Spatial Mechanisms Research Articles

Unsupervised Anomaly Detection via Normal Feature-Enhanced Reverse Teacher–Student Distillation

In modern industrial production, unsupervised anomaly detection methods have gained significant attention due to their ability to address the challenge posed by the scarcity of labeled anomaly samples. Among them, unsupervised anomaly detection methods based on reverse distillation (RD) have become a mainstream choice, which has attracted extensive research due to their excellent anomaly detection performance. However, there is a problem of “feature leakage” in the RD model, which may lead to non-anomalous regions being incorrectly identified as defects. To solve this problem, we propose a Normal Feature-Enhanced Reverse teacher–student Distillation (NFERD) method. Specifically, we designed and incorporated a normal feature bank (NFB) module into the basic RD network. This module stores normal features extracted by the teacher model, assisting the student model in learning normal features more efficiently, thereby addressing the problem of “feature leakage”. In addition, to effectively fuse the feature maps extracted by the student model with the feature maps in NFBs, we designed a Hybrid Attention Fusion Module (HAFM), which ensures the preservation of key information during the feature fusion process by the parallel processing of spatial and channel attention mechanisms. Through experimental verification on two publicly available datasets, i.e., MVTec and KSDD, our method outperformed the existing mainstream methods in both image-level and pixel-level anomaly detection. Specifically, we achieved an average I-AUROC score of 99.32% on MVTec and a 98.75% P-AUROC on the KSDD, showing clearer segmentation results, especially in complex scenarios. Furthermore, our method surpassed the second-best method by over 1.4% PRO on MVTec, demonstrating its effectiveness.

Vehicle Localization Method in Complex SAR Images Based on Feature Reconstruction and Aggregation.

Due to the small size of vehicle targets, complex background environments, and the discrete scattering characteristics of high-resolution synthetic aperture radar (SAR) images, existing deep learning networks face challenges in extracting high-quality vehicle features from SAR images, which impacts vehicle localization accuracy. To address this issue, this paper proposes a vehicle localization method for SAR images based on feature reconstruction and aggregation with rotating boxes. Specifically, our method first employs a backbone network that integrates the space-channel reconfiguration module (SCRM), which contains spatial and channel attention mechanisms specifically designed for SAR images to extract features. The network then connects a progressive cross-fusion mechanism (PCFM) that effectively combines multi-view features from different feature layers, enhancing the information content of feature maps and improving feature representation quality. Finally, these features containing a large receptive field region and enhanced rich contextual information are input into a rotating box vehicle detection head, which effectively reduces false alarms and missed detections. Experiments on a complex scene SAR image vehicle dataset demonstrate that the proposed method significantly improves vehicle localization accuracy. Our method achieves state-of-the-art performance, which demonstrates the superiority and effectiveness of the proposed method.

Towards gamification for spatial digital learning environments

Digital learning environments exhibiting spatial dimensions, such as VR experiences or virtual labs, have become increasingly common in recent years. At the same time, it is known that gamification, i.e., the application of game-like mechanisms, might support the motivational design of learning environments. However, so far, there seems to be no systematic overview on whether and how spatial dimensions are incorporated into gamification mechanisms. Accordingly, we conducted a systematic literature review to identify gamification mechanisms that use spatial dimensions. Out of 849 articles we finally included 11 articles found in the three databases ScienceDirect, IEEE, and ACM DL that conducted gamification in the context of spatial digital learning environments. For the most part, the gamification mechanisms used in these 11 articles did not relate to any spatial dimension. Conclusively, we state that there have been few approaches to gamification exploiting the spatial dimension. Given the limited findings in the literature, we propose developing a model to bridge existing gaps and support a structured approach to incorporating spatial dimensions into gamification mechanisms. Accordingly, we contribute to enhancing spatial digital learning environments with further motivational cues through spatial gamification mechanics.

