Digital Signal Processing最新文献_第3页

EMD-YOLOv8: A road pedestrian detection algorithm based on improved YOLOv8 EMD-YOLOv8：一种基于改进YOLOv8的道路行人检测算法

IF 3 3区工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

Digital Signal Processing

Pub Date : 2026-04-15 Epub Date: 2026-01-19 DOI: 10.1016/j.dsp.2026.105940

Zhuangzhuang Bao, Wenhua Han, Yuchen Pan

As autonomous driving technology advances, pedestrian detection has become a critical task for ensuring road safety. However, in low-light and pedestrian-dense environments, current pedestrian detection algorithms often fail to meet the accuracy requirements for practical applications. To enhance detection accuracy, this paper presents the EMD-YOLOv8, an improved pedestrian detection algorithm. First, to enhance the detail representation of the input images, a Multi-Scale Retinex with Color Restoration algorithm is introduced to optimize the dataset. Next, an enhanced residual block is proposed as a replacement for the redundant BottleNeck structure in the original C2f module, which improves multi-scale object detection capability by integrating high-frequency information with local features. Additionally, a Multi-Scale Spatial Recalibration Network is proposed to dynamically adjust local details and global context features, with the goal of improving feature representation. Finally, a detail enhanced detection head is designed to improve small-object detection performance by shared convolutional parameters and integrating cross-layer feature fusion. Experiments show that the EMD-YOLOv8 algorithm reduces parameters by 47.3% compared to YOLOv8s, while increasing P, R, mAP50, and mAP50-95 by 2.2%, 5.7%, 7.5%, and 4.9%, respectively. The improved algorithm presented in this paper not only effectively addresses the issues of missed detections and false detections but also reduces the parameter count.

随着自动驾驶技术的进步，行人检测已成为确保道路安全的关键任务。然而，在低光照和行人密集的环境下，现有的行人检测算法往往不能满足实际应用的精度要求。为了提高检测精度，本文提出了一种改进的行人检测算法EMD-YOLOv8。首先，为了增强输入图像的细节表现，引入了一种带颜色恢复的多尺度Retinex算法对数据集进行优化。其次，提出了一种增强残差块替代原C2f模块中冗余的瓶颈结构，通过将高频信息与局部特征相结合，提高了多尺度目标检测能力；此外，提出了一个多尺度空间再标定网络，动态调整局部细节和全局上下文特征，以改善特征表示。最后，设计了一种细节增强检测头，通过共享卷积参数和集成跨层特征融合来提高小目标检测性能。实验表明，EMD-YOLOv8算法比yolov8算法减少了47.3%的参数，而P、R、mAP50和mAP50-95分别提高了2.2%、5.7%、7.5%和4.9%。本文提出的改进算法不仅有效地解决了漏检和误检问题，而且减少了参数个数。

{"title":"EMD-YOLOv8: A road pedestrian detection algorithm based on improved YOLOv8","authors":"Zhuangzhuang Bao, Wenhua Han, Yuchen Pan","doi":"10.1016/j.dsp.2026.105940","DOIUrl":"10.1016/j.dsp.2026.105940","url":null,"abstract":"<div><div>As autonomous driving technology advances, pedestrian detection has become a critical task for ensuring road safety. However, in low-light and pedestrian-dense environments, current pedestrian detection algorithms often fail to meet the accuracy requirements for practical applications. To enhance detection accuracy, this paper presents the EMD-YOLOv8, an improved pedestrian detection algorithm. First, to enhance the detail representation of the input images, a Multi-Scale Retinex with Color Restoration algorithm is introduced to optimize the dataset. Next, an enhanced residual block is proposed as a replacement for the redundant BottleNeck structure in the original C2f module, which improves multi-scale object detection capability by integrating high-frequency information with local features. Additionally, a Multi-Scale Spatial Recalibration Network is proposed to dynamically adjust local details and global context features, with the goal of improving feature representation. Finally, a detail enhanced detection head is designed to improve small-object detection performance by shared convolutional parameters and integrating cross-layer feature fusion. Experiments show that the EMD-YOLOv8 algorithm reduces parameters by 47.3% compared to YOLOv8s, while increasing P, R, mAP50, and mAP50-95 by 2.2%, 5.7%, 7.5%, and 4.9%, respectively. The improved algorithm presented in this paper not only effectively addresses the issues of missed detections and false detections but also reduces the parameter count.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"174 ","pages":"Article 105940"},"PeriodicalIF":3.0,"publicationDate":"2026-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146081606","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

Perception-assisted compressed sensing-based channel estimation in massive MIMO systems 海量MIMO系统中基于感知辅助压缩感知的信道估计

IF 3 3区工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

Digital Signal Processing

Pub Date : 2026-04-15 Epub Date: 2026-02-03 DOI: 10.1016/j.dsp.2026.105973

Chaojin Qing, Pu Guo, Zhiying Liu, Ting Qin, Jinxiao Zhang, Pengfei Du

In massive multiple-input multiple-output (mMIMO) systems, compressed sensing (CS)-based channel estimation (CE) methods face significant challenges posed by high computational complexity arising from the iterative reconstruction processing and unknown channel sparsity. To address these issues, inspired by the integration of sensing and communication (ISAC) concept, we propose a perception-assisted CS-based CE method in mMIMO systems. By leveraging perception technology, the sparsity and support set information of the channel vector are first extracted at the user equipment (UE) to form the perceptual prior information. Based on this prior information, a perception-assisted matching pursuit (PaMP) algorithm is developed. In this algorithm, the perceived support set is leveraged to reconstruct the channel vector for addressing the issues of unknown channel sparsity and high computational complexity due to the iterative reconstruction in CS-based CE methods. However, the perceptual information inevitably involves errors due to environmental and modeling errors, resulting in potential issues of false paths and missed paths in the perceived support set. To tackle these challenges, we design an error correction scheme. This scheme incorporates correlation detection and threshold detection mechanisms to suppress false paths and iteratively retrieve missing paths. Experimental results demonstrate that the proposed method outperforms baseline approaches in terms of CE accuracy while reducing computational complexity. Furthermore, the proposed method exhibits strong robustness against parameter variations.

