基于FP-tree的新能源汽车产业国际竞争力影响因素关联挖掘算法

为充分挖掘新能源汽车产业国际竞争力影响因素、探究价值增长点,提出了基于FP-tree的影响因素关联挖掘算法。通过构建国际竞争力各指标影响因素的关联规则,分析任意事务数据集中的关联数据,利用最小支持度参数minsup按照从上到下的方式搜索,确定最长的频繁项目集,采用FP-tree关联频繁项目集,设定分支关联性挖掘标准,实现了新能源汽车产业国际竞争力影响因素的挖掘。测试结果表明,设计算法的最小支持度和数据关联挖掘时间较短,置信度分析具有较高的稳定性。

基于SORT映射的IRCMFDE在旋转机械故障诊断中的应用

针对旋转机械振动信号的强非线性和非平稳性,导致故障特征提取困难的问题,提出了一种基于SORT映射的改进精细复合多尺度波动散布熵(IRCMFDE)和蝙蝠算法优化的相关向量机(BA-RVM)的旋转机械故障诊断方法。首先,利用SORT映射函数替换了精细复合多尺度波动散布熵(RCMFDE)方法的正态累积分布函数,同时对RCMFDE方法的粗粒化方式进行了改进,提出了基于SORT映射的IRCMFDE方法;随后,利用IRCMFDE方法提取了旋转机械振动信号的故障特征,构造了故障特征集;最后,采用BA-RVM分类器对旋转机械的故障类型进行了智能化的识别和分类;将基于IRCMFDE和BA-RVM的故障诊断方法应用于滚动轴承、离心泵和齿轮箱的实验数据分析,并将其与现有故障诊断方法进行了对比分析。研究结果表明:基于IRCMFDE和BA-RVM的故障诊断方法能够有效地识别旋转机械的故障状态,识别准确率分别达到了100%、98%和99%,相比基于RCMFDE、精细复合多尺度熵、精细复合多尺度模糊熵、精细复合多尺度排列熵和精细复合多尺度散布熵的故障特征提取方法,该故障诊断方法的效率和平均识别准确率均优于对比方法,其更适合应用于旋转机械的在线实时故障监测。

An Optimization Approach for Convolutional Neural Network Using Non-Dominated Sorted Genetic Algorithm-Ⅱ

In computer vision,convolutional neural networks have a wide range of uses.Images representmost of today’s data,so it’s important to know how to handle these large amounts of data efficiently.Convolutional neural networks have been shown to solve image processing problems effectively.However,when designing the network structure for a particular problem,you need to adjust the hyperparameters for higher accuracy.This technique is time consuming and requires a lot of work and domain knowledge.Designing a convolutional neural network architecture is a classic NP-hard optimization challenge.On the other hand,different datasets require different combinations of models or hyperparameters,which can be time consuming and inconvenient.Various approaches have been proposed to overcome this problem,such as grid search limited to low-dimensional space and queuing by random selection.To address this issue,we propose an evolutionary algorithm-based approach that dynamically enhances the structure of Convolution Neural Networks(CNNs)using optimized hyperparameters.This study proposes a method using Non-dominated sorted genetic algorithms(NSGA)to improve the hyperparameters of the CNN model.In addition,different types and parameter ranges of existing genetic algorithms are used.Acomparative study was conducted with various state-of-the-art methodologies and algorithms.Experiments have shown that our proposed approach is superior to previous methods in terms of classification accuracy,and the results are published in modern computing literature.

Bubble-sort网络的一类条件连通度

[目的]为评价网络容错性,以具有高对称性和递归结构的Bubble-sort网络为研究对象,确定其h-extra r-分支边连通度.[方法] Bubble-sort网络Bn可以分解成n个子图Bn(i),其中Bn(i)是由点集{x_(1)x_(2)…x_(n):x_(n)=i}(1≤i≤n)导出的子图,并且Bn(i)同构于B_(n-1),利用它的结构特点,用数学归纳法推理证明了主要结果.[结果]确定了bubble-sort网络的h-extra r-分支边连通度cλ2/3(B_(n))=4n-10(n≥4).[结论]研究了bubble-sort网络的一类条件连通度,可用于衡量网络的可靠性.今后将继续深入研究bubble-sort网络的其他条件连通度.