Integration of Generative-Adversarial-Network-Based Data Compaction and Spatial Attention Transductive Long Short-Term Memory for Improved Rainfall–Runoff Modeling

This work presents a novel approach to rainfall–runoff modeling. We incorporate GAN-based data compaction into a spatial-attention-enhanced transductive long short-term memory (TLSTM) network. The GAN component reduces data dimensions while retaining essential features. This compaction enables the TLSTM to capture complex temporal dependencies in rainfall–runoff patterns more effectively. When tested on the CAMELS dataset, the model significantly outperforms benchmark LSTM-based models. For 8-day runoff forecasts, our model achieves an NSE of 0.536, compared to 0.326 from the closest competitor. The integration of GAN-based feature extraction with spatial attention mechanisms improves predictive accuracy, particularly for peak-flow events. This method offers a powerful solution for addressing current challenges in water resource management and disaster planning under extreme climate conditions.

Rolling bearing fault diagnosis method based on MSDCNN in strong noise environment

Aiming at the problems of poor noise resistance, high computational complexity and insufficient generalization performance of traditional bearing fault diagnosis methods based on deep learning, a rolling bearing fault diagnosis method based on multi-scale dynamic convolutional neural network (MSDCNN) is proposed. Firstly, the one-dimensional vibration signal of the rolling bearing is converted to the frequency domain by Fourier transform, and the features are further extracted by wide convolution kernel. Secondly, a multi-scale dynamic convolution structure is proposed, and the feature information extracted by convolution kernels of different sizes is given different weights by using an improved channel attention mechanism. Then, a self-calibrating spatial attention mechanism (SCSAM) is designed, and the extracted feature information is input into the spatial attention mechanism to capture the importance of different regions. Finally, the features are further extracted by small convolution kernels, and the fault category is classified by using Softmax classifier. The fault diagnosis performance of the proposed model is verified by three different data sets. The experimental results show that compared with other intelligent models, the proposed model has higher classification accuracy, better generalization ability and stronger robustness under strong noise background.

CSA-Net: Channel and Spatial Attention-Based Network for Mammogram and Ultrasound Image Classification.

Breast cancer persists as a critical global health concern, emphasizing the advancement of reliable diagnostic strategies to improve patient survival rates. To address this challenge, a computer-aided diagnostic methodology for breast cancer classification is proposed. An architecture that incorporates a pre-trained EfficientNet-B0 model along with channel and spatial attention mechanisms is employed. The efficiency of leveraging attention mechanisms for breast cancer classification is investigated here. The proposed model demonstrates commendable performance in classification tasks, particularly showing significant improvements upon integrating attention mechanisms. Furthermore, this model demonstrates versatility across various imaging modalities, as demonstrated by its robust performance in classifying breast lesions, not only in mammograms but also in ultrasound images during cross-modality evaluation. It has achieved accuracy of 99.9% for binary classification using the mammogram dataset and 92.3% accuracy on the cross-modality multi-class dataset. The experimental results emphasize the superiority of our proposed method over the current state-of-the-art approaches for breast cancer classification.

Deep spatial-temporal information fusion dynamic graph convolutional network for traffic flow prediction

Abstract Predicting traffic flow is vital for advancing intelligent transportation systems, but achieving precise forecasts remains challenging. To deeply explore the complex and diverse dynamic spatio-temporal correlation of traffic data, this study proposes a multi-information fusion spatio-temporal dynamic graph convolution model for traffic flow prediction. In the model construction, firstly, a dynamic graph generation module is designed to fully capture the complex spatial correlation of the traffic network and model the changing spatio-temporal interaction; secondly, the spatial self-attention mechanism is used to dynamically focus on the spatial relationship between different nodes, and combined with the dynamic graph convolution network to extract the spatial dynamic correlation features of the road network to study deep spatiotemporal information; Finally, existing models rarely consider other traffic information related to traffic flow, and a spatiotemporal information fusion module is designed to enhance the ability of the model to perceive information, thereby improving the performance of traffic flow prediction models. Experiments are conducted on four real traffic datasets using three performance evaluation metrics and 18 baseline models. The results show that the model has higher prediction accuracy.