在大规模多输入多输出（mMIMO）系统中，基于压缩感知（CS）的信道估计（CE）方法面临着迭代重建处理带来的高计算复杂度和未知信道稀疏性带来的重大挑战。为了解决这些问题，受传感和通信集成（ISAC）概念的启发，我们提出了一种在mMIMO系统中感知辅助的基于cs的CE方法。利用感知技术，首先在用户设备处提取信道向量的稀疏性和支持集信息，形成感知先验信息。基于这些先验信息，提出了一种感知辅助匹配追踪算法。该算法利用感知支持集重构信道向量，解决了基于cs的CE方法中迭代重构导致的未知信道稀疏性和计算复杂度高的问题。然而，由于环境和建模错误，感知信息不可避免地包含错误，从而导致感知支持集中存在错误路径和遗漏路径的潜在问题。为了应对这些挑战，我们设计了一个纠错方案。该方案结合了相关检测和阈值检测机制来抑制假路径和迭代检索缺失路径。实验结果表明，该方法在降低计算复杂度的同时，在CE精度方面优于基线方法。此外，该方法对参数变化具有较强的鲁棒性。

{"title":"Perception-assisted compressed sensing-based channel estimation in massive MIMO systems","authors":"Chaojin Qing, Pu Guo, Zhiying Liu, Ting Qin, Jinxiao Zhang, Pengfei Du","doi":"10.1016/j.dsp.2026.105973","DOIUrl":"10.1016/j.dsp.2026.105973","url":null,"abstract":"<div><div>In massive multiple-input multiple-output (mMIMO) systems, compressed sensing (CS)-based channel estimation (CE) methods face significant challenges posed by high computational complexity arising from the iterative reconstruction processing and unknown channel sparsity. To address these issues, inspired by the integration of sensing and communication (ISAC) concept, we propose a perception-assisted CS-based CE method in mMIMO systems. By leveraging perception technology, the sparsity and support set information of the channel vector are first extracted at the user equipment (UE) to form the perceptual prior information. Based on this prior information, a perception-assisted matching pursuit (PaMP) algorithm is developed. In this algorithm, the perceived support set is leveraged to reconstruct the channel vector for addressing the issues of unknown channel sparsity and high computational complexity due to the iterative reconstruction in CS-based CE methods. However, the perceptual information inevitably involves errors due to environmental and modeling errors, resulting in potential issues of false paths and missed paths in the perceived support set. To tackle these challenges, we design an error correction scheme. This scheme incorporates correlation detection and threshold detection mechanisms to suppress false paths and iteratively retrieve missing paths. Experimental results demonstrate that the proposed method outperforms baseline approaches in terms of CE accuracy while reducing computational complexity. Furthermore, the proposed method exhibits strong robustness against parameter variations.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"174 ","pages":"Article 105973"},"PeriodicalIF":3.0,"publicationDate":"2026-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146174811","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

DCSDL-MCP: Discriminative supervised dictionary learning with the minimax concave penalty DCSDL-MCP：具有极大极小凹惩罚的判别监督字典学习

IF 3 3区工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

Digital Signal Processing

Pub Date : 2026-04-15 Epub Date: 2026-01-31 DOI: 10.1016/j.dsp.2026.105970

Tianyao Feng , Benying Tan , Muyang Li , Jianpeng Wu , Shuxue Ding

Feature extraction, which focuses on extracting essential characteristics from raw data, plays a critical role in data processing. Although traditional unsupervised dictionary learning is effective in deriving low-dimensional representative features, the learned features often lack discriminative power. To enhance feature discrimination, this paper incorporates label information into the dictionary learning objective function and adopts the Minimax Concave Penalty (MCP) as a regularizer instead of the conventional l₀-norm and l₁-norm, leading to a novel supervised dictionary learning model termed DCSDL-MCP. To address the resulting optimization challenge, the objective function is first consolidated and reformulated into a form akin to traditional unsupervised dictionary learning. Then, a joint optimization strategy combining the Difference of Convex Functions Algorithm (DCA) and the Iterative Soft Thresholding Algorithm (ISTA) is developed to solve the problem efficiently. Extensive experiments on face recognition and object localization datasets demonstrate that the proposed method achieves superior accuracy and robustness. These results underscore its practical value and broad application potential in real-world scenarios.

