Robust Fusion Research Articles

RBCKF-Based Vehicle State Estimation by Adaptive Weighted Fusion Strategy Considering Composite-State Tire Model

The acquisition of vehicle driving status information is a key function of vehicle dynamics systems, and research on high-precision and high-reliability estimation of key vehicle states has significant value. To improve the state observation effect, a vehicle sideslip angle estimation method adopting a robust bias compensation Kalman filter and adaptive weight fusion strategy is proposed. On the basis of the extended Kalman filter algorithm, and with the goals of estimation exactitude and robustness, considering the potential signal deviation, a vehicle state robust deviation compensation Kalman filter estimation algorithm considering bias compensation and residual covariance matrix weighting is proposed. Meanwhile, considering the adaptive and dynamic adjustment capabilities of the observation system in complex state-change scenarios, an estimation strategy based on adaptive weight fusion and a model-based estimator is proposed. The results confirm that the robust bias compensation Kalman filter can ensure estimation exactitude and robustness when the vehicle state fluctuates greatly, and the proposed fusion strategy can ensure that the vehicle maintains optimal estimation performance during operating condition switching.

Multi-scale contextual learning for medical image segmentation via dual distillation.

Recently, many studies have explored fusing features extracted from Convolutional neural networks (CNNs) and transformers to integrate multi-scale representations for better performance in medical image segmentation tasks. Although these hybrid models have achieved better results than previous CNN-based and transformer-based methods, they suffer from high computation and spacecomplexities. The purpose of this research is to address the prohibitive computation and space complexities of hybrid models, which limit their application in clinical practice where computational resources are usuallyconstrained. We propose a novel model equipped with a dual distillation scheme to sufficiently harness the complementary advantages of CNNs and transformers without compromising model efficiency. We further propose a multi-scale prior-knowledge distillation (MPD) module to effectively distill multi-scale knowledge from features extracted from transformers. In addition, to cooperate with the knowledge distillation scheme, we also propose an efficient and robust Selective Fusion module in the studentnetwork. We extensively evaluate the proposed model against fourteen different network frameworks on two representative datasets: SipakMed and ISIC 2017. In the SipakMed dataset, 3037 Pap smear images are used for training and 1012 for testing. In the ISIC 2017 dataset, 2000 dermoscopic images are used for training, 150 for validation, and 600 for testing. Experimental results demonstrate that our method not only surpasses existing methods by a considerable margin with respect to the evaluation metrics of mean Intersection over Union, mean Dice coefficient, mean average symmetric surface distance, but also requires fewer computational resources in terms of model parameters and floating-point operations persecond. Comprehensive comparisons in terms of segmentation accuracy and computational complexity unequivocally confirm that our method effectively and efficiently integrates the advantages of both CNNs and transformers, showing its suitability and significance for clinicalapplications.

An autoencoder-based confederated clustering leveraging a robust model fusion strategy for federated unsupervised learning

Concerns related to data privacy, security, and ethical considerations become more prominent as data volumes continue to grow. In contrast to centralized setups, where all data is accessible at a single location, model-based clustering approaches can be successfully employed in federated settings. However, this approach to clustering in federated settings is still relatively unexplored and requires further attention. As federated clustering deals with remote data and requires privacy and security to be maintained, it poses particular challenges as well as possibilities. While model-based clustering offers promise in federated environments, a robust model aggregation method is essential for clustering rather than the generic model aggregation method like Federated Averaging (FedAvg). In this research, we proposed an autoencoder-based clustering method by introducing a novel model aggregation method FednadamN, which is a fusion of Adam and Nadam optimization approaches in a federated learning setting. Therefore, the proposed FednadamN adopted the adaptive learning rates based on the first and second moments of gradients from Adam which offered fast convergence and robustness to noisy data. Furthermore, FednadamN also incorporated the Nesterov-accelerated gradients from Nadam to further enhance the convergence speed and stability. We have studied the performance of the proposed Autoencoder-based clustering methods on benchmark datasets and using the novel FednadamN model aggregation strategy. It shows remarkable performance gain in federated clustering in comparison to the state-of-the-art.

