Search Results (744)

Dark current represents a fundamental limiting factor in photodiode performance, establishing the noise floor and constraining detectivity in low-light applications. This comprehensive literature review examines publications covering the physical mechanisms underlying dark current generation and diverse techniques employed for its reduction. Covered mechanisms include diffusion current, Shockley–Read–Hall (SRH) generation–recombination, trap-assisted tunneling, band-to-band tunneling, and surface leakage, each examined with respect to its physical origin and characteristic signatures. Reduction strategies are categorized into thermal management approaches, surface passivation techniques including atomic-layer-deposited aluminum oxide (ALD Al₂O₃), guard ring architectures (attached, floating, and combined configurations), gettering and defect engineering methods, doping profile optimization, bias voltage management, and advanced device architectures such as pinned photodiodes and black silicon structures. A classification table organizes all the reviewed literature by material system, reduction technique, and key findings. Special emphasis is placed on silicon, germanium, III–V compounds, and emerging material photodiodes relevant to near-infrared detection, CMOS imaging, single-photon avalanche diodes (SPADs), and Time-of-Flight (ToF) applications. Full article

Early and accurate detection of skin cancer is critical for reducing mortality rates, particularly for malignant melanoma. Automated analysis of dermoscopic images has gained significant attention due to its potential to support clinical diagnosis and overcome the limitations of manual inspection. Motivated by challenges such as image noise, low contrast, lesion variability, and redundant feature representation, this study proposes an optimal deep hybrid framework for skin lesion detection and classification. The objective of this work is to design a robust and efficient system that integrates advanced preprocessing, precise segmentation, optimal feature selection, and accurate classification. Initially, contrast enhancement using Contrast Limited Adaptive Histogram Equalization (CLAHE) and noise reduction using Wiener filtering are applied to improve image quality. Lesion regions are then segmented using a Selective Kernel U-Net (SK-UNet), which adaptively captures multi-scale spatial information. Subsequently, discriminative color, texture, and shape features are extracted and optimized using the Fossa Optimization Algorithm (FOA) to eliminate redundancy. A hybrid one-dimensional Convolutional Neural Network–Gated Recurrent Unit (1D-CNN–GRU) classifier is employed for final classification, learning both spatial and sequential feature patterns. Experimental evaluation on the ISIC and DermMNIST datasets demonstrates that the proposed framework achieves classification accuracies of 97.6% and 95.6%, respectively, outperforming several existing methods. The results confirm that the proposed hybrid framework provides reliable, accurate, and scalable skin cancer diagnosis, highlighting its potential for assisting clinical decision-making and early detection. Full article

Standard Retinex-based algorithms typically rely on gradient constraints to decompose an image, assuming that illumination is spatially smooth while reflectance contains sharp details. However, strictly gradient-based priors frequently produce halo artifacts or over-smoothing because they are unable to differentiate between intrinsic structural edges and high-frequency noise. In this paper, we propose a novel Similarity Image-Guided Retinex (SIMRET) model that fundamentally diverges from traditional derivative-based regularization. We present a color-based pixel-level similarity analysis to build a global guidance matrix rather than merely depending on local gradients. This Similarity Image functions as a reliable weight map during the decomposition process by mathematically encoding the chromatic relationships and spatial coherence between pixels. The model strictly maintains consistency across structural boundaries to avoid halo effects while adaptively enforcing smoothness in homogeneous regions to suppress noise by incorporating this similarity guidance into the optimization objective. We solve the proposed SIMRET model using an alternating optimization framework, where the similarity constraints effectively regularize the ill-posed decomposition problem. Extensive tests on various low-light datasets show that the suggested model successfully overcomes the trade-off between noise reduction and detail preservation, achieving better visual naturalness and signal fidelity than state-of-the-art techniques. Full article

