Target Entrapment Based on Adaptive Transformation of Gene Regulatory Networks

The complexity of unknown scenarios and the dynamics involved in target entrapment make designing control strategies for swarm robots a formidable task,which in turn impacts their efficiency in complex and dynamic settings.To address these challenges,this paper introduces an adaptive swarm robot entrapment control model grounded in the transformation of gene regulatory networks(AT-GRN).This innovative model enables swarm robots to dynamically adjust entrap-ment strategies by assessing current environmental conditions via real-time sensory data.Further-more,an improved motion control model for swarm robots is designed to dynamically shape the for-mation generated by the AT-GRN.Through two sets of rigorous experimental environments,the proposed model significantly enhances the trapping performance of swarm robots in complex envi-ronments,demonstrating remarkable adaptability and stability.

How Software Engineering Transforms Organizations: An Open and Qualitative Study on the Organizational Objectives and Motivations in Agile Transformations

Agile Transformations are challenging processes for organizations that look to extend the benefits of Agile philosophy and methods beyond software engineering.Despite the impact of these transformations on orga-nizations,they have not been extensively studied in academia.We conducted a study grounded in workshops and interviews with 99 participants from 30 organizations,including organizations undergoing transformations(“final organizations”)and companies supporting these processes(“consultants”).The study aims to understand the motivations,objectives,and factors driving and challenging these transformations.Over 700 responses were collected to the question and categorized into 32 objectives.The findings show that organizations primarily aim to achieve customer centricity and adaptability,both with 8%of the mentions.Other primary important objectives,with above 4%of mentions,include alignment of goals,lean delivery,sustainable processes,and a flatter,more team-based organizational structure.We also detect discrepancies in perspectives between the objectives identified by the two kinds of organizations and the existing agile literature and models.This misalignment highlights the need for practitioners to understand with the practical realities the organizations face.

Research on blind source separation of operation sounds of metro power transformer through an Adaptive Threshold REPET algorithm

Purpose–The safe operation of the metro power transformer directly relates to the safety and efficiency of the entire metro system.Through voiceprint technology,the sounds emitted by the transformer can be monitored in real-time,thereby achieving real-time monitoring of the transformer’s operational status.However,the environment surrounding power transformers is filled with various interfering sounds that intertwine with both the normal operational voiceprints and faulty voiceprints of the transformer,severely impacting the accuracy and reliability of voiceprint identification.Therefore,effective preprocessing steps are required to identify and separate the sound signals of transformer operation,which is a prerequisite for subsequent analysis.Design/methodology/approach–This paper proposes an Adaptive Threshold Repeating Pattern Extraction Technique(REPET)algorithm to separate and denoise the transformer operation sound signals.By analyzing the Short-Time Fourier Transform(STFT)amplitude spectrum,the algorithm identifies and utilizes the repeating periodic structures within the signal to automatically adjust the threshold,effectively distinguishing and extracting stable background signals from transient foreground events.The REPET algorithm first calculates the autocorrelation matrix of the signal to determine the repeating period,then constructs a repeating segment model.Through comparison with the amplitude spectrum of the original signal,repeating patterns are extracted and a soft time-frequency mask is generated.Findings–After adaptive thresholding processing,the target signal is separated.Experiments conducted on mixed sounds to separate background sounds from foreground sounds using this algorithm and comparing the results with those obtained using the FastICA algorithm demonstrate that the Adaptive Threshold REPET method achieves good separation effects.Originality/value–A REPET method with adaptive threshold is proposed,which adopts the dynamic threshold adjustment mechanism,adaptively calculates the threshold for blind source separation and improves the adaptability and robustness of the algorithm to the statistical characteristics of the signal.It also lays the foundation for transformer fault detection based on acoustic fingerprinting.

