Multi-scale Correlation Research Articles

Spectrum Sensing in Very Low SNR Environment Using Multi-Scale Temporal Correlation Perception with Residual Attention

Spectrum sensing is recognized as a viable strategy to alleviate the scarcity of spectrum resources and to optimize their usage. In this paper, considering the time-varying characteristics and the dependence on various timescales within a time series of samples composed of in-phase (I) and quadrature (Q) component signals, we propose a multi-scale time-correlated perceptual attention model named MSTC-PANet. The model consists of multiple parallel temporal correlation perceptual attention (TCPA) modules, enabling us to extract features at different timescales and identify dependencies among features across various timescales. Our simulations show that MSTC-PANet significantly improves the detection of channel occupancy at low signal-to-noise ratios (SNR), particularly in untrained scenarios with lower SNR conditions and modulation uncertainties. The analysis of the ROC curve indicates that at an SNR of -20 dB, the proposed MSTC-PANet achieves a detection rate of 98% with a false alarm rate of 10%. Furthermore, MSTC-PANet, which has been trained using digital modulation techniques, also demonstrates applicability to analog modulation.

CG-MCFNet: cross-layer guidance-based multi-scale correlation fusion network for 3D face recognition

CG-MCFNet: cross-layer guidance-based multi-scale correlation fusion network for 3D face recognition

Multi-scale Dynamic Correlation Between Climate Shock and China's Stock Market: Evidence Based on High Frequency Data

Multi-scale Dynamic Correlation Between Climate Shock and China's Stock Market: Evidence Based on High Frequency Data

A Study of Occluded Person Re-Identification for Shared Feature Fusion with Pose-Guided and Unsupervised Semantic Segmentation

The human body is often occluded by a variety of obstacles in the monitoring system, so occluded person re-identification is still a long-standing challenge. Recent methods based on pose guidance or external semantic clues have improved the representation and related performance of features; there are still problems, such as weak model representation and unreliable semantic clues. To solve the above problems, we proposed a feature extraction network, named shared feature fusion with pose-guided and unsupervised semantic segmentation (SFPUS). This network will extract more discriminative features and reduce the occlusion noise on pedestrian matching. Firstly, the multibranch joint feature extraction module (MFE) is used to extract feature sets containing pose information and high-order semantic information. This module not only provides robust extraction capabilities but can also precisely segment occlusion and the body. Secondly, in order to obtain multiscale discriminant features, the multiscale correlation feature matching fusion module (MCF) is used to match the two feature sets, and the Pose–Semantic Fusion Loss is designed to calculate the similarity of the feature sets between different modes and fuse them into a feature set. Thirdly, to solve the problem of image occlusion, we use unsupervised cascade clustering to better prevent occlusion interference. Finally, performances of the proposed method and various existing methods are compared on the Occluded-Duke, Occluded-ReID, Market-1501 and Duke-MTMC datasets. The accuracy of Rank-1 reached 65.7%, 80.8%, 94.8% and 89.6%, respectively, and the mAP accuracy reached 58.8%, 72.5%, 91.8% and 80.1%. The experiment results demonstrate that our proposed SFPUS holds promising prospects and performs admirably compared with state-of-the-art methods.

Multiscale investigations on RDI-SPI teleconnections of Çoruh and Aras Basins, Türkiye using time dependent intrinsic correlation

This paper proposes a multiscale dynamic correlation framework based on time-dependent intrinsic correlation (TDIC) to investigate the teleconnections between reconnaissance drought index (RDI) and standardized precipitation index (SPI). The SPI and RDI indices were calculated at 3-, 6-, and 12-month time scales using data from six stations in the Çoruh and Aras (CA) basins in Turkey for the 1969–2020 period. The spatial variability of the evaluation of the drought class demonstrated that, with the exception of the northwestern region (Bayburt station), the correlation between RDI and SPI exceeded 97% and the difference in occurrence between drought classes is found to be marginal. In the multiscale analysis of the two indices, firstly, the Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (I-CEEMDAN) was employed to decompose the series. The modes of RDI and SPI at different process scales exhibited a strong positive linear correlation, however, the association in their long-term trends may not be of the same nature. The TDIC analysis captured the long-range correlations between the modes of RDI and SPI at diverse process scales, and the newly proposed average significant correlation (ASC) measure could quantify their strength. Within the basin, there exists strong interconnections between SPI and RDI, with the exception of the observed alterations in their nature and strength at short time spells in some inter-annual scales. The proposed TDIC based framework is a generic and robust alternative for multiscale investigation studies while ASC can quantify the strength of non-linear associations at distinct process scales.

