远程塔台由于其低成本高时效远程实时控制技术正越来越受到民航业界的青睐,其中运动目标自动检测和显示是远程塔台的核心技术,作为增强现实技术更好地为管制员提供服务。在分析远程塔台机场场面背景复杂、场面目标多为远场景、小目标等...远程塔台由于其低成本高时效远程实时控制技术正越来越受到民航业界的青睐,其中运动目标自动检测和显示是远程塔台的核心技术,作为增强现实技术更好地为管制员提供服务。在分析远程塔台机场场面背景复杂、场面目标多为远场景、小目标等特点基础上,提出了一种改进的You Only Look Once(YOLO)算法来实现远程塔台运动目标的检测,算法核心思想以Darknet-53为基础网络,多尺度预测边界框,以运动目标图像坐标的偏移量作为边框长宽的线性变换来实现边框的回归,改善了传统YOLO算法损失函数不同大小的边框未做区分的问题,提高了检测准确性和速度。机场真实数据实验表明,该算法能快速、准确的检测出远程塔台的运动目标,并准确的回归运动目标边框及分类。展开更多
The primary objective of this paper is to develop output only modal identification and structural damage detection. Identification of multi-degree of freedom （MDOF） linear time invariant （LTI） and linear time vari...The primary objective of this paper is to develop output only modal identification and structural damage detection. Identification of multi-degree of freedom （MDOF） linear time invariant （LTI） and linear time variant （LTV--due to damage） systems based on Time-frequency （TF） techniques--such as short-time Fourier transform （STFT）, empirical mode decomposition （EMD）, and wavelets--is proposed. STFT, EMD, and wavelet methods developed to date are reviewed in detail. In addition a Hilbert transform （HT） approach to determine frequency and damping is also presented. In this paper, STFT, EMD, HT and wavelet techniques are developed for decomposition of free vibration response of MDOF systems into their modal components. Once the modal components are obtained, each one is processed using Hilbert transform to obtain the modal frequency and damping ratios. In addition, the ratio of modal components at different degrees of freedom facilitate determination of mode shape. In cases with output only modal identification using ambient/random response, the random decrement technique is used to obtain free vibration response. The advantage of TF techniques is that they arc signal based; hence, can be used for output only modal identification. A three degree of freedom 1：10 scale model test structure is used to validate the proposed output only modal identification techniques based on STFT, EMD, HT, wavelets. Both measured free vibration and forced vibration （white noise） response are considered. The secondary objective of this paper is to show the relative ease with which the TF techniques can be used for modal identification and their potential for real world applications where output only identification is essential. Recorded ambient vibration data processed using techniques such as the random decrement technique can be used to obtain the free vibration response, so that further processing using TF based modal identification can be performed.展开更多
The large-sized rammed-earth building foundations on the Panlongcheng site at Huangpi are remains of early Shang period palace complex. The No. 1 Panlongcheng palace consists of four rooms with wooden-framed walls. Th...The large-sized rammed-earth building foundations on the Panlongcheng site at Huangpi are remains of early Shang period palace complex. The No. 1 Panlongcheng palace consists of four rooms with wooden-framed walls. The two rooms in the middle each have two doors on the northern and southern sides respectively；while the two end rooms have only southern doors. So the No. 1 palace must have been in the center of the whole building complex. The roof supported by peripheral columns and wooden-framed walls can be reconstructed to be hipped and single- or double- eaved. The hypothesis that the No. 1 palace may have had projecting-eaves columns has not been confirmed. The No. 2 palace is an open hall without peripheral walls and room division； its roof is supported by peripheral columns only. The idea of reconstructing it as a building with pilasters and multiple rooms seems to lack archaeological evidence. The remaining vestiges show that there were side corridors in the two flanks of the main hall of the No. 2 palace. Referring to the Shang period palace material unearthed from the Shang city-site at Yanshi and other localities, it can be inferred that either of the Nos. 1 and 2 palaces must have had an eastern corridor and a western one, and, in addition, the