Microwave imaging is emerging as an alternative to conventional medical diagnostic techniques. Traditional analytical and numerical methods fail to adequately address these fundamental challenges: they often rely on strict linear approximations or simplified physical models, leading to low reconstruction accuracy, poor robustness, and limited generalization ability in complex clinical scenarios. As a result, they cannot meet the high-precision requirements of practical stroke microwave imaging. To further improve the accuracy of microwave imaging algorithms in recognizing stroke regions and solving the backscattering problem, this study employs a combination of methods with deep learning. It presents the Multi-Scale Attention Network (MSA-Net) for microwave imaging. The network is based on the EGE-UNet network structure with improved multi-axis Hadamard attention, incorporating null-space pyramid pooling and introducing a deep supervisory mechanism to improve the network performance further. To combine microwave imaging with deep learning, firstly, a large amount of microwave data need to be simulated with HFSS, in which the simulation model is a human brain stroke model constructed by an HFSS simulation system. Secondly, the microwave data obtained from the simulation are converted into a tensor format. Then, the tensor data are input into the MSA-Net neural network, which generates a binary mask image that can be used to detect the size and location of the stroke. This study also prompts the model to converge faster by sparsifying the microwave data to improve training efficiency. The method has been tested using simulation data, and based on the comparison experiments with other networks, MSA-Net is more accurate in detecting the location and the bleed size. The experimental results show that the proposed method is superior for stroke imaging. The experimental results show that the proposed model achieves a 1.08 improvement in peak signal-to-noise ratio and a 0.017 reduction in learned perceptual image block similarity, fully validating the effectiveness of the structural optimization strategy proposed in this paper. Full article

This study proposes an integrated software model for automatic document classification and metadata generation based on the Dublin Core standard to address the issue of rapid and consistent management of archival documents in a digital environment. This approach combines the stages of receiving incoming documents, converting them to text using optical character recognition (OCR), image preprocessing (binarization, deskew, noise reduction), and text cleaning and vectorization (TF–IDF) into a single pipeline. In the document classification stage, the Bidirectional Encoder Representations from Transformers (BERT) model with a context-sensitive transformer architecture is used, along with classical machine learning models (Logistic Regression, Naive Bayes, Support Vector Machine) and an ensemble approach (LightGBM), to increase the accuracy by modeling the document content at a deep semantic level. Experiments were conducted on the RVL-CDIP dataset, and the OCR efficiency was evaluated using the Character Error Rate (CER) indicator, and the classification results were evaluated using the accuracy, precision, recall and F1-score metrics. The results confirmed the high stability and generalization ability of the BERT (accuracy, 95.1%; F1, 95.0%) and LightGBM (accuracy, 93.2%; F1, 93.2%) models. In the final stage, OCR, NER, and classification outputs are automatically organized into Dublin Core metadata elements (Title, Creator, Date, Description, Subject, Type, Format, Language) and exported in JSON/XML formats. This automation significantly reduces manual cataloging effort and improves indexing and retrieval efficiency in digital archival systems. Full article

Against the global carbon neutrality target, predictive maintenance (PdM) of automotive engines represents a core technical strategy to advance the sustainable development of the automotive industry. Conventional single-modal diagnostic approaches for engine abnormal sound defects suffer from low accuracy and weak anti-interference capability. Existing multi-modal fusion methods fail to deeply mine the physical coupling between cross-modal features and often entail excessive model complexity, hindering deployment on resource-constrained on-board edge devices. To resolve these limitations, this study proposes a Physical Prior-Embedded Cross-Modal Attention (PPE-CMA) mechanism for lightweight multi-modal fusion diagnosis of engine abnormal sound defects. First, wavelet packet decomposition (WPD) and mel-frequency cepstral coefficients (MFCC) are integrated to extract time-frequency features from engine audio signals, while a channel-pruned ResNet18 is employed to extract spatial features from engine thermal imaging and vibration visualization images. Second, the PPE-CMA module is designed to adaptively assign attention weights to audio and image features by exploiting the physical coupling between engine fault acoustic and visual characteristics, enabling efficient cross-modal feature fusion with redundant information suppression. A rigorous theoretical derivation is provided to link cosine similarity with the physical correlation of engine fault acoustic-visual features, justifying the attention weight constraint (β = 1 − α) from the perspective of fault feature physical coupling. Third, an improved lightweight XGBoost classifier is constructed for fault classification, and a hybrid data augmentation strategy customized for engine multi-modal data is proposed to address the small-sample challenge in industrial applications. Ablation experiments on ResNet18 pruning ratios verify the optimal trade-off between diagnostic performance and computational efficiency, while feature distribution analysis validates the authenticity and effectiveness of the hybrid augmentation strategy. Experimental results on a self-constructed multi-modal dataset show that the proposed method achieves 98.7% diagnostic accuracy and a 98.2% F1-score, retaining 96.5% accuracy under 90 dB high-level environmental noise, with an end-to-end inference speed of 0.8 ms per sample (including preprocessing, feature extraction, and classification). Cross-engine and cross-domain validation on a 2.0T diesel engine small-sample dataset and the open-source SEMFault-2024 dataset yield average accuracies of 94.8% and 95.2%, respectively, demonstrating strong generalization. This method effectively enhances the accuracy and robustness of engine abnormal sound defect diagnosis, offering a lightweight technical solution for on-board real-time fault diagnosis and in-plant online quality inspection. By reducing engine fault-induced energy loss and spare parts waste, it further promotes energy conservation and emission reduction in the automotive industry. Quantified experimental data on fuel efficiency improvement and carbon emission reduction are provided to substantiate the ecological benefits of the proposed framework. Full article