MSCMNet: Multi-scale Semantic Correlation Mining for Visible-Infrared Person Re-Identification

The main challenge in the Visible-Infrared Person Re-Identification (VI-ReID) task lies in extracting discriminative features from different modalities for matching purposes. While existing studies primarily focus on reducing modal discrepancies, the modality information fails to be thoroughly exploited. To solve this problem, the Multi-scale Semantic Correlation Mining network (MSCMNet) is proposed to comprehensively exploit semantic features at multiple scales. The network fuses shallow-level features into the deep network through dimensionality reduction and mapping, and the fused features are utilized to minimize modality information loss in feature extraction. Firstly, considering the effective utilization of modality information, the Multi-scale Information Correlation Mining Block (MIMB) is designed to fuse features at different scales and explore the semantic correlation of fusion features. Secondly, in order to enrich the semantic information that MIMB can utilize, the Quadruple-stream Feature Extractor (QFE) with non-shared parameters is specifically designed to extract information from different dimensions of the dataset. Finally, the Quadruple Center Triplet Loss (QCT) is further proposed to address the information discrepancy in the comprehensive features. Extensive experiments on the SYSU-MM01, RegDB, and LLCM datasets demonstrate that the proposed MSCMNet achieves the greatest accuracy. We release the source code on https://github.com/Hua-XC/MSCMNet.

Multi-scale analysis of nutrient and environmental dynamics in Hongfeng Lake Southwest China

Traditional linear correlation analysis may not fully capture the true relationship between these variables. Therefore, multi-scale running correlation analysis, such as time-dependent intrinsic correlation (TDIC) and continuous wavelet transform based on Hilbert–Huang transform (HHT), provides valuable insights into local correlations and the evolving relationship between nutrients and environmental factors over time. In this study, we investigated seven environmental factors and four water quality nutrient indicators in deep lakes on the Yungui Plateau in southwestern China. The results revealed that there may be strong correlations between environmental factors and nutrient levels during certain periods, while opposite trends may emerge at other times. These variations in correlation could be attributed to uncertain physical processes, spatial heterogeneity, or the impact of different climatic factors on local hydrological processes. Wavelet analysis indicated that changes in environmental factors lag behind those in nutrient levels, particularly on a cycle of about 12 months. This suggests that changes in environmental factors align with natural patterns after the water body has been polluted. These conclusions underscore the complexity and dynamic nature of the relationship between environmental factors and nutrient levels in water bodies, highlighting the importance of employing advanced analysis techniques to capture this complexity.

Unsupervised anomaly detection in shearers via autoencoder networks and multi-scale correlation matrix reconstruction

As the main equipment of coal mining production, the anomaly detection of shearer is important to ensure production efficiency and coal mine safety. One key challenge lies in the limited or even absence of labeled monitoring data for the equipment, coupled with the high costs associated with manual annotation. Another challenge stems from the complex structure of the mining machines, making it difficult to reflect the overall operational state through local anomaly detection. Consequently, the application of decoupled local anomaly detection for mining machines in practical production remains challenging. This paper presents an unsupervised learning-based method for detecting anomalies in shearer. The method includes a module for constructing a Multi-scale Correlation Matrix (MSCM) of mining machine operating conditions, as well as the CNN-ConvLSTM Autoencoder (C-CLA) network. The module for constructing an MSCM enhances the representation of interrelationships between various features of the equipment from different perspectives using multiple correlation analysis methods. The C-CLA network integrates convolutional and convolutional recurrent neural networks, with the convolutional structure extracting local spatial features and the ConvLSTM structure further capturing information from different time scales and feature scales, thereby enhancing the model’s perceptual capabilities towards changes in equipment status. Finally, shearer anomaly detection is achieved through the analysis of reconstructed residual matrices. The rationality and practicality of the proposed method have been validated on our dataset, and the model’s generalization capability has been verified through repeated experiments in similar scenarios. However, due to variations in the working environment of different mining faces and differences in equipment models, implementing detection on other mining faces often requires retraining the model with new data. Furthermore, we compared our method with other anomaly detection techniques, and our detection efficiency was superior by approximately 3%. This method effectively detects anomalies in the shearer.