Influence mechanism of digital economy development on the supply efficiency of ecological products

Improving the supply efficiency of ecological products (EPSE) is of great significance to protect the ecological environment, promote the development of green industry and the sustainable growth of the economy. This paper focuses on constructing an evaluation index system for EPSE and exploring the relationship between the development level of the digital economy (DE) and EPSE in China. To measure the EPSE, the study employs the Undesirable Slacks-Based Measurement (SBM) Model across 30 provinces, cities and districts in China from 2011 to 2022. Furthermore, the research utilizes spatial econometric models, panel threshold effect models, and other methodologies to investigate the impact mechanism and non-linear relationship between DE and EPSE. The research shows that: (1) The overall level of EPSE in the study area is relatively low, with significant development differences observed; (2) The DE has a nota-ble spatial spillover effect on EPSE, with a significantly negative impact in neighboring areas; (3) The development of the DE promotes EPSE through the upgrading of human capital structure, industrial structure, and increases in local government fiscal revenue and corporate operating profits; (4) Heterogeneity analysis shows that the impact of DE on EPSE varies significantly across eastern, central, and western China, with positive effects pronounced in the eastern and western regions; (5) Regression results of the threshold effect indicate a significant single threshold effect on the impact of DE development level on EPSE. Specifically, when the threshold value is less than 0.1232, DE significantly contributes to the improvement of EPSE. This paper contributes new literature evidence and factual references to the understanding of the causal relationship between DE and EPSE. The findings highlight the importance of considering spatial spillover effects, impact mechanisms, and regional heterogeneity in analyzing the relationship between DE and EPSE. The research also suggests that promoting the development of the digital economy could be a viable strategy to enhance EPSE, especially in regions where the threshold value is below 0.1232

Analysis of the Response of Shallow Groundwater Levels to Precipitation Based on Different Wavelet Scales—A Case Study of the Datong Basin, Shanxi

The rise in shallow groundwater levels is typically triggered by precipitation recharge, exhibiting a certain lag relative to precipitation changes. Therefore, identifying the response mechanism of shallow groundwater levels to precipitation is crucial for clarifying the interaction between precipitation and groundwater. However, the response mechanism of groundwater levels to precipitation is complex and variable, influenced by various hydrogeological and geographical conditions, and often exhibits significant nonlinear characteristics. To address this issue, this study employs methods such as continuous wavelet transform, cross wavelet transform, and wavelet coherence to analyze the response patterns of groundwater levels to precipitation at different wavelet scales in the Datong Basin from 2013 to 2022: (i) At short wavelet scales (10.33~61.96 d), the groundwater level dynamics respond almost instantaneously to extreme rainfall; (ii) At medium wavelet scales(61.96~247.83 d), the precipitation-groundwater recharge process shows characteristics of either rapid recovery or significant delay; (iii) At long wavelet scales (247.83~495.67 d), three potential groundwater processes were identified in the Datong Basin, exhibiting long-term lag responses throughout this study period, with lag times of 11.18 days, 148.75 days, and 151.49 days, respectively. Furthermore, the results indicate that the lag response time of shallow groundwater levels to precipitation is not only related to the wavelet scale but also to the identified depth conditions of different groundwater regions, groundwater extraction intensity, precipitation intensity, and aquifer lithology. This study distinguishes the temporal and spatial response mechanisms of shallow groundwater to precipitation at different wavelet scales, and this information may further aid in understanding the interaction between precipitation and groundwater levels.

The Mosque Nearby: Visible Minorities and Far-Right Support in France

How is support for right-wing populist parties affected by exposure to Muslim visibility? Using an original database on French mosques, this article analyzes the relationship between the presence of mosques and support for the Front National at the polling station level in the late 2000s. It finds that the propensity to vote for the Front National increases in polling stations up to intermediate distances from mosques and then decreases, suggesting a spatial mechanism known as the halo effect. The analysis also shows that larger mosques and those with minarets are associated with an accentuated halo effect, suggesting the importance of the salience of minority groups rather than their relative size in influencing political behavior.

Randomised trial of three treatments for amblyopia: Vision therapy and patching, perceptual learning and patching alone.

