-
Wigner distribution associated with linear canonical transform of generalized 2-D analytic signals Digit. Signal Process. (IF 2.9) Pub Date : 2024-03-24 Jian-Yi Chen, Bing-Zhao Li
Analytic complex signals find numerous applications in image and signal processing community. This article focuses on the discussion and definitions of three generalized two-dimensional (2-D) analytic signals (GASs). It begins by explaining the parametric Hilbert transform (PHT) and parametric Riesz transform (PRT), derived from analyzing the irrotational and solenoidal vector fields, respectively
-
Bayesian detection for distributed target with limited training data Digit. Signal Process. (IF 2.9) Pub Date : 2024-03-24 Zhe Zhou, Yuntao Wu, Weijian Liu, Jun Liu, Pengcheng Gong
The issue of subspace-based distributed target detection with limited training data is addressed in this study. We use the Bayesian method to tackle the issue by assuming that the covariance matrix follows an inverse Wishart distribution. According to the generalized likelihood ratio test, Rao test, and Wald test, three Bayesian detectors are designed. Real data and simulations both attest to the usefulness
-
Contribution to historical manuscript dating: A hybrid approach employing hand-crafted features with vision transformers Digit. Signal Process. (IF 2.9) Pub Date : 2024-03-20 Merouane Boudraa, Akram Bennour, Mohammed Al-Sarem, Fahad Ghabban, Omair Ameer Bakhsh
Precisely dating historical manuscripts represents a paramount endeavor in the comprehension and the interpretation of their historical significance as well as in the preservation of our cultural heritage; however, despite the strides made in computer-based dating methodologies, the quest for heightened robustness persists. Recent advancements in vision transformers, renowned for their success across
-
Focus-aware and deep restoration network with transformer for multi-focus image fusion Digit. Signal Process. (IF 2.9) Pub Date : 2024-03-19 Changcheng Wang, Kaixiang Yan, Yongsheng Zang, Dongming Zhou, Rencan Nie
Multi-focus image fusion (MFIF) aims to break the limitations of physical devices and integrate different focused regions from multiple source images into a single fully focused image. However, existing fusion methods suffer from two issues when designing the fusion framework: over-reliance on post-processing operations or underestimation of their impact, leading to unsatisfactory visual results. To
-
Randomized two-sided subspace iteration for low-rank matrix and tensor decomposition Digit. Signal Process. (IF 2.9) Pub Date : 2024-03-19 M.F. Kaloorazi, S. Ahmadi-Asl, S. Rahardja
The low-rank approximation of big data matrices and tensors plays a pivotal role in many modern applications. Although, a truncated version of the singular value decomposition (SVD) furnishes the best approximation, its computation is challenging on modern, multicore architectures. Recently, the randomized subspace iteration has shown to be a powerful tool in approximating large-scale matrices. In
-
FGMNet: Feature grouping mechanism network for RGB-D indoor scene semantic segmentation Digit. Signal Process. (IF 2.9) Pub Date : 2024-03-18 Yuming Zhang, Wujie Zhou, Lv Ye, Lu Yu, Ting Luo
Semantic segmentation is a basic and long-standing research area. Depth images can enrich RGB (red-green-blue) images with their rich geometric information, so as to achieve accurate semantic segmentation. However, redundant information exists in RGB and depth images, and its handling has become an important problem. Filter group convolutions are widely used because they can eliminate redundant information
-
Compressive sensing ISAR imaging with low-rank constraint and anisotropic spatial total variation processing Digit. Signal Process. (IF 2.9) Pub Date : 2024-03-18 Xinlei Jing, Zhongjin Jiang
In order to suppress random noise and remove stripe interference in ISAR imaging, a Compressive Sensing method is proposed for super-resolution ISAR imaging in this paper, which is named LR-ASTV (Low-rank and Anisotropic Spatial Total Variation) algorithm here. In this algorithm, the original echo HRRP is transformed into echo HRRP of MMV model with radial interpolation processing at first. Subsequently
-
Micro LED defect detection with self-attention mechanism-based neural network Digit. Signal Process. (IF 2.9) Pub Date : 2024-03-18 Zebang Zhong, Cheng Li, Meiyun Chen, Heng Wu, Takamasu Kiyoshi