特征提取在数据处理中起着至关重要的作用，其重点是从原始数据中提取基本特征。虽然传统的无监督字典学习在获得低维代表性特征方面是有效的，但学习到的特征往往缺乏判别能力。为了增强特征识别能力，本文将标签信息引入字典学习目标函数中，并采用极大极小凹惩罚（Minimax凹惩罚，MCP）作为正则化器代替传统的10 -范数和11 -范数，建立了一种新的监督式字典学习模型DCSDL-MCP。为了解决由此带来的优化挑战，首先将目标函数整合并重新制定为类似于传统无监督字典学习的形式。然后，提出了一种结合凸函数差分算法（DCA）和迭代软阈值算法（ISTA）的联合优化策略，有效地解决了这一问题。在人脸识别和目标定位数据集上的大量实验表明，该方法具有较好的准确性和鲁棒性。这些结果强调了其在现实场景中的实用价值和广泛的应用潜力。

{"title":"DCSDL-MCP: Discriminative supervised dictionary learning with the minimax concave penalty","authors":"Tianyao Feng , Benying Tan , Muyang Li , Jianpeng Wu , Shuxue Ding","doi":"10.1016/j.dsp.2026.105970","DOIUrl":"10.1016/j.dsp.2026.105970","url":null,"abstract":"<div><div>Feature extraction, which focuses on extracting essential characteristics from raw data, plays a critical role in data processing. Although traditional unsupervised dictionary learning is effective in deriving low-dimensional representative features, the learned features often lack discriminative power. To enhance feature discrimination, this paper incorporates label information into the dictionary learning objective function and adopts the Minimax Concave Penalty (MCP) as a regularizer instead of the conventional <em>l</em><sub>0</sub>-norm and <em>l</em><sub>1</sub>-norm, leading to a novel supervised dictionary learning model termed DCSDL-MCP. To address the resulting optimization challenge, the objective function is first consolidated and reformulated into a form akin to traditional unsupervised dictionary learning. Then, a joint optimization strategy combining the Difference of Convex Functions Algorithm (DCA) and the Iterative Soft Thresholding Algorithm (ISTA) is developed to solve the problem efficiently. Extensive experiments on face recognition and object localization datasets demonstrate that the proposed method achieves superior accuracy and robustness. These results underscore its practical value and broad application potential in real-world scenarios.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"174 ","pages":"Article 105970"},"PeriodicalIF":3.0,"publicationDate":"2026-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146174812","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

Lightweight speech enhancement with state-space model and depthwise separable convolution 基于状态空间模型和深度可分离卷积的轻量级语音增强

IF 3 3区工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

Digital Signal Processing

Pub Date : 2026-04-15 Epub Date: 2026-02-06 DOI: 10.1016/j.dsp.2026.105987

Chen Jiang , Dai Gao , Sirui Wang , Chengxuan Zou , Jie Liu

Speech enhancement is critical for improving intelligibility and quality in real-world noisy environments. However, existing methods often suffer from high computational complexity, limited contextual modeling, and poor robustness to voice interference. To address these challenges, we propose a lightweight framework for efficient and robust speech enhancement. Our approach synergizes a State-Space Model (SSM) with depthwise separable convolutions, combining efficient long-range temporal modeling with precise local feature extraction. We introduce an Auditory Inspired Spectral Compressor (AISC) that mimics the non-linear frequency resolution of the human ear, selectively preserving perceptually relevant information while significantly reducing computational redundancy. Furthermore, an Atrous Spatial Pyramid Pooling (ASPP) module is employed to capture multi-scale context, allowing the model to adapt to diverse noise patterns. To tackle the difficult problem of human-voice interference, we propose a Classifier Loss that enforces feature discriminability to effectively suppress competing human-voice interference. Experiments on multiple datasets demonstrate that our model outperforms state-of-the-art baselines in both objective and subjective evaluations, achieving significantly lower computational cost and better generalization.

语音增强是提高现实世界嘈杂环境中语音清晰度和质量的关键。然而，现有方法往往存在计算复杂度高、上下文建模受限、对语音干扰鲁棒性差等问题。为了解决这些挑战，我们提出了一个轻量级的框架来实现高效和鲁棒的语音增强。我们的方法将状态空间模型（SSM）与深度可分离卷积相结合，将有效的远程时间建模与精确的局部特征提取相结合。我们引入了一种听觉启发频谱压缩器（AISC），它模仿人耳的非线性频率分辨率，选择性地保留感知相关信息，同时显着减少计算冗余。此外，采用空间金字塔池（ASPP）模块捕获多尺度环境，使模型能够适应不同的噪声模式。为了解决人声干扰的难题，我们提出了一种增强特征可判别性的分类器损失，以有效地抑制竞争的人声干扰。在多个数据集上的实验表明，我们的模型在客观和主观评估方面都优于最先进的基线，实现了显着降低的计算成本和更好的泛化。

{"title":"Lightweight speech enhancement with state-space model and depthwise separable convolution","authors":"Chen Jiang , Dai Gao , Sirui Wang , Chengxuan Zou , Jie Liu","doi":"10.1016/j.dsp.2026.105987","DOIUrl":"10.1016/j.dsp.2026.105987","url":null,"abstract":"<div><div>Speech enhancement is critical for improving intelligibility and quality in real-world noisy environments. However, existing methods often suffer from high computational complexity, limited contextual modeling, and poor robustness to voice interference. To address these challenges, we propose a lightweight framework for efficient and robust speech enhancement. Our approach synergizes a State-Space Model (SSM) with depthwise separable convolutions, combining efficient long-range temporal modeling with precise local feature extraction. We introduce an Auditory Inspired Spectral Compressor (AISC) that mimics the non-linear frequency resolution of the human ear, selectively preserving perceptually relevant information while significantly reducing computational redundancy. Furthermore, an Atrous Spatial Pyramid Pooling (ASPP) module is employed to capture multi-scale context, allowing the model to adapt to diverse noise patterns. To tackle the difficult problem of human-voice interference, we propose a Classifier Loss that enforces feature discriminability to effectively suppress competing human-voice interference. Experiments on multiple datasets demonstrate that our model outperforms state-of-the-art baselines in both objective and subjective evaluations, achieving significantly lower computational cost and better generalization.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"174 ","pages":"Article 105987"},"PeriodicalIF":3.0,"publicationDate":"2026-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146174819","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