Multi-scale correlation of impact-induced defects in carbon fiber composites using X-ray scattering and machine learning

Impact-induced defects in carbon fiber-reinforced polymers (CFRPs)-spanning from nanometer to macroscopic length scales-can be monitored using an aggregate of X-ray-based methods, but this is impractical in typical field conditions. We report on a low-velocity impacted CFRP, which is mapped using small- and wide-angle X-ray scattering and X-ray computed tomography, and employ machine learning for correlating material parameterizations derived from these techniques. The observed 1 μ\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\mu$$\\end{document}m to 1 mm-sized defects are parameterized in terms of relative density and fiber orientation indicative of fiber failures (kink bands), and the nanometer sized defects in terms of crystal size and unit cell frustration. The 30 to 300 nm defects are parameterized by a power-law scattering decay, differentiating fractal-like behaviors. We find three spatial domains experimentally and by K-means Clustering: Domains of severe damage (with a visual dent), intact domains (without visual or measurable defects) and a transition domain (defects measurable by X-rays). How the parameters are correlated and how they overlap between the domains are discussed. All parameters are able to point to the detrimental fiber breakage in the severe damage domain, and scattering decay also in the transition domain, for example. How individual parameters determined from one experimental technique can be predicted from that of another is also described.

Multi-scale asynchronous correlation and 2D convolutional autoencoder for adolescent health risk prediction with limited fMRI data.

Adolescence is a fundamental period of transformation, encompassing extensive physical, psychological, and behavioral changes. Effective health risk assessment during this stage is crucial for timely intervention, yet traditional methodologies often fail to accurately predict mental and behavioral health risks due to the intricacy of neural dynamics and the scarcity of quality-annotated fMRI datasets. This study introduces an innovative deep learning-based framework for health risk assessment in adolescents by employing a combination of a two-dimensional convolutional autoencoder (2DCNN-AE) with multi-sequence learning and multi-scale asynchronous correlation information extraction techniques. This approach facilitates the intricate analysis of spatial and temporal features within fMRI data, aiming to enhance the accuracy of the risk assessment process. Upon examination using the Adolescent Risk Behavior (AHRB) dataset, which includes fMRI scans from 174 individuals aged 17-22, the proposed methodology exhibited a significant improvement over conventional models. It attained a precision of 83.116%, a recall of 84.784%, and an F1-score of 83.942%, surpassing standard benchmarks in most pertinent evaluative measures. The results underscore the superior performance of the deep learning-based approach in understanding and predicting health-related risks in adolescents. It underscores the value of this methodology in advancing the precision of health risk assessments, offering an enhanced tool for early detection and potential intervention strategies in this sensitive developmental stage.

A multi-scale dense residual correlation network for remote sensing scene classification

Most existing scene classification methods based on remote sensing images tend to ignore important interactive information at different levels in the image. We propose an effective remote sensing scene classification method named multi-scale dense residual correlation network. The method is divided into three parts. First, the multi-stream feature extraction module is introduced which effectively utilizes features at different scales to extract different levels of information. Secondly, the dense residual connected feature fusion technology is proposed, which allows for a wide range of feature fusion. The Correlation Attention Module learn feature representations at multiple levels. This improves classification performance. The method outperforms existing algorithms in terms of effectiveness and accuracy, achieving state-of-the-art results on widely used remote sensing scene classification benchmarks.

3D Shape Completion on Unseen Categories: A Weakly-Supervised Approach.