We propose a method utilizing a YOLO detector for the precise localization of defective chips and the identification of defect types within multi-scale multi-target images. To address the challenge of optimizing training costs and enhancing model generalization, we introduce an end-to-end deep neural network, CM-YOLOv5, specifically designed for chip detection. We incorporate a novel bottleneck layer
-
Neural operator Res-FNO based on dual-view feature fusion and Fourier transform Digit. Signal Process. (IF 2.9) Pub Date : 2024-03-16 Jinghong Xu, Yuqian Zhou, Qian Liu
In recent years, neural operators have shown great advantages in solving partial differential equations by learning mappings between function spaces. In this paper, we propose a novel neural operator network to further improve the efficiency of operator learning from parameter space to solution space. We introduce Dual-view block in the architecture, which can better extract and fuse potential features
-
Differential evolution VQE for crypto-currency arbitrage. Quantum optimization with many local minima Digit. Signal Process. (IF 2.9) Pub Date : 2024-03-16 Gines Carrascal, Beatriz Roman, Alberto del Barrio, Guillermo Botella
-
A two-parameter extended logistic chaotic map for modern image cryptosystems Digit. Signal Process. (IF 2.9) Pub Date : 2024-03-16 Abdelhakim Latoui, Mohamed El Hossine Daachi
In this paper, a novel Extended Logistic Chaotic Map (ELCM) with two control parameters is proposed. Overcoming the major drawback of the standard Logistic and other existing chaotic maps, the ELCM not only has infinite chaotic range as well as good ergodicity, but also has a simple structure just like the Logistic map, which greatly facilitates its practical implementation and becomes very suitable
-
Industrial defect detection and location based on greedy membrane clustering algorithm Digit. Signal Process. (IF 2.9) Pub Date : 2024-03-15 Yaorui Tang, Bo Yang, Hong Peng, Xiaohui Luo
This paper introduces a related model of membrane calculation in the defect detection and positioning of industrial components. It has the characteristics of distributed and parallel computing, and can efficiently search for better solutions in a given feature space. Inspired by the membrane clustering algorithm, this paper proposes a greedy membrane clustering algorithm and names it GMCA. GMCA is
-
SAGAN: Skip Attention Generative Adversarial Networks for few-shot image generation Digit. Signal Process. (IF 2.9) Pub Date : 2024-03-15 Ali Aldhubri, Jianfeng Lu, Guanyiman Fu
The task of producing high-quality, realistic, and diverse images based on a few instances of newly emerging or long-tail categories is known as few-shot image generation. Despite prior works showing outstanding results, the quality and diversity of the outputs are still limited. In this paper, we tackle this problem by presenting a range of innovative fusion techniques based on attention mechanisms
-
MDJ: A multi-scale difference joint keyframe extraction algorithm for infrared surveillance video action recognition Digit. Signal Process. (IF 2.9) Pub Date : 2024-03-15 Zhiqiang Feng, Xiaogang Wang, Jiayi Zhou, Xin Du
Many action recognition methods require significant computational resources to achieve good results on unedited videos. However, their performance on infrared videos, which contain less information, is often unsatisfactory. In this paper, we propose a multi-scale difference joint key frame extraction algorithm for action recognition in infrared surveillance videos. To evaluate our algorithm, we have
-
Low-light images enhancement via a dense transformer network Digit. Signal Process. (IF 2.9) Pub Date : 2024-03-15 Yi Huang, Gui Fu, Wanchun Ren, Xiaoguang Tu, Ziliang Feng, Bokai Liu, Jianhua Liu, Chao Zhou, Yuang Liu, Xiaoqiang Zhang
This paper proposes a dense network composed of an improved Transformer network, which successfully restores low-light images to high-quality normal-light images, alleviating issues such as low brightness, high noise, and missing critical information in low-light images. The entire network architecture is based on the improved Transformer network and builds a dense network with a combination of long
-
