Digital Signal Processing最新文献

{"title":"Multi-scale Attention Residual Convolution Neural Network for Single Image Super Resolution (MSARCNN)","authors":"Varsha Singh ,&nbsp;Naresh Vedhamuru ,&nbsp;R. Malmathanraj ,&nbsp;P. Palanisamy","doi":"10.1016/j.dsp.2025.105614","DOIUrl":"10.1016/j.dsp.2025.105614","url":null,"abstract":"<div><div>Traditional super-resolution methods often struggle to capture fine details and extract features, especially at higher frequency which leads to poor reconstruction of images. Further some SR methods neglect the significance of complexity while designing deeper networks. Deeper networks are challenging to train and have greater computational load which limits the performance of SR method making it less compatible for other devices. To address this problem, we propose a novel Multi-Scale Attention Residual Convolutional Neural Network(MSARCNN). The model combines eight multi-scale attention residual convolution and a Dilated Convolution Block(DCB). Each MSARCB comprises of a squeeze and excitation block which recalibrates feature maps by emphasizing informative channels and a Pixel Attention Block(PAB) which utilizes attention-based weighting to enhance local feature representation. The MSARCB employs multi-scale hierarchical feature extraction with the help of parallel convolution layers with varying channels and DCB with dilation rates of 1,3,5 and 7 which helps in capturing both spatial features and fine details by enlarging the effective receptive field without increasing the number of learnable parameters. Experiments on four benchmark dataset demonstrate that the proposed model significantly outperforms other state-of-the-art lightweight SR methods, providing a exceptional balance of reconstruction performance, model complexity and parameter count.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"168 ","pages":"Article 105614"},"PeriodicalIF":3.0,"publicationDate":"2025-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145219350","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Gridless SR-STAP algorithm based on square-root lasso for airborne radar clutter suppression","authors":"Junxiang Cao ,&nbsp;Tong Wang ,&nbsp;Weichen Cui","doi":"10.1016/j.dsp.2025.105617","DOIUrl":"10.1016/j.dsp.2025.105617","url":null,"abstract":"<div><div>The grid-based sparse recovery space-time adaptive processing(SR-STAP) algorithm requires the angular Doppler plane to be discretised in order to construct an overcomplete basis matrix. This process can result in off-grid problem and lead to a decline in performance. To address this issue, a gridless SR-STAP algorithm based on square-root Lasso(SRL) is proposed in this paper. Firstly, by introducing auxiliary variables, we obtain an iterative solution to the SRL using alternating optimization. Secondly, substituting the above iterative solution into the original problem transforms the SRL into a trace minimization problem for the clutter covariance matrix(CCM). The trace minimization problem is convex and can be solved globally in the continuous domain, thus avoiding the off-grid problem. Thirdly, to accommodate the noise unknown environment, a closed-form solution for the noise power is derived. Finally, in order to improve the computational efficiency, we solve iteratively for the required parameters in the framework of alternating direction method of multipliers(ADMM). Simulation results demonstrate that the proposed algorithm can overcome the off-grid problem, exhibits better clutter suppression performance and lower computational complexity than typical grid-based SR-STAP algorithm.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"168 ","pages":"Article 105617"},"PeriodicalIF":3.0,"publicationDate":"2025-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145134791","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A physics-guided deep neural network approach for simulating variable-order fractional chaotic financial models","authors":"Maalee Almheidat ,&nbsp;Mohammad Alqudah ,&nbsp;Muhammad Israr ,&nbsp;Assad Ayub","doi":"10.1016/j.dsp.2025.105621","DOIUrl":"10.1016/j.dsp.2025.105621","url":null,"abstract":"<div><h3>Significance</h3><div>Constant fractional-order mathematical models have highlighted key aspects of real-life systems, but a major advancement occurred when scientists began exploring variable-order models. This study holds two main considerable aspects, firstly, it addresses a variable-order fractional financial mathematical system, and secondly, it solves this model using the PINNs methodology.