基于Swin Transformer网络与Adapt-RandAugment数据增强方法的小肠胶囊内镜图像分类方法研究

目的:为提高小肠病变分类识别的准确性,提出一种基于Swin Transformer网络与Adapt-RandAugment数据增强方法的小肠胶囊内镜图像分类方法。方法:基于RandAugment数据增强子策略和增强小肠胶囊内镜图像时不丢失特征、不失真的原则提出Adapt-RandAugment数据增强方法。在公开的小肠胶囊内镜图像Kvasir-Capsule数据集中,基于Swin Transformer网络,采用Adapt-RandAugment数据增强方法进行训练,以卷积神经网络ResNet152、DenseNet161为基准,验证Swin Transformer网络和Adapt-RandAugment数据增强方法组合对小肠胶囊内镜图像分类识别的性能。结果:提出的方法宏平均精度(macro average precision,MAC-PRE)、宏平均召回率(macro average recall,MAC-REC)、宏F1分数(macro average F1 score,MAC-F1-S)分别为0.3832、0.3148、0.2905,微平均精度(micro average precision,MIC-PRE)、微平均召回率(micro average recall,MIC-REC)、微平均F1分数(micro average F1 score,MIC-F1-S)均为0.7553,马修斯相关系数(Matthews correlation coefficient,MCC)为0.4523,均优于ResNet152和DenseNet161网络。结论:基于Swin Transformer网络与Adapt-RandAugment数据增强方法的小肠胶囊内镜图像分类方法具有较好的小肠胶囊内镜图像分类识别效果和较高的识别准确率。

基于改进Transformer的滚动轴承剩余寿命预测方法

针对现有的滚动轴承剩余使用寿命预测方法存在预测准确度不足、训练效率不高等问题,提出一种时频分析结合改进Transformer的轴承剩余使用寿命预测方法。首先用短时傅里叶变换提取轴承的时频特征,为了改善Transformer的特征提取能力,研究了基于膨胀因果卷积的可变长度数据分析结构,并设计了自适应位置编码模块替代Transformer的传统编码方式,改进的模型增强了对时频数据的分析能力,实现了高效、准确的端到端的滚动轴承剩余寿命预测。在PHM2012轴承数据集上的实验结果表明提出的方法的效率比LSTM高20%,同时预测精度相比于多种现有传统方法均具有16%以上的提升。

基于多模态融合和自适应剪枝Transformer的脑肿瘤图像分割算法

脑肿瘤是目前世界上最致命的肿瘤之一,所以脑肿瘤图像的自动分割在临床诊疗中变得日益重要.近年来,基于CNN和Transformer的脑肿瘤分割方法在医学图像分割领域取得了令人欣喜的成就.然而,大多数方法没有充分利用脑肿瘤多模态间的互补性和差异性,并且模型中的Transformer在捕获远程依赖性的同时,忽略了其较大的计算复杂性、冗余依赖性等问题.针对此问题,提出一种基于多模态融合和自适应剪枝Transformer的脑肿瘤图像分割方法(MF-MAPT Swin UNETR),其中多模态融合模块可以充分学习性质相近的模态间信息和不同模态不同尺度的特征变化,为后续分割提供了充分的准备;基于多模态的自适应剪枝Transformer可以降低计算复杂度,对提升性能有一定的帮助,将MF-MAPT Swin UNETR模型在两个公共数据集上进行了实验验证,结果表明,该模型较最先进的方法整体具有突出的分割性能.

Multiconstraint adaptive three-dimensional guidance law using convex optimization

The traditional guidance law only guarantees the accuracy of attacking a target. However, the look angle and acceleration constraints are indispensable in applications. A new adaptive three-dimensional proportional navigation(PN) guidance law is proposed based on convex optimization. Decomposition of the three-dimensional space is carried out to establish threedimensional kinematic engagements. The constraints and the performance index are disposed by using the convex optimization method. PN guidance gains can be obtained by solving the optimization problem. This solution is more rapid and programmatic than the traditional method and provides a foundation for future online guidance methods, which is of great value for engineering applications.

