Digital Signal Processing最新文献_第8页

Direct target localization in USNs with hybrid quantized multi-snapshot measurements: A geometric structure-aided approach 基于混合量化多快照测量的USNs直接目标定位：几何结构辅助方法

IF 3 3区工程技术

Digital Signal Processing Pub Date : 2025-08-27 DOI: 10.1016/j.dsp.2025.105552

Chunjin Jiang , Shefeng Yan , Linlin Mao , Shoude Jiang , Wei Wang , Jiaping Yu

{"title":"Direct target localization in USNs with hybrid quantized multi-snapshot measurements: A geometric structure-aided approach","authors":"Chunjin Jiang , Shefeng Yan , Linlin Mao , Shoude Jiang , Wei Wang , Jiaping Yu","doi":"10.1016/j.dsp.2025.105552","DOIUrl":"10.1016/j.dsp.2025.105552","url":null,"abstract":"<div><div>In this article, a multi-snapshot hybrid quantization algorithm designed to enhance target localization accuracy is proposed for an underwater sensor network system, comprising an active acoustic source, multiple distributed passive sensors, and a fusion center. Within this framework, a direct target localization algorithm with particle dimension reduction is introduced. The proposed method considers channel transmission errors and allows for varying quantization depths at each sensor. The Cramer-Rao lower bound (CRLB) for the target localization with multi-snapshot hybrid quantization is derived, demonstrating that increasement of signal snapshots significantly reduces target localization error. The optimal quantization threshold is obtained by maximizing the objective function concerning the determinant of the Fisher information matrix, aiming to maximize localization performance. Leveraging the geometric structure of the model, a genetic algorithm embedded with particle dimension reduction (GA-PDR) is proposed to locate the target directly. Numerical results demonstrate that the proposed multi-snapshot hybrid quantization algorithm significantly improves overall localization performance, while the GA-PDR locates the target precisely and achieves convergence more quickly.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"168 ","pages":"Article 105552"},"PeriodicalIF":3.0,"publicationDate":"2025-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144925673","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}

引用次数: 0

Joint blind calibration for multichannel compressed sampling systems 多通道压缩采样系统的联合盲校正

IF 3 3区工程技术

Digital Signal Processing Pub Date : 2025-08-27 DOI: 10.1016/j.dsp.2025.105555

Yinuo Su, Jingchao Zhang, Liyan Qiao

{"title":"Joint blind calibration for multichannel compressed sampling systems","authors":"Yinuo Su, Jingchao Zhang, Liyan Qiao","doi":"10.1016/j.dsp.2025.105555","DOIUrl":"10.1016/j.dsp.2025.105555","url":null,"abstract":"<div><div>With the rapid advancement of radar and communication systems, multichannel compressed sampling architectures have emerged as a pivotal solution for high-frequency and wideband signal acquisition. However, practical implementations are inevitably plagued by non-ideal factors, particularly unknown gain-phase errors and inter-channel mutual coupling, which severely degrade signal reconstruction accuracy. Existing calibration methods primarily focus on array manifolds or depend on prior knowledge, often proving inadequate for addressing the distinct measurement matrix structures in compressed sampling systems. To address this limitation, we propose a joint blind calibration framework that enables simultaneous sparse signal recovery and system error correction in multichannel compressed sampling systems, eliminating the need for dedicated test signals or auxiliary calibration equipment. We reformulated the joint calibration problem as a multilinear inverse problem, which is further transformed into an eigenvector/eigenvalue optimization task solvable via a dual-projection gradient descent algorithm. The main work of this paper lies in providing a theoretical analysis of eigenvalue distribution ranges and perturbation bounds for proposed eigenvector-solving problem. These analyses reveal that the eigenvalue gap is governed by mutual coupling attenuation coefficients, ensuring algorithmic convergence under practical noise conditions. Extensive numerical experiments validate the method's superiority. Notably, the theoretical bounds on mutual coupling effects align closely with empirical results, demonstrating the framework's reliability.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"168 ","pages":"Article 105555"},"PeriodicalIF":3.0,"publicationDate":"2025-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144908685","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}

