Search Results (4,754)

Background. To investigate the behavior of spiking neural networks (SNNs), the sensitivity of input perturbation serves as an effective metric for assessing the influence on the network output. However, existing methods fall short in evaluating the sensitivity of SNNs featuring biologically plausible leaky integrate-and-fire (LIF) neurons due to the intricate neuronal dynamics during the feedforward process. Methods. This paper first defines the sensitivity of a temporal-coded spiking neuron (SN) as the deviation between the perturbed and unperturbed output under a given input perturbation with respect to overall inputs. Then, the sensitivity algorithm of an entire SNN is derived iteratively from the sensitivity of each individual neuron. Instead of using the actual firing time, the desired firing time is employed to derive a more precise analytical expression of the sensitivity. Moreover, the expectation of the membrane potential difference is utilized to quantify the magnitude of the input deviation. Results/Conclusions. The theoretical results achieved with the proposed algorithm are in reasonable agreement with the simulation results obtained with extensive input data. The sensitivity also varies monotonically with changes in other parameters, except for the number of time steps, providing valuable insights for choosing appropriate values to construct the network. Nevertheless, the sensitivity exhibits a piecewise decreasing tendency with respect to the number of time steps, with the length and starting point of each piece contingent upon the specific parameter values of the neuron. Full article

Understanding the complex variations in water use efficiency (WUE) is critical for optimizing agricultural productivity and resource management. Traditional analytical methods often fail to capture the nonlinear and multiscale variations inherent in WUE, where multifractal theory offers distinct advantages. Given its limited application in WUE studies, this paper analyzes the spatiotemporal characteristics and influencing factors of the WUE in Anhui Province from 2001 to 2022 using a multifractal, multiscale approach. The results indicated that the WUE exhibited significant interannual variation, peaking in summer, especially in August (2.4552 gC·mm⁻¹·m⁻²), with the monthly average showing an inverted “V” shape. Across different spatial and temporal scales, the WUE displayed clear multifractal characteristics. Temporally, the variation in fractal features between years was not prominent, while inter-seasonal variation was most complex in August during summer. Spatially, the most distinct multifractal patterns were observed in hilly and mountainous areas, particularly in regions with brown soil distribution. Rainfall was identified as the primary natural driver influencing regional WUE changes. This study aims to promote the sustainable use of water resources while ensuring the stability of agricultural production within protected farmlands. Full article

Index modulation (IM) is considered a promising approach for fifth-generation wireless systems due to its spectral efficiency and reduced complexity compared to conventional modulation techniques. However, IM faces difficulties in environments with unpredictable channel conditions, particularly in accurately detecting index values and dynamically adjusting index assignments. Deep learning (DL) offers a potential solution by improving detection performance and resilience through the learning of intricate patterns in varying channel conditions. In this paper, we introduce a robust detection method based on a hybrid DL (HDL) model designed specifically for orthogonal frequency-division multiplexing with IM (OFDM-IM) in challenging channel environments. Our proposed HDL detector leverages a one-dimensional convolutional neural network (1D-CNN) for feature extraction, followed by a bidirectional long short-term memory (Bi-LSTM) network to capture temporal dependencies. Before feeding data into the network, the channel matrix and received signals are preprocessed using domain-specific knowledge. We evaluate the bit error rate (BER) performance of the proposed model using different optimizers and equalizers, then compare it with other models. Moreover, we evaluate the throughput and spectral efficiency across varying SNR levels. Simulation results demonstrate that the proposed hybrid detector surpasses traditional and other DL-based detectors in terms of performance, underscoring its effectiveness for OFDM-IM under uncertain channel conditions. Full article

In response to the shortcomings of current spatiotemporal prediction models, which frequently encounter difficulties in temporal feature extraction and the forecasting of medium to high echo intensity regions over extended sequences, this study presents a novel model for radar echo extrapolation that combines a translator encoder-decoder architecture with a spatiotemporal dual-discriminator conditional generative adversarial network (STD-TranslatorNet). Initially, an image reconstruction network is established as the generator, employing a combination of a temporal attention unit (TAU) and an encoder–decoder framework. Within this architecture, both intra-frame static attention and inter-frame dynamic attention mechanisms are utilized to derive attention weights across image channels, thereby effectively capturing the temporal evolution of time series images. This approach enhances the network’s capacity to comprehend local spatial features alongside global temporal dynamics. The encoder–decoder configuration further bolsters the network’s proficiency in feature extraction through image reconstruction. Subsequently, the spatiotemporal dual discriminator is crafted to encapsulate both temporal correlations and spatial attributes within the generated image sequences. This design serves to effectively steer the generator’s output, thereby augmenting the realism of the produced images. Lastly, a composite multi-loss function is proposed to enhance the network’s capability to model intricate spatiotemporal evolving radar echo data, facilitating a more comprehensive assessment of the quality of the generated images, which in turn fortifies the network’s robustness. Experimental findings derived from the standard radar echo dataset (SRAD) reveal that the proposed radar echo extrapolation technique exhibits superior performance, with average critical success index (CSI) and probability of detection (POD) metrics per frame increasing by 6.9% and 7.6%, respectively, in comparison to prior methodologies. Full article

