IEEE Signal Processing Letters最新文献_第10页

TJCMNet: An Efficient Vision-Text Joint Identity Clues Mining Network for Visible-Infrared Person Re-Identification TJCMNet：一种高效的视觉-文本联合身份线索挖掘网络，用于可见-红外人物再识别

IF 3.2 2区工程技术

IEEE Signal Processing Letters Pub Date : 2025-04-01 DOI: 10.1109/LSP.2025.3556784

ZhuXuan Cheng;ZhiJia Zhang;Huijie Fan;XingQi Na

{"title":"TJCMNet: An Efficient Vision-Text Joint Identity Clues Mining Network for Visible-Infrared Person Re-Identification","authors":"ZhuXuan Cheng;ZhiJia Zhang;Huijie Fan;XingQi Na","doi":"10.1109/LSP.2025.3556784","DOIUrl":"https://doi.org/10.1109/LSP.2025.3556784","url":null,"abstract":"Retrieving images for Visible-Infrared Person Re-identification task is challenging, because of the huge modality discrepancy caused by the different imaging principle of RGB and infrared cameras. Existing approaches rely on seeking distinctive information within unified visual feature space, ignoring the stable identity information brought by textual description. To overcome these problems, this letter propose a novel Text-vision Joint Clue Mining (TJCM) network to aggregate vision and text features, then distill the joint knowledge for enhancing the modality-shared branch. Specifically, we first extract modality-shared and textual features using a parameter-shared vision encoder and a text encoder. Then, a text-vision co-refinement module is proposed to refine the implicit information within vision feature and text feature, then aggregate them into joint feature. Finally, introduce the heterogeneous distillation alignment loss provides enhancement for modality-shared feature through joint knowledge distillation at feature-level and logit-level. Our TJCMNet achieves significant improvements over the state-of-the-art methods on three mainstream datasets.","PeriodicalId":13154,"journal":{"name":"IEEE Signal Processing Letters","volume":"32 ","pages":"1615-1619"},"PeriodicalIF":3.2,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143871028","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Feature Bank-Guided Reconstruction for Anomaly Detection 特征库引导下的异常检测重建

IF 3.2 2区工程技术

IEEE Signal Processing Letters Pub Date : 2025-03-31 DOI: 10.1109/LSP.2025.3555544

Sihan He;Tao Zhang;Wei Song;Hongbin Yu

引用次数: 0

Secure Wireless Communications via Frequency Diverse Arrays 通过频变阵列的安全无线通信

IF 3.2 2区工程技术

IEEE Signal Processing Letters Pub Date : 2025-03-30 DOI: 10.1109/LSP.2025.3575330

Zhenqiao Cheng;Chongjun Ouyang;Xingqi Zhang

引用次数: 0

Reducing Mutual Coupling in Circular Motion-Based DOA Estimation: An Iterative Array Configuration Method 基于圆运动的DOA估计中减少相互耦合的迭代阵列配置方法

IF 3.2 2区工程技术

IEEE Signal Processing Letters Pub Date : 2025-03-29 DOI: 10.1109/LSP.2025.3575018

Jianhong Chu;Zhi Zhang;Chengjin Kang;Min Liang;Zhe Fu

{"title":"Reducing Mutual Coupling in Circular Motion-Based DOA Estimation: An Iterative Array Configuration Method","authors":"Jianhong Chu;Zhi Zhang;Chengjin Kang;Min Liang;Zhe Fu","doi":"10.1109/LSP.2025.3575018","DOIUrl":"https://doi.org/10.1109/LSP.2025.3575018","url":null,"abstract":"Mutual coupling (MC) has a severe effect on the direction-of-arrival (DOA) estimation and even leads to estimation failures. The emerging array motion techniques, which construct flexible virtual arrays with larger apertures by moving an initial array and combining phase-corrected signals sampled at designed time delays, have shown potential in mitigating the effect of MC by constructing virtual sensors. However, the virtual array remains susceptible to MC. To handle this problem, in this letter, a general synthetic signal model with MC is derived and an iterative method on the array configuration is proposed to form a virtual array with less MC for the circular motion-based DOA estimation. Specifically, a virtual uniform concentric circular array (VUCCA) is constructed by synthesizing all the observations at different time delays with uniform spacing, and the equation between MC matrix of the VUCCA and the initial array is derived, proving that the MC of VUCCA is solely related to the configuration of initial array. Thus, an iterative method is proposed to yield an optimal configuration with less MC by minimizing the norm close with MC. A higher degree-of-freedom and better estimation accuracy are achieved effectively.","PeriodicalId":13154,"journal":{"name":"IEEE Signal Processing Letters","volume":"32 ","pages":"2314-2318"},"PeriodicalIF":3.2,"publicationDate":"2025-03-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144308390","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Motion Intent Analysis-Based Full-Frame Video Stabilization 基于动作意图分析的全帧视频防抖

IF 3.2 2区工程技术

IEEE Signal Processing Letters Pub Date : 2025-03-28 DOI: 10.1109/LSP.2025.3555492