Indoor occupancy monitoring using environmental feature fusion and semi-supervised machine learning models

Smart buildings optimize energy consumption and occupant comfort through Heating, Ventilation and Air Conditioning, and lighting management. Nevertheless, large venues require data fusion techniques to improve analysis and forecasting. This study aims to evaluate the effectiveness of using different feature fusion techniques, environmental sensors, and semi-supervised learning to estimate indoor occupancy in a 230 m2 office. Using five Internet of Things devices measuring air temperature, relative humidity, and barometric pressure, data was collected for 99 days with 6800 entries (on average) and only 14% labeled. Eight feature selection methods were evaluated along with three supervised and two semi-supervised classification methods. Results indicate that the Chi-squared-based approach for feature fusion outperformed others. Similarly, the semi-supervised Self-Training model achieved better performance than the supervised methods. This research shows that combining semi-supervised learning and data fusion allows for estimating the occupancy level in large indoor spaces with high accuracy and low labeling costs. Highlights This study pioneers in exploring semi-supervised learning and distinct feature fusion methods for estimating indoor occupancy levels in a 230 m 2 open office using only Internet of Things (IoT) environmental sensors (air temperature, relative humidity, and barometric pressure). A comprehensive comparison of statistical methods, feature selection, and dimensionality reduction techniques are conducted to determine their ability to generate robust feature fusion sets. The feature fusion selected through the Chi-squared test stood out with a high accuracy F1-score (average of 0.95) and an average accuracy of 0.99. The Self-Training model reached the best performance from semi-supervised learning, with an average F1-Score of 0.90 and an average accuracy of 0.97, based on a dataset with a large proportion of unlabelled data (16,847 entries) and only 9367 labels. For supervised learning, Random Forest achieved a high accuracy (average of 0.98) and F1-score (average of 0.93) across various feature sets.

Remote sensing image classification based on non-linear enhanced attention mechanism

Abstract In recent years, Transformer technology has gradually shown promising applications in the field of computer vision, becoming a research hotspot. However, traditional vision Transformers suffer from significant computational burdens. Although previous studies have attempted to alleviate this issue by reducing the computational load of attention mechanisms, their methods still require improvement. Additionally, past research often overlooked the non-linear representation of values within attention mechanisms, posing another challenge to address. Therefore, this paper proposes a novel attention mechanism aimed at reducing the computational burden of traditional attention mechanisms while enhancing their ability to express non-linearities in values. Furthermore, conventional feedforward neural networks typically output results directly after passing through an activation function, which may limit the model’s representational capacity due to the simplistic structure leading to singular feature information. To address this issue, this study introduces a more robust feature fusion convolutional feedforward network. Through these methods, the aim is to enhance the accuracy of Transformers in the field of remote-sensing image classification.

Fiber-to-Chip Packaging With Robust Fiber Fusion Splicing for Low-Temperature Applications

Fiber-to-Chip Packaging With Robust Fiber Fusion Splicing for Low-Temperature Applications

Robust fusion filter for networked uncertain descriptor systems with colored noise and cyber-attacks

The robust fusion filtering problem of multi-sensor networked uncertain descriptor systems (NUDSs) with colored noise, uncertain noise variances and cyber-attacks is investigated. During data transmission in unreliable communication networks, the data can be maliciously attacked by attackers. In other words, the local filters (LFs) may receive false data or may not receive data because of the cyber-attacks. By adopting the singular value decomposition (SVD) method, the original NUDSs can be converted into two reduced-order subsystems with uncertain correlated fictitious white noises, and the cyber-attacks are transformed into the fictitious noises. Cross-covariance matrices between local filtering errors are derived. The robust LFs are obtained according to the minimax robust estimation principle. Under the linear unbiased minimum variance criterion, three weighted fusion algorithms are applied to fuse the LFs. For all allowable uncertainties of noise variances and cyber-attacks, the minimal upper bounds of covariance matrices of the local and distributed fusion filters are guaranteed. The proof of their robustness is established through the minimax estimation principle and Lyapunov equation method. Finally, the correctness and effectiveness of the proposed algorithms are verified by a circuit system example.