基于SORT算法的图像轨迹跟踪混合控制方法

当目标物体被其他物体部分或完全遮挡时,目标的有效特征点数量会逐渐减少,跟踪器无法继续准确地锁定目标,导致目标轨迹中断。为此,文中研究基于SORT算法的图像轨迹跟踪混合控制方法。选取FCOS算法,利用特征金字塔结构,依据检测头层输出的目标分类得分、位置回归结果以及中心度检测图像目标。将目标检测结果作为卡尔曼滤波器的输入,利用离散控制过程系统描述视频图像中的目标运动状态,预测目标轨迹。利用SORT算法控制图像目标检测结果与目标轨迹预测结果进行级联匹配与IoU匹配,输出匹配成功的目标,即图像目标轨迹跟踪结果。实验结果表明,该方法可有效地跟踪视频图像目标轨迹,未出现ID变更情况,轨迹中断占比低于0.2%。

Comparative analysis shows high level of lineage sorting in genomic regions with low recombination in the extended Picea likiangensis species complex

Genome-scale data,while promising for illuminating phylogenetic relationships,frequently pose a conundrum by yielding conflicting topologies and highly variable gene tree distributions(Pease et al.,2016).This complexity likely arises from the reticulate evolution observed in many taxa,where genetic information exchange occurs through diverse biological processes.

Digital Twin Technology of Human-Machine Integration in Cross-Belt Sorting System

The Chinese express delivery industry processes nearly 110 billion items in 2022,averaging an annual growth rate of 200%.Among the various types of sorting systems used for handling express items,cross-belt sorting systems stand out as the most crucial.However,despite their high degree of automation,the workload for operators has intensified owing to the surging volume of express items.In the era of Industry 5.0,it is imperative to adopt new technologies that not only enhance worker welfare but also improve the efficiency of cross-belt systems.Striking a balance between efficiency in handling express items and operator well-being is challenging.Digital twin technology offers a promising solution in this respect.A realization method of a human-machine integrated digital twin is proposed in this study,enabling the interaction of biological human bodies,virtual human bodies,virtual equipment,and logistics equipment in a closed loop,thus setting an operating framework.Key technologies in the proposed framework include a collection of heterogeneous data from multiple sources,construction of the relationship between operator fatigue and operation efficiency based on physiological measurements,virtual model construction,and an online optimization module based on real-time simulation.The feasibility of the proposed method was verified in an express distribution center.

A novel integrated microfluidic chip for on-demand electrostatic droplet charging and sorting

On-demand droplet sorting is extensively applied for the efficient manipulation and genome-wide analysis of individual cells.However,state-of-the-art microfluidic chips for droplet sorting still suffer from low sorting speeds,sample loss,and labor-intensive preparation procedures.Here,we demonstrate the development of a novel microfluidic chip that integrates droplet generation,on-demand electrostatic droplet charging,and high-throughput sorting.The charging electrode is a copper wire buried above the nozzle of the microchannel,and the deflecting electrode is the phosphate buffered saline in the microchannel,which greatly simplifies the structure and fabrication process of the chip.Moreover,this chip is capable of high-frequency droplet generation and sorting,with a frequency of 11.757 kHz in the drop state.The chip completes the selective charging process via electrostatic induction during droplet generation.On-demand charged microdroplets can arbitrarilymove to specific exit channels in a three-dimensional(3D)-deflected electric field,which can be controlled according to user requirements,and the flux of droplet deflection is thereby significantly enhanced.Furthermore,a lossless modification strategy is presented to improve the accuracy of droplet deflection or harvest rate from 97.49% to 99.38% by monitoring the frequency of droplet generation in real time and feeding it back to the charging signal.This chip has great potential for quantitative processing and analysis of single cells for elucidating cell-to-cell variations.

