IEEE Signal Processing Letters最新文献

Lightweight Spectral Super-Resolution Network for Hyperspectral Image Compression 用于高光谱图像压缩的轻型光谱超分辨率网络

IF 3.2 2区工程技术

IEEE Signal Processing Letters Pub Date : 2025-06-03 DOI: 10.1109/LSP.2025.3576177

Wei Zhang;Pengpeng Yu;Yueru Chen;Dingquan Li;Wen Gao

{"title":"Lightweight Spectral Super-Resolution Network for Hyperspectral Image Compression","authors":"Wei Zhang;Pengpeng Yu;Yueru Chen;Dingquan Li;Wen Gao","doi":"10.1109/LSP.2025.3576177","DOIUrl":"https://doi.org/10.1109/LSP.2025.3576177","url":null,"abstract":"The growing use of hyperspectral images demands efficient compression techniques to handle their extensive spectral data. However, current methods are constrained by their inability to adapt to high bit depth and effectively utilize the spectral characteristics, leading to suboptimal compression ratios. This paper presents a novel hyperspectral compression framework that employs a lightweight spectral super-resolution network to address these limitations. The proposed approach divides the hyperspectral image into two sub-images, comprising two distinct groups of bands: a base image consisting of anchor bands and a supplementary image comprising non-anchor bands. The base image is compressed losslessly using a conventional codec, thereby ensuring the preservation of essential information. In contrast, the supplementary image is compressed efficiently by overfitting a lightweight super-resolution network to predict the non-anchor bands during encoding. The optimized network parameters are encoded as side information to ensure high-quality spectral super-resolution during decoding. Experimental results on the ARAD hyperspectral image dataset demonstrate that our approach significantly outperforms state-of-the-art methods, effectively meeting the demand for efficient hyperspectral image compression while maintaining acceptable processing speeds.","PeriodicalId":13154,"journal":{"name":"IEEE Signal Processing Letters","volume":"32 ","pages":"2339-2343"},"PeriodicalIF":3.2,"publicationDate":"2025-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144308503","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

Dynamic Fusion for Generating High-Quality Labels in Low-Light Image Enhancement 低光图像增强中生成高质量标签的动态融合

IF 3.2 2区工程技术

IEEE Signal Processing Letters Pub Date : 2025-06-02 DOI: 10.1109/LSP.2025.3575608

Zhuo-Ming Du;Hong-An Li;Fei-long Han

{"title":"Dynamic Fusion for Generating High-Quality Labels in Low-Light Image Enhancement","authors":"Zhuo-Ming Du;Hong-An Li;Fei-long Han","doi":"10.1109/LSP.2025.3575608","DOIUrl":"https://doi.org/10.1109/LSP.2025.3575608","url":null,"abstract":"Generating high-quality labels is crucial for self-supervised learning in low-light conditions, where traditional enhancement methods often struggle to balance detail enhancement and color fidelity. This paper presents a traditional image fusion approach that dynamically combines Multi-Scale Retinex (MSR) and Adaptive Histogram Equalization (AHE) outputs with the original image using an adaptive weighting strategy. The primary goal is not to compete with state-of-the-art deep learning-based enhancement methods but to produce intermediate images that can serve as effective labels for training self-supervised models without requiring ground-truth datasets. By dynamically fusing MSR and AHE outputs with the original image using adaptive brightness and color weights, the method improves structural integrity while enhancing brightness and color consistency. Experiments on standard low-light datasets demonstrate significant improvements in PSNR and SSIM compared to traditional enhancement methods. However, a visual analysis of the generated labels reveals differences in color saturation when compared to ground truth, providing insights into designing a suitable loss function for future self-supervised learning applications. It is important to note that this work does not include experiments or methods related to self-supervised learning itself; instead, it focuses on preparing high-quality labels for such approaches. Additionally, our method strikes a balance between computational efficiency and visual quality, making it suitable for real-time applications and paving the way for more robust and versatile learning frameworks.","PeriodicalId":13154,"journal":{"name":"IEEE Signal Processing Letters","volume":"32 ","pages":"2324-2328"},"PeriodicalIF":3.2,"publicationDate":"2025-06-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144308508","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