</div></div><div><h3>Purpose</h3><div>This study explores the numerical solution of variable order fractional financial mathematical model through Limited-memory Broyden–Fletcher–Goldfarb–Shanno (L-BFGS) optimizer-based Physics-Informed Neural Networks (PINNs). This framework captures the nonlinear, complex and memory-dependent features inherent in fractional financial model with Caputo variable-order fractional differential operator. This scheme ensures fast convergence and provides robust approximations of chaotic trajectories under varying fractional orders. This study shows better representation of the intricacies of financial systems. There are discussed two main cases of integer and variable order fractional system. Furthermore, there are three subcases of variable case with different functions. For each case different plots are displayed, which show chaotic behavior.</div></div><div><h3>Findings</h3><div>In this study, it is focused on a generalized variable-order fractional financial system that incorporates key economic components: interest rate, investment demand, and price index. Results confirm that the method achieves a convergence order consistent with theoretical expectations. Our findings suggest that the function employed in Case 2c, <span><math><mrow><mo>(</mo><mrow><msub><mi>O</mi><mn>3</mn></msub><mrow><mo>(</mo><mi>t</mi><mo>)</mo></mrow><mo>=</mo><mrow><mo>(</mo><mfrac><mn>1</mn><mn>15</mn></mfrac><mo>)</mo></mrow><msup><mrow><mi>e</mi></mrow><mrow><mi>sin</mi><mo>(</mo><mfrac><mn>1</mn><mn>25</mn></mfrac><mo>)</mo><mi>t</mi></mrow></msup><mo>+</mo><mn>0.76</mn></mrow><mo>)</mo></mrow></math></span> is the most effective in capturing the complex and non-stationary nature of real financial dynamics.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"168 ","pages":"Article 105621"},"PeriodicalIF":3.0,"publicationDate":"2025-09-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145157744","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Scaling anomaly detection with segmentation models","authors":"Stefano Samele,&nbsp;Francesco Attorre,&nbsp;Matteo Matteucci","doi":"10.1016/j.dsp.2025.105613","DOIUrl":"10.1016/j.dsp.2025.105613","url":null,"abstract":"<div><div>Detecting anomalies in product images is a critical task in industrial quality control, where even subtle defects can have operational and financial impact. However, deploying anomaly detection algorithms in real-world industrial scenarios remains challenging, particularly when products are large, complex, or captured at high resolution. Many existing methods struggle to scale effectively while maintaining precision. This work aims to develop an algorithm that can effectively scale to larger, more complex objects. The method, <em>SADSeM</em> (Scaling Anomaly Detection with Segmentation Models), is based on classic convolutional neural networks for segmentation, such as Mask-RCNN. Thanks to these models’ ability to learn and encode an object’s structure, we can design a pipeline that uses both their segmentation maps and feature embeddings to carry out unsupervised anomaly detection. As the segmentation task is effectively solved by these models independently of image size, we scale to higher-resolution images with more effectiveness than competitors, while maintaining competitive results in simpler scenarios.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"168 ","pages":"Article 105613"},"PeriodicalIF":3.0,"publicationDate":"2025-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145157746","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Water surface object detection in river channels based on feature cross-layer fusion and reconstruction","authors":"Enze Zhang ,&nbsp;Yecai Guo ,&nbsp;Songbin Li","doi":"10.1016/j.dsp.2025.105618","DOIUrl":"10.1016/j.dsp.2025.105618","url":null,"abstract":"<div><div>Water surface object detection in river channels is essential for effective river monitoring systems. However, existing object detection techniques are frequently inadequate for perceiving objects of varying sizes in complex and varying backgrounds. To address this issue, firstly, we propose a Feature Cross-Layer Fusion and Reconstruction Module, which effectively fuses multi-scale features through adaptive weights (used to dynamically adjust the importance of features at different layers) and employs a spatial-channel reconstruction mechanism (by separately learning and reconstructing spatial and channel dimension features) to reduce background feature redundancy, achieving a 5.1% improvement in Precision over the baseline model. Furthermore, we introduce a Feature Extraction Module based on structural reparameterization, which enhances the feature representation capability while maintaining computational efficiency, resulting in a 1% improvement in mAP @0.5 compared to the baseline. Building on these improvements, we develop a water surface object detection algorithm that incorporates an improved loss function for better accuracy. To comprehensively evaluate its performance, we constructed a dedicated UARODD dataset, which includes 16 categories of commonly observed water surface objects in river channels. The dataset consists of 9500 images collected from real aerial photography and the internet, with a total of 24534 annotated instances, covering a wide range of river scenes worldwide. Experimental results indicate that the proposed algorithm achieves a mean Average Precision (mAP@ 0.5) of 79.6% on this real-world dataset, representing a 5% improvement in mAP@ 0.5 compared to the YOLOv11 baseline. The detailed program code and weight files have been made publicly available at ht tps://github.com/zhangenze1016/FCFR-yolo.