Phylotranscriptomic discordance is best explained by incomplete lineage sorting within Allium subgenus Cyathophora and thus hemiplasy accounts for interspecific trait transition

The transition of traits between genetically related lineages is a fascinating topic that provides clues to understanding the drivers of speciation and diversification.Much can be learned about this process from phylogeny-based trait evolution.However,such inference is often plagued by genome-wide gene-tree discordance(GTD),mostly due to incomplete lineage sorting(ILS)and/or introgressive hybridization,especially when the genes underlying the traits appear discordant.Here,by collecting transcriptomes,whole chloroplast genomes(cpDNA),and population genetic datasets,we used the coalescent model to turn GTD into a source of information for ILS and employed hemiplasy to explain specific cases of apparent“phylogenetic discordance”between different morphological traits and probable species phylogeny in the Allium subg.Cyathophora.Both concatenation and coalescence methods consistently showed the same phylogenetic topology for species tree inference based on single-copy genes(SCGs),as supported by the KS distribution.However,GTD was high across the genomes of subg.Cyathophora:~27%e38.9%of the SCG trees were in conflict with the species tree.Plasmid and nuclear incongruence was also present.Our coalescent simulations indicated that such GTD was mainly a product of ILS.Our hemiplasy risk factor calculations supported that random fixation of ancient polymorphisms in different populations during successive speciation events along the subg.Cyathophora phylogeny may have caused the character transition,as well as the anomalous cpDNA tree.Our study exemplifies how phylogenetic noise can be transformed into evolutionary information for understanding character state transitions along species phylogenies.

改进Bot-SORT的边坡落石监测方法

针对边坡落石监测中存在的目标尺寸小、石块与背景特征差距小、落石目标运动速度快等问题,提出一种基于检测的改进Bot-SORT多目标跟踪算法。在检测部分对YOLOv7模型进行改进,引入注意力机制,提升模型对石块特征的提取能力,并使用归一化高斯Wasserstein距离作为真值框与预测框的距离度量方式,降低模型对小目标的漏检率;在跟踪部分引入GIoU距离匹配方式,有效跟踪快速运动的落石。通过实景拍摄及Unity仿真方式建立训练及测试数据集,消融实验和对比实验结果表明,本文改进算法能够有效提高落石的检测率和跟踪精度。

Enhancing XRF sensor-based sorting of porphyritic copper ore using particle swarm optimization-support vector machine(PSO-SVM)algorithm

X-ray fluorescence(XRF)sensor-based ore sorting enables efficient beneficiation of heterogeneous ores,while intraparticle heterogeneity can cause significant grade detection errors,leading to misclassifications and hindering widespread technology adoption.Accurate classification models are crucial to determine if actual grade exceeds the sorting threshold using localized XRF signals.Previous studies mainly used linear regression(LR)algorithms including simple linear regression(SLR),multivariable linear regression(MLR),and multivariable linear regression with interaction(MLRI)but often fell short attaining satisfactory results.This study employed the particle swarm optimization support vector machine(PSO-SVM)algorithm for sorting porphyritic copper ore pebble.Lab-scale results showed PSO-SVM out-performed LR and raw data(RD)models and the significant interaction effects among input features was observed.Despite poor input data quality,PSO-SVM demonstrated exceptional capabilities.Lab-scale sorting achieved 93.0%accuracy,0.24%grade increase,84.94%recovery rate,57.02%discard rate,and a remarkable 39.62 yuan/t net smelter return(NSR)increase compared to no sorting.These improvements were achieved by the PSO-SVM model with optimized input combinations and highest data quality(T=10,T is XRF testing times).The unsuitability of LR methods for XRF sensor-based sorting of investigated sample is illustrated.Input element selection and mineral association analysis elucidate element importance and influence mechanisms.

Improving path planning efficiency for underwater gravity-aided navigation based on a new depth sorting fast search algorithm

This study focuses on the improvement of path planning efficiency for underwater gravity-aided navigation.Firstly,a Depth Sorting Fast Search(DSFS)algorithm was proposed to improve the planning speed of the Quick Rapidly-exploring Random Trees*(Q-RRT*)algorithm.A cost inequality relationship between an ancestor and its descendants was derived,and the ancestors were filtered accordingly.Secondly,the underwater gravity-aided navigation path planning system was designed based on the DSFS algorithm,taking into account the fitness,safety,and asymptotic optimality of the routes,according to the gravity suitability distribution of the navigation space.Finally,experimental comparisons of the computing performance of the ChooseParent procedure,the Rewire procedure,and the combination of the two procedures for Q-RRT*and DSFS were conducted under the same planning environment and parameter conditions,respectively.The results showed that the computational efficiency of the DSFS algorithm was improved by about 1.2 times compared with the Q-RRT*algorithm while ensuring correct computational results.