No. 2 palace must have had a southern corridor with a gate house. The Nos. 1--3 palaces of the Panlongcheng site formed three compounds located one behind another, and belonged to the type of court-and-living building complex. Among them the No. 2 palace was the outer court for holding great ceremonies, the No. 1 palace was the inner court for handling daily administrative affairs, and the No. 3 palace was the king and queen's living place. To the southeast of the No. 2 palace remains a group of rammed-earth house-foundations, which must have been left over from another type of palace building, possibly an ancestral temple. The remaining city-walls at Panlongcheng must have belonged to the peripheral city-walls. The palace area is in the northeast of the encl展开更多
For the detection of marine ship objects in radar images, large-scale networks based on deep learning are difficult to be deployed on existing radar-equipped devices. This paper proposes a lightweight convolutional ne...For the detection of marine ship objects in radar images, large-scale networks based on deep learning are difficult to be deployed on existing radar-equipped devices. This paper proposes a lightweight convolutional neural network, LiraNet, which combines the idea of dense connections, residual connections and group convolution, including stem blocks and extractor modules.The designed stem block uses a series of small convolutions to extract the input image features, and the extractor network adopts the designed two-way dense connection module, which further reduces the network operation complexity. Mounting LiraNet on the object detection framework Darknet, this paper proposes Lira-you only look once(Lira-YOLO), a lightweight model for ship detection in radar images, which can easily be deployed on the mobile devices. Lira-YOLO's prediction module uses a two-layer YOLO prediction layer and adds a residual module for better feature delivery. At the same time, in order to fully verify the performance of the model, mini-RD, a lightweight distance Doppler domain radar images dataset, is constructed. Experiments show that the network complexity of Lira-YOLO is low, being only 2.980 Bflops, and the parameter quantity is smaller, which is only 4.3 MB. The mean average precision(mAP) indicators on the mini-RD and SAR ship detection dataset(SSDD) reach 83.21% and 85.46%, respectively,which is comparable to the tiny-YOLOv3. Lira-YOLO has achieved a good detection accuracy with less memory and computational cost.展开更多
The reform of the housing system in Shanghai has unexpectedly given rise to a self-governed property owners’ collective supervisory system, primarily in the form of Property Owners’ Supervisory Council (POSC), which...The reform of the housing system in Shanghai has unexpectedly given rise to a self-governed property owners’ collective supervisory system, primarily in the form of Property Owners’ Supervisory Council (POSC), which has picked up some of the government’s administrative functions. Although this new, institutionalized management model has theoretically made democratic managerial participation at the grassroots level possible, it has brought about endless problems, overt and covert, shortly after its appearance, some of which are even alarming. A comprehensive analysis of the data collected over a long period of time has led to the conclusion that this system is a failure, attributable to the overt factor related to skills in reality and the covert “priori” factor that is masked by the former. The existence of such “priori” factor once again demonstrates the deep-rooted, traditional managerial logic: Positive operations is society need only to depend upon individuals’ unstable self-disciplined morality rather than to build a system. The current paper points out that any change in the socioeconomic structure that has long been subject to the power of politics is to inevitably incur a corresponding global structural accommodation, including politics itself. To respond to the two factors for the failure, system building in the two overlapping areas is a must.展开更多