Video Complicated-Information Extraction and Filtering Network for Weakly-Supervised Temporal Action Localization 面向弱监督时间动作定位的视频复杂信息提取与过滤网络

IF 3.2 2区工程技术

IEEE Signal Processing Letters Pub Date : 2025-06-02 DOI: 10.1109/LSP.2025.3575626

Jiaxuan Li;Tiancheng Ma;Xiaohui Yang;Lijun Yang;Chen Zheng

{"title":"Video Complicated-Information Extraction and Filtering Network for Weakly-Supervised Temporal Action Localization","authors":"Jiaxuan Li;Tiancheng Ma;Xiaohui Yang;Lijun Yang;Chen Zheng","doi":"10.1109/LSP.2025.3575626","DOIUrl":"https://doi.org/10.1109/LSP.2025.3575626","url":null,"abstract":"Weakly-supervised temporal action localiza- tion aims to identify action instances using only video-level labels, and localize the action position in untrimmed videos. Due to the temporal continuity of video data, most methods that use single scale convolution kernel cannot model against the characterization of video data effectively, and lead to a decrease in accuracy. However, simply using multi-scale features can introduce redundant information and noise, reducing model efficiency while also affecting the accurate judgement of the model during training process. To alleviate this problem, a video complicated-information extraction and filtering network (VCEF-Net) is proposed. It contains two main modules. The first multi-scale feature extraction module is developed to enrich the information that model received. The second pseudo-label filtering module inhibits redundant information interference. VCEF-Net introduces these two modules for achieving a better utilization of video information. Experiments tested on THUMOS14 and ActivityNet1.2 demonstrate better performances of the proposed VCEF-Net and validate its effectiveness.","PeriodicalId":13154,"journal":{"name":"IEEE Signal Processing Letters","volume":"32 ","pages":"2334-2338"},"PeriodicalIF":3.2,"publicationDate":"2025-06-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144308391","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

Sequential Number-Theoretic Optimization for High-Dimensional Channel Equalization in Coherent Optical PDM Systems 相干光PDM系统中高维信道均衡的序贯数论优化

IF 3.2 2区工程技术

IEEE Signal Processing Letters Pub Date : 2025-06-02 DOI: 10.1109/LSP.2025.3575611

Shuai Liu;Yangfan Xu;Xinwei Du

{"title":"Sequential Number-Theoretic Optimization for High-Dimensional Channel Equalization in Coherent Optical PDM Systems","authors":"Shuai Liu;Yangfan Xu;Xinwei Du","doi":"10.1109/LSP.2025.3575611","DOIUrl":"https://doi.org/10.1109/LSP.2025.3575611","url":null,"abstract":"Polarization-division multiplexing (PDM) in coherent optical communications enhances system capacity but is vulnerable to various channel distortions including transmitter and receiver in-phase/quadrature (IQ) mismatch, rotation of state of polarization (RSOP), frequency offset (FO) and phase noise (PN), which significantly degrade the system performance. To address these challenges, we propose a novel approach using sequential number-theoretic optimization (SNTO) for the joint estimation of these distortions. We further introduce a decision-aided scheme with a window-split structure to accurately track and compensate for time-varying RSOP and PN, thereby implementing signal detection. Through comprehensive mean squared error (MSE) and bit error rate (BER) analysis under different signal-to-noise ratio (SNR) conditions and varying RSOP speeds, our method demonstrates high precision and effectiveness. The SNTO-based algorithm maintains robust performance with superior estimation accuracy and resilience against ultra-fast RSOP. This work introduces an innovative solution for high-dimensional channel equalization in coherent optical PDM systems with not only effectiveness but also robustness.","PeriodicalId":13154,"journal":{"name":"IEEE Signal Processing Letters","volume":"32 ","pages":"2329-2333"},"PeriodicalIF":3.2,"publicationDate":"2025-06-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144308502","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