Accelerating Large-Scale Sorting through Parallel Algorithms

This study explores the application of parallel algorithms to enhance large-scale sorting, focusing on the QuickSort method. Implemented in both sequential and parallel forms, the paper provides a detailed comparison of their performance. This study investigates the efficacy of both techniques through the lens of array generation and pivot selection to manage datasets of varying sizes. This study meticulously documents the performance metrics, recording 16,499.2 milliseconds for the serial implementation and 16,339 milliseconds for the parallel implementation when sorting an array by using C++ chrono library. These results suggest that while the performance gains of the parallel approach over its serial counterpart are not immediately pronounced for smaller datasets, the benefits are expected to be more substantial as the dataset size increases.

基于关联分析FP-Tree算法的企业风险信息数据在线挖掘方法

现阶段的数据挖掘方法缺少对数据关联分析的过程,挖掘效果较差,故文章提出基于关联分析频繁模式树(FrequentPattern Tree, FP-Tree)算法的企业风险信息数据在线挖掘方法。选取与企业风险相关的信息指标,收集有关数据并进行预处理操作后,设计一种考虑关联分析的FP-Tree算法,生成FP-Tree节点的条件模式树挖掘频繁项集,计算满足最小置信度的频繁项集,实现企业风险信息数据在线挖掘。实验结果表明,所用方法挖掘量和挖掘效率较高。

基于YOLO v5s和改进SORT算法的黑水虻幼虫计数方法

目前农业环境下的无序目标的精确计数有很高的应用需求,这种计数对其生物量、生物密度管理起到了重要的指导作用。如黑水虻幼虫目标追踪过程中,追踪对象具有高速和非线性的特征,常规算法存在追踪目标速度不足和丢失目标后的再识别困难等问题。针对以上问题,本文提出了一种改进SORT算法,通过改进卡尔曼滤波模型的方式提升目标追踪算法的快速性和准确性,提升了计数的精度。另外,针对黑水虻幼虫目标识别过程中幼虫性状的多样性和混料导致的复杂背景问题,本文通过实验对比多种深度学习网络性能选定YOLO v5s算法提取图像多维度特征,提升了目标识别精度。实验结果表明:在划线计数方面,本文提出的改进SORT算法与原模型相比,平均精度从91.36%提升到95.55%,提升4.19个百分点,通过仿真和实际应用,证明了本文模型的有效性;在目标识别方面,使用YOLO v5s模型在训练集上帧率为156 f/s,mAP@0.5为99.10%,精度为90.11%,召回率为99.22%,综合性能优于其他网络。

基于改进Faster R-CNN和Deep Sort的棉铃跟踪计数

棉铃作为棉花重要的产量与品质器官,单株铃数、铃长、铃宽等相关表型性状一直是棉花育种的重要研究内容。为解决由于叶片遮挡导致传统静态图像检测方法无法获取全部棉铃数量的问题,提出了一种以改进Faster R-CNN、Deep Sort和撞线匹配机制为主要算法框架的棉铃跟踪计数方法,以实现在动态视频输入情况下对盆栽棉花棉铃的数量统计。采用基于特征金字塔的Faster R-CNN目标检测网络,融合导向锚框、Soft NMS等网络优化方法,实现对视频中棉铃目标更精确的定位;使用Deep Sort跟踪器通过卡尔曼滤波和深度特征匹配实现前后帧同一目标的相互关联,并为目标进行ID匹配;针对跟踪过程ID跳变问题设计了掩模撞线机制以实现动态旋转视频棉铃数量统计。试验结果表明:改进Faster R-CNN目标检测结果最优,平均测量精度mAP75和F1值分别为0.97和0.96,较改进前分别提高0.02和0.01;改进Faster R-CNN和Deep Sort跟踪结果最优,多目标跟踪精度为0.91,较Tracktor和Sort算法分别提高0.02和0.15;单株铃数计数结果决定系数、均方误差、平均绝对误差和平均绝对百分比误差分别为0.96、1.19、0.81和5.92%,与人工值具有较高一致性,开发的棉铃跟踪软件可以实现对棉铃的有效跟踪和计数。