译入母语，还是译出母语？对一般译者而言，答案不言而喻。这是因为，翻译界一直以来存在这样一个约定俗成的观念，即“只有译入自己惯常使用的语言，才能确保翻译表达自然、准确，获得最佳翻译效果”（Translating into your language ...译入母语，还是译出母语？对一般译者而言，答案不言而喻。这是因为，翻译界一直以来存在这样一个约定俗成的观念，即“只有译入自己惯常使用的语言，才能确保翻译表达自然、准确，获得最佳翻译效果”（Translating into your language of habitual use is the only way you can translate naturally and accurately and with maximum effectiveness．）（Newmark，2001：3）。展开更多
微动脉瘤是糖尿病视网膜病变的初期症状,消除该病灶可在早期非常有效地预防糖尿病视网膜病变。但由于视网膜结构复杂,同时眼底图像的成像由于患者、环境、采集设备等因素的不同会存在不同的亮度和对比度,现有的微动脉瘤检测算法难以实...微动脉瘤是糖尿病视网膜病变的初期症状,消除该病灶可在早期非常有效地预防糖尿病视网膜病变。但由于视网膜结构复杂,同时眼底图像的成像由于患者、环境、采集设备等因素的不同会存在不同的亮度和对比度,现有的微动脉瘤检测算法难以实现该病灶的精确检测和定位,为此本文提出嵌入SENet(squeeze-andexcitation networks)的改进YOLO(you only look once)v4自动检测算法。该算法在YOLOv4网络基础上,首先通过使用一种改进的快速模糊C均值聚类算法对目标样本进行先验框参数优化,以提高先验框与特征图的匹配度;然后,在主干网络嵌入SENet模块,通过强化关键信息,抑制背景信息,提高微动脉瘤的置信度;此外,还在网络颈部增加空间金字塔池化结构以增强主干网络输出特征的接受域,从而有助于分离出重要的上下文信息;最后,在Kaggle数据集上进行模型验证,并与其他方法进行对比。实验结果表明,与其他各种结构的YOLOv4网络模型相比,所提出的嵌入SENet的改进YOLOv4网络模型能显著提高检测结果(与原始YOLOv4相比Fscore提升了12.68%);与其他网络模型以及方法相比,所提出的嵌入SENet的改进YOLOv4网络模型的自动检测精度明显更优,且可实现精准定位。故本文所提出的嵌入SENet的改进YOLOv4算法性能较优,能准确、有效地检测并定位出眼底图像中的微动脉瘤。展开更多
目标检测是计算机视觉领域的研究热点和基础任务,其中基于锚点(Anchor)的目标检测已在众多领域得到广泛应用。当前锚点选取方法主要面临两个问题:基于特定数据集的先验取值尺寸固定、面对不同场景泛化能力弱。计算锚框的无监督K-means算...目标检测是计算机视觉领域的研究热点和基础任务,其中基于锚点(Anchor)的目标检测已在众多领域得到广泛应用。当前锚点选取方法主要面临两个问题:基于特定数据集的先验取值尺寸固定、面对不同场景泛化能力弱。计算锚框的无监督K-means算法,受初始值影响较大,对目标尺寸较单一的数据集聚类产生的锚点差异较小,无法充分体现网络多尺度输出的特点。针对上述问题,本文提出一种基于多尺度的目标检测锚点构造方法(multi-scale-anchor,MSA),将聚类产生的锚点根据数据集本身的特性进行尺度的缩放和拉伸,优化的锚点即保留原数据集的特点也体现了模型多尺度的优势。另外,本方法应用在训练的预处理阶段,不增加模型推理时间。最后,选取单阶段主流算法YOLO(You Only Look Once),在多个不同场景的红外或工业场景数据集上进行丰富的实验。结果表明,多尺度锚点优化方法MSA能显著提高小样本场景的检测精度。展开更多
针对无人机航拍图像目标检测中视野变化大、时空信息复杂等问题,文中基于YOLOv5(You Only Look Once Version5)架构,提出基于图像低维特征融合的航拍小目标检测模型.引入CA(Coordinate Attention),改进MobileNetV3的反转残差块,增加图...针对无人机航拍图像目标检测中视野变化大、时空信息复杂等问题,文中基于YOLOv5(You Only Look Once Version5)架构,提出基于图像低维特征融合的航拍小目标检测模型.引入CA(Coordinate Attention),改进MobileNetV3的反转残差块,增加图像空间维度信息的同时降低模型参数量.改进YOLOv5特征金字塔网络结构,融合浅层网络中的特征图,增加模型对图像低维有效信息的表达能力,进而提升小目标检测精度.同时为了降低航拍图像中复杂背景带来的干扰,引入无参平均注意力模块,同时关注图像的空间注意力与通道注意力;引入VariFocal Loss,降低负样本在训练过程中的权重占比.在VisDrone数据集上的实验验证文中模型的有效性,该模型在有效提升检测精度的同时明显降低复杂度.展开更多
针对现有混凝土构件裂缝人工检测操作不仅费时、费力,而且易出现错检、误检、漏检,以及部分位置难以开展检测的问题,提出一种基于深度学习YOLOX(You Only Look Once)算法的混凝土构件裂缝智能化检测方法;首先采集、整理包含各类混凝土...针对现有混凝土构件裂缝人工检测操作不仅费时、费力,而且易出现错检、误检、漏检,以及部分位置难以开展检测的问题,提出一种基于深度学习YOLOX(You Only Look Once)算法的混凝土构件裂缝智能化检测方法;首先采集、整理包含各类混凝土构件的典型裂缝图像,并通过图像数据增强建立Pascal VOC数据集,然后基于Facebook公司开发的深度学习框架Pytorch,利用数据集训练YOLOX算法,并进行裂缝识别和验证;将训练完成后YOLOX算法移植至搭载安卓系统的手机端,进行现场实时检测操作。结果表明:在迭代次数为700时,混凝土构件裂缝识别精度可达88.84%,能有效筛分混凝土构件表面裂缝,并排除其他干扰项,证明了所提出的方法对裂缝具有较高的识别精度和广泛的适用性;经试验测试,移植至手机端的YOLOX算法能在提升便携性的同时保证高效、准确的检测效果,具有良好的应用前景。展开更多
Drone or unmanned aerial vehicle(UAV)technology has undergone significant changes.The technology allows UAV to carry out a wide range of tasks with an increasing level of sophistication,since drones can cover a large ...Drone or unmanned aerial vehicle(UAV)technology has undergone significant changes.The technology allows UAV to carry out a wide range of tasks with an increasing level of sophistication,since drones can cover a large area with cameras.Meanwhile,the increasing number of computer vision applications utilizing deep learning provides a unique insight into such applications.The primary target in UAV-based detection applications is humans,yet aerial recordings are not included in the massive datasets used to train object detectors,which makes it necessary to gather the model data from such platforms.You only look once(YOLO)version 4,RetinaNet,faster region-based convolutional neural network(R-CNN),and cascade R-CNN are several well-known detectors that have been studied in the past using a variety of datasets to replicate rescue scenes.Here,we used the search and rescue(SAR)dataset to train the you only look once version 5(YOLOv5)algorithm to validate its speed,accuracy,and low false detection rate.In comparison to YOLOv4 and R-CNN,the highest mean average accuracy of 96.9%is obtained by YOLOv5.For comparison,experimental findings utilizing the SAR and the human rescue imaging database on land(HERIDAL)datasets are presented.The results show that the YOLOv5-based approach is the most successful human detection model for SAR missions.展开更多
