Symmetric Fusion Research Articles

The association scheme on the set of flags of a finite generalized quadrangle

In this paper, the association scheme defined on the flags of a finite generalized quadrangle is considered. All possible fusions of this scheme are listed, and a full description for those of classes 2 and 3 is given.Furthermore, it is showed that an association scheme with appropriate parameters must arise from the flags of a generalized quadrangle. The same is done for one of its 4-class symmetric fusion.

MV-SHIF: Multi-view symmetric hypothesis inference fusion network for emotion-cause pair extraction in documents

Emotion-cause pair extraction (ECPE) is a challenging task that aims to automatically identify pairs of emotions and their causes from documents. The difficulty of ECPE lies in distinguishing valid emotion-cause pairs from many irrelevant ones. Most previous methods have primarily focused on utilizing multi-task learning to extract semantic information solely from documents without explicitly encoding the relations between clauses. We propose a new approach that incorporates textual entailment paradigm aiming to infer the entailment relationship between the original document as the premise and the clauses or pairs described as the hypothesis. Our approach designs label-view hypothesis templates to improve ECPE by filtering out irrelevant emotion and cause clauses. Furthermore, we formulate candidate emotion-cause pairs as hypothesis statements, and define explicit multi-view symmetric templates to capture the emotion-cause relation semantics. The text entailment recognition for ECPE is finally implemented by fusing multi-view semantic information using a simplified capsule network. Our proposed model achieves state-of-the-art performance on ECPE compared to previous baselines. More importantly, this work demonstrates a novel effective way of applying the textual entailment paradigm to ECPE or clause-level causal discovery by designing multi-view hypothesis inference and information fusion.

SBIT-Fuse: Infrared and visible image fusion based on Symmetrical Bilateral interaction and Transformer

Infrared and visible image fusion, as a typical image enhancement task, aims to generate an image containing complementary information. Existing deep learning-based methods usually use a dual-stream model to extract features, which suffers from insufficient feature interaction in the feature extraction process. In order to effectively utilize information, this work proposes a Symmetrical Bilateral Interaction and Transformer fusion method (SBIT-Fuse) that is simple and efficient for constructing a dual-stream interaction network. Specifically, inspired by the problem of dead ReLU, we propose a Symmetrical Bilateral Interaction (SBI) module which consists of a certain number of cascaded cross domain activation interaction (CDAI) layers, where the deactivation information of the ReLU rectifier is transferred from one flow to another instead of being discarded. The proposed module exploits the drawbacks of ReLU to efficiently build the interaction of two streams and combine spatial and channel attention mechanisms, achieving the enhancement of feature extraction. Besides, our network is end-to-end and a transformer module is employed for extracting long-range semantic information. Moreover, combining a novel designed loss function, the proposed method can generate higher-quality fusion images. The comparative experiments with existing advanced methods and ablation studies on publicly datasets demonstrate the effectiveness of the proposed approach. Our code can be found at https://github.com/ljx111790/SBIT-Fuse.

SC-Net: Symmetrical conical network for colorectal pathology image segmentation

Background and ObjectiveImage segmentation of histopathology of colorectal cancer is a core task of computer aided medical image diagnosis system. Existing convolutional neural networks generally extract multi-scale information in linear flow structures by inserting multi-branch modules, which is difficult to extract heterogeneous semantic information under multi-level and different receptive field and tough to establish context dependency among different receptive field features. MethodsTo address these issues, we propose a symmetric spiral progressive feature fusion encoder-decoder network called the Symmetric Conical Network (SC-Net). First, we design a Multi-scale Feature Extraction Block (MFEB) matching with the Symmetric Conical Network to obtain multi-branch heterogeneous semantic information under different receptive fields, so as to enrich the diversity of extracted feature information. The encoder is composed of MFEB through spiral and multi-branch arrangement to enhance context dependence between different information flow. Secondly, the information loss of contour, color and others in high-level semantic information through causally stacking MFEB, the Feature Mapping Layer (FML) is designed to map low-level features to high-level semantic features along the down-sampling branch and solve the problem of insufficient global feature extraction in deep levels. ResultsThe SC-Net was evaluated on our self-constructed colorectal cancer dataset, a publicly available breast cancer dataset and a polyp dataset. The results revealed that the mDice of segmentation reached 0.8611, 0.7259 and 0.7144. We compare our model with the state-of-art semantic segmentation UNet++, PSPNet, Attention U-Net, R2U-Net and other advanced segmentation networks. The experimental results demonstrate that we achieve the most advanced performance. ConclusionsThe results indicate that the proposed SC-Net excels in segmenting H&E stained pathology images, effectively preserving morphological features and spatial information even in scenarios with weak texture, poor contrast, and variations in appearance.