引入轻量级Transformer的自适应窗口立体匹配算法

现有端到端的立体匹配算法为了减轻显存消耗和计算量而预设固定视差范围,在匹配精度和运行效率上难以平衡。提出一种基于轻量化Transformer的自适应窗口立体匹配算法。利用具有线性复杂度的坐标注意力层对低分辨率特征图进行位置编码,减轻计算量并增强相似特征的辨别力;设计轻量化Transformer特征描述模块,转换上下文相关的特征,并引入可分离多头自注意力层对Transformer进行轻量化改进,降低Transformer的延迟性;用可微匹配层对特征进行匹配,设计自适应窗口匹配细化模块进行亚像素级的匹配细化,在提高匹配精度的同时减少显存消耗;经视差回归后生成无视差范围的视差图。在KITTI2015、KITTI2012和SceneFlow数据集上的对比实验表明,该算法比基于标准Transformer的STTR在匹配效率上快了近4.7倍,具有更快的运行速度和更友好的存储性能;比基于3D卷积的PSMNet误匹配率降低了18%,运行时间快了5倍,实现了更好的速度和精度的平衡。

一种通道自适应与局部增强的Transformer术中血压预测方法

准确预测术中患者的血压状态来预防术中低血压,对提高手术安全性和降低术后并发症有积极作用,以往的低血压预测方法主要视为二分类任务,忽略了患者血压变化的过程,从而限制了干预策略的制定。因此提前预测血压的变化趋势,具有更重要的临床研究和应用价值。针对以上问题,对通过监测的术中生理序列实时预测未来5min、10min、15min血压的连续值展开研究,并提出了一种通道自适应与局部增强Transformer模型,该模型采用卷积注意力机制捕捉血压序列的局部相似性,同时提出一种通道自适应模块嵌入模型来建模生理序列潜在交互关系。结果表明,该模型相比于基准模型在5min、10min、15min预测精度分别提升4.88%、8.2%和8.42%,预测的平均动脉压的MAE分别为2.997、3.393、3.743,且显著优于其余对比模型,为术中血压预测提供新的解决方案。

时空自适应图卷积与Transformer结合的动作识别网络

在一个以人为中心的智能工厂中,感知和理解工人的行为是至关重要的,不同工种类别往往与工作时间和工作内容相关。该文通过结合自适应图和Transformer两种方式使模型更关注骨架的时空信息来提高模型识别的准确率。首先,采用一个自适应的图方法去关注除人体骨架之外的连接关系。进一步,采用Transformer框架去捕捉工人骨架在时间维度上的动态变化信息。为了评估模型性能,制作了智能生产线装配任务中6种典型的工人动作数据集,并进行验证,结果表明所提模型在Top-1精度上与主流动作识别模型相当。最后,在公开的NTURGBD和Skeleton-Kinetics数据集上,将该文模型与一些主流方法进行对比,实验结果表明,所提模型具有良好鲁棒性。

基于全局自适应宽度注意力改进的Transformer

Transformer在自然语言处理中运用广泛,但存在文本长度过长带来的输入信息被切割、显存占用太大的问题,已有的解决方法是让模型动态决定每层注意力宽度,可以在控制计算量和显存开销的前提下关联最优序列长度,但存在每层最优的注意力宽度并不能达到模型最优注意力宽度的缺点。为此,提出一种全层自适应宽度注意力模型(GAA)。让每层的注意力范围和全局关联,实现模型全局注意力范围最优,还将模型的前馈层修改为带门控单元的前馈层(FFN_(GLU))。在数据集enwiki8和text-8上的验证表明,该方法仅使用25%的训练计算成本,即可达到比基线更好的性能。

Comparison of Adaptive Simulation Observation Experiments of the Heavy Rainfall in South China and Sichuan Basin