With the increasing number of vehicles,manual security inspections are becoming more laborious at road checkpoints.To address it,a specialized Road Checkpoints Robot(RCRo)system is proposed,incorporated with enhanced ...With the increasing number of vehicles,manual security inspections are becoming more laborious at road checkpoints.To address it,a specialized Road Checkpoints Robot(RCRo)system is proposed,incorporated with enhanced You Only Look Once(YOLO)and a 6-degree-of-freedom(DOF)manipulator,for autonomous identity verification and vehicle inspection.The modified YOLO is characterized by large objects’sensitivity and faster detection speed,named“LF-YOLO”.The better sensitivity of large objects and the faster detection speed are achieved by means of the Dense module-based backbone network connecting two-scale detecting network,for object detection tasks,along with optimized anchor boxes and improved loss function.During the manipulator motion,Octree-aided motion control scheme is adopted for collision-free motion through Robot Operating System(ROS).The proposed LF-YOLO which utilizes continuous optimization strategy and residual technique provides a promising detector design,which has been found to be more effective during actual object detection,in terms of decreased average detection time by 68.25%and 60.60%,and increased average Intersection over Union(Io U)by 20.74%and6.79%compared to YOLOv3 and YOLOv4 through experiments.The comprehensive functional tests of RCRo system demonstrate the feasibility and competency of the multiple unmanned inspections in practice.展开更多
针对部署于有限算力平台的YOLOv3(you only look once v3)算法对电容器外观缺陷存在检测速度较慢的问题,提出了基于YOLOv3算法改进的轻量化算法MQYOLOv3。首先采用轻量化网络MobileNet v2作为特征提取模块,通过利用深度可分离式卷积替...针对部署于有限算力平台的YOLOv3(you only look once v3)算法对电容器外观缺陷存在检测速度较慢的问题,提出了基于YOLOv3算法改进的轻量化算法MQYOLOv3。首先采用轻量化网络MobileNet v2作为特征提取模块,通过利用深度可分离式卷积替换一般卷积操作,使得模型的参数量大幅度降低进而提高模型的检测速度,同时也带来了检测精度的降低;然后在网络结构中嵌入空间金字塔池化结构实现局部特征与全局特征的融合、引入距离交并比(distance intersection over union,DIoU)损失函数优化交并比(intersection over union,IoU)损失函数以及使用Mish激活函数优化Leaky ReLU激活函数来提高模型的检测精度。本文采用自制的电容器外观缺陷数据集进行实验,轻量化MQYOLOv3算法的平均精度均值(mean average precision,mAP)为87.96%,较优化前降低了1.16%,检测速度从1.5 FPS提升到7.7 FPS。实验表明,本文设计的轻量化MQYOLOv3算法在保证检测精度的同时,提高了检测速度。展开更多
BACKGROUND Studies have revealed that Children's psychological,behavioral,and emotional problems are easily influenced by the family environment.In recent years,the family structure in China has undergone signific...BACKGROUND Studies have revealed that Children's psychological,behavioral,and emotional problems are easily influenced by the family environment.In recent years,the family structure in China has undergone significant changes,with more families having two or three children.AIM To explore the relationship between emotional behavior and parental job stress in only preschool and non-only preschool children.METHODS Children aged 3-6 in kindergartens in four main urban areas of Shijiazhuang were selected by stratified sampling for a questionnaire and divided into only and nononly child groups.Their emotional behaviors and parental pressure were compared.Only and non-only children were paired in a 1:1 ratio by class and age(difference less than or equal to 6 months),and the matched data were compared.The relationship between children's emotional behavior and parents'job stress before and after matching was analyzed.RESULTS Before matching,the mother's occupation,children's personality characteristics,and children's rearing patterns differed between the groups(P<0.05).After matching 550 pairs,differences in the children's parenting styles remained.There were significant differences in children's gender and parents'attitudes toward children between the two groups.The Strengths and Difficulties Questionnaire(SDQ)scores of children in the only child group and the Parenting Stress Index-Short Form(PSI-SF)scores of parents were significantly lower than those in the non-only child group(P<0.05).Pearson’s correlation analysis showed that after matching,there was a positive correlation between children's parenting style and parents'attitudes toward their children(r=0.096,P<0.01),and the PSI-SF score was positively correlated with children's gender,parents'attitudes toward their children,and SDQ scores(r=0.077,0.193,0.172,0.222).CONCLUSION Preschool children's emotional behavior and parental pressure were significantly higher in multi-child families.Parental pressure in differently structured families was associated with many f展开更多