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"168 ","pages":"Article 105618"},"PeriodicalIF":3.0,"publicationDate":"2025-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145117915","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Leveraging stochastic filtering approach in fractional order Colpitts oscillator circuit","authors":"Rahul Bansal ,&nbsp;Sudipta Majumdar","doi":"10.1016/j.dsp.2025.105610","DOIUrl":"10.1016/j.dsp.2025.105610","url":null,"abstract":"<div><div>Chaos detection in noisy framework is a crucial issue that is significant in several engineering domains. In recent state-of-the-art work, authors proposed different methods for chaos detection; however, they lack reliability, flexibility, and have a greater computational burden. To get rid of the aforementioned limitations, this paper presents the Bayesian filtering-based bifurcation analysis of the Colpitts oscillator to show regular and irregular (chaotic) oscillations. Initially, we formulate fractional-order derivative (FOD) based stochastic differential equations (SDEs) using Kirchhoff’s law by introducing Gaussian noise to circuit elements, then we predicted the chaos using FOC-based adaptive iterated extended Kalman filter (AIEKF) method and compared with the FOC-based extended Kalman filter (EKF) method, FOC-based wavelet transform (WT) method. We also compare the estimated output with PSPICE simulated values and illustrate the efficacy of the proposed approach with respect to the FOC-based conventional method.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"168 ","pages":"Article 105610"},"PeriodicalIF":3.0,"publicationDate":"2025-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145219440","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"BCGW-YOLO: A lightweight network for road damage detection using enhanced feature fusion and dynamically adjusted gradient loss","authors":"Youlin Chen ,&nbsp;Yuqi Wang ,&nbsp;Huakun Luo ,&nbsp;Xi Li ,&nbsp;Jianhui Zhan ,&nbsp;Weichao Chen","doi":"10.1016/j.dsp.2025.105609","DOIUrl":"10.1016/j.dsp.2025.105609","url":null,"abstract":"<div><div>Achieving accurate road damage detection while maintaining low computational cost remains a key challenge, particularly for deployment on the edge devices with limited resources. To address this, we propose BCGW-YOLO, a lightweight yet practical detection framework based on YOLOv8s, tailored for road damage detection in diverse and dynamic environments. We propose the Bidirectional Ghost-based Hierarchical Feature Fusion Network (BG-HFFN), which effectively aggregates features across shallow to deep layers (P2 to P5). By leveraging lightweight Ghost convolutions, the network preserves fine-grained spatial information while significantly reducing computational overhead. In addition, a Content-Shape Feature Enhancement (CSFE) module is specifically designed to improve the model’s ability to extract and fuse shape-specific and contextual features, thereby enhancing the recognition of various crack types. To further improve robustness and convergence, we design a Weighted Focal IoU (WFIoU) loss function that integrates WIoU and Focaler-IoU to address class imbalance and anchor quality issues in complex scenarios. Extensive experiments on the RoadDamageDataset, RDD2020, and RDD2022 validate the effectiveness of the proposed framework, demonstrating a 4.9 % increase in precision and a 2.0 % improvement in [email protected] over the YOLOv8s baseline, along with a 33.9 % reduction in parameters, 9.9 % lower computation, and a 33.3 % smaller model size. These results indicate that BCGW-YOLO provides a practical and efficient solution for real-time road damage inspection in real-world applications.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"168 ","pages":"Article 105609"},"PeriodicalIF":3.0,"publicationDate":"2025-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145117916","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A sparse Bayesian learning based network for energy-efficient ECG compressed sensing","authors":"Le Qin ,&nbsp;Yukang Xu ,&nbsp;Yuan Wang ,&nbsp;Zenan Xiong ,&nbsp;Yugen Yi","doi":"10.1016/j.dsp.2025.105608","DOIUrl":"10.1016/j.dsp.2025.105608","url":null,"abstract":"<div><div>In recent years, wireless body-area networks (WBANs) have become prevalent for remote electrocardiogram (ECG) monitoring. However, the long-term operation of these systems demands significant energy from sensors. To address this, it is essential to streamline signal acquisition and reduce signal dimensionality, thereby decreasing communication bandwidth and on-chip power usage. Compressed sensing (CS), an emerging sampling technique, has been increasingly adopted for remote ECG monitoring. While traditional CS