Design and Biomechanical Properties of Symmetrical Lumbar Fusion Cage Based on Lightweight Titanium Alloy Flexible Microporous Metal Rubber

In recent years, the incidence rate of lumbar diseases has been progressively increasing. The conventional lumbar fusion cages used in existing lumbar interbody fusion surgery are not able to take into account the multiple characteristics of cushioning, vibration reduction, support, cell adhesion, and bone tissue growth. Therefore, in this work, based on the CT data of a lumbar intervertebral disc plain scan, a combined symmetric lumbar fusion cage structure was innovatively designed. The core was made of lightweight TC4 medical titanium alloy flexible microporous metal rubber (LTA-FMP MR), and the outer frame was made of cobalt–chromium–molybdenum alloy. Its comprehensive biomechanical performance was comprehensively evaluated through finite element simulation, static and dynamic mechanics, and impact resistance tests. The three-dimensional model of the L3/L4 lumbar segment was established by reverse engineering, and a Mises stress analysis was conducted on the lumbar fusion cage by importing it into Ansys to understand its structural advantages compared to the traditional lumbar fusion cage. Through static experiments, the influence of the internal nucleus of a symmetrical lumbar fusion cage with different material parameters on its static performance was explored. At the same time, to further explore the superior characteristics of this symmetrical structure in complex human environments, a biomechanical test platform was established to analyze its biomechanical performance under sinusoidal excitation of different amplitudes and frequencies, as well as impact loads of different amplitudes and pulse widths. The results show that under different amplitudes and frequencies, the lumbar fusion cage with a symmetrical structure has a small loss factor, a high impact isolation coefficient, and a maximum energy consumption of 422.8 N·mm, with a maximum kinetic energy attenuation rate of 0.43. Compared to existing traditional lumbar fusion cages in clinical practice, it not only has sufficient stiffness, but also has good vibration damping, support, and impact resistance performance, and has a lower probability of postoperative settlement, which has broad application prospects.

A framework for synthetic diagnostics using energetic-particle orbits in tokamaks

In fusion plasma physics, the large-scale trajectories of energetic particles in magnetic confinement devices are known as orbits. To effectively and efficiently be able to work with orbits, the Orbit Weight Computational Framework (OWCF) was developed. The OWCF constitutes a set of scripts, functions and applications capable of computing, visualizing and working with quantities related to fast-ion (FI) orbits in toroidally symmetric fusion devices. The current version is highly integrated with the DRESS code, which enables the OWCF to compute and analyze the orbit sensitivity for arbitrary neutron- and gamma-diagnostics. However, the framework is modular in the sense that any future codes (e.g. FIDASIM) can be easily integrated. The OWCF can also compute projected velocity spectra for FI orbits, which play a key role in many FI diagnostics. Via interactive applications, the OWCF can function both as a tool for investigative research but also for teaching. The OWCF will be used to analyze and simulate the diagnostic results of current and future fusion experiments such as ITER. The orbit weight functions computed with the OWCF can be used to reconstruct the FI distribution in terms of FI orbits from experimental measurements using tomographic inversion.