A multimodal cross-transformer-based model to predict mild cognitive impairment using speech, language and vision

Mild Cognitive Impairment (MCI) is an early stage of memory loss or other cognitive ability loss in individuals who maintain the ability to independently perform most activities of daily living. It is considered a transitional stage between normal cognitive stage and more severe cognitive declines like dementia or Alzheimer’s. Based on the reports from the National Institute of Aging (NIA), people with MCI are at a greater risk of developing dementia, thus it is of great importance to detect MCI at the earliest possible to mitigate the transformation of MCI to Alzheimer’s and dementia. Recent studies have harnessed Artificial Intelligence (AI) to develop automated methods to predict and detect MCI. The majority of the existing research is based on unimodal data (e.g., only speech or prosody), but recent studies have shown that multimodality leads to a more accurate prediction of MCI. However, effectively exploiting different modalities is still a big challenge due to the lack of efficient fusion methods. This study proposes a robust fusion architecture utilizing an embedding-level fusion via a co-attention mechanism to leverage multimodal data for MCI prediction. This approach addresses the limitations of early and late fusion methods, which often fail to preserve inter-modal relationships. Our embedding-level fusion aims to capture complementary information across modalities, enhancing predictive accuracy. We used the I-CONECT dataset, where a large number of semi-structured conversations via internet/webcam between participants aged 75+ years old and interviewers were recorded. We introduce a multimodal speech-language-vision Deep Learning-based method to differentiate MCI from Normal Cognition (NC). Our proposed architecture includes co-attention blocks to fuse three different modalities at the embedding level to find the potential interactions between speech (audio), language (transcribed speech), and vision (facial videos) within the cross-Transformer layer. Experimental results demonstrate that our fusion method achieves an average AUC of 85.3% in detecting MCI from NC, significantly outperforming unimodal (60.9%) and bimodal (76.3%) baseline models. This superior performance highlights the effectiveness of our model in capturing and utilizing the complementary information from multiple modalities, offering a more accurate and reliable approach for MCI prediction.

SBD-Net: Incorporating Multi-Level Features for an Efficient Detection Network of Student Behavior in Smart Classrooms

Detecting student behavior in smart classrooms is a critical area of research in educational technology that significantly enhances teaching quality and student engagement. This paper introduces an innovative approach using advanced computer vision and artificial intelligence technologies to monitor and analyze student behavior in real time. Such monitoring assists educators in adjusting their teaching strategies effectively, thereby optimizing classroom instruction. However, the application of this technology faces substantial challenges, including the variability in student sizes, the diversity of behaviors, and occlusions among students in complex classroom settings. Additionally, the uneven distribution of student behaviors presents a significant hurdle. To overcome these challenges, we propose Student Behavior Detection Network (SBD-Net), a lightweight target detection model enhanced by the Focal Modulation module for robust multi-level feature fusion, which augments feature extraction capabilities. Furthermore, the model incorporates the ESLoss function to address the imbalance in behavior sample detection effectively. The innovation continues with the Dyhead detection head, which integrates three-dimensional attention mechanisms, enhancing behavioral representation without escalating computational demands. This balance achieves both a high detection accuracy and manageable computational complexity. Empirical results from our bespoke student behavior dataset, Student Classroom Behavior (SCBehavior), demonstrate that SBD-Net achieves a mean Average Precision (mAP) of 0.824 with a low computational complexity of just 9.8 G. These figures represent a 4.3% improvement in accuracy and a 3.8% increase in recall compared to the baseline model. These advancements underscore the capability of SBD-Net to handle the skewed distribution of student behaviors and to perform high-precision detection in dynamically challenging classroom environments.

Collaborative graph neural networks for augmented graphs: A local-to-global perspective

In the field of graph neural networks (GNNs) for representation learning, a noteworthy highlight is the potential of embedding fusion architectures for augmented graphs. However, prevalent GNN embedding fusion architectures mainly focus on handling graph combinations from a global perspective, often ignoring their collaboration with the information of local graph combinations. This inherent limitation constrains the ability of the constructed models to handle multiple input graphs, particularly when dealing with noisy input graphs collected from error-prone sources or those resulting from deficiencies in graph augmentation methods. In this paper, we propose an effective and robust embedding fusion architecture from a local-to-global perspective termed collaborative graph neural networks for augmented graphs (LoGo-GNN). Essentially, LoGo-GNN leverages a pairwise graph combination scheme to generate local perspective inputs. Together with the global graph combination, this serves as the basis to generate a local-to-global perspective. Specifically, LoGo-GNN employs a perturbation augmentation strategy to generate multiple augmentation graphs, thereby facilitating collaboration and embedding fusion from a local-to-global perspective through the use of graph combinations. In addition, LoGo-GNN incorporates a novel loss function for learning complementary information between different perspectives. We also conduct theoretical analysis to assess its expressive power under ideal conditions, demonstrating the effectiveness of LoGo-GNN. Our experiments, focusing on node classification and clustering tasks, highlight the superior performance of LoGo-GNN compared to state-of-the-art methods. Additionally, robustness analysis further confirms its effectiveness in addressing uncertainty challenges.