In the background of f(R,L_(m))gravity,this work investigates three distinct dark matter halo profiles to test the possibility of generalised wormhole geometry within the galactic halo regions.The current study aims t...In the background of f(R,L_(m))gravity,this work investigates three distinct dark matter halo profiles to test the possibility of generalised wormhole geometry within the galactic halo regions.The current study aims to accomplish these goals by examining various dark matter profiles including universal rotation curves(URC),Navarro-Frenk-White(NFW)model-Ⅰ,and NFW model-Ⅱinside two distinct f(R,L_(m))gravity models.According to the f(R,L_(m))=R/2+L^(a)_(m)model,the dark matter(DM)halo density profiles produce suitable shape functions that meet all the necessary requirements for exhibiting the wormhole geometries with appropriate choice of free parameters.In addition,to examine DM profiles under the f(R,L_(m))=R/2+(1+λR)L_(m) model,we consider a specific shape function.Further,we observed that the derived solution from both two models violates the null energy constraints,confirming that the DM supports wormholes to maintain in the galactic halo.展开更多
文摘远程塔台由于其低成本高时效远程实时控制技术正越来越受到民航业界的青睐,其中运动目标自动检测和显示是远程塔台的核心技术,作为增强现实技术更好地为管制员提供服务。在分析远程塔台机场场面背景复杂、场面目标多为远场景、小目标等特点基础上,提出了一种改进的You Only Look Once(YOLO)算法来实现远程塔台运动目标的检测,算法核心思想以Darknet-53为基础网络,多尺度预测边界框,以运动目标图像坐标的偏移量作为边框长宽的线性变换来实现边框的回归,改善了传统YOLO算法损失函数不同大小的边框未做区分的问题,提高了检测准确性和速度。机场真实数据实验表明,该算法能快速、准确的检测出远程塔台的运动目标,并准确的回归运动目标边框及分类。
基金National Science Foundation Grant NSF CMS CAREER Under Grant No.9996290NSF CMMI Under Grant No.0830391
文摘The primary objective of this paper is to develop output only modal identification and structural damage detection. Identification of multi-degree of freedom （MDOF） linear time invariant （LTI） and linear time variant （LTV--due to damage） systems based on Time-frequency （TF） techniques--such as short-time Fourier transform （STFT）, empirical mode decomposition （EMD）, and wavelets--is proposed. STFT, EMD, and wavelet methods developed to date are reviewed in detail. In addition a Hilbert transform （HT） approach to determine frequency and damping is also presented. In this paper, STFT, EMD, HT and wavelet techniques are developed for decomposition of free vibration response of MDOF systems into their modal components. Once the modal components are obtained, each one is processed using Hilbert transform to obtain the modal frequency and damping ratios. In addition, the ratio of modal components at different degrees of freedom facilitate determination of mode shape. In cases with output only modal identification using ambient/random response, the random decrement technique is used to obtain free vibration response. The advantage of TF techniques is that they arc signal based; hence, can be used for output only modal identification. A three degree of freedom 1：10 scale model test structure is used to validate the proposed output only modal identification techniques based on STFT, EMD, HT, wavelets. Both measured free vibration and forced vibration （white noise） response are considered. The secondary objective of this paper is to show the relative ease with which the TF techniques can be used for modal identification and their potential for real world applications where output only identification is essential. Recorded ambient vibration data processed using techniques such as the random decrement technique can be used to obtain the free vibration response, so that further processing using TF based modal identification can be performed.