Spatio-Temporal Two-stage Fusion for video question answering

Video question answering (VideoQA) has attracted much interest from of scholars as one of the most representative multimodal tasks in recent years. The task requires the model to interact and reason between the video and the question. Most known approaches use pre-trained networks to extract complex embeddings of videos and questions independently before performing multimodal fusion. However, they overlook two factors: (1) These feature extractors are pre-trained for the image or video classification task without taking the question into consideration, therefore may not be suitable for VideoQA task. (2) Using multiple feature extractors to extract features at different levels introduce more irrelevant information to some extent, thus making the task more difficult. For the above reasons, we propose a new model named Spatio-Temporal Two-Stage Fusion, which ties together multiple levels of feature extraction processes and divides them into two distinct stages: spatial fusion and temporal fusion. Specifically, in the spatial fusion stage, we use Vision Transformer to integrate the intra-frame information to generate frame-level features. At the same time, we design a multimodal temporal fusion module that enables the video to fuse textual information and assign different levels of attention to each frame. Then the obtained frame-level features are used to generate global video features by another Vision Transformer. In order to efficiently generate modal interaction information, we design a video–text symmetric fusion module to retain the most relevant information by mutual guidance between the two modalities. Our method is evaluated on three benchmark datasets: MSVD-QA, MSRVTT-QA and TGIF-QA, and achieves competitive results.

Ternary symmetric fusion network for camouflaged object detection

Ternary symmetric fusion network for camouflaged object detection

Remarks on Symmetric Fusion Categories of Low Rank in Positive Characteristic

Abstract We give lower bounds for the rank of a symmetric fusion category in characteristic $p\geq 5$ in terms of $p$. We prove that the second Adams operation $\psi _{2}$ is not the identity for any non-trivial symmetric fusion category, and that symmetric fusion categories satisfying $\psi _{2}^{a}=\psi _{2}^{a-1}$ for some positive integer $a$ are super-Tannakian. As an application, we classify all symmetric fusion categories of rank 3 and those of rank 4 with exactly two self-dual simple objects.

MFNet: Mutual Feature-Aware Networks for Remote Sensing Change Detection

Remote sensing change detection involves detecting pixels that have changed from a bi-temporal image of the same location. Current mainstream change detection models use encoder-decoder structures as well as Siamese networks. However, there are still some challenges with this: (1) Existing change feature fusion approaches do not take into account the symmetry of change features, which leads to information loss; (2) The encoder is independent of the change detection task, and feature extraction is performed separately for dual-time images, which leads to underutilization of the encoder parameters; (3) There are problems of unbalanced positive and negative samples and bad edge region detection. To solve the above problems, a mutual feature-aware network (MFNet) is proposed in this paper. Three modules are proposed for the purpose: (1) A symmetric change feature fusion module (SCFM), which uses double-branch feature selection without losing feature information and focuses explicitly on focal spatial regions based on cosine similarity to introduce strong a priori information; (2) A mutual feature-aware module (MFAM), which introduces change features in advance at the encoder stage and uses a cross-type attention mechanism for long-range dependence modeling; (3) A loss function for edge regions. After detailed experiments, the F1 scores of MFNet on SYSU-CD and LEVIR-CD were 83.11% and 91.52%, respectively, outperforming several advanced algorithms, demonstrating the effectiveness of the proposed method.

Symmetric data-driven fusion of diffusion tensor MRI: Age differences in white matter.

In the past 20 years, white matter (WM) microstructure has been studied predominantly using diffusion tensor imaging (DTI). Decreases in fractional anisotropy (FA) and increases in mean (MD) and radial diffusivity (RD) have been consistently reported in healthy aging and neurodegenerative diseases. To date, DTI parameters have been studied individually (e.g., only FA) and separately (i.e., without using the joint information across them). This approach gives limited insights into WM pathology, increases the number of multiple comparisons, and yields inconsistent correlations with cognition. To take full advantage of the information in a DTI dataset, we present the first application of symmetric fusion to study healthy aging WM. This data-driven approach allows simultaneous examination of age differences in all four DTI parameters. We used multiset canonical correlation analysis with joint independent component analysis (mCCA + jICA) in cognitively healthy adults (age 20-33, n = 51 and age 60-79, n = 170). Four-way mCCA + jICA yielded one high-stability modality-shared component with co-variant patterns of age differences in RD and AD in the corpus callosum, internal capsule, and prefrontal WM. The mixing coefficients (or loading parameters) showed correlations with processing speed and fluid abilities that were not detected by unimodal analyses. In sum, mCCA + jICA allows data-driven identification of cognitively relevant multimodal components within the WM. The presented method should be further extended to clinical samples and other MR techniques (e.g., myelin water imaging) to test the potential of mCCA+jICA to discriminate between different WM disease etiologies and improve the diagnostic classification of WM diseases.

