Fusion Approach Research Articles

A Novel Fusion of Machine Learning Methods for Enhancing Named Entity Recognition in Indonesian Language Text

One of the important implementations in machine learning is Named Entity Recognition (NER), which is used to process text and extract entities such as people, organizations, laws, religions, and locations. NER for the Indonesian language still faces significant challenges due to the lack of high-quality labelled datasets, which limits the development of more advanced models. To address this issue, we utilized several pre-trained BERT models (bert-base-uncased, indobenchmark/indobert-base-p1, indolem/indobert-base-uncased) and datasets (NERGRIT-IndoNLU, NERGRIT-Corpus, NERUGM, and NERUI). This study proposes a novel fusion approach by integrating deep learning architectures such as CNN, Bi-LSTM, Bi-GRU, and CRF to detect 19 entities. This approach enhances BERT’s sequence modelling and feature extraction capabilities, while CRF improves entity prediction by enforcing global word-sequence constraints. Experimental results demonstrate that the fusion approach outperforms previous methods. On the bert-base-uncased dataset, accuracy reached 94.75%, while indobenchmark/indobert-base-p1 achieved 95.75%, and indolem/indobert-base-uncased achieved 95.85%. This study emphasizes the effectiveness of combining deep learning architectures with pre-trained transformers to improve NER performance in the Indonesian language. The proposed methodology offers significant advancements in entity extraction for languages with limited datasets, such as Indonesian.

Dense projection fusion for 3D object detection

Fusing information from LiDAR and cameras can effectively enhance the overall perceptivity of autonomous vehicles in various scenarios. Despite the relatively good results achieved by point-wise fusion and Bird’s-Eye-View (BEV) fusion, they still cannot fully leverage the image information and lack of effective depth information. For any fusion methods, the multi-modal features first need to be concatenated along the channel, and then the fused features are extracted using convolutional layers. However, this type of fusion methods is effective, but too coarse which causes that the fused features cannot pay more attention to the regions with important features and suffer from severe noise. To tackle these issues, we propose in this paper a Dense Projection Fusion (DPFusion) approach. It consists of two new modules: dense depth map guided BEV transform (DGBT) module and multi-modal feature adaptive fusion (MFAF) module. The DGBT module first quickly estimates the depth of each pixel and then projects all image features to the BEV space, making full use of the image information. The MFAF module computes the image weights and point cloud weights for each channel in each BEV grid and then adaptively weights and fuses the image BEV features with the point cloud BEV features. It is worth pointing out that the MFAF module makes the fused features pay more attention to background outline and object outline. Our proposed DPFusion demonstrates competitive results in 3D object detection, achieving a mean Average Precision (mAP) of 70.4 and a nuScenes detection score (NDS) of 72.3 on the nuScenes validation set.

Fuzzy Linear Temporal Logic with Quality Constraints

As an extension of quantitative temporal logic, uncertain temporal logic essentially describes the temporal behavior of uncertain and incomplete systems, thus better solving search and decision-making problems in such systems. Fuzzy linear temporal logic (FLTL) is a focal point in uncertain temporal logic research. However, there are evident shortcomings in the current research outcomes. First, in previous FLTL studies, the practice of obtaining path reachability and formula satisfaction values independently and subsequently selecting the smaller of the two as the satisfaction value metric led to information loss. Furthermore, this simplistic information fusion approach fails to reflect the varying importance of these two types of information to the requirements. Second, computing path reachability and temporal logic formula satisfaction values separately may result in a mismatch between the two pieces of information with respect to the same path segment. Thus, the primary challenge lies in accurately integrating the satisfaction values of temporal logic formulas with the path reachability of the segments that yields these satisfaction values, utilizing various reasonable information synthesis methods, to ensure synchronization between path reachability and formula satisfaction values without incurring information loss. Additionally, it is crucial to reflect the different preference requirements for these two types of information. Moreover, the temporal logic formula characterizes system properties, with its sub-formulas delineating distinct sub-properties. Consequently, considering the varying importance preferences of sub-formulas is also significant. To address these deficiencies, we introduced quality constraint operators into FLTL, resulting in quality-constrained fuzzy linear temporal logic (QFLTL). This incorporation enables the synchronization and comprehensive fusion of path-reachability information and formula satisfaction values within the final semantic metric, thereby resolving the issues related to information synchronization and loss. Furthermore, it can accommodate the differing preference requirements between the two types of information and sub-properties during the information synthesis process. We defined the syntax and semantics of QFLTL and examined its expressive power and properties. Notably, we investigated the decidability of logical decision problems in QFLTL, encompassing validity, satisfiability, and model-checking issues. We proposed corresponding solution algorithms and analyzed their complexities.