文摘The large-sized rammed-earth building foundations on the Panlongcheng site at Huangpi are remains of early Shang period palace complex. The No. 1 Panlongcheng palace consists of four rooms with wooden-framed walls. The two rooms in the middle each have two doors on the northern and southern sides respectively；while the two end rooms have only southern doors. So the No. 1 palace must have been in the center of the whole building complex. The roof supported by peripheral columns and wooden-framed walls can be reconstructed to be hipped and single- or double- eaved. The hypothesis that the No. 1 palace may have had projecting-eaves columns has not been confirmed. The No. 2 palace is an open hall without peripheral walls and room division； its roof is supported by peripheral columns only. The idea of reconstructing it as a building with pilasters and multiple rooms seems to lack archaeological evidence. The remaining vestiges show that there were side corridors in the two flanks of the main hall of the No. 2 palace. Referring to the Shang period palace material unearthed from the Shang city-site at Yanshi and other localities, it can be inferred that either of the Nos. 1 and 2 palaces must have had an eastern corridor and a western one, and, in addition, the No. 2 palace must have had a southern corridor with a gate house. The Nos. 1--3 palaces of the Panlongcheng site formed three compounds located one behind another, and belonged to the type of court-and-living building complex. Among them the No. 2 palace was the outer court for holding great ceremonies, the No. 1 palace was the inner court for handling daily administrative affairs, and the No. 3 palace was the king and queen's living place. To the southeast of the No. 2 palace remains a group of rammed-earth house-foundations, which must have been left over from another type of palace building, possibly an ancestral temple. The remaining city-walls at Panlongcheng must have belonged to the peripheral city-walls. The palace area is in the northeast of the encl
基金supported by the Joint Fund of Equipment Pre-Research and Aerospace Science and Industry (6141B07090102)。
文摘For the detection of marine ship objects in radar images, large-scale networks based on deep learning are difficult to be deployed on existing radar-equipped devices. This paper proposes a lightweight convolutional neural network, LiraNet, which combines the idea of dense connections, residual connections and group convolution, including stem blocks and extractor modules.The designed stem block uses a series of small convolutions to extract the input image features, and the extractor network adopts the designed two-way dense connection module, which further reduces the network operation complexity. Mounting LiraNet on the object detection framework Darknet, this paper proposes Lira-you only look once(Lira-YOLO), a lightweight model for ship detection in radar images, which can easily be deployed on the mobile devices. Lira-YOLO's prediction module uses a two-layer YOLO prediction layer and adds a residual module for better feature delivery. At the same time, in order to fully verify the performance of the model, mini-RD, a lightweight distance Doppler domain radar images dataset, is constructed. Experiments show that the network complexity of Lira-YOLO is low, being only 2.980 Bflops, and the parameter quantity is smaller, which is only 4.3 MB. The mean average precision(mAP) indicators on the mini-RD and SAR ship detection dataset(SSDD) reach 83.21% and 85.46%, respectively,which is comparable to the tiny-YOLOv3. Lira-YOLO has achieved a good detection accuracy with less memory and computational cost.
文摘The reform of the housing system in Shanghai has unexpectedly given rise to a self-governed property owners’ collective supervisory system, primarily in the form of Property Owners’ Supervisory Council (POSC), which has picked up some of the government’s administrative functions. Although this new, institutionalized management model has theoretically made democratic managerial participation at the grassroots level possible, it has brought about endless problems, overt and covert, shortly after its appearance, some of which are even alarming. A comprehensive analysis of the data collected over a long period of time has led to the conclusion that this system is a failure, attributable to the overt factor related to skills in reality and the covert “priori” factor that is masked by the former. The existence of such “priori” factor once again demonstrates the deep-rooted, traditional managerial logic: Positive operations is society need only to depend upon individuals’ unstable self-disciplined morality rather than to build a system. The current paper points out that any change in the socioeconomic structure that has long been subject to the power of politics is to inevitably incur a corresponding global structural accommodation, including politics itself. To respond to the two factors for the failure, system building in the two overlapping areas is a must.