引用次数: 0

Convergence-enhanced 1-bit DACs precoding for massive MIMO-OFDM systems via Anderson acceleration 基于Anderson加速的大规模MIMO-OFDM系统的收敛增强1位dac预编码

IF 3 3区工程技术

Digital Signal Processing Pub Date : 2025-08-27 DOI: 10.1016/j.dsp.2025.105545

Guodong Xue, Hui Li, Rui Liang

{"title":"Convergence-enhanced 1-bit DACs precoding for massive MIMO-OFDM systems via Anderson acceleration","authors":"Guodong Xue, Hui Li, Rui Liang","doi":"10.1016/j.dsp.2025.105545","DOIUrl":"10.1016/j.dsp.2025.105545","url":null,"abstract":"<div><div>In the downlink of massive multiple-input multiple-output (MIMO) systems, high-resolution digital-to-analog converters (DACs) are a major source of power consumption. This paper mainly focuses on the precoding design for the downlink of massive MIMO-orthogonal frequency division multiplexing (OFDM) systems employing 1-bit DACs to reduce power consumption. We propose a Douglas-Rachford splitting (DRS)-based 1-bit precoding algorithm, where operator linearization is adopted during iterations to avoid matrix inversion, thereby reducing computational complexity. To enhance convergence efficiency, we demonstrate that the proposed precoding algorithm is a fixed-point iteration and introduce an Anderson acceleration module, developing an Anderson acceleration linearized DRS (LDRS-AA) precoding algorithm. Notably, we introduce a decision function to improve the stability of classical Anderson acceleration. A detailed analysis of the convergence and computational complexity for the proposed algorithms is also provided. Simulation results show that the proposed Anderson acceleration scheme achieves significant convergence speed improvement without performance loss. In addition, the proposed precoding algorithm achieves superior bit error rate (BER) performance, exhibiting approximately 1.25 dB and 1 dB performance gains over existing advanced algorithms under quadrature phase shift keying (QPSK) modulation and 16-ary quadrature amplitude modulation (QAM), respectively.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"168 ","pages":"Article 105545"},"PeriodicalIF":3.0,"publicationDate":"2025-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144913983","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}

引用次数: 0

Multi-weak targets detection and measurement from range-Doppler maps in GNSS-based passive bistatic radar 基于gnss的无源双基地雷达距离-多普勒图的多弱目标探测与测量

IF 3 3区工程技术

Digital Signal Processing Pub Date : 2025-08-26 DOI: 10.1016/j.dsp.2025.105566

Zuhan Cheng , Jun Wang , Te Zhao , Jinxin Sui , Ziqian Huang , Hui Ma , Luo Zuo

{"title":"Multi-weak targets detection and measurement from range-Doppler maps in GNSS-based passive bistatic radar","authors":"Zuhan Cheng , Jun Wang , Te Zhao , Jinxin Sui , Ziqian Huang , Hui Ma , Luo Zuo","doi":"10.1016/j.dsp.2025.105566","DOIUrl":"10.1016/j.dsp.2025.105566","url":null,"abstract":"<div><div>This paper explores the GNSS satellite system as illumination in passive radar for maneuvering target detection. The main difficulty of this technology is the limited arrival power from navigation satellites. To address this, a long-time integration algorithm for multi-weak targets detection is proposed. It begins with the segmented signal model to obtain the range-compressed data, followed by the second-keystone transform to correct the intra-frame quadratic range migration. Then, a multi-target motion parameter estimation method based on the modified variational Bayesian framework is employed on the azimuth signal. Lastly, a proper compensation strategy is applied to align the targets’ positions in all integrated maps. It estimates the multiple targets’ motion measurements from the bistatic range-Doppler maps accurately, and compensates for the complicated range and Doppler migrations to improve the detection capability. This technique is validated through theoretical analysis with simulations, as well as experiments. Both simulated and real-measured data experiments prove the remarkable multi-weak targets detection and measurement performance than the existing methods.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"168 ","pages":"Article 105566"},"PeriodicalIF":3.0,"publicationDate":"2025-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145046126","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}