Stability and chaotic dynamic analysis of Li-doped fullerene-IRMOF composite materials for hydrogen storage

Isoreticular metal organic framework (IRMOF) material is a type of ideal hydrogen storage material for its porosity and large surface area. Lithium doping and fullerene impregnation can enhance the hydrogen storage capacity of IRMOF efficiently. In this paper, the nonlinear dynamic characteristics of Li-doped fullerene-IRMOF composite material are qualitatively discussed in theory, which determine its thermal stability. A type of Li-doped fullerene-IRMOF composite material is proposed for hydrogen storage, and a new constitutive model is proposed to describe the hysteretic property of IRMOF. The system's nonlinear dynamic model is developed, and the fast-slow dynamic analysis method is introduced to interpret not only the system's jumping phenomena between the multi-pulse homoclinic and heteroclinic orbits but also its chaotic dynamics. The nonlinear dynamic responses of the system under deterministic and stochastic excitation are obtained respectively, and symmetrical and asymmetric orbital fusion phenomena have been discovered. The results of theoretical analysis and numerical simulation both show that the system has novel nonlinear dynamic characteristics such as double periodic motion, multi-pulse jumping orbits, and chaos; multi-pulse jumping phenomena are caused by the stiffness difference between fullerene and IRMOF, which form a fast-slow system; symmetrical fusion and asymmetrical fusion of system's orbits occur under different parameters. Safe, efficient and rapid hydrogen storage and release can be achieved through controlling the system's parameters to adjust the dynamic behavior of the system, which will be a development direction in the application of hydrogen energy.

Super-Resolution Reconstruction of Remote Sensing Images Based on Symmetric Local Fusion Blocks

In view of the rich information and strong autocorrelation of remote sensing images, a super-resolution reconstruction algorithm based on symmetric local fusion blocks is proposed using a convolutional neural network based on local fusion blocks, which improves the effect of high-frequency information reconstruction. By setting local fusion in the residual block, the problem of insufficient high-frequency feature extraction is alleviated, and the reconstruction accuracy of remote sensing images of deep networks is improved. To improve the utilization of global features and reduce the computational complexity of the network, a residual method is used to set the symmetric jump connection between the local fusion blocks to form the symmetry between them. Experimental results show that the reconstruction results of 2-, 3-, and 4-fold sampling factors on the UC Merced and nwpu-resisc45 remote sensing datasets are better than those of comparison algorithms in image clarity and edge sharpness, and the reconstruction results are better in objective evaluation and subjective vision.

A normative model for Bayesian combination of subjective probability estimates

AbstractCombining experts’ subjective probability estimates is a fundamental task with broad applicability in domains ranging from finance to public health. However, it is still an open question how to combine such estimates optimally. Since the beta distribution is a common choice for modeling uncertainty about probabilities, here we propose a family of normative Bayesian models for aggregating probability estimates based on beta distributions. We systematically derive and compare different variants, including hierarchical and non-hierarchical as well as asymmetric and symmetric beta fusion models. Using these models, we show how the beta calibration function naturally arises in this normative framework and how it is related to the widely used Linear-in-Log-Odds calibration function. For evaluation, we provide the new Knowledge Test Confidence data set consisting of subjective probability estimates of 85 forecasters on 180 queries. On this and another data set, we show that the hierarchical symmetric beta fusion model performs best of all beta fusion models and outperforms related Bayesian fusion models in terms of mean absolute error.