Video AI recognition is crucial for coal mine safety, but complex environments often yield low-quality images, hindering intelligent monitoring. Existing enhancement methods typically focus on image quality alone, lacking adaptability to specific tasks. Therefore, we propose Mine-DCE-YDT: a task-aware low-light image enhancement model that jointly optimizes enhancement with downstream object detection, ensuring enhanced images are both visually clearer and more conducive to accurate detection. Firstly, an improved Zero-DCE algorithm (Mine-DCE) is presented by introducing a Brightness-aware Mask Coordinate Attention (BMCA) module to improve illumination balance in the Value channel of the HSV image and a Multi-scale Detail Enhancement (MDE) module to reinforce textures and suppress noise. Then, Mine-DCE is co-modeled with YOLOv11n by training end-to-end via a joint loss fusing detection and enhancement quality losses to form Mine-DCE-YDT, which can enhance specific details containing image detection targets. Experimental results show that compared with Zero-DCE, Mine-DCE-YDT achieves reductions of 9.5% in NIQE and 35.5% in BRISQUE on the custom-constructed MineDataset and exhibits great enhancement performance on the public dataset LOL-V1. For the miner detection task in MineDataset, the integration of Mine-DCE-YDT with YOLOv11n achieves increases of 2.8% and 8.3% in mAP@0.5 and mAP@0.5:0.95, demonstrating its effectiveness in enhancing task-critical image features. Full article

High-quality Fourier Transform Infrared (FTIR) imaging usually needs extensive signal averaging to reduce noise and drift, which severely limits clinical speed. Deep learning can accelerate imaging by reconstructing spectra from rapid, single-scan inputs. However, separating noise and baseline drift simultaneously without ground truth is an ill-posed inverse problem. Standard black-box architectures often rely on statistical approximations that introduce spectral hallucinations or fail to generalize to unstable atmospheric conditions. To solve these issues, we propose a physics-informed cascade Unet that separates denoising and baseline correction tasks using a new, deterministic Physics Bridge. This architecture forces the network to separate random noise from chemical signals using an embedded SNIP layer to enforce spectroscopic constraints instead of learning statistical approximations. We benchmarked this approach against a standard single Unet and a traditional Savitzky–Golay smoothing followed by SNIP baseline correction workflow. We used a dataset of human hypopharyngeal carcinoma cells (FaDu). The cascade model outperformed all other methods, achieving a 51.3% reduction in RMSE compared to raw single-scan inputs, surpassing both the single Unet (40.2%) and the traditional workflow (33.7%). Peak-aware metrics show that the cascade architecture eliminates spectral hallucinations found in standard deep learning. It also preserves peak intensity with much higher fidelity than traditional smoothing. These results show that the cascade Unet is a robust solution for diagnostic-grade FTIR imaging. It enables imaging speeds 32 times faster than current methods. Full article

Background/Objectives: Brain tumors are among the most severe neurological disorders, and their variability in size, morphology, and anatomical location complicates early and accurate diagnosis. Although magnetic resonance imaging (MRI) is the most reliable non-invasive modality for tumor detection, manual interpretation remains time-consuming, subjective, and susceptible to human error. This study aims to develop an optimization-driven hybrid machine learning framework for accurate and computationally efficient automatic brain tumor classification. Methods: The dataset includes 834 MRI images (583-training, 123-validation, 128-independent test). Because YOLOv11 detects tumor and non-tumor regions separately, the sample size doubled during region-based analysis, and all subsequent stages were conducted at the regions of interest (ROI) level. On the independent test set, YOLOv11 achieved 98.87% mAP@50, 98.54% precision, and 98.21% recall. The proposed framework combines automated tumor localization with image standardization using Gaussian noise reduction and bilinear interpolation. From the processed MR images, 39 entropy-based features were extracted. To enhance diagnostic performance and eliminate redundant information, the superb fairy-wren optimization algorithm (SFOA) was applied for feature selection and compared with particle swarm optimization (PSO), Harris hawk optimization (HHO), and puma optimization (PO). Final classification was primarily performed using k-nearest neighbors (kNN), while support vector machines (SVM) were used for comparative evaluation. Results: SFOA reduced the feature dimensionality from 39 to 5 features while achieving 99.20% classification accuracy on the independent test set. In comparison, PSO selected 10 features, HHO selected 6 features and PO selected 10 features, all achieving 98.45% accuracy. The best performance obtained with SVM was 98.45% accuracy (HHO-SVM), which remained lower than the 99.20% achieved by the proposed SFOA-kNN model. Conclusions: The results indicate that combining entropy-based feature extraction with SFOA-driven feature selection and kNN classification significantly enhances diagnostic accuracy while reducing computational complexity, highlighting the strong potential of the proposed framework for integration into computer-aided diagnosis systems to support clinical decision-making. Full article