引用次数: 0

CCTFaceNet: Enhancing face super-resolution with cascaded CNN-transformer and dual-path feature fusion CCTFaceNet：利用级联CNN-transformer和双路特征融合增强人脸超分辨率

IF 3 3区工程技术

Digital Signal Processing Pub Date : 2025-08-26 DOI: 10.1016/j.dsp.2025.105557

Naveen Kumar Tiwari , Shyam Singh Rajput , Raj Patel

{"title":"CCTFaceNet: Enhancing face super-resolution with cascaded CNN-transformer and dual-path feature fusion","authors":"Naveen Kumar Tiwari , Shyam Singh Rajput , Raj Patel","doi":"10.1016/j.dsp.2025.105557","DOIUrl":"10.1016/j.dsp.2025.105557","url":null,"abstract":"<div><div>The convolutional neural networks (CNNs) have significantly advanced face super-resolution techniques, enabling the restoration of degraded facial details. However, these methods often encounter limitations related to computational cost. Additionally, the limited receptive fields of CNNs can hinder the realistic and natural reconstruction of facial images. Transformer-based models counter this issue by global feature learning using multi-head self attention, but lack the incremental feature learning capabilities of CNNs. This paper proposes a Cascaded CNN-Transformer-based Face image super-resolution Network (CCTFaceNet) to deal with the above-mentioned issues. In the proposed network, to enrich input low-resolution face images, a Preliminary Super-Resolution, <em>i.e.</em>, PreSR network is placed at the beginning of CCTFaceNet. The output of PreSR is then fed to a deep feature extraction block consisting of a dual-path feature fusion block (DPFF). This block internally has two paths, one for CNN and the other for the cascaded attention transformer (CAT). DPFF also has a context-resolving unit responsible for filtering out redundant information. CAT consists of a shifted window multi-head self-attention, a multi scale edge attention, and a channel importance recalibration module; they are assembled in a cascaded manner. This assembly can reconstruct highly accurate details spatially and along the channel with crisp edges. A feed-forward layer is also sandwiched between the above attention cascade. The extracted deep features are upsampled using pixel-shuffle and sub-pixel convolutional layers. Extensive experiments conducted on several benchmark datasets affirm the supremacy of the proposed network.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"168 ","pages":"Article 105557"},"PeriodicalIF":3.0,"publicationDate":"2025-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144913984","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}

引用次数: 0

Image splicing localization based on Hellinger distance and noise estimation through convolutional neural network and vision transformer 基于海灵格距离和噪声估计的卷积神经网络图像拼接定位

IF 3 3区工程技术

Digital Signal Processing Pub Date : 2025-08-26 DOI: 10.1016/j.dsp.2025.105559

Debjit Das , Ruchira Naskar , Rajat Subhra Chakraborty

{"title":"Image splicing localization based on Hellinger distance and noise estimation through convolutional neural network and vision transformer","authors":"Debjit Das , Ruchira Naskar , Rajat Subhra Chakraborty","doi":"10.1016/j.dsp.2025.105559","DOIUrl":"10.1016/j.dsp.2025.105559","url":null,"abstract":"<div><div>With the proliferation of readily available image-tampering tools, image forgery has become widespread. <em>Image Splicing</em>, where multiple portions of different source images are combined to synthesize an artificial or forged image, is a powerful image forgery technique that can lead to various malicious activities and therefore mislead common masses. In this work, we propose a two-stage image splicing localization method, where the first stage is based on noise estimate variation between image blocks and inter-block horizontal and vertical Hellinger distances computed from block-wise pixel probability distributions to mark suspicious image blocks. At the final stage of our method, we perform finer classification of suspicious image blocks using two different deep neural network models: first, a transfer learning based extended residual dense neural network model and second, a modified large vision transformer. We achieve a significant reduction in the training data requirement as compared to the state-of-the-art. Extensive experiments on five benchmark image forgery datasets demonstrate that the localization accuracy of the proposed model is above 90%. We also prove the proposed method's resilience to common post-processing attacks.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"168 ","pages":"Article 105559"},"PeriodicalIF":3.0,"publicationDate":"2025-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144902355","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}