On the Utilization of Emoji Encoding and Data Preprocessing with a Combined CNN-LSTM Framework for Arabic Sentiment Analysis

Social media users often express their emotions through text in posts and tweets, and these can be used for sentiment analysis, identifying text as positive or negative. Sentiment analysis is critical for different fields such as politics, tourism, e-commerce, education, and health. However, sentiment analysis approaches that perform well on English text encounter challenges with Arabic text due to its morphological complexity. Effective data preprocessing and machine learning techniques are essential to overcome these challenges and provide insightful sentiment predictions for Arabic text. This paper evaluates a combined CNN-LSTM framework with emoji encoding for Arabic Sentiment Analysis, using the Arabic Sentiment Twitter Corpus (ASTC) dataset. Three experiments were conducted with eight-parameter fusion approaches to evaluate the effect of data preprocessing, namely the effect of emoji encoding on their real and emotional meaning. Emoji meanings were collected from four websites specialized in finding the meaning of emojis in social media. Furthermore, the Keras tuner optimized the CNN-LSTM parameters during the 5-fold cross-validation process. The highest accuracy rate (91.85%) was achieved by keeping non-Arabic words and removing punctuation, using the Snowball stemmer after encoding emojis into Arabic text, and applying Keras embedding. This approach is competitive with other state-of-the-art approaches, showing that emoji encoding enriches text by accurately reflecting emotions, and enabling investigation of the effect of data preprocessing, allowing the hybrid model to achieve comparable results to the study using the same ASTC dataset, thereby improving sentiment analysis accuracy.

A novel approach to aeroengine performance diagnosis based on physical model coupling data-driven using low-rank multimodal fusion method

Aeroengine health assessment plays a pivotal role in ensuring flight safety and reliability. Traditionally, this process involves diagnosing the performance of the aeroengine gas path. However, owing to the intricacies of operating conditions, non-linear performance, and the interplay of gas path performance fault characteristics, determining the aeroengine health condition directly from engine monitoring information poses a significant challenge, particularly in cases of insufficient sensor data. To address these challenges, a novel digital twin method for aeroengine performance diagnosis has been proposed. This method integrates data-driven and performance models, employing a low-rank multimodal fusion approach. By digitizing the physical system or process through mathematical models and simulation technology, this approach presents distinct advantages compared to previous methods relying solely on models or data. At the aeroengine component level, an adaptive model was implemented, and the data-driven model was constructed using flight data. Gas path fault classification employed support vector machines. The engine digital twin was established through low-order multimodal fusion. Results indicate that the proposed method attains excellent diagnostic accuracy under both steady and transient conditions. It can be harnessed to enhance engine performance monitoring and evaluation, thereby improving the reliability, availability, and efficiency of the engine.