Active vision therapy for amblyopia shows good results, but there is no standard vision therapy protocol. This study compared the results of three treatments, two combining patching with active therapy and one with patching alone, in a sample of children with amblyopia. Two protocols have been developed: (a) perceptual learning with a computer game designed to favour the medium-to-high spatial frequency-tuned achromatic mechanisms of parvocellular origin and (b) vision therapy with a specific protocol and 2-h patching. The third treatment group used patching only. Fifty-two amblyopic children (aged 4-12 years), were randomly assigned to three monocular treatment groups: 2-h patching (n = 18), monocular perceptual learning (n = 17) and 2-h patching plus vision therapy (n = 17). Visual outcomes were analysed after 3 months and compared with a control group (n = 36) of subjects with normal vision. Visual acuity (VA) and stereoacuity (STA) improved significantly after treatment for the three groups with the best results for patching plus vision therapy, followed by monocular perceptual learning, with patching only least effective. Change in the interocular difference in VA was significant for monocular perceptual learning, followed by patching. Differences in STA between groups were not significant. For VA and interocular differences, the final outcomes were influenced by the baseline VA and interocular difference, respectively, with greater improvements in subjects with poorer initial values. Visual acuity and STA improved with the two most active treatments, that is, vision therapy followed by perceptual learning. Patching alone showed the worst outcome. These results suggest that vision therapy should include monocular accommodative exercises, ocular motility and central fixation exercises where the fovea is more active.

Fast Hyperspectral Image Classification with Strong Noise Robustness Based on Minimum Noise Fraction

A fast hyperspectral image classification algorithm with strong noise robustness is proposed in this paper, aiming at the hyperspectral image classification problems under noise interference. Based on the Fast 3D Convolutional Neural Network (Fast-3DCNN), this algorithm enables the classification model to have good tolerance for various types of noise by using a Minimum Noise Fraction (MNF) as dimensionality reduction module for hyperspectral image input data. In addition, by introducing lightweight hybrid attention modules with the spatial and the channel information, the deep features extracted by the Convolutional Neural Network are further refined, ensuring that the model has high classification accuracy. Public dataset experiments have shown that compared to traditional methods, the MNF in this algorithm reduces the dimensionality of input spectral data, preserves information with higher signal-to-noise ratio(SNR) in the spectral bands, and aggregates spectral features into class feature vectors, greatly improving the noise robustness of the model. At the same time, based on a lightweight spectral–spatial hybrid attention mechanism, combined with fewer spectral dimensions, the model effectively avoids overfitting. With less loss in model training speed, it achieved better classification accuracy in small-scale training sample experiments, fully demonstrating the good generalization ability of this algorithm.

Calcium signaling in mitochondrial intermembrane space.

The mitochondrial intermembrane space (IMS) is a highly protected compartment, second only to the matrix. It is a crucial bridge, coordinating mitochondrial activities with cellular processes such as metabolites, protein, lipid, and ion exchange. This regulation influences signaling pathways for metabolic activities and cellular homeostasis. The IMS harbors various proteins critical for initiating apoptotic cascades and regulating reactive oxygen species production by controlling the respiratory chain. Calcium (Ca2+), a key intracellular secondary messenger, enter the mitochondrial matrix via the IMS, regulating mitochondrial bioenergetics, ATP production, modulating cell death pathways. IMS acts as a regulatory site for Ca2+ entry due to the presence of different Ca2+ sensors such as MICUs, solute carriers (SLCs); ion exchangers (LETM1/SCaMCs); S100A1, mitochondrial glycerol-3-phosphate dehydrogenase, and EFHD1, each with unique Ca2+ binding motifs and spatial localizations. This review primarily emphasizes the role of these IMS-localized Ca2+ sensors concerning their spatial localization, mechanism, and molecular functions. Additionally, we discuss how these sensors contribute to the progression and pathogenesis of various human health conditions and diseases.