RFPA-keystone transform: A search-free coherent integration method for random frequency and PRI agile radar RFPA-keystone变换：随机频率和PRI敏捷雷达的无搜索相干积分方法

IF 3 3区工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

Digital Signal Processing

Pub Date : 2026-04-15 Epub Date: 2026-01-30 DOI: 10.1016/j.dsp.2026.105969

Yiheng Liu , Hua Zhang , Xuemei Wang , Qinghai Dong , Xiaode Lyu

Random frequency and pulse repetition interval agile (RFPA) signals are valued for their low probability of intercept (LPI) performance, yet their effective exploitation hinges on efficient coherent integration. Existing methods for RFPA coherent integration, however, typically rely on exhaustive parameter searches, leading to an inherent difficulty in balancing detection performance with computational efficiency. To overcome this limitation, we propose an RFPA-Keystone transform (RFPA-KT) algorithm. The algorithm first performs a search-free phase compensation to correct hopping-induced phase offsets without the need for exhaustive range search, then mitigates Doppler ambiguity through a pre-compensation framework, and finally employs the resampling KT to correct range cell migration (RCM) and achieve signal coherence. Both simulations and semi-physical experiments show that the proposed method achieves a probability of detection (Pd) approaching the maximum-likelihood (ML) performance benchmark under varying noise levels and parameter agility conditions, while significantly reducing the complexity from

O (N_{r} M_{v} M)

to

O (N_{r} M_{v} lo g_{2} M)

. These advantages highlight its potential as an efficient and robust solution for real-time coherent integration in RFPA radar systems.

随机频率和脉冲重复间隔敏捷（RFPA）信号因其低截获概率（LPI）性能而受到重视，但其有效利用取决于有效的相干积分。然而，现有的RFPA相干积分方法通常依赖于穷举参数搜索，导致在平衡检测性能和计算效率方面存在固有的困难。为了克服这一限制，我们提出了一种RFPA-Keystone变换（RFPA-KT）算法。该算法首先进行无搜索相位补偿来校正跳频引起的相位偏移，而不需要进行详尽的距离搜索，然后通过预补偿框架减轻多普勒模糊，最后采用重采样KT来校正距离单元迁移（RCM）并实现信号相干性。仿真和半物理实验表明，在不同噪声水平和参数敏捷性条件下，该方法的检测概率（Pd）接近最大似然（ML）性能基准，同时将复杂度从0 （NrMvM）显著降低到O（NrMvlog2M）。这些优点突出了它作为RFPA雷达系统中实时相干集成的高效鲁棒解决方案的潜力。

{"title":"RFPA-keystone transform: A search-free coherent integration method for random frequency and PRI agile radar","authors":"Yiheng Liu , Hua Zhang , Xuemei Wang , Qinghai Dong , Xiaode Lyu","doi":"10.1016/j.dsp.2026.105969","DOIUrl":"10.1016/j.dsp.2026.105969","url":null,"abstract":"<div><div>Random frequency and pulse repetition interval agile (RFPA) signals are valued for their low probability of intercept (LPI) performance, yet their effective exploitation hinges on efficient coherent integration. Existing methods for RFPA coherent integration, however, typically rely on exhaustive parameter searches, leading to an inherent difficulty in balancing detection performance with computational efficiency. To overcome this limitation, we propose an RFPA-Keystone transform (RFPA-KT) algorithm. The algorithm first performs a search-free phase compensation to correct hopping-induced phase offsets without the need for exhaustive range search, then mitigates Doppler ambiguity through a pre-compensation framework, and finally employs the resampling KT to correct range cell migration (RCM) and achieve signal coherence. Both simulations and semi-physical experiments show that the proposed method achieves a probability of detection (Pd) approaching the maximum-likelihood (ML) performance benchmark under varying noise levels and parameter agility conditions, while significantly reducing the complexity from <span><math><mrow><mi>O</mi><mo>(</mo><mrow><msub><mi>N</mi><mi>r</mi></msub><msub><mi>M</mi><mi>v</mi></msub><mi>M</mi></mrow><mo>)</mo></mrow></math></span> to <span><math><mrow><mi>O</mi><mo>(</mo><mrow><msub><mi>N</mi><mi>r</mi></msub><msub><mi>M</mi><mi>v</mi></msub><mtext>lo</mtext><msub><mi>g</mi><mn>2</mn></msub><mi>M</mi></mrow><mo>)</mo></mrow></math></span>. These advantages highlight its potential as an efficient and robust solution for real-time coherent integration in RFPA radar systems.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"174 ","pages":"Article 105969"},"PeriodicalIF":3.0,"publicationDate":"2026-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146174906","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

Distributed multi-Sensor multi-Target track matching algorithm based on LMB filter 基于LMB滤波器的分布式多传感器多目标航迹匹配算法