FNNGM: A neural-driven fractional-derivative multivariate fusion model for interpretable real-time CPI forecasts

Integrating models from diverse sources has attracted substantial interest in developing advanced time series forecasting technologies. However, current research lacks a comprehensive and deep fusion model to integrate multiple forecasting methodologies. To this end, this paper proposes a neural-driven fractional-derivative multivariate fusion model (FNNGM (p, n)) to assimilate the fractional-derivative dynamical system, the driving factor in grey multivariate models, and the neural network into a cohesive framework. Consequently, this fusion architecture can benefit from the synergy of the target system's dynamics, extensive exogenous information, and non-linear transformation. Additionally, FNNGM (p, n) fosters extra functionalities through its inherent memory layer and sequence decomposition, bolstering model interpretability with the visible memory mechanism and understandable model workflows. To showcase the utility of FNNGM (p, n), this paper conducts real-time monthly consumer price index (CPI) forecasts that span ten years (from 2013:08 to 2023:07), analyzing the interpretable results from FNNGM (p, n) and contrasting it against many prevailing benchmark models. The comparison results reveal FNNGM (p, n)’s highly concentrated error distributions and the minimum mean absolute percentage forecasting error (APFE), squared forecasting error (SFE), and absolute forecasting error (AFE) values of 0.22 %, 0.59, and 0.56, respectively. Furthermore, the ablation experiments are performed to explore the specific effects and compatibilities of the fusion components, validating the effectiveness of the proposed fusion approach.

Feature-level Fusion vs. Score-level Fusion for Image Retrieval Based on Pre-trained Deep Neural Networks

Today’s complex multimedia content made retrieving images similar to the user’s query from the database a challenging task. The performance of a Content-Based Image Retrieval System (CBIR) system highly depends on the image representation in a form of low-level features and similarity measurement. The traditional visual descriptors that do not provide good prior domain knowledge could lead to poor performance retrieval results. On the other hand, Deep Convolutional Neural Networks (DCNNs) have recently achieved a remarkable success as methods for image classification in various domains. Recently, pre-trained deep convolution neural networks on thousands of classes have the ability to extract very accurate and representative features which, in addition to classification, can also be successfully used in image retrieval systems. ResNet152, GoogLeNet and InceptionV3 are some of the effective and successful examples of pre-trained DCNNs recently applied in a computer vision tasks such as object recognition, clustering, and classification. In this paper, two approaches for a CBIR system, namely early fusion and late fusion, have been presented and compared. The early fusion utilizes concatenation of the features extracted by each possible pair of DCNNs, that is ResNet152-GoogLeNet, ResNet152-InceptionV3, and GoogLeNet-InceptionV3, and the late fusion apply CombSum method with Z-Score standardization to combine the score results provided by each DCNN of the aforementioned pairs. In the experiments on a popular WANG dataset it has been shown that late fusion approach slightly outperforms early fusion approach. The best performance of our experiments in terms of Average Precision (AP) for the top 20 results reaches 96.82%.

A New Coupled Biogeochemical Modeling Approach Provides Accurate Predictions of Methane and Carbon Dioxide Fluxes Across Diverse Tidal Wetlands

AbstractTidal wetlands provide valuable ecosystem services, including storing large amounts of carbon. However, the net exchanges of carbon dioxide (CO2) and methane (CH4) in tidal wetlands are highly uncertain. While several biogeochemical models can operate in tidal wetlands, they have yet to be parameterized and validated against high‐frequency, ecosystem‐scale CO2 and CH4 flux measurements across diverse sites. We paired the Cohort Marsh Equilibrium Model (CMEM) with a version of the PEPRMT model called PEPRMT‐Tidal, which considers the effects of water table height, sulfate, and nitrate availability on CO2 and CH4 emissions. Using a model‐data fusion approach, we parameterized the model with three sites and validated it with two independent sites, with representation from the three marine coasts of North America. Gross primary productivity (GPP) and ecosystem respiration (Reco) modules explained, on average, 73% of the variation in CO2 exchange with low model error (normalized root mean square error (nRMSE) <1). The CH4 module also explained the majority of variance in CH4 emissions in validation sites (R2 = 0.54; nRMSE = 1.15). The PEPRMT‐Tidal‐CMEM model coupling is a key advance toward constraining estimates of greenhouse gas emissions across diverse North American tidal wetlands. Further analyses of model error and case studies during changing salinity conditions guide future modeling efforts regarding four main processes: (a) the influence of salinity and nitrate on GPP, (b) the influence of laterally transported dissolved inorganic C on Reco, (c) heterogeneous sulfate availability and methylotrophic methanogenesis impacts on surface CH4 emissions, and (d) CH4 responses to non‐periodic changes in salinity.