From Play to Understanding: Large Language Models in Logic and Spatial Reasoning Coloring Activities for Children

Visual thinking leverages spatial mechanisms in animals for navigation and reasoning. Therefore, given the challenge of abstract mathematics and logic, spatial reasoning-based teaching strategies can be highly effective. Our previous research verified that innovative box-and-ball coloring activities help teach elementary school students complex notions like quantifiers, logical connectors, and dynamic systems. However, given the richness of the activities, correction is slow, error-prone, and demands high attention and cognitive load from the teacher. Moreover, feedback to the teacher should be immediate. Thus, we propose to provide the teacher with real-time help with LLMs. We explored various prompting techniques with and without context—Zero-Shot, Few-Shot, Chain of Thought, Visualization of Thought, Self-Consistency, logicLM, and emotional —to test GPT-4o’s visual, logical, and correction capabilities. We obtained that Visualization of Thought and Self-Consistency techniques enabled GPT-4o to correctly evaluate 90% of the logical–spatial problems that we tested. Additionally, we propose a novel prompt combining some of these techniques that achieved 100% accuracy on a testing sample, excelling in spatial problems and enhancing logical reasoning.

Spatial Pattern and Influence Mechanisms of Forest Land Quality under the Background of Carbon Peaking and Carbon Neutrality: A Case Study in Kaizhou District, Chongqing, China

Since the goals of carbon peaking and carbon neutrality have been established, forest carbon sinks have garnered significant attention. As a fundamental component of forest carbon sinks, the quality of forest land significantly influences the carbon sink capacity of forests. This study utilized Kaizhou District, Chongqing City, a typical forest area as a case study, and used the correction method, factor method, CASA model, landscape ecology indexes, and canonical correlation analysis to evaluate the level of forest land quality and reveal the spatial distribution pattern and influencing mechanisms of forest land quality. The results showed that: (i) The quality index of public welfare forest land was distributed in [37.89, 148.15], and each quality level was diversified in space. The quality index of commodity forest land was distributed in [40.00, 92.67], and some high-quality forest land appeared in the transition zone of each region; (ii) The forest land quality index and the amount of net primary productivity passed the correlation test. Primary net productivity was higher on forest land with a high-quality index and lower on forest area with a low-quality index; (iii) public welfare forest land was mainly positively affected by community structure, average annual precipitation, average annual temperature, and soil moisture. Commodity forest land was mainly positively affected by average annual temperature, soil moisture, and slope aspect. However, landform had a significant negative impact on the two types of forest land. Given these findings, we also proposed a series of measures aimed at promoting the sustainable development of research on regional forest land.

Geometric mechanics of kiri-origami-based bifurcated mechanical metamaterials.

We explore a new design strategy of leveraging kinematic bifurcation in creating origami/kirigami-based three-dimensional (3D) hierarchical, reconfigurable, mechanical metamaterials with tunable mechanical responses. We start from constructing three basic, thick, panel-based structural units composed of 4, 6 and 8 rigidly rotatable cubes in close-looped connections. They are modelled, respectively, as 4R, 6R and 8R (R stands for revolute joint) spatial looped kinematic mechanisms, and are used to create a library of reconfigurable hierarchical building blocks that exhibit kinematic bifurcations. We analytically investigate their reconfiguration kinematics and predict the occurrence and locations of kinematic bifurcations through a trial-correction modelling method. These building blocks are tessellated in 3D to create various 3D bifurcated hierarchical mechanical metamaterials that preserve the kinematic bifurcations in their building blocks to reconfigure into different 3D architectures. By combining the kinematics and considering the elastic torsional energy stored in the folds, we develop the geometric mechanics to predict their tunable anisotropic Poisson's ratios and stiffnesses. We find that kinematic bifurcation can significantly effect mechanical responses, including changing the sign of Poisson's ratios from negative to positive beyond bifurcation, tuning the anisotropy, and overcoming the polarity of structural stiffness and enhancing the number of deformation paths with more reconfigured shapes.This article is part of the theme issue 'Origami/Kirigami-inspired structures: from fundamentals to applications'.

It's about (taking up) space: Discreteness of individuals and the strength of spatial coexistence mechanisms.