引用次数: 0

An EEG-based seizure detection method with multiband guided fusion and cross-frequency interaction 一种基于脑电图的多波段引导融合和交叉频率交互检测方法

IF 3 3区工程技术

Digital Signal Processing Pub Date : 2025-08-25 DOI: 10.1016/j.dsp.2025.105553

Liuliang Chen, Yang Tian, Tao Deng

{"title":"An EEG-based seizure detection method with multiband guided fusion and cross-frequency interaction","authors":"Liuliang Chen, Yang Tian, Tao Deng","doi":"10.1016/j.dsp.2025.105553","DOIUrl":"10.1016/j.dsp.2025.105553","url":null,"abstract":"<div><div>The automated detection of epilepsy is of great importance for improving the efficiency of clinical diagnosis. However, existing methods are insufficient in modeling cross-frequency interactions in multiband electroencephalogram (EEG) signals. To address this limitation, we propose an epilepsy detection model based on multiband guided fusion. The goal is to enhance the modeling capacity of cross-frequency dynamic features through band division and a hierarchical feature integration mechanism. A multi-scale temporal encoding module (MTEM) is designed for each specific frequency band, enabling the extraction of local dynamic features from low-, mid-, and high-frequency signals. Furthermore, a hierarchical fusion framework is developed, which incorporates a cross-frequency guidance mechanism and a frequency-specific statistical attention module (FSAM). This design effectively captures cross-frequency interactions across different frequency bands, which are often overlooked by previous methods that fail to model cross-frequency dependencies. Our model enhances these interactions, outperforming existing methods with an F1-score of 99.42% on the TUSZ dataset, 99.9% on the CHB-MIT dataset, and 95.46% on the TUEV dataset. The results indicate that the strategy of frequency band division and cross-frequency fusion can effectively capture multiscale dynamic patterns related to epilepsy, offering a more reliable automated solution for clinical EEG analysis.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"168 ","pages":"Article 105553"},"PeriodicalIF":3.0,"publicationDate":"2025-08-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144907988","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}

引用次数: 0

Improved cylindrical displaced coprime conformal array for 2-D DOA and polarization estimation 用于二维DOA和偏振估计的改进型圆柱位移质素共形阵

IF 3 3区工程技术

Digital Signal Processing Pub Date : 2025-08-25 DOI: 10.1016/j.dsp.2025.105551

Mingcheng Fu , Zhi Zheng , Wen-Qin Wang

{"title":"Improved cylindrical displaced coprime conformal array for 2-D DOA and polarization estimation","authors":"Mingcheng Fu , Zhi Zheng , Wen-Qin Wang","doi":"10.1016/j.dsp.2025.105551","DOIUrl":"10.1016/j.dsp.2025.105551","url":null,"abstract":"<div><div>Recently, sparse conformal arrays have received lots of attention due to its increased array aperture, improved degrees-of-freedom (DOFs) and reduced mutual coupling compared to uniform conformal arrays. In this article, we devise a new sparse conformal array, referred to as the improved cylindrical displaced coprime conformal array (ICDCCA) for two-dimensional (2-D) direction-of-arrival (DOA) and polarization estimation. Unlike the existing sparse conformal arrays, the ICDCCA consists of multiple two-parallel linear arrays, which are constructed by displacing in parallel one subarray in the prototype coprime array and enlarging its inter-sensor spacings. For the ICDCCA configuration, there are closed-form expressions for the sensor positions, and its number of achievable DOFs can be analytically computed, from which the optimum configurations are obtained. Compared with the existing sparse conformal arrays, the proposed array configuration provides a larger array aperture, a higher number of DOFs, as well as less mutual coupling effects. Numerical results demonstrate the superiority of the ICDCCA over several existing sparse conformal arrays.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"168 ","pages":"Article 105551"},"PeriodicalIF":3.0,"publicationDate":"2025-08-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144902354","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}