IFNet: Data-driven multisensor estimate fusion with unknown correlation in sensor measurement noises

In recent years, multisensor fusion for state estimation has gained considerable attention. The effectiveness of the optimal fusion estimation method heavily relies on the correlation among sensor measurement noises. To enhance estimate fusion performance by mining unknown correlation in the data, this paper introduces a novel multisensor fusion approach using an information filtering neural network (IFNet) for discrete-time nonlinear state space models with cross-correlated measurement noises. The method presents three notable advantages: First, it offers a data-driven perspective to tackle uncertain correlation in multisensor estimate fusion while preserving the interpretability of the information filtering. Second, by harnessing the RNN’s capability to manage data streams, it can dynamically update the fusion weights between sensors to improve fusion accuracy. Third, this method has a lower complexity than the state-of-the-art KalmanNet measurement fusion method when dealing with the fusion problem involving a large number of sensors. Numerical simulations demonstrate that IFNet exhibits better fusion accuracy than traditional filtering methods and KalmanNet fusion filtering when correlation among measurement noises is unknown.

A black tea quality testing method for scale production using CV and NIRS with TCN for spectral feature extraction

To rigorously assess black tea quality in large-scale production, this study introduces a multi-modal fusion approach integrating computer vision (CV) with Near-Infrared Spectroscopy (NIRS). CV technology is first applied to evaluate the tea's appearance quality, while NIRS quantifies key chemical components, including tea polyphenols (TP), free amino acids (FAA), and caffeine (CAF). Additionally, different methods are employed to extract potential quality features from NIR spectra. The information are then fused, and a classifier is utilized to accurately identify tea quality. Results show that the Temporal Convolutional Network (TCN) fused model achieves a 98.2 % accuracy rate, surpassing both the Convolutional Neural Network (CNN) fused model and traditional methods. This study demonstrates that TCNs effectively extract spectral features and that data fusion significantly enhances tea quality testing, offering valuable insights for production optimization.

FusionNGFPE: An Image Fusion Approach Driven by Non-global Fuzzy Pre-enhancement Framework

The majority of prevailing image fusion methods employ a global strategy, often resulting in a reduction of contrast. This study addresses this issue by proposing a novel image fusion approach called FusionNGFPE, specifically designed for the structural characteristics of infrared (IR) imagery. The approach introduces a contrast equalization algorithm based on the Fourth-order Partial Differential Equation (FPDE) to enhance background regions effectively. Considering the inherent differences between IR and visible (VIS) images, we developed a hybrid fusion strategy that combines the Expectation Maximization (EM) algorithm and Principal Component Analysis (PCA). Comparative analysis with state-of-the-art fusion methods shows that our proposed algorithm achieves superior performance in both qualitative and quantitative evaluations. To further demonstrate the practical significance of FusionNGFPE, we integrated this fusion framework into the RGBT target tracking task using the VOT-RGBT and OTCBVS datasets. Extensive comparative experiments confirm that the FusionNGFPE framework integrates seamlessly with the tracking task, significantly improving tracking accuracy across diverse scenarios.