文摘译入母语，还是译出母语？对一般译者而言，答案不言而喻。这是因为，翻译界一直以来存在这样一个约定俗成的观念，即“只有译入自己惯常使用的语言，才能确保翻译表达自然、准确，获得最佳翻译效果”（Translating into your language of habitual use is the only way you can translate naturally and accurately and with maximum effectiveness．）（Newmark，2001：3）。
文摘微动脉瘤是糖尿病视网膜病变的初期症状,消除该病灶可在早期非常有效地预防糖尿病视网膜病变。但由于视网膜结构复杂,同时眼底图像的成像由于患者、环境、采集设备等因素的不同会存在不同的亮度和对比度,现有的微动脉瘤检测算法难以实现该病灶的精确检测和定位,为此本文提出嵌入SENet(squeeze-andexcitation networks)的改进YOLO(you only look once)v4自动检测算法。该算法在YOLOv4网络基础上,首先通过使用一种改进的快速模糊C均值聚类算法对目标样本进行先验框参数优化,以提高先验框与特征图的匹配度;然后,在主干网络嵌入SENet模块,通过强化关键信息,抑制背景信息,提高微动脉瘤的置信度;此外,还在网络颈部增加空间金字塔池化结构以增强主干网络输出特征的接受域,从而有助于分离出重要的上下文信息;最后,在Kaggle数据集上进行模型验证,并与其他方法进行对比。实验结果表明,与其他各种结构的YOLOv4网络模型相比,所提出的嵌入SENet的改进YOLOv4网络模型能显著提高检测结果(与原始YOLOv4相比Fscore提升了12.68%);与其他网络模型以及方法相比,所提出的嵌入SENet的改进YOLOv4网络模型的自动检测精度明显更优,且可实现精准定位。故本文所提出的嵌入SENet的改进YOLOv4算法性能较优,能准确、有效地检测并定位出眼底图像中的微动脉瘤。
文摘目标检测是计算机视觉领域的研究热点和基础任务,其中基于锚点(Anchor)的目标检测已在众多领域得到广泛应用。当前锚点选取方法主要面临两个问题:基于特定数据集的先验取值尺寸固定、面对不同场景泛化能力弱。计算锚框的无监督K-means算法,受初始值影响较大,对目标尺寸较单一的数据集聚类产生的锚点差异较小,无法充分体现网络多尺度输出的特点。针对上述问题,本文提出一种基于多尺度的目标检测锚点构造方法(multi-scale-anchor,MSA),将聚类产生的锚点根据数据集本身的特性进行尺度的缩放和拉伸,优化的锚点即保留原数据集的特点也体现了模型多尺度的优势。另外,本方法应用在训练的预处理阶段,不增加模型推理时间。最后,选取单阶段主流算法YOLO(You Only Look Once),在多个不同场景的红外或工业场景数据集上进行丰富的实验。结果表明,多尺度锚点优化方法MSA能显著提高小样本场景的检测精度。
文摘针对无人机航拍图像目标检测中视野变化大、时空信息复杂等问题,文中基于YOLOv5(You Only Look Once Version5)架构,提出基于图像低维特征融合的航拍小目标检测模型.引入CA(Coordinate Attention),改进MobileNetV3的反转残差块,增加图像空间维度信息的同时降低模型参数量.改进YOLOv5特征金字塔网络结构,融合浅层网络中的特征图,增加模型对图像低维有效信息的表达能力,进而提升小目标检测精度.同时为了降低航拍图像中复杂背景带来的干扰,引入无参平均注意力模块,同时关注图像的空间注意力与通道注意力;引入VariFocal Loss,降低负样本在训练过程中的权重占比.在VisDrone数据集上的实验验证文中模型的有效性,该模型在有效提升检测精度的同时明显降低复杂度.
文摘针对现有混凝土构件裂缝人工检测操作不仅费时、费力,而且易出现错检、误检、漏检,以及部分位置难以开展检测的问题,提出一种基于深度学习YOLOX(You Only Look Once)算法的混凝土构件裂缝智能化检测方法;首先采集、整理包含各类混凝土构件的典型裂缝图像,并通过图像数据增强建立Pascal VOC数据集,然后基于Facebook公司开发的深度学习框架Pytorch,利用数据集训练YOLOX算法,并进行裂缝识别和验证;将训练完成后YOLOX算法移植至搭载安卓系统的手机端,进行现场实时检测操作。结果表明:在迭代次数为700时,混凝土构件裂缝识别精度可达88.84%,能有效筛分混凝土构件表面裂缝,并排除其他干扰项,证明了所提出的方法对裂缝具有较高的识别精度和广泛的适用性;经试验测试,移植至手机端的YOLOX算法能在提升便携性的同时保证高效、准确的检测效果,具有良好的应用前景。
文摘Drone or unmanned aerial vehicle(UAV)technology has undergone significant changes.The technology allows UAV to carry out a wide range of tasks with an increasing level of sophistication,since drones can cover a large area with cameras.Meanwhile,the increasing number of computer vision applications utilizing deep learning provides a unique insight into such applications.The primary target in UAV-based detection applications is humans,yet aerial recordings are not included in the massive datasets used to train object detectors,which makes it necessary to gather the model data from such platforms.You only look once(YOLO)version 4,RetinaNet,faster region-based convolutional neural network(R-CNN),and cascade R-CNN are several well-known detectors that have been studied in the past using a variety of datasets to replicate rescue scenes.Here,we used the search and rescue(SAR)dataset to train the you only look once version 5(YOLOv5)algorithm to validate its speed,accuracy,and low false detection rate.In comparison to YOLOv4 and R-CNN,the highest mean average accuracy of 96.9%is obtained by YOLOv5.For comparison,experimental findings utilizing the SAR and the human rescue imaging database on land(HERIDAL)datasets are presented.The results show that the YOLOv5-based approach is the most successful human detection model for SAR missions.