Graph-topology-learning-based IoT positioning under incomplete measurement data Digit. Signal Process. (IF 2.9) Pub Date : 2024-03-15 Mengya Xie, Feng Li, Shikun Qiao
Obtaining location information of the multi-sensor internet of things (IoT) is a fundamental requirement. But, two kinds of the incomplete measurement data enhance the difficulty of positioning: 1) The location information is partially missing. 2) The errors of the measurement. Furthermore, the complex environmental impact and the characters of the measurement methods make the error model more changeable
-
Geometric Algebra based 2D-DOA Estimation for Non-circular Signals with an Electromagnetic Vector Array Digit. Signal Process. (IF 2.9) Pub Date : 2024-03-14 Xiangyang Wang, Yichen Feng, Xiaolu Lv, Rui Wang
This paper presents two novel methods based on geometric algebra (GA) to estimate two-dimensional (2D) direction-of-arrival (DOA) of non-circular (NC) signals for uniform rectangular array (URA). Traditional methods treat the received NC signals as a long vector which will inevitably lose orthogonality inside each electromagnetic vector sensor (EMVS) and thus miss some information of second-order statistical
-
A novel aspect of automatic vlog content creation using generative modeling approaches Digit. Signal Process. (IF 2.9) Pub Date : 2024-03-10 Lalit Kumar, Dushyant Kumar Singh
Generative models have emerged as potential tools for creating high-quality images, videos, and text. This paper explores the application of generative models in automating vlog content creation. It addresses both static and dynamic visual elements, eliminating the need for human intervention. Traditional vlogs often require specific environmental conditions and proper lighting for the vlog creation
-
Persymmetric design of jointly detection and bearing estimation for a 2D array radar in training demanding scenarios Digit. Signal Process. (IF 2.9) Pub Date : 2024-03-08 Kexuan Cui, Yongchan Gao, Zekang Zhang, Lei Zuo
The problem of joint detection and bearing estimation for a two-dimensional (2D) array radar in training demanding scenarios is addressed. Regardless of the cause of the angle deviation, we model the 2D steering vector as a fully incremental form. Thus, the 2D array requires the optimization of two target cosine offsets. To relax the requirement of sufficient training data, we incorporate persymmetric
-
Time-difference reassigned transform with application to time difference of arrival for impulsive signal Digit. Signal Process. (IF 2.9) Pub Date : 2024-03-06 Peng Zhang, Hongyuan Wen, Zhao Zhao, Zhiyong Xu
Time difference of arrival (TDOA) is essential in localization, communication, and navigation. Under ambient noise interference, the impulsive signal is transmitted over a long distance and reaches the sensor with a low signal-to-noise ratio (SNR). Aiming to achieve a precise time delay estimation in scenarios with low SNR, this paper extends the time-reassigned extracting transform (TRET) theory to
-
Optimal synchronization with binary marker for segmented burst deletion errors Digit. Signal Process. (IF 2.9) Pub Date : 2024-03-05 Chen Yi, Jihua Zhou, Tao Zhao, Baoze Ma, Yong Li, Francis C.M. Lau
In some telecommunication and magnetic/digital recording applications, bits/symbols tend to be lost in the transmission due to the interference. In this paper, we consider a segmented burst deletion channel where in a block of consecutive bits at most a single burst deletion of length up to bits exists. Existing synchronization approaches either provide a poor synchronization performance or suffer
-
Bayesian inference for amplitude distribution with application to radar clutter Digit. Signal Process. (IF 2.9) Pub Date : 2024-03-05 Mahdi Teimouri, Seyed Mehdi Hoseini, Maria Sabrina Greco
The performance of telecommunication systems is significantly subject to scattered signals superposed at the receivers. Notably, if the superposed scattered signals are impulsive in nature, this leads to burst effect on the communicated symbols. Thus, attaining an accurate estimate of the parameters of the underlying statistical model of the aggregate scattered signals becomes more important. More
-
A 1-bit DACs precoding for MU-MIMO based on binary equilibrium constraint: An alternating direction method Digit. Signal Process. (IF 2.9) Pub Date : 2024-03-04 Guodong Xue, Hui Li, Rui Liang