An Event-Triggered Hybrid Consensus Filter for Distributed Extended Object Tracking 一种用于分布式扩展对象跟踪的事件触发混合一致性过滤器

IF 3.2 2区工程技术

IEEE Signal Processing Letters Pub Date : 2025-04-29 DOI: 10.1109/LSP.2025.3565369

Runyan Lyu;Yunze Cai;Lixiu Yao

引用次数: 0

Steady-State Performance Analysis of the Nearest Kronecker Product Decomposition Based LMS Adaptive Algorithm 基于最近邻Kronecker积分解的LMS自适应算法稳态性能分析

IF 3.2 2区工程技术

IEEE Signal Processing Letters Pub Date : 2025-04-29 DOI: 10.1109/LSP.2025.3565395

Lei Li;Yunfei Zheng;Zhongyuan Guo;Guobing Qian;Shiyuan Wang

引用次数: 0

An Energy-Concentrated Transform for Improved Time-Frequency Representation of Seismic Signals 一种改进地震信号时频表示的能量集中变换

IF 3.2 2区工程技术

IEEE Signal Processing Letters Pub Date : 2025-04-29 DOI: 10.1109/LSP.2025.3565164

Siyuan Wang;Ying Hu;Hui Chen;Xuping Chen

{"title":"An Energy-Concentrated Transform for Improved Time-Frequency Representation of Seismic Signals","authors":"Siyuan Wang;Ying Hu;Hui Chen;Xuping Chen","doi":"10.1109/LSP.2025.3565164","DOIUrl":"https://doi.org/10.1109/LSP.2025.3565164","url":null,"abstract":"Time-frequency (TF) analysis is a useful tool for seismic signal processing, where reliably representing the signal's TF distribution is essential for geological interpretation. However, the traditional short-time Fourier transform (STFT) offers an ambiguous TF representation. While synchroextracting methods improve TF localization, they distort the time-width and bandwidth of seismic signals, which are crucial for applications. To address this issue, we propose a novel TF analysis method, the energy-concentrated transform (ECT), aimed at enhancing TF localization while preserving the essential time-width and bandwidth features of seismic signals. First, we analyze the limitations of the STFT and synchroextracting methods. Next, we introduce an energy suppression operator that concentrates STFT's diffused spectral energy, aligning it with the signal's intrinsic time-width and bandwidth. Additionally, an energy recovery operator is proposed to ensure the consistency of spectral energy with the STFT spectrum. Numerical examples demonstrate the effectiveness of the ECT in enhancing TF localization, improving noise immunity, and preserving critical TF features, making it a promising tool for TF representation in seismic signals.","PeriodicalId":13154,"journal":{"name":"IEEE Signal Processing Letters","volume":"32 ","pages":"2084-2088"},"PeriodicalIF":3.2,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144117350","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

Dual Space Representation Learning for Skeleton-Based Action Recognition 基于骨架的动作识别的双空间表示学习

IF 3.2 2区工程技术

IEEE Signal Processing Letters Pub Date : 2025-04-28 DOI: 10.1109/LSP.2025.3564883

Yuheng Yang;Haipeng Chen;Zhenguang Liu;Sihao Hu;Yingying Jiao

引用次数: 0

DOA Estimation for Coherent and Non-Coherent Mixed Signals Using Toeplitz Diagonal Diffusion 基于Toeplitz对角扩散的相干和非相干混合信号DOA估计

IF 3.2 2区工程技术

IEEE Signal Processing Letters Pub Date : 2025-04-28 DOI: 10.1109/LSP.2025.3565163

Wenlong Wang;Lei Zhang

引用次数: 0

A Non-Asymptotic Analysis on the Additional Bias of Capon's Method Capon方法附加偏差的非渐近分析