Yu Zhang;Pengcheng Guo;Moran Ju;Qing Hu

{"title":"Motion Intent Analysis-Based Full-Frame Video Stabilization","authors":"Yu Zhang;Pengcheng Guo;Moran Ju;Qing Hu","doi":"10.1109/LSP.2025.3555492","DOIUrl":"https://doi.org/10.1109/LSP.2025.3555492","url":null,"abstract":"Video stabilization aims to eliminate random jitter in video sequences, but most methods result in stabilized video with degraded resolution and content loss. In this letter, we propose a full-frame video stabilization algorithm based on motion intent analysis. The algorithm consists of three main steps: motion estimation, motion smoothing, and video completion. First, robust keypoints are extracted using the improved SuperPoint network and refined with the suppression via square covering (SSC) algorithm to obtain stable and reliable keypoints. Then, the Lucas-Kanade algorithm is applied for motion estimation of inter-frame matched feature points. Second, motion smoothing is achieved using the Kalman filtering algorithm to remove the high-frequency jitter component from the trajectory, and motion compensation is applied to the original video sequence to generate a stable image sequence. Finally, to preserve the original video resolution, we propose a video completion method based on motion intent analysis. Experimental results demonstrate that our method achieves higher stability while maintaining the original video resolution compared to the current state-of-the-art video stabilization algorithms.","PeriodicalId":13154,"journal":{"name":"IEEE Signal Processing Letters","volume":"32 ","pages":"1685-1689"},"PeriodicalIF":3.2,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143913472","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Transient Segment Detection Based on Energy Spectrum Coefficient Entropy Posterior Probability Density 基于能谱系数熵后验概率密度的暂态片段检测

IF 3.2 2区工程技术

IEEE Signal Processing Letters Pub Date : 2025-03-27 DOI: 10.1109/LSP.2025.3555421

Jingbo Zhang;Rongwen Lin;Qingshuo Liu

{"title":"Transient Segment Detection Based on Energy Spectrum Coefficient Entropy Posterior Probability Density","authors":"Jingbo Zhang;Rongwen Lin;Qingshuo Liu","doi":"10.1109/LSP.2025.3555421","DOIUrl":"https://doi.org/10.1109/LSP.2025.3555421","url":null,"abstract":"In wireless communication devices, the power envelope during the transient segment of energy-limited pulse signals effectively captures the physical characteristics of the radio frequency circuit. This unique signature can serve as a radio frequency fingerprinting, enhancing the security of the wireless communication system. A key issue in this context is accurately detecting transient signals. The existing detection algorithms can only detect the start time of transient segments, making them unable to characterize the complete transient segments. This study proposes a transient segment detection algorithm based on energy spectral coefficient entropy posterior probability density (ESCE-PPD), which can simultaneously estimate the start and end time of transient segments without relying on prior information. The effectiveness of the proposed algorithm is verified using an open source Bluetooth dataset, and its performance is compared with existing algorithms. The results demonstrate that the ESCE-PPD algorithm adds the capability to detect the end time of transient segments without increasing computational complexity and reducing anti-noise performance.","PeriodicalId":13154,"journal":{"name":"IEEE Signal Processing Letters","volume":"32 ","pages":"1505-1509"},"PeriodicalIF":3.2,"publicationDate":"2025-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143830478","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Alternating Offer-Based Payment Allocation for Privacy Non-Disclosure in Federated Learning 联邦学习中基于交替出价的隐私保密支付分配

IF 3.2 2区工程技术

IEEE Signal Processing Letters Pub Date : 2025-03-27 DOI: 10.1109/LSP.2025.3555386

Suyeon Jin;Chaeyeon Cha;Hyunggon Park

引用次数: 0

Geometry-Aware RWKV for Heterogeneous Light Field Spatial Super-Resolution 基于几何感知的非均匀光场空间超分辨率RWKV

IF 3.2 2区工程技术

IEEE Signal Processing Letters Pub Date : 2025-03-27 DOI: 10.1109/LSP.2025.3555445

Zean Chen;Yeyao Chen;Linwei Zhu;Haiyong Xu;Gangyi Jiang

引用次数: 0

Watermark Removal Attack Against Text-to-Image Generative Model Watermarking 针对文本到图像生成模型水印的水印去除攻击

IF 3.2 2区工程技术

IEEE Signal Processing Letters Pub Date : 2025-03-26 DOI: 10.1109/LSP.2025.3554514

Zihan Yuan;Li Li;Zichi Wang;Jingyuan Jiang;Xinpeng Zhang

{"title":"Watermark Removal Attack Against Text-to-Image Generative Model Watermarking","authors":"Zihan Yuan;Li Li;Zichi Wang;Jingyuan Jiang;Xinpeng Zhang","doi":"10.1109/LSP.2025.3554514","DOIUrl":"https://doi.org/10.1109/LSP.2025.3554514","url":null,"abstract":"The artist's style can be quickly imitated by fine-tuning a text-to-image model using artist's artworks, which raises serious copyright concerns. Scholars have proposed many watermarking methods to protect the artists' copyright. To evaluate the security and enhance the performance of existing watermarking, this paper proposes a watermark removal attack for text-to-image generative model watermarking for the first time. This attack aims to invalidate watermarking designed to detect art theft mimicry in text-to-image models. In this method, a watermark recognition network and a watermark removal network are designed. The watermark recognition network identifies whether an artwork contains watermark, and the watermark removal network is used to remove it. Consequently, text-to-image models fine-tuned with watermark-removed artworks can reproduce an artist's style while evading watermark detection. This makes the copyright authentication of artworks ineffective. Experiments show that the proposed attack can effectively remove watermarks, with watermark extraction accuracy dropping below 48.64%. Additionally, the images after watermark removal retain high similarity to the original images, with PSNR exceeding 27.96 and SSIM exceeding 0.92.","PeriodicalId":13154,"journal":{"name":"IEEE Signal Processing Letters","volume":"32 ","pages":"1470-1474"},"PeriodicalIF":3.2,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143809027","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Hybrid Multi-Class Token Vision Transformer Convolutional Network for DOA Estimation DOA估计的混合多类令牌视觉变换卷积网络

IF 3.2 2区工程技术

IEEE Signal Processing Letters Pub Date : 2025-03-26 DOI: 10.1109/LSP.2025.3573949

Yuxuan Xie;Aifei Liu;Xinyu Lu;Dufei Chong

引用次数: 0