High power consumption is one of the main problems in the practical application of massive multi-user multi-input-multi-output (MU-MIMO) systems, and using 1-bit digital-to-analog converters (DACs) instead of high-resolution DACs is an effective way to reduce the power consumption of massive MU-MIMO systems. How to design efficient precoding schemes to improve the performance of 1-bit MU-MIMO systems
-
Decoding of auditory surprise in adult magnetoencephalography data using Bayesian models Digit. Signal Process. (IF 2.9) Pub Date : 2024-03-01 Parya Tavoosi, Ghasem Azemi, Paul F. Sowman
The Bayesian brain framework has been proposed to explain how the brain processes and interprets sensory information. Magnetoencephalography (MEG) and electroencephalography (EEG) are two neuroimaging techniques commonly used with decoding models to study neural responses to auditory, visual and somatosensory stimuli. Our study aims to investigate neural responses to auditory stimuli using MEG data
-
Classification of imagined speech of vowels from EEG signals using multi-headed CNNs feature fusion network Digit. Signal Process. (IF 2.9) Pub Date : 2024-03-01 Smita Tiwari, Shivani Goel, Arpit Bhardwaj
Brain-computer interface (BCI) provides a platform for humans to communicate using Electroencephalogram (EEG) signals by converting them into commands that can be used by the output device to perform the desired tasks. This paper focuses on the identification of vowels from EEG signals. First, a dataset of EEG signals has been created for the identification of vowels by collecting data using a 14-channel
-
Parameter identification algorithm for ship manoeuvrability and wave peak model based multi-innovation stochastic gradient algorithm use data filtering technique Digit. Signal Process. (IF 2.9) Pub Date : 2024-03-01 Yang Liu, Shun An, Longjin Wang, Yan He, Zhimin Fan
This paper addresses the issue of identifying ship motion parameters and wave peak frequency. Utilising the Euler discretisation principle, we establish a discrete-time auto-regressive moving average model with exogenous input (ARMAX) for the ship-wave system. Furthermore, we develop a filtering-based stochastic gradient algorithm for the system by applying filtering techniques and auxiliary model
-
Persymmetric detection based on asymptotically optimal convex linear combination Digit. Signal Process. (IF 2.9) Pub Date : 2024-03-01 Jie Lin, Chaoshu Jiang, Haohao Ren, Yuanhua Fu, Keyan Qi
Persymmetric structure has been utilized in space-time adaptive processing for heterogeneous environment, which leads to some detection methods based on persymmetric structure, such as persymmetric adaptive matched filter (PS-AMF). However, when the sample support is extremely limited, these methods still suffer the serious degradation in detection performance due to the large error in estimating covariance
-
Enhancing seismic data by edge-preserving geometrical mode decomposition Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-29 Tara P. Banjade, Cong Zhou, Hui Chen, Hongxing Li, Juzhi Deng
Real-time seismic signals are intertwined with different types of noises during the generation, acquisition, and transmission process. The enhanced data with high resolution assists to interpret and analyze records more accurately. In this paper, we propose a mathematical approach based on recently developed geometrical mode decomposition (GMD) and adaptive self-guided filter (ASGF) to attenuate noise
-
Occlusion-aware visual object tracking based on multi-template updating Siamese network Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-28 Lifan Sun, Jiayi Zhang, Dan Gao, Bo Fan, Zhumu Fu
Visual object tracking is a crucial area of computer vision research. It aims to accurately track objects in videos with challenges such as occlusion, deformation, and lighting variations. Existing algorithms face difficulties when objects leave the camera or reappear after being occluded, and they also struggle to track objects with significant appearance changes. To address these issues, this study
-
Sub-Nyquist sensing of Gaussian pulse streams with unknown shape factor based on information fitting Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-28 Shuangxing Yun, Ning Fu, Liyan Qiao