3D shapes captured by scanning devices are often incomplete due to occlusion. 3D shape completion methods have been explored to tackle this limitation. However, most of these methods are only trained and tested on a subset of categories, resulting in poor generalization to unseen categories. In this paper, we propose a novel weakly-supervised framework to reconstruct the complete shapes from unseen categories. We first propose an end-to-end prior-assisted shape learning network that leverages data from the seen categories to infer a coarse shape. Specifically, we construct a prior bank consisting of representative shapes from the seen categories. Then, we design a multi-scale pattern correlation module for learning the complete shape of the input by analyzing the correlation between local patterns within the input and the priors at various scales. In addition, we propose a self-supervised shape refinement model to further refine the coarse shape. Considering the shape variability of 3D objects across categories, we construct a category-specific prior bank to facilitate shape refinement. Then, we devise a voxel-based partial matching loss and leverage the partial scans to drive the refinement process. Extensive experimental results show that our approach is superior to state-of-the-art methods by a large margin. We will make the source code publicly available at https://github.com/ltwu6/WSSC.

A long-term dissolved oxygen prediction model in aquaculture using transformer with a dynamic adaptive mechanism

The prediction of Dissolved Oxygen (DO) in oceans is crucial for the survival of marine life and the management of marine farms. However, due to the inherent instability and uncertainty of DO, prediction models frequently necessitate parameter adjustments, posing challenges to maintaining reliable predictive performance and generalization capabilities. Based on this, a long-term dissolved oxygen prediction model in aquaculture using Transformer with a dynamic adaptive mechanism is proposed in the paper. A multi-scale latent correlation decomposition strategy was devised, merging both time and frequency domain decompositions. This approach proficiently untangles time series exhibiting intricate non-linear cycles, thereby delving deeper into and furnishing a more comprehensive array of feature data. Furthermore, an uncertainty-aware attention mechanism was posited, ensuring accurate pinpointing of pivotal data amidst a highly volatile data environment, thus cementing prediction accuracy. Finally, the incorporation of an adaptive hyperparameter adjustment mechanism enables the model to recalibrate in line with feature insights, bolstering its generalization potential across marine farms in diverse regions. Empirical results show that the model proposed in this study can effectively accomplish high-precision long-term forecasting. When juxtaposed with other models under comparable conditions, its mean squared error significantly diminished by 40.2%, the mean absolute error contracted by 55.4%, and concurrently, the R2 value ascended by 7%. This research proffers a fresh vantage point for long-term marine chemical data predictions and extends robust technical bolstering for refining and augmenting aquacultural strategies.

Multi-scale correlation analysis between geometric parameters and solar radiation in high density urban environment——Case study in Nanjing

This study explores the impact of multi-scale urban morphology on solar radiation in high-density environments, focusing on Nanjing as a case study. Using Grasshopper, the research decomposed urban models into four scales—district, block group, block, and plot—and analyzed eight morphological indicators, including site coverage (SC), floor area ratio (FAR), and block surface ratio (BSR). Solar radiation was simulated for the entire year, the summer solstice, and the winter solstice. Key findings reveal that different scales exhibit varying influences on solar radiation. At the district scale, BSR shows the highest correlation with annual solar radiation (R2 = 0.96), while at the neighborhood cluster scale, the minimum distance between buildings (DBmin) is most correlated (R2 = 0.7). At the block scale, BSR significantly correlates with summer solar radiation (R2 = 0.55). At the plot scale, no single indicator strongly correlates with solar radiation, but a combination of indicators is more relevant. These findings enable rapid solar performance assessments in urban design, promoting efficient solar energy use.

MPNet: A lightweight fault diagnosis network for rotating machinery

Rotating machinery is prone to faults, especially bearing faults. Existing machinery fault diagnosis methods suffer from low accuracy and poor robustness under actual complex working conditions. To address these problems, this paper proposes a sound signal-based rotating machinery fault diagnosis method and designs a lightweight fault diagnosis network called MPNet. A multi-branch feature fusion module is constructed to capture the multi-scale correlation information of the sound features. A residual attention pyramid module is designed to adaptively learn abstract fault information at different levels, then the feature-enhanced attention maps at multi-scale generated by hierarchical fusion. Experimental results on a public bearing dataset and a self-made idler dataset reveal that the proposed method achieves 95.83% and 94.81% accuracy, respectively, meeting the requirements of lightweight and real-time detection. Compared with conventional methods, the proposed method has better diagnostic precision and robustness under strong noise. The code library is available at: https://github.com/xgli411/MPNet.