IF 3 3区工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

Digital Signal Processing

Pub Date : 2026-04-15 Epub Date: 2026-01-29 DOI: 10.1016/j.dsp.2026.105953

Kuiwu Wang , Qin Zhang , Xiaolong Hu , Pengfei Wan , Zhenlu Jin

This paper delves into the label matching problem within the Label Multi-Bernoulli framework for multi-target tracking tasks under a distributed multi-sensor system, emphasizing its pivotal role in the domain of multi-sensor multi-target tracking. The paper elucidates that within the LMB fusion process, label matching and data fusion can be effectively addressed as distinct, independent stages, thereby enhancing system modularity and processing efficiency. Regarding the innovation in fusion strategies, this paper proposes a pruning and merging approach based on the dual correlation between Gaussian component distance and motion direction. By precisely identifying and consolidating redundant information that potentially signifies the same target, this method not only optimizes fusion outcomes but also mitigates the computational burden on the sensor network. To tackle the label matching challenge, this paper devises a statistic rooted in label history similarity for two prevalent communication protocols. This statistic comprehensively considers the survival history of labels, offering a more reliable criterion for assessing label matching quality. Furthermore, to address the global label matching problem, this paper introduces the genetic algorithm as an intelligent optimization tool. Leveraging the iterative search mechanism of the genetic algorithm, this paper achieves optimal or near-optimal solutions for global label matching, significantly boosting the overall system performance. Simulation results demonstrate that the method exhibits strong performance in complex environments characterized by reduced detection probabilities and dense clutter, showcasing robustness and adaptability.

本文研究了分布式多传感器系统下多目标跟踪任务的标签多伯努利框架下的标签匹配问题，强调了其在多传感器多目标跟踪领域中的关键作用。本文阐述了在LMB融合过程中，标签匹配和数据融合可以作为不同的、独立的阶段进行有效处理，从而提高系统的模块化和处理效率。在融合策略方面，本文提出了一种基于高斯分量距离与运动方向对偶相关性的剪枝合并方法。通过精确识别和整合可能表示相同目标的冗余信息，该方法不仅优化了融合结果，还减轻了传感器网络的计算负担。为了解决标签匹配问题，本文设计了一种基于标签历史相似度的统计方法，用于两种流行的通信协议。该统计综合考虑了标签的生存历史，为评估标签匹配质量提供了更可靠的标准。此外，为了解决全局标签匹配问题，本文引入了遗传算法作为智能优化工具。利用遗传算法的迭代搜索机制，实现了全局标签匹配的最优或近最优解，显著提高了系统的整体性能。仿真结果表明，该方法在检测概率低、杂波密集的复杂环境中表现出较强的性能，具有鲁棒性和自适应性。

{"title":"Distributed multi-Sensor multi-Target track matching algorithm based on LMB filter","authors":"Kuiwu Wang , Qin Zhang , Xiaolong Hu , Pengfei Wan , Zhenlu Jin","doi":"10.1016/j.dsp.2026.105953","DOIUrl":"10.1016/j.dsp.2026.105953","url":null,"abstract":"<div><div>This paper delves into the label matching problem within the Label Multi-Bernoulli framework for multi-target tracking tasks under a distributed multi-sensor system, emphasizing its pivotal role in the domain of multi-sensor multi-target tracking. The paper elucidates that within the LMB fusion process, label matching and data fusion can be effectively addressed as distinct, independent stages, thereby enhancing system modularity and processing efficiency. Regarding the innovation in fusion strategies, this paper proposes a pruning and merging approach based on the dual correlation between Gaussian component distance and motion direction. By precisely identifying and consolidating redundant information that potentially signifies the same target, this method not only optimizes fusion outcomes but also mitigates the computational burden on the sensor network. To tackle the label matching challenge, this paper devises a statistic rooted in label history similarity for two prevalent communication protocols. This statistic comprehensively considers the survival history of labels, offering a more reliable criterion for assessing label matching quality. Furthermore, to address the global label matching problem, this paper introduces the genetic algorithm as an intelligent optimization tool. Leveraging the iterative search mechanism of the genetic algorithm, this paper achieves optimal or near-optimal solutions for global label matching, significantly boosting the overall system performance. Simulation results demonstrate that the method exhibits strong performance in complex environments characterized by reduced detection probabilities and dense clutter, showcasing robustness and adaptability.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"174 ","pages":"Article 105953"},"PeriodicalIF":3.0,"publicationDate":"2026-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146174912","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

End-to-end target speaker speech recognition with voice activity detection fusion 端到端目标说话人语音识别与语音活动检测融合

IF 3 3区工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

Digital Signal Processing

Pub Date : 2026-04-15 Epub Date: 2026-01-29 DOI: 10.1016/j.dsp.2026.105966

Zhentao Lin , Bi Zeng , Song Wen , Zihao Chen , Huiting Hu

Traditional Voice Activity Detection (VAD)-based systems frequently encounter challenges in handling speaker overlap within multi-speaker environments, particularly in the context of target speaker Automatic Speech Recognition (ASR). This difficulty arises predominantly from the limitations of front-end VAD modules, which are independently trained to distinguish noise from speech but often introduce insertion and deletion errors, adversely affecting the overall performance of the ASR system. To address this coupling deficiency, we propose an End-to-End Streaming Personal target speaker ASR (SP-ASR) framework that achieves fusion of VAD and ASR components in a streaming style. Our architecture introduces two key innovations: Initially, a Streaming Personal VAD (SP-VAD) module functions as a neural gatekeeper, segmenting audio streams while emphasizing target speaker characteristics through its Contextual Attention and Target Speaker Attention (CA-TSA) mechanism. Subsequently, a Streaming Mask-based ASR (SM-ASR) model is employed, which is integrated with SP-VAD and fine-tuned using both coarse-grained and fine-grained speaker information to extract speaker-specific transcriptions. Our experiments reveal a remarkable reduction in the concatenated target-speaker Word Error Rate (ctWER), showcasing the superiority of the End-to-End SP-ASR fusion system over conventional ASR systems, especially under conditions with significant speech overlap and noise.