Biometrics play an important role in modern access control and security systems. The need of novel biometrics to complement traditional biometrics has been at the forefront of research. The Facial Blood Flow (FBF) biometric trait, recently proposed by our team, is a spatio-temporal representation of facial blood flow, constructed using motion magnification from facial areas where skin is visible. Due to its design and construction, the FBF does not need information from the eyes, nose, or mouth, and, therefore, it yields a versatile biometric of great potential. In this work, we evaluate the effectiveness of novel temporal partitioning and Fast Fourier Transform-based features that capture the temporal evolution of facial blood flow. These new features, along with a “time-distributed” Convolutional Neural Network-based deep learning architecture, are experimentally shown to increase the performance of FBF-based person identification compared to our previous efforts. This study provides further evidence of FBF’s potential for use in biometric identification. Full article

Reliable environmental perception capabilities are a prerequisite for achieving autonomous driving. Cameras and LiDAR are sensitive to illumination and weather conditions, while millimeter-wave radar avoids these issues. Existing models rely heavily on image-based approaches, which may not be able to fully characterize radar sensor data or efficiently further utilize them for perception tasks. This paper rethinks the approach to modeling radar signals and proposes a novel U-shaped multilayer perceptron network (U-MLPNet) that aims to enhance the learning of omni-dimensional spatio-temporal dependencies. Our method involves innovative signal processing techniques, including a 3D CNN for spatio-temporal feature extraction and an encoder–decoder framework with cross-shaped receptive fields specifically designed to capture the sparse and non-uniform characteristics of radar signals. We conducted extensive experiments using a diverse dataset of urban driving scenarios to characterize the sensor’s performance in multi-view semantic segmentation and object detection tasks. Experiments showed that U-MLPNet achieves competitive performance against state-of-the-art (SOTA) methods, improving the mAP by 3.0% and mDice by 2.7% in RD segmentation and AR and AP by 1.77% and 2.03%, respectively, in object detection. These improvements signify an advancement in radar-based perception for autonomous vehicles, potentially enhancing their reliability and safety across diverse driving conditions. Full article

In recent years, mobility as a service (MaaS) has been thought as one of the opportunities for shifting towards shared travel solutions with respect to private transport modes, particularly owned cars. Although many MaaS aspects have been explored in the literature, there are still issues, such as platform implementations, travel solution generation, and the user’s role for making an effective system, that require more research. This paper extends and improves a previous study carried out by the authors by providing more details and experiments. The paper proposes a diachronic network model for representing travel services available in a given MaaS platform by using an agent-based approach to simulate the interactions between travel operators and travelers. Particularly, the diachronic network model allows the consideration of both the spatial and temporal features of the available transport services, while the agent-based framework allows the representation of how shared services might be used and which effects, in terms of modal split, could be expected. The final aim is to provide insights for setting the architecture of an agent-based MaaS platform where transport operators would share their data for providing seamless travel opportunities to travelers. The results obtained for a simulated test case are promising. Particularly, there are interesting findings concerning the traffic congestion boundary values that would move users towards shared travel solutions. Full article

Hydropower generation is influenced by various factors such as precipitation, temperature, and installed capacity, with hydrometeorological factors exhibiting significant temporal variability. This study proposes a hydropower generation forecasting method based on Transformer and SE-Attention for different provinces. In the model, the outputs of the Transformer and SE-Attention modules are fed into an LSTM layer to capture long-term data dependencies. The SE-Attention module is reintroduced to enhance the model’s focus on important temporal features, and a linear layer maps the hidden state of the last time step to the final output. The proposed Transformer-LSTM-SE model was tested using provincial hydropower generation data from Yunnan, Sichuan, and Chongqing. The experimental results demonstrate that this model achieves high accuracy and stability in medium- and long-term hydropower forecasting at the provincial level, with an average accuracy improvement of 33.79% over the LSTM model and 24.30% over the Transformer-LSTM model. Full article