Improving heat transfer prediction across catalytic SiC interface through a multiscale framework leveraging machine learning and data fusion techniques

Accurate aerothermal prediction under hypersonic flow conditions is crucial for any thermal protection materials design and engineering. The surface catalytic effect, where dissociated atoms recombine into their molecular forms at the air-solid interface, is an exothermic reaction and plays an important role in high-enthalpy aerodynamic environment prediction. Taking SiC, a widely recognized high-temperature thermal protection material as an example, a multiscale fusion approach for aerothermal prediction is proposed. The approach utilizes reactive molecular dynamics method to construct an interface catalysis model, and the Bayesian maximum entropy to integrate experimental and simulational data into an optimized database. Subsequently, using the radial basis function neural network algorithm, a machine learning-based reaction kinetics model with precise analysis of surface catalysis is trained. The obtained catalytic recombination efficiency is used as a boundary input for computational fluid dynamics simulation, enabling rapid and accurate prediction of hypersonic thermal environment. The result shows that the predicted surface heat flux by this novel approach is consistent with the benchmark studies, but comes with significantly reduced computation time. The stagantion heat flux predictions based on the assumptions of full catalytic wall, finite rate catalytic wall (from the proposed multiscale fusion method), and non-catalytic wall are determined to be 8.65×106 W/m2, 7.51×106 W/m2, and 4.02×106 W/m2, respectively. Comparing these with the wind tunnel benchmark value of 7.48×106 W/m2, the error from the proposed multiscale fusion method is 0.30 %, indicating significant enhancement in prediction accuracy through the proposed upscaling method. The new approach could not only reveal complex exothermic reaction processes at the interface, but also enhance the prediction accuracy at much higher computational efficiency, providing an alternative for multiscale modeling of complex flow and heat transfer at the interface.

Noninvasive Blood Glucose Monitoring Using Spatiotemporal ECG and PPG Feature Fusion and Weight-Based Choquet Integral Multimodel Approach.

change of blood glucose (BG) level stimulates the autonomic nervous system leading to variation in both human's electrocardiogram (ECG) and photoplethysmogram (PPG). In this article, we aimed to construct a novel multimodal framework based on ECG and PPG signal fusion to establish a universal BG monitoring model. This is proposed as a spatiotemporal decision fusion strategy that uses weight-based Choquet integral for BG monitoring. Specifically, the multimodal framework performs three-level fusion. First, ECG and PPG signals are collected and coupled into different pools. Second, the temporal statistical features and spatial morphological features in the ECG and PPG signals are extracted through numerical analysis and residual networks, respectively. Furthermore, the suitable temporal statistical features are determined with three feature selection techniques, and the spatial morphological features are compressed by deep neural networks (DNNs). Lastly, weight-based Choquet integral multimodel fusion is integrated for coupling different BG monitoring algorithms based on the temporal statistical features and spatial morphological features. To verify the feasibility of the model, a total of 103 days of ECG and PPG signals encompassing 21 participants were collected in this article. The BG levels of participants ranged between 2.2 and 21.8 mmol/L. The results obtained show that the proposed model has excellent BG monitoring performance with a root-mean-square error (RMSE) of 1.49 mmol/L, mean absolute relative difference (MARD) of 13.42%, and Zone A + B of 99.49% in tenfold cross-validation. Therefore, we conclude that the proposed fusion approach for BG monitoring has potentials in practical applications of diabetes management.

Mid-level data fusion strategy based on urinary nucleosides SERS spectra and blood CEA levels for enhanced preoperative detection of lymph node metastasis in colorectal cancer