传统的基于语音活动检测（VAD）的系统在处理多说话人环境中的说话人重叠时经常遇到挑战，特别是在目标说话人自动语音识别（ASR）的背景下。这种困难主要来自前端VAD模块的局限性，这些模块被独立训练以区分噪声和语音，但经常引入插入和删除错误，对ASR系统的整体性能产生不利影响。为了解决这一耦合缺陷，我们提出了一个端到端流式个人目标扬声器ASR （SP-ASR）框架，该框架以流方式实现了VAD和ASR组件的融合。我们的架构引入了两个关键的创新：最初，一个流式个人VAD （SP-VAD）模块作为一个神经看门人，分割音频流，同时通过其上下文注意和目标说话人注意（CA-TSA）机制强调目标说话人的特征。随后，采用基于流掩码的ASR （SM-ASR）模型，该模型与SP-VAD集成，并使用粗粒度和细粒度的说话人信息进行微调，以提取说话人特定的转录。我们的实验显示，连接目标说话人的单词错误率（ctWER）显著降低，展示了端到端SP-ASR融合系统比传统ASR系统的优势，特别是在语音重叠和噪声严重的情况下。

{"title":"End-to-end target speaker speech recognition with voice activity detection fusion","authors":"Zhentao Lin , Bi Zeng , Song Wen , Zihao Chen , Huiting Hu","doi":"10.1016/j.dsp.2026.105966","DOIUrl":"10.1016/j.dsp.2026.105966","url":null,"abstract":"<div><div>Traditional Voice Activity Detection (VAD)-based systems frequently encounter challenges in handling speaker overlap within multi-speaker environments, particularly in the context of target speaker Automatic Speech Recognition (ASR). This difficulty arises predominantly from the limitations of front-end VAD modules, which are independently trained to distinguish noise from speech but often introduce <em>insertion and deletion errors</em>, adversely affecting the overall performance of the ASR system. To address this coupling deficiency, we propose an End-to-End Streaming Personal target speaker ASR (SP-ASR) framework that achieves fusion of VAD and ASR components in a streaming style. Our architecture introduces two key innovations: Initially, a Streaming Personal VAD (SP-VAD) module functions as a neural gatekeeper, segmenting audio streams while emphasizing target speaker characteristics through its Contextual Attention and Target Speaker Attention (CA-TSA) mechanism. Subsequently, a Streaming Mask-based ASR (SM-ASR) model is employed, which is integrated with SP-VAD and fine-tuned using both coarse-grained and fine-grained speaker information to extract speaker-specific transcriptions. Our experiments reveal a remarkable reduction in the concatenated target-speaker Word Error Rate (ctWER), showcasing the superiority of the End-to-End SP-ASR fusion system over conventional ASR systems, especially under conditions with significant speech overlap and noise.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"174 ","pages":"Article 105966"},"PeriodicalIF":3.0,"publicationDate":"2026-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146081607","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

AIDENet: Adaptive illumination decomposition enhancement network for robust low-light image restoration AIDENet：用于弱光图像鲁棒恢复的自适应光照分解增强网络

IF 3 3区工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

Digital Signal Processing

Pub Date : 2026-04-15 Epub Date: 2026-02-03 DOI: 10.1016/j.dsp.2026.105980

Dong Guo, Yi Han, Yang Cui

Low-light imaging is constrained by insufficient scene illumination, often accompanied by issues such as brightness attenuation, contrast reduction, noise amplification, and color bias, which severely affect subsequent visual tasks. Traditional Retinex methods rely on handcrafted priors and struggle to adapt to real-world sudden illumination changes. Although existing deep models have made progress in brightness enhancement, they lack coordination in estimating non-uniform illumination, removing content-coupled noise, reconstructing multi-scale details, and maintaining color fidelity, resulting in lost details in dark areas or residual artifacts. This paper proposes the Adaptive Illumination Decomposition Enhancement Network (AIDENet), which integrates decomposition, denoising, restoration, and color correction to address these challenges simultaneously. The network dynamically estimates and corrects spatially varying illumination through a lightweight illumination decomposition branch; introduces a noise-aware feature learning module to adaptively suppress noise and preserve texture based on content; designs a hierarchical detail recovery branch to fuse salient structures under multi-receptive fields; and constructs a multi-modal color correction sub-network to achieve natural color bias correction using joint global-local statistics. All modules contain only 0.27 M parameters and can be jointly optimized end-to-end. Experiments on LOLv1, LOLv2_real, and LOLv2_synthetic demonstrate that AIDENet outperforms current mainstream methods in terms of PSNR, SSIM, and LPIPS metrics. The visual results show clearer dark area details, lower noise levels, and more accurate color reproduction, verifying its effectiveness in robust low-light image recovery.