Background/Objectives: The objective of this study was to ascertain whether photon-counting CT angiography (PCD-CTA) can optimize image quality for the visualization of perforating arteries for planning fibular transplant procedures in comparison to energy-integrating CT angiography (EID-CTA). Methods: In this retrospective single-center study, all patients who underwent preoperative CT of the peripheral runoff for planning between October 2021 and July 2023 were consecutively included. PCD-CTA was performed in standard resolution mode as 55 keV images with 90 mL of iodine-containing contrast agent or alternatively, an EID-CTA as a low-kV scan with 110 mL of contrast agent. The raw data were reformatted using comparable soft vascular and sharp regular convolution kernels, slice thickness/increment, and field of view. Contrast-to-noise ratio was calculated for objective image quality. Subjective evaluation was based on a rating by three radiologists using a five-point Likert scale (criteria: overall image quality, luminal attenuation, vessel sharpness, and perforator depiction). Results: Of the 26 patients who were screened, 9 could be included in each group, while 8 were excluded due to incomplete reconstructions. The reduction in contrast agent dose resulted in a non-significant decrease in luminal attenuation on PCD-CTA (452.5 ± 53.6 HU vs. 465.5 ± 99.6 HU; p = 0.375). The image noise was considerably lower for PCD-CTA (21.1 ± 1.0 HU vs. 32.9 ± 1.6 HU; p < 0.001). This resulted in a significantly higher contrast-to-noise ratio (CNR) for sharp kernel reconstructions (22.4 ± 3.5 vs. 14.5 ± 3.8; p < 0.001). No significant differences were observed for the soft vascular kernel. Subjective evaluation revealed a significant enhancement in overall image quality, vascular sharpness, and perforator depiction for PCD-CTA with sharp reconstructions. In contrast, soft kernel reconstructions and luminal attenuation demonstrated no substantial difference. Interrater agreement was good to excellent. Conclusions: PCD-CTA with sharp kernel reformatting has been demonstrated to yield superior image quality and perforator delineation of the fibular artery in comparison to standard EID-CTA. Full article

Electrophoretic Displays (EPDs) are widely adopted in e-readers and portable devices due to their ultra-low power consumption and eye-friendly reflective characteristics. However, inherent hardware limitations, such as low resolution, slow response speed, and display degradation, frequently result in blurred strokes and degraded text readability. While traditional driving waveform optimizations can mitigate these issues, they are device-dependent and require extensive manual calibration. To address these challenges, this paper proposes an Improved Whale Optimization Algorithm-based Multi-scale Fusion Attention-enhanced SwinIR (IWOA-MFA-SwinIR) model for super-resolution and recognition of text images on EPDs. Structurally, the model incorporates a multi-scale fused attention (MFA) module that synergistically integrates channel, spatial, and gated attention mechanisms to precisely capture high-frequency text details while suppressing background noise within the SwinIR architecture. Furthermore, to enhance model robustness and eliminate manual tuning, an Improved Whale Optimization Algorithm (IWOA) is employed to adaptively optimize critical hyperparameters, including embedding dimension (d), attention head count (h), learning rate ($lr$), and dimensionality reduction coefficient (r). Experiments conducted on the TextZoom and EPD datasets demonstrate that the proposed model achieves state-of-the-art performance. In the ablation study, it attains a Peak Signal-to-Noise Ratio (PSNR) of 24.406, a Structural Similarity Index (SSIM) of 0.8837, and a Character Recognition Accuracy (CRA) of 89.81%. In the comparative evaluation, the proposed model consistently outperforms the second-best comparison model across three difficulty levels, yielding approximately a 1% improvement in PSNR, a 0.8% improvement in SSIM, and an 8% improvement in CRA. This confirms the proposed model’s superiority over mainstream comparative models in restoring text fidelity and improving recognition rates. Full article

