure precision in flood delineation. The Random Forest classifier achieved an overall flood mapping accuracy exceeding 94%, with Cohen’s kappa coefficients of approximately 0.75–0.80, demonstrating the approach’s reliability in distinguishing transient floodwaters from permanent water bodies. The spatial distribution of flooding was strongly influenced by topography and land cover. Analysis of Shuttle Radar Topography Mission (SRTM) digital elevation data revealed that low-lying, flat terrain was most vulnerable to inundation; correspondingly, the land cover types most affected were hay/pasture, cultivated land, and emergent wetlands. Additionally, urban areas with low-intensity development experienced extensive flooding, attributed to impervious surfaces exacerbating runoff. A strong, statistically significant correlation (R² = 0.87, p < 0.01) was observed between precipitation and flood extent, indicating that heavier rainfall led to greater inundation; accordingly, the areas with the highest rainfall totals (e.g., Jefferson and Chambers counties) experienced the most extensive flooding, as confirmed by SAR-based change detection. The proposed approach eliminates the need for manual threshold selection, thereby reducing misclassification errors due to speckle noise and land cover heterogeneity. Harnessing globally available Sentinel-1 data with near-real-time processing and a robust classifier, this approach provides a scalable solution for rapid flood monitoring. These findings underscore the potential of SAR-based flood mapping under adverse weather conditions, thereby contributing to improved disaster preparedness and resilience in flood-prone regions. Full article

The grayscale images of passenger aircraft targets obtained via Synthetic Aperture Radar (SAR) have problems such as complex airport backgrounds, significant speckle noise, and variable scales of targets. Most of the existing deep learning-based target recognition algorithms for SAR images are transferred from optical images, and it is difficult for them to extract the multi-dimensional features of targets comprehensively. To overcome these challenges, we proposed three enhanced methods for interpreting aircraft targets based on YOLOv8. First, we employed the Shi–Tomasi corner detection algorithm and the Enhanced Lee filtering algorithm to convert grayscale images into RGB images, thereby improving detection accuracy and efficiency. Second, we augmented the YOLOv8 model with an additional detection branch, which includes a detection head featuring the Coordinate Attention (CA) mechanism. This enhancement boosts the model’s capability to detect small and multi-scale aircraft targets. Third, we integrated the Swin Transformer mechanism into the YOLOv8 backbone, forming the C2f-SWTran module that better captures long-range dependencies in the feature map. We applied these improvements to two datasets: the ISPRS-SAR-aircraft dataset and the SAR-Aircraft-1.0 dataset. The experimental results demonstrated that our methods increased the mean Average Precision (mAP_50~95) by 2.4% and 3.4% over the YOLOv8 baseline, showing competitive advantages over other deep learning-based object detection algorithms. Full article

Side-Scan Sonar (SSS) is widely used in underwater rescue operations and the detection of seabed targets, such as shipwrecks, drowning victims, and aircraft. However, the quality of sonar images is often degraded by noise sources like reverberation and speckle noise, which complicate the extraction of effective features. Additionally, challenges such as limited sample sizes and class imbalances are prevalent in side-scan sonar image data. These issues directly impact the accuracy of deep learning-based target classification models for SSS images. To address these challenges, we propose a side-scan sonar image classification model based on joint image deblurring–denoising and a pre-trained feature fusion attention network. Firstly, by employing transform domain filtering in conjunction with upsampling and downsampling techniques, the joint image deblurring–denoising approach effectively reduces image noise while preserving and enhancing edge and texture features. Secondly, a feature fusion attention network based on transfer learning is employed for image classification. Through the transfer learning approach, a feature extractor based on depthwise separable convolutions and densely connected networks is trained to effectively address the challenge of limited training samples. Subsequently, a dual-path feature fusion strategy is utilized to leverage the complementary strengths of different feature extraction networks. Furthermore, by incorporating channel attention and spatial attention mechanisms, key feature channels and regions are adaptively emphasized, thereby enhancing the accuracy and robustness of image classification. Finally, the Gradient-weighted Class Activation Mapping (Grad-CAM) technique is integrated into the proposed model to ensure interpretability and transparency. Experimental results show that our model achieves a classification accuracy of 96.80% on a side-scan sonar image dataset, confirming the effectiveness of this method for SSS image classification. Full article

Background/Objectives: Ultrasound (US) imaging plays a crucial role in the early detection and treatment of hepatocellular carcinoma (HCC). However, challenges such as speckle noise, low contrast, and diverse lesion morphology hinder its diagnostic accuracy. Methods: To address these issues, we propose CSM-FusionNet, a novel framework that integrates clustering, SoftMax-weighted Box Fusion (SM-WBF), and padding. Using raw US images from a leading hospital, Samsung Medical Center (SMC), we applied intensity adjustment, adaptive histogram equalization, low-pass, and high-pass filters to reduce noise and enhance resolution. Data augmentation generated ten images per one raw US image, allowing the training of 10 YOLOv8 networks. The mAP@0.5 of each network was used as SoftMax-derived weights in SM-WBF. Threshold-lowered bounding boxes were clustered using Density-Based Spatial Clustering of Applications with Noise (DBSCAN), and outliers were managed within clusters. SM-WBF reduced redundant boxes, and padding enriched features, improving classification accuracy. Results: The accuracy improved from 82.48% to 97.58% with sensitivity reaching 100%. The framework increased lesion detection accuracy from 56.11% to 95.56% after clustering and SM-WBF. Conclusions: CSM-FusionNet demonstrates the potential to significantly improve diagnostic reliability in US-based lesion detection, aiding precise clinical decision-making. Full article

The speckle is a granular undesired pattern present in Synthetic-Aperture Radar (SAR) imagery. Despeckling has been an active field of research during the last decades, with approaches from local filters to non-local filters that calculate the new value of a pixel according to characteristics of other pixels that are not close, the more advanced paradigms based on deep learning, and the newer based on generative artificial intelligence. For the latter, it is necessary to have a large enough labeled dataset for training and validation. In this study, we propose using a dataset designed entirely from actual SAR imagery, calculated by multitemporal fusion operations to generate a ground truth reference, which will yield the models to be trained with the actual speckle patterns in the noisy images. Then, a comparative analysis of the impacts of including the generative capacity in the models is performed through visual and quantitative assessment. From the findings, it is concluded that the use of generative artificial intelligence with actual speckle exhibits notable efficiency compared to other approaches, which makes this a promising path for research in the context of SAR imagery. Full article