This study examines the effectiveness of adaptive observation experiments using the ensemble transformation sensitivity(ETS) method to improve precipitation forecasts during heavy rainfall events in South China and the Sichuan Basin. High-resolution numerical models are employed to simulate adaptive observations. By identifying the sensitive areas of key weather system positions 42 hours before heavy rainfall events, the adaptive observations improve the prediction of jet streams, strong winds, and shear lines, which are essential for accurate heavy rainfall forecasting. This improvement is reflected in both the precipitation structure and location accuracy within the verification region. In South China, targeted observations enhance rainfall predictions by improving water vapor transport. In the Sichuan Basin, adaptive observations refine water vapor transport and adjust vortex dynamics. This research highlights the importance of accurately predicting shear lines and jet streams for forecasting heavy rainfall in these areas. Overall, this study found that adaptive observation enhances the precipitation forecast skills of the structure and location for heavy rainfall in South China and the Sichuan Basin, emphasizing their potential utility in operational numerical weather prediction.

Phase Error Compensation of Three-Dimensional Reconstruction Combined with Hilbert Transform

Nonlinear response is an important factor affecting the accuracy of three-dimensional image measurement based on the fringe structured light method.A phase compensation algorithm combined with a Hilbert transform is proposed to reduce the phase error caused by the nonlinear response of a digital projector in the three-dimensional measurement system of fringe structured light.According to the analysis of the influence of Gamma distortion on the phase calculation,the algorithm establishes the relationship model between phase error and harmonic coefficient,introduces phase shift to the signal,and keeps the signal amplitude constant while filtering out the DC component.The phase error is converted to the transform domain,and compared with the numeric value in the space domain.The algorithm is combined with a spiral phase function to optimize the Hilbert transform,so as to eliminate external noise,enhance the image quality,and get an accurate phase value.Experimental results show that the proposed method can effectively improve the accuracy and speed of phase measurement.By performing phase error compensation for free-form surface objects,the phase error is reduced by about 26%,and about 27%of the image reconstruction time is saved,which further demonstrates the feasibility and effectiveness of the method.

APK-CNN和Transformer增强的多域虚假新闻检测模型

为解决社交媒体新闻中的领域转移、领域标签不完整问题,以及探索更高效的多域新闻文本特征提取和融合网络,提出一种基于APK-CNN(Adaptive Pooling Kernel Convolutional Neural Network)和Transformer增强的多域虚假新闻检测模型Transm3。首先,设计三通道网络对文本的语义、情感和风格信息进行特征提取和表示,并利用多粒度跨域交互器对这些特征进行视图组合;其次,通过优化的软共享内存网络和域适配器来完善新闻领域标签;再次,将Transformer与多粒度跨域交互器结合,使用更先进的融合网络动态加权聚合不同领域的交互特征;最后,将融合特征输入分类器中用于真/假新闻判别。实验结果表明,Transm3与M3FEND(Memory-guided Multi-view Multi-domain FakE News Detection)和EANN(Event Adversarial Neural Networks for multi-modal fake news detection)相比,综合F1值在中文数据集上分别提高了3.68%和6.46%,在英文数据集上分别提高了6.75%和11.93%,在各分领域上F1值也有明显的提高,充分验证了Transm3在多域虚假新闻检测工作上的有效性。

Transformative Learning:Intercultural Adaptation of Chinese Teachers at the Confucius Institute in Spain

In recent years there has been a proliferation of studies demonstrating the value of teaching abroad as much for its benefits for the training and professional development of these teachers,as for its impact and beneficial effects on students.This article uses transformative learning as a theoretical framework to interpret the achievements associated with the experience of teaching abroad,and to identify and analyze different motivational factors,adjustments,changes,challenges,and perspectives of Chinese teachers linked to a Confucius Institute in Spain,through a qualitative analysis of narratives elicited through in-depth interviews and focus groups.It highlights the role and potential of the transnational,intercultural experience of these teachers as authentic actors in the part played by the Confucius Institutes in language teaching and the promotion of Chinese culture,indicating many issues including language difficulties,professional adjustment,ideas about education,beliefs of teachers and the management of the program.