Hierarchical channel-spatial interaction network for partial-to-partial point cloud registration

ABSTRACT Partial-to-partial point cloud registration is a fundamental and challenging task in the three-dimensional computer vision. In this letter, we propose a novel Hierarchical Channel-Spatial Interaction Network (HCSINet), which can hierarchically explore information interaction in terms of the channel and spatial dimensions, enabling well partial point cloud perception for partial-to-partial registration. Specifically, the Channel-wise Feature Interaction (CFI) module is regarded as the basic unit of the encoder sub-network, which achieves feature hierarchical interaction to learn the overlapping scores and intuitively improve the importance of overlapping areas. Furthermore, to improve the discrimination of the overlapping points, the Co-Attention-based Spatial Association (CA-SA) module progressively excavates multi-scale spatial correlation of overlapping points by conjointly computing collaborative attention. Experimental results verify the validity of our HCSINet, which achieves competitive partial-to-partial registration performances compared to existing traditional and deep learning-based methods. The source code is publicly available https://github.com/lxnudt/HCSINet.

Well logging restoration based on spatial-temporal information mining from muti-scale correlation graph representation

Geological reservoir characterization using well-loggingg data often faces challenges, such as data distortion or even missing values. To tackle this, traditional signal restoration theories and machine learning techniques, such as neural networks, have been commonly employed. However, they struggle to fully explore correlation information among logging curves within the same well and lack adaptability across different wells. In this paper, a novel Logging Curve Restoration method based on Spatial-Temporal Information Mining (LCR-STIM) using multi-scale correlation graph representation is proposed. Specially, a novel multi-scale graph representation technique is employed to build spatial (i.e., horizontal) nonlinear relationships among logging curves. Subsequently, deep forest model is introduced to generate multi-scale predictions to uncover the rich spatial information hidden in the multi-scale graph representations. Finally, to fully utilize temporal information in the obtained different multi-scale predictions, Long Short-Term Memory (LSTM) network is employed to combine the multi-scale restoration results to predict target logging data accurately. Taking the well logging data of 41 wells from Daqing oil field as training data and the target logging curves from 3 other wells as the test data, experimental results shown that the restoration performance can be improved by at least 20% compared with the state-of-the-art comparison methods.

Multiscale Correlation Network and Geodesic Distance for Remote Passive Ship Detection in Marine Environment

Multiscale Correlation Network and Geodesic Distance for Remote Passive Ship Detection in Marine Environment

Next generation fatigue crack growth experiments of aerospace materials

Today’s societal challenges require rapid response and smart materials solutions in almost all technical areas. Driven by these needs, data-driven research has emerged as an enabler for faster innovation cycles. In fields such as chemistry, materials science and life sciences, automatic and even autonomous data generation and processing is already accelerating knowledge discovery. In contrast, in experimental mechanics, complex investigations like studying fatigue crack growth in structural materials have traditionally adhered to standardized procedures with limited adoption of the digital transformation. In this work, we present a novel infrastructure for data-centric experimental mechanics in the field of fatigue crack growth. Our methodology incorporates a robust code base that complements a multi-scale digital image correlation and robot-assisted test rig. Using this approach, the information-to-cost ratio of fatigue crack growth experiments in aerospace materials is significantly higher compared to traditional experiments. Thus, serves as a catalyst for discovering new scientific knowledge in the field of structural materials and structures.

MGAtt-LSTM: A multi-scale spatial correlation prediction model of PM2.5 concentration based on multi-graph attention

The increase in air pollution has posed numerous new challenges for human society, making the exploration of an effective method for predicting air pollutant concentrations highly significant. The current research faces several primary challenges: the neglect of non-Euclidean characteristics of site distribution on data and the strong spatiotemporal dependencies in the dispersion process of pollutants. To address these issues, this paper constructs a spatiotemporal hybrid prediction model – the MGAtt-LSTM method – for predicting PM2.5 concentrations, which employs the dynamic multi-graph attention module (MGAtt) to tackle spatial dependencies and Long Short-Term Memory networks (LSTM) to address temporal dependencies. Additionally, extensive experiments are conducted by using historical air pollutant monitoring data and meteorological data from the Beijing-Tianjin-Hebei region. The results demonstrate that the proposed MGAtt-LSTM model achieved superior performance in concentration prediction compared to existing benchmark models.