BackgroundPreoperative prediction of lymph node metastasis (LNM) plays a crucial role in the treatment and prognosis of colorectal cancer (CRC). The traditional histopathological examination is invasive and time-consuming, providing pathological features only postoperatively. Preoperative serum carcinoembryonic antigen (CEA) is strongly correlated with postoperative LN status. However, the detection accuracy of LNM based on a single preoperative CEA level is low. Therefore, developing a more powerful and sensitive diagnostic tool would be of great clinical value for improving the accurate preoperative prediction of LNM in CRC patients. ResultsThis study aimed to develop a mid-level fusion approach using urinary nucleosides Raman spectra and blood CEA data to enhance the preoperative discrimination of CRC patients with and without LNM. Surface-enhanced Raman scattering (SERS) spectra of urinary modified nucleosides, isolated by affinity chromatography, were first acquired from 48 patients with LNM and 49 patients without LNM. The principal component analysis (PCA) scores obtained from the SERS spectra were then combined with preoperative blood CEA values to create a fused data array. The discriminant accuracy based on either dataset alone or the fused data was evaluated using three machine learning algorithms: linear discriminant analysis, k-nearest neighbors, and support vector machine. Results showed that the fused data could discriminate between the two groups with an accuracy of up to 91 %, outperforming SERS alone (86 %) and CEA alone (69 %). SignificanceTo our knowledge, this is the first report of mid-level data fusion of urinary nucleosides SERS spectra with blood CEA levels for the preoperative prediction of LNM in CRC. This work demonstrates that the mid-level data fusion strategy aided by SVM algorithm can greatly improve the preoperative prediction accuracy of LNM. This is crucial for therapeutic decision-making and prognostic assessment in CRC.

Comparison of a Novel Posterior Integrated Transfixation Sacroiliac Joint Fusion Approach to the Posterolateral and Lateral Approaches: A Cadaveric Biomechanical and Computational Analysis of the Fixation, Invasiveness, and Fusion Area.

To concurrently assess and compare the fixation efficacy, invasiveness, and fusion potential of a posterior integrated transfixation cage system to the posterolateral threaded implant and lateral triangular rod systems, in a cadaveric model. Twelve (12) cadaveric sacroiliac joint specimens were utilized and tested within the single-leg stance multidirectional pure moment bending model. Each specimen was tested in the intact, destabilized, treated (using posterior, posterolateral, and lateral systems), and post-fatigue conditions by applying 0 to ± 7.5 Nm of moment in flexion-extension, axial rotation, and lateral bending while measuring the angular range of motion between the sacrum and ilium. Computational models were reconstructed from Computed Tomography (CT) scans and manufacturer surgical technique guides. The models were utilized to quantify the volume of bone removed during implantation and the surface area available for fusion. The posterior integrated transfixation cage system and the lateral triangular rods produced equivalent motion reduction in all motion planes (P > 0.583). The posterolateral cylindrical threaded implant produced less motion reductions than the posterior and lateral implants in flexion-extension (6% ± 3% vs 37% ± 10% and 33% ± 11%, respectively, P <0.05). The posterior system removed 22%-60% less bone volume from the sacrum and ilium (P<0.10), introduced 200%-270% more implant surface to the joint space (P<0.01) and decorticated 75%-375% more joint surface area (P<0.01). The posterior integrated transfixation single-implant cage system is superior to the posterolateral cylindrical threaded single-implant system. Its performance in osteopenic bone is equivalent to the lateral triangular rod system in healthy bone; however, the posterior integrated transfixation cage system requires a single implant, while the lateral triangular rod system requires three. The posterior implant removes the least bone volume and has the most surface area for fusion, providing a significantly better opportunity for robust sacroiliac joint arthrodesis.

Research on the fusion positioning system of UWB/LiDAR based on the algorithm of SPF and KF

In order to overcome the disadvantages of single sensor localization in complex indoor environments, this paper proposes a design scheme for the ultra-wide band (UWB) and LiDAR co-location fusion system. To attain the ideal positioning effect of this system, a segmented point fusion approach based on Euclid's theorem is put forward to optimize the conventional extended Kalman filter and unscented Kalman filter algorithms. It can effectively reduce the positioning error of UWB in non-line-of-sight environments and under the multipath effect, and also assist LiDAR in correcting trajectory drift in sparsely textured scenes. Through both simulations and experiments, the feasibility of the proposed algorithm in the UWB/LiDAR positioning system is verified. The results reveal that the positioning error of the enhanced multi-sensor fusion algorithm is reduced by 22% compared with the original algorithm and 25.7% compared with the single sensor positioning method.