Precise object tracking of space objects is an image acquisition method that uses the mount of the telescope to orient the instrument in real time towards the target to be tracked, compensating for the target’s motion. Using this method, the object of interest will appear as a circular or point-like shape in the acquired image, while the background stars will appear as streaks. Using precise object tracking, the light from a faint object accumulates in the same region of the image, increasing the chance of observation, but longer exposures also increase the length of the background star streaks and makes the astrometric calibration difficult. This paper presents a method for the automatic processing of image sequences acquired in precise object tracking mode. Our proposed method includes a filtering mechanism that will ensure local maxima in the center of star streaks in order to allow for a publicly available astrometric calibration software to work even if the stars are not point-like, a weighted stacking mechanism to increase the signal-to-noise ratio for faint targets while excluding the stars, an automatic object detection and astrometric reduction mechanism and a constraint-based filtering of outliers for the final generation of the tracklet. The method was tested on multiple observation sessions for surveying the CLUSTER II highly eccentric orbit satellites, including the CLUSTER II FM5 satellite (Rumba) on its final passes before reentry, and the accuracy of the measurements was estimated based on ground truth from ESA’s reentry team. The method was also tested on lower orbit objects and found to be accurate for objects with ranges of more than 1300 km from the observer. Full article

Finger vein recognition has emerged as a highly robust and intrinsically stable biometric technology, demonstrating great potential in identity authentication and intelligent security applications. However, conventional methods still suffer from constraints in feature representation and computational efficiency, particularly under challenging conditions such as illumination variation, pose deviation, and noise interference. To address these challenges, this study presents an efficient finger vein recognition approach based on a lightweight convolutional neural network (LCNN) architecture. The proposed framework integrates a multi-stage image preprocessing pipeline for automatic vein region detection, advanced denoising, and refined texture enhancement, which is subsequently followed by compact feature modeling within a lightweight deep network. Extensive experiments on the public Shandong University Machine Learning and Applications-Homologous Multi-Modal Traits (SDUMLA-HMT) dataset and a self-acquired Laboratory Finger-Vein (Lab-Vein) dataset validate the superiority of the proposed method, achieving recognition accuracies of 97.1% and 98.3%, respectively, surpassing existing benchmark models. Moreover, the model demonstrates notable reductions in parameter complexity and computational cost, achieving an average inference time of only 12.6 ms, which confirms its strong real-time capability and suitability for embedded deployment. Overall, the proposed approach attains a desirable trade-off between accuracy and efficiency, offering meaningful implications for the advancement of lightweight biometric recognition systems. Full article

With the rapid development of intelligent manufacturing and industrial automation, defect recognition and detection of hot-rolled strip steel have become crucial to ensuring both production efficiency and product quality. However, existing hot-rolled strip steel detection systems often rely on expensive, energy-intensive, stationary equipment, making them unsuitable for mobile applications, such as outdoor use. To address this challenge, this paper proposes and designs a lightweight dual-surface defect recognition model for hot-rolled steel strips that can be implemented on mobile low-power devices (e.g., Raspberry Pi). First, to train the lightweight model, the NEU-CLS dataset is augmented through image generation via StyleGAN3, denoising with a water-wave-like noise removal algorithm, and super-resolution with Real-ESRGAN. Then, MMAM-EfficientNet-B0 is pruned during training, and the Network Slimming algorithm is applied to optimize it on the expanded NEU-CLS dataset, removing 70% of the network structure. Finally, the pruned recognition model is deployed on a Raspberry Pi, achieving an accuracy of 96.333%, with a classification time of 1.527 s per image, a reduction of 155.010% compared to the original model. Our experiments confirm the real-time effectiveness and practical application value of the model. Full article

Objectives: The aim of this systematic review was to evaluate the current evidence on photon-counting detector computed tomography (PCCT) in hepatocellular carcinoma (HCC) imaging. Methods: A systematic literature search was performed in PubMed and Scopus, and five articles were finally included. Results: Four studies focused on the optimization of acquisition and reconstruction parameters such as slice thickness, kernels, virtual monoenergetic imaging (VMI), and quantum iterative reconstruction (QIR), with 50 keV reconstructions consistently associated with improved lesion conspicuity. QIR demonstrated significant noise reduction compared with filtered back projection, enhancing overall image quality, while one proof-of-concept study investigated dual-contrast PCCT, showing feasibility for simultaneous arterial and portal-phase acquisition. According to QUADAS-2, most studies presented a low or unclear risk of bias, with only one study rated at high risk for patient selection. Conclusions: In conclusion, PCCT shows promising technical advances and potential for improved HCC detection and characterization. Current evidence remains preliminary and focused on image quality rather than clinical outcomes; PCCT applications in routine practice are still largely unexplored. Full article