One strand of modern coexistence theory (MCT) partitions invader growth rates (IGR) to quantify how different mechanisms contribute to species coexistence, highlighting fluctuation-dependent mechanisms. A general conclusion from the classical analytic MCT theory is that coexistence mechanisms relying on temporal variation (such as the temporal storage effect) are generally less effective at promoting coexistence than mechanisms relying on spatial or spatiotemporal variation (primarily growth-density covariance). However, the analytic theory assumes continuous population density, and IGRs are calculated for infinitesimally rare invaders that have infinite time to find their preferred habitat and regrow, without ever experiencing intraspecific competition. Here we ask if the disparity between spatial and temporal mechanisms persists when individuals are, instead, discrete and occupy finite amounts of space. We present a simulation-based approach to quantifying IGRs in this situation, building on our previous approach for spatially non-varying habitats. As expected, we found that spatial mechanisms are weakened; unexpectedly, the contribution to IGR from growth-density covariance could even become negative, opposing coexistence. We also found shifts in which demographic parameters had the largest effect on the strength of spatial coexistence mechanisms. Our substantive conclusions are statements about one model, across parameter ranges that we subjectively considered realistic. Using the methods developed here, effects of individual discreteness should be explored theoretically across a broader range of conditions, and in models parameterized from empirical data on real communities.

Weak Edge Target Segmentation Network Based on Dual Attention Mechanism

Segmentation of weak edge targets such as glass and plastic poses a challenge in the field of target segmentation. The detection process is susceptible to background interference and various external factors due to the transparent nature of these materials. To address this issue, this paper introduces a segmentation network for weak edge target objects (WETS-Net). To effectively extract edge information of such objects and eliminate redundant information during feature extraction, a dual-attention mechanism is employed, including the Edge Attention Extraction Module (EAEM) and the Multi-Scale Information Fusion Module (MIFM). Specifically, the EAEM combines improved edge feature extraction kernels to selectively enhance the importance of edge features, aiding in more precise target region extraction. The MIFM utilizes spatial attention mechanisms to fuse multi-scale features, reducing background and external interference. These innovations enhance the performance of WETS-Net, offering a new direction for weak edge target segmentation research. Finally, through ablation experiments, the effectiveness of each module is effectively validated. Moreover, the proposed algorithm achieves an average detection accuracy of 95.83% and 96.13% on the dataset and a self-made dataset, respectively, outperforming similar U-Net-improved networks.

Solar Radiation Prediction Model Based on Spatial Attention Mechanisms and Sun Position Feature Maps

This study introduces a novel method for predicting solar radiation by focusing on the sun's position and cloud conditions in the sky. The proposed approach utilizes sun position feature maps and spatial attention mechanisms to improve the accuracy of solar irradiance prediction. Key contributions include the development of sun position feature maps to precisely locate the sun in all-sky images and the integration of spatial attention mechanisms with convolutional neural networks to enhance predictive modeling. Comparative analysis with three other models demonstrates the effectiveness of the proposed approach, particularly in overcast conditions. These innovative techniques significantly enhance the accuracy of solar radiation prediction and offer valuable insights for the design and operation of renewable energy systems.

Remote sensing scene classification based on contextual attention mechanism of lie group space

ABSTRACT High-resolution remote sensing (HRRS) scene classification is based on the semantic information in the image and its contextual spatial correlation, labelling different semantic categories for different HRRS. Convolutional neural networks (CNN) have been widely studied and applied in remote sensing scene classification in recent years. However, most of the existing CNN models focus on the global and high-level semantic features, ignoring the shallower feature information. In addition, complex background information and variable scales lead to a series of problems with large intra-class differences and high inter-class similarities, which also brings challenges to scene classification. To address the above problems and challenges, a scenario classification model based on the contextual spatial attention and channel attention mechanism of Lie Group manifold space learning is proposed in this study. In this model, we fully explore the multi-scale features of the scene (shallower-level and high-level) and propose a novel contextual spatial attention mechanism and channel attention mechanism. Extensive experimentation was carried out on the Union Remote Sensing Image Data Set (URSIS), which improved by 7.13% compared with the classical model. The experimental results showed that compared with other state-of-the-art remote sensing scene classification models, our proposed method has achieved significant improvement in classification accuracy and performance.