As urbanization accelerates globally, urban areas have become major sources of greenhouse gas emissions. In this context, urban parks are crucial as significant components of carbon sinks. Using Shanghai Century Park as a case study, this study aims to develop an applicable and reliable workflow to accurately assess the carbon sequestration capacity of urban parks from a spatial–temporal perspective. Firstly, the random forest model is employed for biotope classification and mapping in the park based on multi-source data, including raw spectral bands, vegetation indices, and texture features. Subsequently, the Net Primary Productivity and biomass of different biotope types are calculated, enabling dynamic monitoring of the park’s carbon sequestration capacity from 2018 to 2023. Moreover, the study explores the main factors influencing changes in carbon sequestration capacity from the management perspective. The findings reveal: (1) The application of multi-source imagery data enhances the accuracy of biotope mapping, with winter imagery proving more precise in classification. (2) From 2018 to 2023, Century Park’s carbon sequestration capacity showed a fluctuating upward trend, with significant variations in the carbon sequestration abilities of different biotope types within the park. (3) Renovation and construction work related to biotope types significantly impacted the park’s carbon sequestration capacity. Finally, the study proposes optimization strategies focused on species selection and layout, planting density, and park management. Full article

The proliferation of AI-generated content (AIGC) has empowered non-experts to create highly realistic Deepfake images and videos using user-friendly software, posing significant challenges to the legal system, particularly in criminal investigations, court proceedings, and accident analyses. The absence of reliable Deepfake verification methods threatens the integrity of legal processes. In response, researchers have explored deep forgery detection, proposing various forensic techniques. However, the swift evolution of deep forgery creation and the limited generalizability of current detection methods impede practical application. We introduce a new deep forgery detection method that utilizes image decomposition and lighting inconsistency. By exploiting inherent discrepancies in imaging environments between genuine and fabricated images, this method extracts robust lighting cues and mitigates disturbances from environmental factors, revealing deeper-level alterations. A crucial element is the lighting information feature extractor, designed according to color constancy principles, to identify inconsistencies in lighting conditions. To address lighting variations, we employ a face material feature extractor using Pattern of Local Gravitational Force (PLGF), which selectively processes image patterns with defined convolutional masks to isolate and focus on reflectance coefficients, rich in textural details essential for forgery detection. Utilizing the Lambertian lighting model, we generate lighting direction vectors across frames to provide temporal context for detection. This framework processes RGB images, face reflectance maps, lighting features, and lighting direction vectors as multi-channel inputs, applying a cross-attention mechanism at the feature level to enhance detection accuracy and adaptability. Experimental results show that our proposed method performs exceptionally well and is widely applicable across multiple datasets, underscoring its importance in advancing deep forgery detection. Full article

The increase in motorized vehicles in cities and the inefficient use of parking spaces have exacerbated parking difficulties in cities. To effectively improve the utilization rate of parking spaces, it is necessary to accurately predict future parking demand. This paper proposes a deep learning model based on multi-graph convolutional Transformer, which captures geographic spatial features through a Multi-Graph Convolutional Network (MGCN) module and mines temporal feature patterns using a Transformer module to accurately predict future multi-step parking demand. The model was validated using historical parking transaction volume data from all on-street parking lots in Nanshan District, Shenzhen, from September 2018 to March 2019, and its superiority was verified through comparative experiments with benchmark models. The results show that the MGCN–Transformer model has a MAE, RMSE, and R² error index of 0.26, 0.42, and 95.93%, respectively, in the multi-step prediction task of parking demand, demonstrating its superior predictive accuracy compared to other benchmark models. Full article