引用次数: 0

Fuzzy adaptive bounded fraction non-Gaussian filter 模糊自适应有界分数非高斯滤波器

IF 3 3区工程技术

Digital Signal Processing Pub Date : 2025-08-23 DOI: 10.1016/j.dsp.2025.105538

Xiaoliang Feng, Shuo Wang, Chuanbo Wen

{"title":"Fuzzy adaptive bounded fraction non-Gaussian filter","authors":"Xiaoliang Feng, Shuo Wang, Chuanbo Wen","doi":"10.1016/j.dsp.2025.105538","DOIUrl":"10.1016/j.dsp.2025.105538","url":null,"abstract":"<div><div>In this paper, a novel non-Gaussian filter named the fuzzy adaptive bounded fraction non-Gaussian filter (FABFF) is proposed for addressing the filtering problem of linear non-Gaussian systems. The proposed filter integrates a robust loss function with a fuzzy membership function. First, a bounded fraction loss function is designed, which exhibits high robustness and numerical stability, effectively mitigating the impact of outliers. In addition, an adaptive parameterization scheme is developed based on the sample error for the bounded fraction loss function, which achieves a balance between the filtering accuracy and real-time performance compared to approaches using fixed parameters. Second, a novel weighted cost function is designed by incorporating sample weights, thereby improving the filtering accuracy compared to the cost function using average weights. The sample weights are determined based on the degrees of abnormality of each sample, which is quantified through a fuzzy membership function. By applying the fixed-point iterative method, the new cost function is solved, and FABFF is obtained. Subsequently, the performance of the bounded fraction loss function, the computational complexity, and the convergence of the proposed algorithm are analyzed. Finally, simulation results are presented to validate the effectiveness of the proposed algorithm.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"168 ","pages":"Article 105538"},"PeriodicalIF":3.0,"publicationDate":"2025-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144914058","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}

引用次数: 0

DD-YOLO: A dual-channel dual-path YOLO network for target detection of blurred vehicles DD-YOLO：一种用于模糊车辆目标检测的双通道双路径YOLO网络

IF 3 3区工程技术

Digital Signal Processing Pub Date : 2025-08-22 DOI: 10.1016/j.dsp.2025.105565

Chaoqun Duan , Yuhan Guo , Xuelian Duan , Guoqiang Li , Bo Sheng

{"title":"DD-YOLO: A dual-channel dual-path YOLO network for target detection of blurred vehicles","authors":"Chaoqun Duan , Yuhan Guo , Xuelian Duan , Guoqiang Li , Bo Sheng","doi":"10.1016/j.dsp.2025.105565","DOIUrl":"10.1016/j.dsp.2025.105565","url":null,"abstract":"<div><div>The detection of blurred vehicle targets is essential for maintaining traffic efficiency and ensuring road safety. Although various you-only-look-once (YOLO)-based models exist, few studies have focused on blurred vehicle detection under real-world traffic conditions. To fill this gap, we propose a novel dual-channel dual-path YOLO (DD-YOLO) network, featuring dual-channel feature extraction and dual-path feature fusion. The network comprises a hybrid pooling pyramid (HPP) module, a dual-channel feature extraction backbone, and a dual-path fusion pooling neck (DPFP-neck). Within the DD-YOLO network, we first introduce the HPP module to reduce dependence on key features by combining max and average pooling, incorporating background information to mitigate false positives. Subsequently, the dual-channel backbone is designed to enhance DD-YOLO’s sensitivity for blurred vehicle targets by integrating multiple convolution and attention mechanisms, including the convolutional block attention module (CBAM), simple and parameter-free attention module (SIMAM), standard convolution, and ghost convolution, to capture richer features and improve recall. Finally, the DPFP-neck is developed to fuse diverse information and expand the receptive field across network depths, providing a satisfactory balance between precision and recall. Experiments on the BDD100K and KITTI datasets show that DD-YOLO improves detection accuracy by 4.9% and 4.0%, respectively, with [email protected] gains of 2.4% and 2.7% over the baseline, demonstrating its effectiveness and real-time capability in detecting blurred vehicle targets.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"168 ","pages":"Article 105565"},"PeriodicalIF":3.0,"publicationDate":"2025-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144988361","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}

引用次数: 0