低照度成像受场景照度不足的制约，往往伴随着亮度衰减、对比度降低、噪声放大、色差等问题，严重影响后续的视觉任务。传统的Retinex方法依赖于手工制作的先验，难以适应现实世界的突然照明变化。现有的深度模型虽然在亮度增强方面取得了一定的进展，但在估计非均匀照度、去除内容耦合噪声、重建多尺度细节、保持色彩保真度等方面缺乏协调性，导致暗区细节丢失或残留伪影。本文提出了自适应光照分解增强网络（AIDENet），该网络集成了分解、去噪、恢复和色彩校正来同时解决这些挑战。该网络通过一个轻量级的光照分解分支来动态估计和校正空间变化的光照；引入噪声感知特征学习模块，实现基于内容的自适应噪声抑制和纹理保存；设计层次细节恢复分支，融合多感受野下的突出结构；构建了多模态色彩校正子网络，利用全局-局部联合统计实现自然色彩偏差校正。所有模块只有0.27 M个参数，可以端到端联合优化。在LOLv1、LOLv2_real和LOLv2_synthetic上的实验表明，AIDENet在PSNR、SSIM和LPIPS指标方面优于当前主流方法。视觉结果显示出更清晰的暗区细节，更低的噪声水平，更准确的色彩再现，验证了其在鲁棒低光图像恢复中的有效性。

{"title":"AIDENet: Adaptive illumination decomposition enhancement network for robust low-light image restoration","authors":"Dong Guo, Yi Han, Yang Cui","doi":"10.1016/j.dsp.2026.105980","DOIUrl":"10.1016/j.dsp.2026.105980","url":null,"abstract":"<div><div>Low-light imaging is constrained by insufficient scene illumination, often accompanied by issues such as brightness attenuation, contrast reduction, noise amplification, and color bias, which severely affect subsequent visual tasks. Traditional Retinex methods rely on handcrafted priors and struggle to adapt to real-world sudden illumination changes. Although existing deep models have made progress in brightness enhancement, they lack coordination in estimating non-uniform illumination, removing content-coupled noise, reconstructing multi-scale details, and maintaining color fidelity, resulting in lost details in dark areas or residual artifacts. This paper proposes the Adaptive Illumination Decomposition Enhancement Network (AIDENet), which integrates decomposition, denoising, restoration, and color correction to address these challenges simultaneously. The network dynamically estimates and corrects spatially varying illumination through a lightweight illumination decomposition branch; introduces a noise-aware feature learning module to adaptively suppress noise and preserve texture based on content; designs a hierarchical detail recovery branch to fuse salient structures under multi-receptive fields; and constructs a multi-modal color correction sub-network to achieve natural color bias correction using joint global-local statistics. All modules contain only 0.27 M parameters and can be jointly optimized end-to-end. Experiments on LOLv1, LOLv2_real, and LOLv2_synthetic demonstrate that AIDENet outperforms current mainstream methods in terms of PSNR, SSIM, and LPIPS metrics. The visual results show clearer dark area details, lower noise levels, and more accurate color reproduction, verifying its effectiveness in robust low-light image recovery.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"174 ","pages":"Article 105980"},"PeriodicalIF":3.0,"publicationDate":"2026-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146174813","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

Zero-reference illumination estimation model for image enhancement in underground mines 地下矿山图像增强的零参考照度估计模型

IF 3 3区工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

Digital Signal Processing

Pub Date : 2026-04-15 Epub Date: 2026-02-06 DOI: 10.1016/j.dsp.2026.105982

Xingyu Gong, Yu Guan, Yi Yang, Rongkun Jiang, Na Li

Due to artificial lighting and suspended dust, images captured in underground mines often suffer from low illumination, uneven lighting, and significant noise, which severely degrade image quality and hinder downstream vision tasks. To address these challenges, we propose a zero-reference illumination estimation model called AM-ZIE for real-world image enhancement under non-uniform lighting conditions. Specifically, the attention-guided illumination estimation network (AIE-Net) employs cascaded channel and spatial attention modules to extract illumination features and suppress noise. The multi-scale light adjustment module (MLA-Mod) performs pixel-wise illumination correction across varying receptive fields to achieve brightness enhancement and detail preservation. The segment-compensated exposure loss is designed to adjust image brightness adaptively according to the illumination distribution, enhancing dark regions while suppressing overexposed areas. Extensive experiments conducted on three underground mine datasets and two public datasets demonstrate that AM-ZIE achieves strong generalization and cross-scene adaptability, excelling in brightness balancing, noise suppression, and detail preservation. Our method outperforms eight unsupervised baselines in both no-reference (NIQE, BRISQUE) and full-reference (PSNR, SSIM) metrics, and improves the accuracy of mining personnel detection by 16.01%, highlighting its potential for industrial safety monitoring. The code will be released at https://github.com/Lucky-Guan/AM-ZIE.

由于人工照明和悬浮粉尘的影响，地下矿山拍摄的图像照度低、光照不均匀、噪声明显，严重降低了图像质量，阻碍了下游的视觉任务。为了解决这些挑战，我们提出了一种称为AM-ZIE的零参考照度估计模型，用于非均匀照明条件下的真实图像增强。其中，注意引导照明估计网络（AIE-Net）采用级联通道和空间注意模块提取照明特征并抑制噪声。多尺度光调节模块（MLA-Mod）在不同的接受域上执行逐像素的照明校正，以实现亮度增强和细节保留。部分补偿曝光损失的设计是根据光照分布自适应调整图像亮度，增强暗区，抑制过度曝光区域。在3个地下矿山数据集和2个公开数据集上进行的大量实验表明，AM-ZIE具有较强的泛化能力和跨场景适应性，在亮度平衡、噪声抑制和细节保留方面表现优异。该方法在无参考基准（NIQE， BRISQUE）和全参考基准（PSNR， SSIM）指标上均优于8条无监督基准，并将采矿人员检测的准确率提高了16.01%，突出了其在工业安全监测中的潜力。代码将在https://github.com/Lucky-Guan/AM-ZIE上发布。