SuperYOLO: Super Resolution Assisted Object Detection in Multimodal Remote Sensing Imagery

Accurately and timely detecting multiscale small objects that contain tens of pixels from remote sensing images (RSI) remains challenging. Most of the existing solutions primarily design complex deep neural networks to learn strong feature representations for objects separated from the background, which often results in a heavy computation burden. In this article, we propose an accurate yet fast object detection method for RSI, named SuperYOLO, which fuses multimodal data and performs high-resolution (HR) object detection on multiscale objects by utilizing the assisted super resolution (SR) learning and considering both the detection accuracy and computation cost. First, we utilize a symmetric compact multimodal fusion (MF) to extract supplementary information from various data for improving small object detection in RSI. Furthermore, we design a simple and flexible SR branch to learn HR feature representations that can discriminate small objects from vast backgrounds with low-resolution (LR) input, thus further improving the detection accuracy. Moreover, to avoid introducing additional computation, the SR branch is discarded in the inference stage, and the computation of the network model is reduced due to the LR input. Experimental results show that, on the widely used VEDAI RS dataset, SuperYOLO achieves an accuracy of 75.09% (in terms of mAP50 ), which is more than 10% higher than the SOTA large models, such as YOLOv5l, YOLOv5x, and RS designed YOLOrs. Meanwhile, the parameter size and GFLOPs of SuperYOLO are about 18 times and 3.8 times less than YOLOv5x. Our proposed model shows a favorable accuracy and speed tradeoff compared to the state-of-the-art models. The code will be open-sourced at https://github.com/icey-zhang/SuperYOLO.

Conjugate Symmetric Data Transmission Control Method based on Machine Learning

In conjugate symmetric data transmission, due to insufficient judgment of congestion during transmission, the amount of data is large and the transmission rate is low. In order to improve the data transmission rate, a conjugate symmetric data transmission control method based on machine learning is designed. Firstly, the data to be transmitted is tracked and determined, and then conjugate symmetric data fusion is completed according to the calculation result of the best tracking signal. According to the fusion results, the framework of the conjugate symmetric data coding system is established, and the data coding is completed. The average congestion mark value is calculated by the machine learning method, and the congestion judgment of data transmission is completed. On the basis of congestion determination, the efficient transmission control of conjugate symmetric data is realized by specifying the conjugate symmetric data transmission protocol. Experimental results show that compared with traditional control methods, this control method has the advantages of a high delivery rate, low message transmission overhead and low data transmission delay. Compared to the traditional two-way path model, the scheduling method proposed in this study increases the transmission delivery rate by 5%, while reducing the transmission cost and delay by 0.7 cost index and 1.1 min delay, respectively. In comparison with the performance of accurate error tracking equalization, the transmission delivery rate of the research method increased by 21%, and in transmission cost index and delay analysis, it also decreased by 3.1 and 2.9 minutes. Based on the above performance comparison analysis, it can be concluded that the machine learning method has more superior transmission control performance.

SuperFusion: A Versatile Image Registration and Fusion Network with Semantic Awareness

Image fusion aims to integrate complementary information in source images to synthesize a fused image comprehensively characterizing the imaging scene. However, existing image fusion algorithms are only applicable to strictly aligned source images and cause severe artifacts in the fusion results when input images have slight shifts or deformations. In addition, the fusion results typically only have good visual effect, but neglect the semantic requirements of high-level vision tasks. This study incorporates image registration, image fusion, and semantic requirements of high-level vision tasks into a single framework and proposes a novel image registration and fusion method, named SuperFusion. Specifically, we design a registration network to estimate bidirectional deformation fields to rectify geometric distortions of input images under the supervision of both photometric and end-point constraints. The registration and fusion are combined in a symmetric scheme, in which while mutual promotion can be achieved by optimizing the naive fusion loss, it is further enhanced by the mono-modal consistent constraint on symmetric fusion outputs. In addition, the image fusion network is equipped with the global spatial attention mechanism to achieve adaptive feature integration. Moreover, the semantic constraint based on the pre-trained segmentation model and Lovasz-Softmax loss is deployed to guide the fusion network to focus more on the semantic requirements of high-level vision tasks. Extensive experiments on image registration, image fusion, and semantic segmentation tasks demonstrate the superiority of our SuperFusion compared to the state-of-the-art alternatives. The source code and pre-trained model are publicly available at https://github.com/Linfeng-Tang/SuperFusion.