methods enhance reconstruction precision by using signal features as prior knowledge, they do not fully exploit the potential of these priors. This paper introduces a hybrid approach, PC-BCSNet, which combines the CS-based framework of pattern-coupled sparse Bayesian learning (PC-SBL) with a data-driven deep learning method. This dual-driven architecture develops a generalized prior model for post-sparsification ECG signals, employing the generalized approximate message passing (GAMP) algorithm for rapid reconstruction. Furthermore, an interpretable deep iterative neural network is designed to execute the full iterative Bayesian inference process. The scale parameters of the prior model serve as trainable weights, capturing features specific to ECG signals. Experiments demonstrate that PC-BCSNet significantly outperforms other state-of-the-art algorithms in reconstruction accuracy and speed, as evaluated on the European ST-T and MIT-BIT Arrhythmia databases. Notably, our network design adapts readily to changes in measurement matrices, providing enhanced flexibility and robustness for practical applications.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"168 ","pages":"Article 105608"},"PeriodicalIF":3.0,"publicationDate":"2025-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145117914","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Perception-driven underwater image enhancement via multi-level feature fusion","authors":"Hu Qiang ,&nbsp;Quan Xiao ,&nbsp;Xingxing You ,&nbsp;Yuzhong Zhong ,&nbsp;Songyi Dian","doi":"10.1016/j.dsp.2025.105606","DOIUrl":"10.1016/j.dsp.2025.105606","url":null,"abstract":"<div><div>In recent years, underwater image enhancement technology has attracted increasing attention due to its significant contribution to high-level visual tasks. However, existing methods fail to balance enhancement performance and computational cost, limiting their practical application. To address this issue, we propose a perception-driven underwater image enhancement framework based on multi-level feature fusion. Specifically, we employ depthwise and pointwise convolutions to build a lightweight backbone. To compensate for the limitations of the lightweight network in feature extraction, we design a multi-scale texture-structure auxiliary network to extract texture and structure features at different levels, and incorporate these features as perceptual information into the backbone network. Furthermore, to effectively mitigate the color distortion and low contrast in degraded images, we propose a histogram distribution loss function and an adaptive hybrid color space loss function. Extensive experiments demonstrate that the proposed perception-driven framework outperforms existing state-of-the-art methods, achieving superior enhancement quality at lower computational costs. Additionally, underwater object detection experiments validate that our method significantly improves performance in high-level visual tasks. The code is available at <span><span>https://github.com//PDMF</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"168 ","pages":"Article 105606"},"PeriodicalIF":3.0,"publicationDate":"2025-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145095020","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enhanced 2D DOA and polarization estimation for sparse distributed orthogonal loop and dipole planar array based on fourth-order cumulant","authors":"Jun Pan ,&nbsp;Yaxing Yue ,&nbsp;Min Tian ,&nbsp;Fuquan Nie ,&nbsp;Dawei Gao ,&nbsp;Guisheng Liao ,&nbsp;Zhiguo Shi","doi":"10.1016/j.dsp.2025.105607","DOIUrl":"10.1016/j.dsp.2025.105607","url":null,"abstract":"<div><div>Two-dimensional (2D) direction-of-arrival (DOA) and polarization estimation using sparse polarimetric array shows advantages in increasing degrees-of-freedom (DoFs) and reducing hardware costs. However, most relevant studies still rely on second-order statistics, which constrain the achievable DoFs. To overcome such limitations, we propose a fourth-order cumulant-based approach for multi-parameter estimation in joint spatial-polarimetric domains. Via such an approach, a covariance-like standard cumulant matrix corresponding to a virtual uniform counterpart of the considered sparse distributed orthogonal loop and dipole planar array is constructed, where we have defined the involved selection matrices in the data reordering process. A virtual spatial-polarimetric rotational-invariance procedure is then presented to obtain an efficient estimation of 2D DOA and polarization in closed form. Simulation results are then included to verify the performance advantages of the proposed approach in terms of identifiability, estimation accuracy, probability of successful resolution, and computational efficiency.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"168 ","pages":"Article 105607"},"PeriodicalIF":3.0,"publicationDate":"2025-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145157747","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}