基于VFNet-Improved和Deep Sort的棉花黄萎病病情分级

棉花是全球最重要的经济作物之一,而黄萎病是世界主要棉花生产区的第一大病害,黄萎病病原菌通过感染棉花的根部使叶片萎蔫、褪色以致脱落,导致棉花质量和产量严重下降。国家标准将患黄萎病叶片划分为5个等级,传统检测方法主要依赖人工,存在主观、低效、重复性差等问题,因此提出一种以VFNet-Improved、Deep Sort和撞线匹配机制为主要算法框架的棉花黄萎病病情分级方法,实现在旋转视频输入情况下对患病叶片的数量统计和病情等级的划分。研究首先基于VFNet目标检测网络,融合多尺度训练、动态卷积等优化方法,实现对旋转视频中患病叶片的精准定位;然后采用Deep Sort跟踪器实现前后帧同一叶片的相互关联,并针对跟踪过程ID跳变问题设计了掩膜撞线匹配机制;最后使用OpenCV对经过掩膜线的叶片进行特征提取与患病分级的划分。试验结果表明,VFNet-Improved可以有效改善棉花患病叶片识别精度,mAP75达到0.906,较改进前VFNet模型提升了0.012,帧率FPS为12.9帧/s;Deep Sort跟踪器跟踪效果MOTA为0.835,对患病叶片数量统计结果R2、RMSE、MAE与MAPE分别为0.890、5.138、4.300和14.967%,与人工统计值具有较高一致性。本研究为棉花黄萎病病情精准、高效鉴定提供一种新的科学工具,对棉花抗病品种筛选和遗传机制解析具有重要意义。

基于YOLOv5与Deep-SORT的机场跑道侵入告警技术研究

针对传统的跑道侵入告警设备自动化水平低、安装维护成本较高的问题,通过机场视频系统获取机场场面图像信息,采用YOLOv5对机场场面航空器进行检测;使用轻量化网络ShuffleNetv2对Deep-SORT算法进行优化,实现对机场场面航空器的跟踪;根据单目视频采集系统建立坐标转换和测距模型,对机场场面航空器与跑道中线的距离进行准确测量,根据地面保护区设置合适的阈值实现跑道侵入告警。实验结果表明,优化后的模型平均处理时间降低了25.64%,模拟环境下航空器距跑道中心线11、18和43 cm的测距平均误差分别为0.02、0.01和0.01 cm,跑道侵入告警准确率为95.86%,该模型实时性好、准确率高,能够有效预防跑道侵入事件的发生。

Real-time ore sorting using color and texture analysis

Sensor-based ore sorting is a technology used to classify high-grade mineralized rocks from low-grade waste rocks to reduce operation costs.Many ore-sorting algorithms using color images have been proposed in the past,but only some validate their results using mineral grades or optimize the algorithms to classify rocks in real-time.This paper presents an ore-sorting algorithm based on image processing and machine learning that is able to classify rocks from a gold and silver mine based on their grade.The algorithm is composed of four main stages:(1)image segmentation and partition,(2)color and texture feature extraction,(3)sub-image classification using neural networks,and(4)a voting system to determine the overall class of the rock.The algorithm was trained using images of rocks that a geologist manually classified according to their mineral content and then was validated using a different set of rocks analyzed in a laboratory to determine their gold and silver grades.The proposed method achieved a Matthews correlation coefficient of 0.961 points,higher than other classification algorithms based on support vector machines and convolutional neural networks,and a processing time under 44 ms,promising for real-time ore sorting applications.

New event detection based on sorted subtopic matching algorithm

How to quickly and accurately detect new topics from massive data online becomes a main problem of public opinion monitoring in cyberspace. This paperpresents a new event detection method for the current new event detection system, based on sorted subtopic matching algorithm and constructs the entire design framework. In this p^per, the subtopics contained in old topics （or news stories） are sorted in descending order according to their importance to the topic（or news stories）, and form a sorted subtopic sequence. In the process of subtopic matching, subtopic scoring matrix is used to determine whether a new story is reporting a new event. Experimental results show that the sorted subtopic matching model improved the accuracy and effectiveness ofthenew event detection system in cyberspace.