Dynamics of asymmetric membranes and interleaflet coupling as intermediates in membrane fusion

Membrane fusion is a tool to increase the complexity of model membrane systems. Here, we use silica nanoparticles to fuse liquid-disordered DOPC giant unilamellar vesicles (GUVs) and liquid-ordered DPPC:cholesterol (7:3) GUVs. After fusion, GUVs display large membrane domains as confirmed by fluorescence confocal microscopy. Laurdan spectral imaging of the membrane phases in the fused GUVs shows differences compared with the initial vesicles indicating some lipid redistribution between phase domains as dictated by the tie lines of the phase diagram. Remarkably, using real-time confocal microscopy we were able to record the dynamics of formation of asymmetric membrane domains in hemifused GUVs and detected interleaflet coupling phenomena by which the DOPC-rich liquid-disordered domains in outer monolayer modulates the phase state of the DPPC:cholesterol inner membrane leaflet which transitions from liquid-ordered to liquid-disordered phase. We find that internal membrane stresses generated by membrane asymmetry enhance the efficiency of full fusion compared with our previous studies on symmetric vesicle fusion. Furthermore, under these conditions, the liquid-disordered monolayer dictates the bilayer phase state of asymmetric membrane domains in >90% of observed cases. By comparison to the findings of previous literature, we suggest that the monolayer phase that dominates the bilayer properties could be a mechanoresponsive signaling mechanism sensitive to the local membrane environment.

Computing the Group of Minimal Non-degenerate Extensions of a Super-Tannakian Category

We prove an analog of the K\"unneth formula for the groups of minimal non-degenerate extensions arXiv:1602.05936 of symmetric fusion categories. We describe in detail the structure of the group of minimal extensions of a pointed super-Tannakian fusion category. This description resembles that of the third cohomology group of a finite abelian group. We explicitly compute this group in several concrete examples.

The relationship between non-aneurysmal spontaneous subarachnoid hemorrhage and basilar tip anatomy.

Objective Non-aneurysmal spontaneous subarachnoid hemorrhage (NASAH) has a good prognosis, but its cause has not been clearly identified. In this study, we assessed the clinical and radiological features of NASAH and suggested an anatomical relationship between the basilar tip anatomy and NASAH.Methods From August 2013 to May 2020, 21 patients were diagnosed with NASAH at our institution. We evaluated the clinical features of NASAH. NASAH was classified into a perimesencephalic pattern and aneurysmal pattern according to the distribution of hemorrhage based on initial brain computed tomography. Digital subtraction angiography was used to classify the basilar tip anatomy into symmetric cranial fusion, symmetric caudal fusion, or asymmetric fusion types.Results Of the 21 patients, twenty patients had a good clinical outcome (modified Rankin Scale (mRS) 1–2; Glasgow Outcome Scale (GOS) 4–5). These patients showed improvement in mRS and Glasgow Coma Scale (GCS) at the last follow-up (P=.003 and P=.016, respectively). Eighteen patients with NASAH (85.7%) had the caudal fusion type, and only three patients with NASAH (14.3%) had the cranial fusion type. Seven patients with the perimesencephalic pattern (77.8%) had the caudal fusion type, and eleven patients with the aneurysmal pattern (91.7%) had the caudal fusion type.Conclusions In NASAH patients, the caudal fusion tends to occur frequently among patients with basilar tip anatomy. In the case of the caudal fusion, the perforators around the basilar tip would be more susceptible to hemodynamic stress, which could contribute to the occurrence of NASAH.