Gaussian pulse streams can be characterized by a finite number of unit-time parameters, and classical Finite Rate of Innovation (FRI) sampling enables sub-Nyquist sensing of these signals. However, prior knowledge of its shape factor is required, limiting FRI's applicability. This paper proposes a solution to the FRI sampling problem of Gaussian pulse streams with an unknown pulse shape factor. We
-
Information criteria for structured parameter selection in high-dimensional tree and graph models Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-27 Maarten Jansen
Parameter selection in high-dimensional models is typically fine-tuned in a way that keeps the (relative) number of false positives under control. This is because otherwise the few true positives may be dominated by the many possible false positives. This happens, for instance, when the selection follows from a naive optimisation of an information criterion, such as AIC or Mallows's . It can be argued
-
Deep-feature-based asymmetrical background-aware correlation filter for object tracking Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-27 Yingpin Chen, Huanyu Wu, Zhaojun Deng, Jun Zhang, Hui Wang, Lingzhi Wang, Wentong Huang
Correlation filter-based video object-tracking algorithms have gained widespread attention due to their efficiency and excellent tracking performance. However, traditional correlation filtering tracking algorithms possess several limitations. (1) They extract image features only by using rule sampling, which ignores the shape information of the target, resulting in insufficient discriminative power
-
High-quality matched transfer generation adversarial network for synthetic cross-material surface defect images Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-24 Xikun Xie, Changjiang Li, Rui Qing, Chuande Zhou, Zhong Zhang
Generating Adversarial Network based on style transfer is an effective method to expand sample data. Nonetheless, an urgent issue that demands resolution is the fusion of cross-material defects and backgrounds to generate high-quality defect samples. In this paper, we propose the High-quality Matching Transfer Generative Adversarial Network (HMTGAN), an innovative framework. This network is based on
-
An efficient non-negative least mean squares algorithm based on q-gradient for system identification Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-23 Yikun Yang, Bintang Yang
The system identification under non-negative constraints problem is a common and important one in the real-life problems. A non-negative least mean squares algorithm was proposed to address such problems. However, it suffers from slow and unbalanced convergence. Motivated by the introduction of non-Newtonian gradient in least mean squares algorithm can accelerate the convergence. In this paper, an
-
HFENet: Hybrid feature encoder network for detecting salient objects in RGB-thermal images Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-23 Fan Sun, Wujie Zhou, Weiqing Yan, Yulai Zhang
Deep convolutional neural networks (CNNs) have gained prominence in computer vision applications, including RGB salient object detection (SOD), owing to the advancements in deep learning. Nevertheless, the majority of deep CNNs employ either VGGNet or ResNet as their backbone architecture for extracting image information. This approach may lead to the following problems. 1) Variations between imaging
-
Application of second order multi-synchrosqueezing transform for seismic data analysis Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-20 Si-Yi Chen, Ya-Juan Xue, Lin Huang
Time-frequency methods, which allow us to observe the seismic signal in time and frequency domains simultaneously, can efficiently discern the seismic response features of geological bodies in different frequency bands. In this paper, a newly developed time-frequency method named second order multi-synchrosqueezing transform (SMSST) is introduced to analyze seismic data. SMSST combines the second-order
-
Automatic learning-based data optimization method for autonomous driving Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-20 Yang Wang, Jin Zhang, Yihao Chen, Hao Yuan, Cheng Wu
-
DOA estimation and signal sorting methods of multi-baseline polarized interferometer Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-20 Mingchao Qu, Ruizhi Liu, Yue Zhang, Weijian Si
-
Effect of reconstruction error in subtractive dither structure Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-19 Yanrui Su, Yu Bai, Zhao Wu, Yupeng Shen, Hongqiang Song, Fabao Yan