Application of multimodal deep learning and multi-instance learning fusion techniques in predicting STN-DBS outcomes for Parkinson's disease patients

Parkinson's Disease (PD) is a progressive neurodegenerative disorder with substantial impact on patients' quality of life. Subthalamic nucleus deep brain stimulation (STN-DBS) is an effective treatment for advanced PD, but patient responses vary, necessitating predictive models for personalized care. Recent advancements in medical imaging and machine learning offer opportunities to enhance predictive accuracy, particularly through deep learning and multi-instance learning (MIL) techniques. This retrospective study included 127 PD patients undergoing STN-DBS. Medical records and imaging data were collected, and patients were categorized based on treatment outcomes. Advanced segmentation models were trained for automated region of interest (ROI) delineation. A novel 2.5D deep learning approach incorporating multi-slice representation was developed to extract detailed ROI features. Multi-instance learning fusion techniques integrated predictions across multiple slices, combining radiomics and deep learning features to enhance model performance. Various machine learning algorithms were evaluated, and model robustness was assessed using cross-validation and hyperparameter optimization. The MIL model achieved an area under the curve (AUC) of 0.846 for predicting STN-DBS outcomes, surpassing the radiomics model's AUC of 0.825. Integration of MIL and radiomics features in the DLRad model further improved discriminative ability to an AUC of 0.871. Calibration tests showed good model reliability, and decision curve analysis demonstrated clinical utility, affirming the model's predictive advantage. This study demonstrates the efficacy of integrating MIL, radiomics, and deep learning techniques to predict STN-DBS outcomes in PD patients. The multimodal fusion approach enhances predictive accuracy, supporting personalized treatment planning and advancing patient care.

A Task-Guided, Implicitly-Searched and Meta-Initialized Deep Model for Image Fusion.

Image fusion plays a key role in a variety of multi-sensor-based vision systems, especially for enhancing visual quality and/or extracting aggregated features for perception. However, most existing methods just consider image fusion as an individual task, thus ignoring its underlying relationship with these downstream vision problems. Furthermore, designing proper fusion architectures often requires huge engineering labor. It also lacks mechanisms to improve the flexibility and generalization ability of current fusion approaches. To mitigate these issues, we establish a Task-guided, Implicit-searched and Meta-initialized (TIM) deep model to address the image fusion problem in a challenging real-world scenario. Specifically, we first propose a constrained strategy to incorporate information from downstream tasks to guide the unsupervised learning process of image fusion. Within this framework, we then design an implicit search scheme to automatically discover compact architectures for our fusion model with high efficiency. In addition, a pretext meta initialization technique is introduced to leverage divergence fusion data to support fast adaptation for different kinds of image fusion tasks. Qualitative and quantitative experimental results on different categories of image fusion problems and related downstream tasks (e.g., visual enhancement and semantic understanding) substantiate the flexibility and effectiveness of our TIM.

Modeling of zirconium alloy cladding corrosion behavior based on neural ordinary differential equation

Current zirconium alloy cladding corrosion models are mainly semi-empirical and show significant dispersion when compared to measured data. This study introduces neural ordinary differential equation (neural ODE) to model corrosion behavior, utilizing data-model fusion approach for network training. Initially, a semi-empirical model for zirconium alloy cladding corrosion is established through differential evolution algorithm, generating a large dataset for pre-training the neural network. The network is then fine-tuned using measured data. These methods effectively address the challenges of sparse cladding corrosion data and data available only at fixed time points, resulting in a more accurate model. The results shows that the differential evolution algorithm can identify a set of appropriate parameters for the semi-empirical model, achieving a standard deviation of 0.040. The neural ODE model demonstrates even higher accuracy, reducing the standard deviation to 0.031 and improving accuracy by approximately 25%. Additionally, the model demonstrates excellent generalization capacity on other time points and new power histories.

A novel steady-state visually evoked potential-based brain-computer interfaces using trans-subject feature fusion approach

A novel steady-state visually evoked potential-based brain-computer interfaces using trans-subject feature fusion approach