{"title":"Zero-reference illumination estimation model for image enhancement in underground mines","authors":"Xingyu Gong, Yu Guan, Yi Yang, Rongkun Jiang, Na Li","doi":"10.1016/j.dsp.2026.105982","DOIUrl":"10.1016/j.dsp.2026.105982","url":null,"abstract":"<div><div>Due to artificial lighting and suspended dust, images captured in underground mines often suffer from low illumination, uneven lighting, and significant noise, which severely degrade image quality and hinder downstream vision tasks. To address these challenges, we propose a zero-reference illumination estimation model called AM-ZIE for real-world image enhancement under non-uniform lighting conditions. Specifically, the attention-guided illumination estimation network (AIE-Net) employs cascaded channel and spatial attention modules to extract illumination features and suppress noise. The multi-scale light adjustment module (MLA-Mod) performs pixel-wise illumination correction across varying receptive fields to achieve brightness enhancement and detail preservation. The segment-compensated exposure loss is designed to adjust image brightness adaptively according to the illumination distribution, enhancing dark regions while suppressing overexposed areas. Extensive experiments conducted on three underground mine datasets and two public datasets demonstrate that AM-ZIE achieves strong generalization and cross-scene adaptability, excelling in brightness balancing, noise suppression, and detail preservation. Our method outperforms eight unsupervised baselines in both no-reference (NIQE, BRISQUE) and full-reference (PSNR, SSIM) metrics, and improves the accuracy of mining personnel detection by 16.01%, highlighting its potential for industrial safety monitoring. The code will be released at <span><span>https://github.com/Lucky-Guan/AM-ZIE</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"174 ","pages":"Article 105982"},"PeriodicalIF":3.0,"publicationDate":"2026-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146174814","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

No-reference magnetic resonance image quality assessment via local-global feature integration 基于局部-全局特征集成的无参考磁共振图像质量评价

IF 3 3区工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

Digital Signal Processing

Pub Date : 2026-04-15 Epub Date: 2026-02-05 DOI: 10.1016/j.dsp.2026.105985

Xuejin Wang , Zhenhui Zhong , Jinbin Hu

The accuracy of Magnetic Resonance (MR) image quality assessment (IQA) directly impacts the time that radiologists spend on image acquisition and interpretation, playing a crucial role in disease diagnosis. However, existing IQA methods evaluate the quality either from the global or local perspective, failing to achieve the comprehensive and accurate quality evaluation of MR images. To consider both local lesion detail distortion and global anatomical structure while leveraging the prior knowledge of vision-language models, we propose a novel no-reference MR IQA method, which is composed of a dual-branch network based on the Contrastive Language-Image Pre-Training (CLIP) model, namely a local distortion perception network and a global quality perception network. For the local branch, an edge-based preprocessing module is employed to enhance the local details of the MR image, facilitating the differentiation between high-quality and low-quality images. Additionally, to fully exploit both text and image features, a multi-modal feature fusion module is introduced to integrate semantic features from different modalities, thereby obtaining the semantic quality. Finally, the overall quality score is derived by integrating global quality, local quality, and semantic quality. Extensive experiments conducted on the publicly available datasets demonstrate that the proposed method outperforms existing MR IQA approaches.

磁共振（MR）图像质量评估（IQA）的准确性直接影响到放射科医生在图像采集和解释上花费的时间，在疾病诊断中起着至关重要的作用。然而，现有的IQA方法要么从全局角度评估质量，要么从局部角度评估质量，无法实现对MR图像质量的全面准确评估。为了在充分利用视觉语言模型的先验知识的前提下兼顾病灶局部细节畸变和全局解剖结构，我们提出了一种新的无参考MR IQA方法，该方法由基于对比语言图像预训练（CLIP）模型的双分支网络组成，即局部畸变感知网络和全局质量感知网络。对于局部分支，采用基于边缘的预处理模块增强MR图像的局部细节，便于区分高质量和低质量图像。此外，为了充分利用文本和图像的特征，引入了多模态特征融合模块，将不同模态的语义特征进行融合，从而获得语义质量。最后，综合全局质量、局部质量和语义质量得出总体质量分数。在公开数据集上进行的大量实验表明，所提出的方法优于现有的MR IQA方法。

{"title":"No-reference magnetic resonance image quality assessment via local-global feature integration","authors":"Xuejin Wang , Zhenhui Zhong , Jinbin Hu","doi":"10.1016/j.dsp.2026.105985","DOIUrl":"10.1016/j.dsp.2026.105985","url":null,"abstract":"<div><div>The accuracy of Magnetic Resonance (MR) image quality assessment (IQA) directly impacts the time that radiologists spend on image acquisition and interpretation, playing a crucial role in disease diagnosis. However, existing IQA methods evaluate the quality either from the global or local perspective, failing to achieve the comprehensive and accurate quality evaluation of MR images. To consider both local lesion detail distortion and global anatomical structure while leveraging the prior knowledge of vision-language models, we propose a novel no-reference MR IQA method, which is composed of a dual-branch network based on the Contrastive Language-Image Pre-Training (CLIP) model, namely a local distortion perception network and a global quality perception network. For the local branch, an edge-based preprocessing module is employed to enhance the local details of the MR image, facilitating the differentiation between high-quality and low-quality images. Additionally, to fully exploit both text and image features, a multi-modal feature fusion module is introduced to integrate semantic features from different modalities, thereby obtaining the semantic quality. Finally, the overall quality score is derived by integrating global quality, local quality, and semantic quality. Extensive experiments conducted on the publicly available datasets demonstrate that the proposed method outperforms existing MR IQA approaches.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"174 ","pages":"Article 105985"},"PeriodicalIF":3.0,"publicationDate":"2026-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146174818","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