This paper studies the causes and distribution of reconstruction errors in subtractive dither structure and provides solution to minimize these errors. We present the generation of reconstruction errors in two types of quantization, namely mid-riser and mid-tread, with their corresponding distributions. We then construct a 12-bit pipeline analog to digital converters (ADC) simulation model and use
-
On the performance of power-sensing RIS-SM: Effects of improper Gaussian noise and Nakagami-m fading channel Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-13 A, y, s, e, , E, ., , C, a, n, b, i, l, e, n
The innovative idea of data transmission through a reconfigurable intelligent surface (RIS) is a promising and exciting solution to the ever-present problems in wireless communications, such as energy consumption and hardware cost. The soft-controlled dynamic shaping of the electro-magnetic waves, which enables utilizing the randomness of the propagation medium to achieve more efficacious wireless
-
Anti-interrupted sampling repeater jamming based on intra-pulse frequency modulation slope agile radar waveform joint FrFT Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-13 Xiaoge Wang, Binbin Li, Weijian Liu, Hui Chen, Yongzhe Zhu, Mengyu Ni
Interrupted sampling repeater jamming (ISRJ) is coherent with the radar-transmitted signal, which can form multi-false targets at the radar receiver. Therefore, ISRJ suppression is a compelling task in electronic counter-countermeasure. Intra-pulse frequency coded signals are not well energy-concentrated in the time-frequency (TF) domain, resulting in interference and target overlapping in the TF domain
-
CARgram: CNN-based accident recognition from road sounds through intensity-projected spectrogram analysis Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-13 Alessandro Sebastian Podda, Riccardo Balia, Livio Pompianu, Salvatore Carta, Gianni Fenu, Roberto Saia
Road surveillance systems play an important role in traffic monitoring and detecting hazardous events. In recent years, several artificial intelligence-based approaches have been proposed for this purpose, typically based on the analysis of the acquired video streams. However, occlusions, poor lighting conditions, and heterogeneity of the events may often reduce their effectiveness and reliability
-
Exploring varying color spaces through representative forgery learning to improve deepfake detection Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-12 Muhammad Ahmad Amin, Yongjian Hu, Yu Guan, Muhammad Zain Amin
In the digital age, the rise of deepfake technology has brought unprecedented challenges to multimedia content authentication. The existing deepfake detection methods generally perform well in known settings. However, generalization and robustness are still challenging tasks. Observing that most conventional methods adopt the RGB color space, we introduce a novel deepfake detection approach by utilizing
-
Graph Signal Reconstruction Based on Spatio-temporal Features Learning Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-09 Jie Yang, Ce Shi, Yueyan Chu, Wenbin Guo
This paper presents a new algorithm for reconstructing time-varying graph signals using spatiotemporal feature learning. We introduce a time series analysis method to capture the temporal stationarity of graph signals and propose a reconstruction model based on spatiotemporal coupling features. However, prior knowledge of the temporal stationarity of graph signals is required for this task. To address
-
Interrupted-sampling multi-strategy forwarding jamming with amplitude constraints based on simultaneous transmission and reception technology Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-09 Jie Xiao, Xizhang Wei, Jia Sun
The interrupted-sampling repeater jamming (ISRJ) is a mainstream jamming method in electronic warfare, simple yet implemented. With the rapid development of anti-ISRJ, due to the defects of ISRJ and the difficulty in ensuring real-time response, there is an urgent need to propose a new ISRJ-based jamming method based on simultaneous transmission and reception (STAR) technology. An interrupted-sampling
-
-
DRANet: A semantic segmentation network for Chinese landscape paintings Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-09 QiYao Hu, Wanlin Zhou, Xianlin Peng, Xiang Zhang, Penglin Xie, Yuzhe Liu, Jinye Peng, Jianping Fan