How Does the Stress in the Fixation Device Change during Different Stages of Bone Healing in the Treatment of Fractures? A Finite Element Study of External Fixation for Tibial Fractures.

Although the specific relationship between the stress changes in the external fixator during tibial fracture treatment and the bone healing process remains unclear, it is believed that stress variations in the external fixator scaffold can, to a certain extent, reflect the progress of tibial healing. This study aims to propose a non-invasive method for assessing the degree of fracture healing by monitoring the changes in stress transmission, the locations of stress-sensitive points, and displacement in the external fixator-tibia system during the healing process of tibial fractures. In this study, finite element models of tibial fractures at various healing stages were developed. Physiological conditions, including axial, torsional, and bending loads on the tibia, were simulated to evaluate stress and strain within the external scaffold-tibia system under normal physiological loading conditions. The results indicate variations in the stress distribution between the external fixator and the tibia during different stages of healing. In the early phase of fracture healing, the external fixator plays a crucial role as the primary load-bearing unit under all three loading conditions. As the fracture healing progresses, the stress on the tibia gradually increases, concentrating on the medial part of the tibia under axial and torsional loading, and at the upper and lower ends, as well as the central part of the anterior and posterior tibia during bending loading. The stress at the callus gradually increases, while micro-movements decrease. The stress within the external bracket gradually decreases, with a tendency for the connecting rod to transfer stress towards the screws. Throughout the fracture healing process, the location of maximum stress in the external fixator remains unchanged. Under axial and torsional loading, the maximum stress is located at the intersection of the lowest screw and the bone cortex, while under bending loading, it is at the intersection of the second screw and the connecting rod. During the bone healing process, stress is transferred between the external fixation frame and the bone. As bone healing advances, the stress on the connecting rods and screws of the external fixation frame decreases, and the amplitude of stress changes diminishes. When complete and robust fusion is achieved, stress variations stabilize, and the location of maximum stress on the external fixation frame remains unchanged. The intersections of the lowest screw and the bone cortex, as well as the second screw and the connecting rod, can serve as sensitive points for monitoring the degree of bone healing.

Boosting Photocatalytic CO2 Methanation through Interface Fusion over CdS Quantum Dot Aerogels.

In the field of photocatalytic CO2 reduction, quantum dot (QD) assemblies have emerged as promising candidate photocatalysts due to their superior light absorption and better substrate adsorption. However, the poor contacts within QD assemblies lead to low interfacial charge transfer efficiency, making QD assemblies suffer from unsatisfactory photocatalytic performance. Herein, a novel approach is presented involving the construction of strongly interfacial fused CdS QD assemblies (CdS QD gel) for CO2 reduction. The novel CdS QD gel demonstrates outstanding photocatalytic performance for CO2 methanation, achieving a CH4 generation rate of ≈296 µmol g-1 h-1, with a selectivity surpassing 76% and an apparent quantum yield (AQY) of 1.4%. Further investigations reveal that the robust interfacial fusion in these CdS QDs not only boosts their ability to absorb visible light but also significantly promotes charge separation. The present work paves the way for utilizing QD gel photocatalysts in realizing efficient CO2 reduction and highlights the critical role of interfacial engineering in photocatalysts.