A systematic understanding of the spatial-temporal evolution patterns of cultivated land fragmentation (CLF), its driving factors, and its relationship with the expansion of urban and rural construction land is essential for identifying strategies to mitigate CLF in rapidly urbanizing regions. This study combined landscape fragmentation with ownership fragmentation, analyzing CLF through three dimensions: resource endowment, spatial concentration, and convenience of utilization, with eight selected indicators. By comparing village-level data from 2013 to 2022, we explored the key drivers of CLF and its conflicts with urban and rural construction land expansion. The findings indicate a clear spatial variation in village-level CLF in Ezhou, characterized by low fragmentation in the northwest and northeast, and high fragmentation in the southwest and central regions. This pattern is in contrast to Ezhou’s economic development, which decreased progressively from east to north and south. Over the study period, village-level CLF in Ezhou evolved from being primarily moderately and relatively severely fragmented to predominantly severely and relatively severely fragmented, with an overall declining trend and more pronounced polarization. At the same time, the CLF within the village region demonstrated notable spatial clustering features, with a rapid increase observed between 2013 and 2022. It was also discovered that CLF is driven by various factors, with the main influences being the proportion of construction land, land use intensity, and population density. Cultivated land is the main source of both urban construction land (UCL) and rural construction land (RCL), with average contribution rates of 46.47% and 62.62%, respectively. This research offers empirical evidence for rapid urbanization and serves as a critical reference for rural revitalization and coordinated urban–rural development, with potential guidance for future policy formulation and implementation. Full article

Background/Objective: Muscle synergy analysis based on machine learning has significantly advanced our understanding of the mechanisms underlying the central nervous system motor control of gait and has identified abnormal gait synergies in stroke patients through various analytical approaches. However, discrepancies in experimental conditions and computational methods have limited the clinical application of these findings. This review seeks to integrate the results of existing studies on the features of muscle synergies in stroke-related gait abnormalities and provide clinical and research insights into gait rehabilitation. Methods: A systematic search of Web of Science, PubMed, and Scopus was conducted, yielding 10 full-text articles for inclusion. Results: By comprehensively reviewing the consistencies and differences in the study outcomes, we emphasize the need to segment the gait cycle into specific phases (e.g., weight acceptance, push-off, foot clearance, and leg deceleration) during the treatment process of gait rehabilitation and to develop rehabilitation protocols aimed at restoring normal synergy patterns in each gait phase and fractionating reduced synergies. Conclusions: Future research should focus on validating these protocols to improve clinical outcomes and introducing indicators to assess abnormalities in the temporal features of muscle synergies. Full article

Driven by the widespread adoption of deep learning (DL) in crop mapping with satellite image time series (SITS), this study was motivated by the recent success of temporal attention-based approaches in crop mapping. To meet the needs of beekeepers, this study aimed to develop DL-based classification models for mapping five essential crops in pollination services in Quebec province, Canada, by using Sentinel-2 SITS. Due to the challenging task of crop mapping using SITS, this study employed three DL-based models, namely one-dimensional temporal convolutional neural networks (CNNs) (1DTempCNNs), one-dimensional spectral CNNs (1DSpecCNNs), and long short-term memory (LSTM). Accordingly, this study aimed to capture expert-free temporal and spectral features, specifically targeting temporal features using 1DTempCNN and LSTM models, and spectral features using the 1DSpecCNN model. Our findings indicated that the LSTM model (macro-averaged recall of 0.80, precision of 0.80, F1-score of 0.80, and ROC of 0.89) outperformed both 1DTempCNNs (macro-averaged recall of 0.73, precision of 0.74, F1-score of 0.73, and ROC of 0.85) and 1DSpecCNNs (macro-averaged recall of 0.78, precision of 0.77, F1-score of 0.77, and ROC of 0.88) models, underscoring its effectiveness in capturing temporal features and highlighting its suitability for crop mapping using Sentinel-2 SITS. Furthermore, applying one-dimensional convolution (Conv1D) across the spectral domain demonstrated greater potential in distinguishing land covers and crop types than applying it across the temporal domain. This study contributes to providing insights into the capabilities and limitations of various DL-based classification models for crop mapping using Sentinel-2 SITS. Full article

Graph convolutional networks (GCN) are an important research method for intelligent transportation systems (ITS), but they also face the challenge of how to describe the complex spatio-temporal relationships between traffic objects (nodes) more effectively. Although most predictive models are designed based on graph convolutional structures and have achieved effective results, they have certain limitations in describing the high-order relationships between real data. The emergence of hypergraphs breaks this limitation. A dynamic spatio-temporal hypergraph convolutional network (DSTHGCN) model is proposed in this paper. It models the dynamic characteristics of traffic flow graph nodes and the hyperedge features of hypergraphs simultaneously, achieving collaborative convolution between graph convolution and hypergraph convolution (HGCN). On this basis, a hyperedge outlier removal mechanism (HOR) is introduced during the process of node information propagation to hyper-edges, effectively removing outliers and optimizing the hypergraph structure while reducing complexity. Through in-depth experimental analysis on real-world datasets, this method has better performance compared to other methods. Full article