With large-scale human-annotated pixel-level annotations, many deep learning-based methods have achieved impressive innovations in the field of semantic image segmentation. However, the lack of publicly available Chinese landscape paintings with pixel-level annotations seriously hinders the process of segmentation of Chinese landscape paintings. In addition, a prominent feature of Chinese landscape
-
A secure image authentication technique based on sparse approximation and quantum mechanism Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-08 Rohit Agrawal, Kuldeep Narayan Tripathi, Ranjeet Kumar Singh, Nitin Arvind Shelke, Umesh Gupta
The security and protection of digital content transmitted over a network are critical concerns due to the ease with which various types of data can be transformed into text, images, audio, and video forms. This vulnerability makes them susceptible to unauthorized modifications by hackers. To address this issue, several techniques have been developed, and one such approach is image encryption and watermarking
-
Single-layer folded RNN for time series prediction and classification under a non-Von Neumann architecture Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-07 Zhou Wenjun, Zhu Chuan, Ma Jianmin
A delay dynamical system can fold a feedforward neural network into one nonlinear neuron and multiple delay loops under the non-Von Neumann structure, greatly decreasing the hardware requirements. In this paper, we transform the folded-in-time DNN (Fit-DNN) into a folded-in-time RNN (Fit-RNN) and derive the backpropagation algorithm for it. The performance of the folded reservoir computing (DRC), Fit-DNN
-
Frequency-invariant direction-of-arrival estimation of circular acoustic vector sensor array for wideband signals on the cylindrical baffle Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-06 Xiaochun Zhu, Xu Zhang, Chenyang Gui, Fujia Xu, Nan Shen, Shengguo Shi
This passage discusses a method for estimating the DOA (direction-of-arrival) of broadband signal targets on the surface of a cylindrical baffle using a circular acoustic vector sensor array and based on the Minimum Variance Distortionless Response (MVDR) algorithm with joint processing of sound pressure and particle velocity. The approach employs a cylindrical baffle scattering sound field model and
-
Angle-Weighted trilateration method fusing multi-data processing technologies in indoor scene Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-06 Liguo Zang, Jing Jiao, Jie Wang, Ning Ding, Zizhou Wang, Xinyi Min
The trilateration method based on Ultra-Wideband (UWB) technology is widely used in indoor positioning. However, the complexity of indoor environment will affect the transmission of UWB signals, reducing accuracy of trilateration method. Therefore, an Angle-Weighted trilateration method fusing multi-data processing technologies is utilized. By analyzing the relationship between positioning accuracy
-
FractalRG: Advanced fractal region growing using Gaussian mixture models for left atrium segmentation Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-06 Marjan Firouznia, Javad Alikhani Koupaei, Karim Faez, Aziza Saber Jabdaragh, Cigdem Gunduz-Demir
This paper presents an advanced region growing method for precise left atrium (LA) segmentation and estimation of atrial wall thickness in CT/MRI scans. The method leverages a Gaussian mixture model (GMM) and fractal dimension (FD) analysis in a three-step procedure to enhance segmentation accuracy. The first step employs GMM for seed initialization based on the probability distribution of image intensities
-
In-sensor nonlinear convolutional processing based on hybrid MTJ/CMOS arrays Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-06 Minhui Ji, Liyuan Yang, Mengchun Pan, Xinmiao Zhang, Jiayuan Wang, Yueguo Hu, Qingfa Du, Jiafei Hu, Weicheng Qiu, Junping Peng, Peisen Li
In-sensor computing implemented by novel neuromorphic devices has been regarded as the potential technology to break the acquisition wall. Moreover, the nonlinear convolution inspired by the biological neural system outperforms the traditional linear convolution. Therefore, realizing the in-sensor nonlinear convolutional processing with the intrinsic nonlinearity of novel neuromorphic devices would