Multi Parameter Adaptive Estimation of Reaction-Diffusion Equation

This study addresses the problem of parameter estimation for a one-dimensional reaction-diffusion equation, involving both unknown domain parameters and unknown boundary parameters. The proposed approach utilizes the least-squares method to design an adaptive law for parameter estimation. The convergence analysis demonstrates that under persistent excitation conditions, the adaptive law converges exponentially to zero, indicating that the estimated parameters converge exponentially to their true values. Numerical simulations confirm the effectiveness. Furthermore, it is shown that within a certain range of the reaction coefficient, the auxiliary system acts as a state observer, providing an accurate estimate of the system state at an exponential rate. .

Transformer在域适应中的应用研究综述

作为迁移学习的重要分支,域适应旨在解决传统机器学习算法在训练样本和测试样本服从不同数据分布时性能急剧下降的问题。Transformer是基于自注意力机制的深度学习框架,具有强大的全局特征提取能力和建模能力,近年来Transformer与域适应相结合也成为研究的热点。虽然已有大量相关方法问世,但Transformer应用在域适应的综述却未见报道。为了填补这个领域的空白,为相关研究提供借鉴和参考,对近年来出现的一些基于Transformer的典型域适应方法进行归纳总结与分析,概述域适应的相关概念与Transformer的基本结构,从图像分类、图像语义分割、目标检测、医学图像分析四个应用梳理了各种基于Transformer的域适应方法,对图像分类下的域适应方法进行比较,总结当前域适应Transformer模型存在的挑战并探讨未来可行的研究方向。

Multi-Modal Medical Image Fusion Based on Improved Parameter Adaptive PCNN and Latent Low-Rank Representation

Multimodal medical image fusion can help physicians provide more accurate treatment plans for patients, as unimodal images provide limited valid information. To address the insufficient ability of traditional medical image fusion solutions to protect image details and significant information, a new multimodality medical image fusion method(NSST-PAPCNNLatLRR) is proposed in this paper. Firstly, the high and low-frequency sub-band coefficients are obtained by decomposing the source image using NSST. Then, the latent low-rank representation algorithm is used to process the low-frequency sub-band coefficients;An improved PAPCNN algorithm is also proposed for the fusion of high-frequency sub-band coefficients. The improved PAPCNN model was based on the automatic setting of the parameters, and the optimal method was configured for the time decay factor αe. The experimental results show that, in comparison with the five mainstream fusion algorithms, the new algorithm has significantly improved the visual effect over the comparison algorithm,enhanced the ability to characterize important information in images, and further improved the ability to protect the detailed information;the new algorithm has achieved at least four firsts in six objective indexes.

Performance of Wavelet-Transform-Domain Adaptive Equalizers

In this paper performances of wavelet transform domain (WTD) adaptive equalizers based on the least mean￣square (LMS) algorithm are analyzed. The optimum Wiener solution, the condition of convergence, the minimum mean square error (MSE) and the steady state excess MSE of the WTD adaptive equalizer are obtained. Constant and time varying convergence factor adaptive algorithms are studied respectively. Computational complexities of WTD LMS equalizers are given. The equalizer in WTD shows much better convergence performance than that of the conventional in time domain.

The optimal fractional Gabor transform based on the adaptive window function and its application

We designed the window function of the optimal Gabor transform based on the time-frequency rotation property of the fractional Fourier transform. Thus, we obtained the adaptive optimal Gabor transform in the fractional domain and improved the time-frequency concentration of the Gabor transform. The algorithm first searches for the optimal rotation factor, then performs the p-th FrFT of the signal and, finally, performs time and frequency analysis of the FrFT result. Finally, the algorithm rotates the plane in the fractional domain back to the normal time-frequency plane. This promotes the application of FrFT in the field of high-resolution reservoir prediction. Additionally, we proposed an adaptive search method for the optimal rotation factor using the Parseval principle in the fractional domain, which simplifies the algorithm. We carried out spectrum decomposition of the seismic signal, which showed that the instantaneous frequency slices obtained by the proposed algorithm are superior to the ones obtained by the traditional Gabor transform. The adaptive time frequency analysis is of great significance to seismic signal processing.