RViT: Robust Fusion Vision Transformer with Variational Hierarchical Denoising Process for Image Classification

Transformers designed for natural language processing have originally been explored for computer vision in recent research. Various Vision Transformers (ViTs) play an increasingly important role in the field of image tasks such as computer vision, multimodal fusion and multimedia analysis. However, to obtain promising performance, most existing ViTs usually rely on artificially filtered high-quality images, which may suffer from inherent noise risk. Generally, such well-constructed images are not always available in every situation. To this end, we propose a Robust ViT (RViT) to focus on the relevant and robust representation learning for image classification tasks. Specifically, we first develop a novel Denoising VTUnet module, where we conceptualize the nonrobust noise as the uncertainty under the variational conditions. Furthermore, we design a fusion transformer backbone with a tailored fusion attention mechanism to perform image classification based on the extracted robust representations effectively. To demonstrate the superiority of our model, the compared experiments are conducted on several popular datasets. Benefiting from the sequence regularity of the Transformer and captured robust feature, the proposed method exceeds compared Transformer-based models with superior performance in visual tasks.

Analysis of Failed Atlantoaxial Reduction: Causes of Failure and Strategies for Revision.

The craniovertebral junction (CVJ) presents intricate anatomical challenges. In severe or irreducible malformations, complications such as reduction loss and fixation failure may occur, necessitating revision surgery. The posterior facet joint distraction and fusion (PFDF) technique, offers a solely posterior approach for revisions. Hence, we delineate varied revision scenarios, proposing surgical strategies and technical details to enhance outcomes and mitigate risks, thereby enriching the neurosurgical community's repertoire. This was a retrospective cohort study, analyzed patient data from Xuanwu Hospital, between 2017 to 2023. All patients had a history of surgical treatment for CVJ malformations, and experienced failure or loss of reduction. The distance from the odontoid process tip to the Chamberlain's line (DCL), the atlantodental interval (ADI), clivus-canal angle, cervicomedullary angle, width of subarachnoid space, CVJ area, and width of syrinx were used for radiographic assessment. Japanese Orthopaedic Association (JOA) scores and SF-12 scores were used for clinical assessment. Independent sample t-tests were employed. A significance level of p < 0.05 indicates statistically significant differences. We analyzed data from 35 patients. For patients who underwent PFDF, the postoperative DCL, ADI, and clivus-canal angle significantly improved. For all patients, the postoperative cervicomedullary angle, width of subarachnoid space, CVJ area, and width of syrinx all demonstrated significant improvement, indicating the relief of neural compression. All patients showed significant improvement in both symptoms and clinical assessments. Severe atlantoaxial joint locking or ligament contracting are the fundamental cause of reduction and fixation failure. Anterior odontoidectomy is indicated for patients with robust bony fusion of the atlantoaxial joint in an unreduced position. The PFDF technique is safe and effective for patients with incomplete atlantoaxial bony fusion. Preoperative assessment of surgical feasibility and vertebral artery status ensures surgical safety and efficacy.

A robust fusion of STEAM and edutainment for holistic public health rooted in early childhood education.

A robust fusion of STEAM and edutainment for holistic public health rooted in early childhood education.

The Design and Analysis of the Helical Gear in the Car Gear Box for the Slag Port Transfer

The design and analysis of helical gears with involutes profiles is the primary focus of this study. The most widely utilized gearing system in use today is the involutes gear profile. Power and torque are typically transmitted through them. When compared to other forms of transmission, the efficiency of power transmission via gears is extremely high. Bending and contact stress are the main causes of gear tooth failure. The current study examines helical gear, which has to do with slag transfer in the steel production sector. The most affected stress concentrated location of any set of gears in the slag port transfer car (SPTC) gear box is identified by examining the contact stresses for the current design set of gears. The high stress concentration gear set was redesigned with a constant (45mm) face width at various Helix angles (14, 15, 16, 17, 19^, 20 ̊, 25^, and 30 ̊) and a constant pressure angle (20). To calculate the contact stresses between two gears that are mating. The robust and cutting-edge solid modeling program AUTODESK FUSION 360 creates a three-dimensional solid model. The ANSYS Workbench module was utilized to do the numerical analysis. The outcomes of the workbench module for ANSYS. The current study is helpful in quantifying the previously mentioned factors, which aids in the safe and effective design of the helical gear in the SPTC Gearbox.

Infrared and Harsh Light Visible Image Fusion Using an Environmental Light Perception Network.