-
Eigenstructure methods for DOA estimation of circular acoustic vector sensor array with axial angle bias in nonuniform noise Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-02 Shengguo Shi, Fujia Xu, Xu Zhang, Xiaochun Zhu, Nan Shen, Chenyang Gui
To improve the performance of direction of arrival (DOA) estimation under the coexistence of nonuniform noise and axial angle bias, we propose new DOA estimation methods based on the eigendecomposition of a covariance matrix constructed by the analytic velocity and acoustic pressure. The analytical velocity model can convert axial angle bias into phase error to facilitate its estimation. Meanwhile
-
Efficient self-calibrated and hierarchical refinement network for lightweight super-resolution Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-02 Wenbo Zhang, Lulu Pan, Ke Xu, Guo Li, Yanheng Lv
Recently, deep learning methods have achieved excellent performance in single image super-resolution (SISR), but most existing methods suffer from heavy computational costs and memory storage. To address this problem, some lightweight SR methods have been proposed, among which convolutional neural network (CNN) with attention mechanisms has received increasing attention. However, the existing CNN-based
-
Time domain speech enhancement with CNN and time-attention transformer Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-02 Nasir Saleem, Teddy Surya Gunawan, Sami Dhahbi, Sami Bourouis
Speech enhancement in the time domain involves improving the quality and intelligibility of noisy speech by processing the waveform directly without the need for explicit feature extraction or domain transformation. Deep learning is a powerful approach for time domain speech enhancement, offering significant improvements over traditional techniques. Formulating a resource-efficient deep neural model
-
An analysis of data leakage and generalizability in MRI based classification of Parkinson's Disease using explainable 2D Convolutional Neural Networks Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-02 Iswarya Kannoth Veetil, Divi Eswar Chowdary, Paleti Nikhil Chowdary, Sowmya V., E.A. Gopalakrishnan
Parkinson's Disease (PD) is a progressive neurological disorder caused by the death of dopamine producing neurons. Neuroimaging techniques such as Magnetic Resonance Imaging (MRI) allows the visualization of the structural changes in the brain due to PD. Advances in computer vision has led to a new area of research that combines the expertise of deep learning (DL) tools such as Convolutional Neural
-
Fast 3-D millimeter-wave MIMO array imaging algorithms based on the CF-DFrFT Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-01 Qirun Li, Xinbo Li, Ziyi Chen, Liangxu Jiang, Yingwei Wang
In near-field applications, 3-D imaging with the millimeter-wave multiple-input-multiple-output (MIMO) array provides accurate reconstruction with high dynamic range. However, current algorithms make it difficult to process multidimensional echo data in real-time with ordinary computational power. To improve the imaging speed, the property of the focus position control by closed-form discrete fractional
-
Extended two-dimensional separable sensing matrix in compressive sensing Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-01 Xiao Xue, Song Xiao, Wenqian Dong
Recently, many types of compressive sensing (CS) matrices have been constructed. Most of these constructions focus on the measurement of one-dimensional (1-D) signals. However, many practical applications involve the acquisition of two-dimensional (2-D) signals. One of the main difficulties in the direct application of a 1-D CS measurement system to the measurement of a 2-D signal is that the computational
-
Journey into gait biometrics: Integrating deep learning for enhanced pattern recognition Digit. Signal Process. (IF 2.9) Pub Date : 2024-02-01 Anubha Parashar, Apoorva Parashar, Imad Rida
Exploring Gait Biometrics within the domain of deep learning offers a potent fusion that significantly enhances pattern recognition capabilities. Over the past decade, the evolution of deep learning (DL) pipelines has showcased their effectiveness in overcoming complex challenges within image and signal processing applications. Constructing these pipelines requires a deep understanding of the diverse