The complementary combination of emphasizing target objects in infrared images and rich texture details in visible images can effectively enhance the information entropy of fused images, thereby providing substantial assistance for downstream composite high-level vision tasks, such as nighttime vehicle intelligent driving. However, mainstream fusion algorithms lack specific research on the contradiction between the low information entropy and high pixel intensity of visible images under harsh light nighttime road environments. As a result, fusion algorithms that perform well in normal conditions can only produce low information entropy fusion images similar to the information distribution of visible images under harsh light interference. In response to these problems, we designed an image fusion network resilient to harsh light environment interference, incorporating entropy and information theory principles to enhance robustness and information retention. Specifically, an edge feature extraction module was designed to extract key edge features of salient targets to optimize fusion information entropy. Additionally, a harsh light environment aware (HLEA) module was proposed to avoid the decrease in fusion image quality caused by the contradiction between low information entropy and high pixel intensity based on the information distribution characteristics of harsh light visible images. Finally, an edge-guided hierarchical fusion (EGHF) module was designed to achieve robust feature fusion, minimizing irrelevant noise entropy and maximizing useful information entropy. Extensive experiments demonstrate that, compared to other advanced algorithms, the method proposed fusion results contain more useful information and have significant advantages in high-level vision tasks under harsh nighttime lighting conditions.

A Novel Real-Time Detection and Classification Method for ECG Signal Images Based on Deep Learning.

In this paper, a novel deep learning method Mamba-RAYOLO is presented, which can improve detection and classification in the processing and analysis of ECG images in real time by integrating three advanced modules. The feature extraction module in our work with a multi-branch structure during training can capture a wide range of features to ensure efficient inference and rich feature extraction. The attention mechanism module utilized in our proposed network can dynamically focus on the most relevant spatial and channel-wise features to improve detection accuracy and computational efficiency. Then, the extracted features can be refined for efficient spatial feature processing and robust feature fusion. Several sets of experiments have been carried out to test the validity of the proposed Mamba-RAYOLO and these indicate that our method has made significant improvements in the detection and classification of ECG images. The research offers a promising framework for more accurate and efficient medical ECG diagnostics.

Feature-consistent coplane-pair correspondence- and fusion-based point cloud registration

It is an important and challenging task to register two point clouds, and the estimated registration solution can be applied in 3D vision. In this paper, an outlier removal method is first proposed to delete redundant coplane-pair correspondences for constructing three feature-consistent coplane-pair correspondence subsets. Next, Rodrigues’ formula and a scoring-based method are adopted to solve the representative registration solution of each correspondence subset. Then, a robust fusion method is proposed to fuse the three representative solutions as the final registration solution. Based on typical testing datasets, comprehensive experimental results demonstrated that with good registration accuracy, our registration algorithm achieves significant execution time reduction effect when compared with the state-of-the-art methods.

Quantifying Posttraumatic Stress Disorder Symptoms During Traumatic Memories Using Interpretable Markers of Respiratory Variability.

Posttraumatic stress disorder (PTSD) causes heightened fight-or-flight responses to traumatic memories (i.e., hyperarousal). Although hyperarousal is hypothesized to cause irregular breathing (i.e., respiratory variability), no quantitative markers of respiratory variability have been shown to correspond with PTSD symptoms in humans. In this study, we define interpretable markers of respiration pattern variability (RPV) and investigate whether these markers respond during traumatic memories, correlate with PTSD symptoms, and differ in patients with PTSD. We recruited 156 veterans from the Vietnam-Era Twin Registry to participate in a trauma recall protocol. From respiratory effort and electrocardiogram measurements, we extracted respiratory timings and rate using a robust quality assessment and fusion approach. We then quantified RPV using the interquartile range and compared RPV between baseline and trauma recall conditions, correlated PTSD symptoms to the difference between trauma recall and baseline RPV (i.e., ∆RPV), and compared ∆RPV between patients with PTSD and trauma-exposed controls. Leveraging a subset of 116 paired twins, we then uniquely controlled for factors shared by co-twins via within-pair analysis for further validation. We found RPV was increased during traumatic memories (p .001), ∆ RPV was positively correlated with PTSD symptoms (p .05), and patients with PTSD exhibited higher ∆ RPV than trauma-exposed controls (p . 05). This paper is the first to elucidate RPV markers that respond during traumatic memories, especially in patients with PTSD, and correlate with PTSD symptoms. These findings encourage future studies outside the clinic, where interpretable markers of respiratory variability are used to track hyperarousal.