基金supported by the National Key Research and Development Program of China(grant number:2017YFC0806503)。
文摘With the increasing number of vehicles,manual security inspections are becoming more laborious at road checkpoints.To address it,a specialized Road Checkpoints Robot(RCRo)system is proposed,incorporated with enhanced You Only Look Once(YOLO)and a 6-degree-of-freedom(DOF)manipulator,for autonomous identity verification and vehicle inspection.The modified YOLO is characterized by large objects’sensitivity and faster detection speed,named“LF-YOLO”.The better sensitivity of large objects and the faster detection speed are achieved by means of the Dense module-based backbone network connecting two-scale detecting network,for object detection tasks,along with optimized anchor boxes and improved loss function.During the manipulator motion,Octree-aided motion control scheme is adopted for collision-free motion through Robot Operating System(ROS).The proposed LF-YOLO which utilizes continuous optimization strategy and residual technique provides a promising detector design,which has been found to be more effective during actual object detection,in terms of decreased average detection time by 68.25%and 60.60%,and increased average Intersection over Union(Io U)by 20.74%and6.79%compared to YOLOv3 and YOLOv4 through experiments.The comprehensive functional tests of RCRo system demonstrate the feasibility and competency of the multiple unmanned inspections in practice.
文摘针对部署于有限算力平台的YOLOv3(you only look once v3)算法对电容器外观缺陷存在检测速度较慢的问题,提出了基于YOLOv3算法改进的轻量化算法MQYOLOv3。首先采用轻量化网络MobileNet v2作为特征提取模块,通过利用深度可分离式卷积替换一般卷积操作,使得模型的参数量大幅度降低进而提高模型的检测速度,同时也带来了检测精度的降低;然后在网络结构中嵌入空间金字塔池化结构实现局部特征与全局特征的融合、引入距离交并比(distance intersection over union,DIoU)损失函数优化交并比(intersection over union,IoU)损失函数以及使用Mish激活函数优化Leaky ReLU激活函数来提高模型的检测精度。本文采用自制的电容器外观缺陷数据集进行实验,轻量化MQYOLOv3算法的平均精度均值(mean average precision,mAP)为87.96%,较优化前降低了1.16%,检测速度从1.5 FPS提升到7.7 FPS。实验表明,本文设计的轻量化MQYOLOv3算法在保证检测精度的同时,提高了检测速度。
基金Shijiazhuang City Science and Technology Research and Development Self Raised Plan,No.221460383。
文摘BACKGROUND Studies have revealed that Children's psychological,behavioral,and emotional problems are easily influenced by the family environment.In recent years,the family structure in China has undergone significant changes,with more families having two or three children.AIM To explore the relationship between emotional behavior and parental job stress in only preschool and non-only preschool children.METHODS Children aged 3-6 in kindergartens in four main urban areas of Shijiazhuang were selected by stratified sampling for a questionnaire and divided into only and nononly child groups.Their emotional behaviors and parental pressure were compared.Only and non-only children were paired in a 1:1 ratio by class and age(difference less than or equal to 6 months),and the matched data were compared.The relationship between children's emotional behavior and parents'job stress before and after matching was analyzed.RESULTS Before matching,the mother's occupation,children's personality characteristics,and children's rearing patterns differed between the groups(P<0.05).After matching 550 pairs,differences in the children's parenting styles remained.There were significant differences in children's gender and parents'attitudes toward children between the two groups.The Strengths and Difficulties Questionnaire(SDQ)scores of children in the only child group and the Parenting Stress Index-Short Form(PSI-SF)scores of parents were significantly lower than those in the non-only child group(P<0.05).Pearson’s correlation analysis showed that after matching,there was a positive correlation between children's parenting style and parents'attitudes toward their children(r=0.096,P<0.01),and the PSI-SF score was positively correlated with children's gender,parents'attitudes toward their children,and SDQ scores(r=0.077,0.193,0.172,0.222).CONCLUSION Preschool children's emotional behavior and parental pressure were significantly higher in multi-child families.Parental pressure in differently structured families was associated with many f
基金University Grant Commission(UGC),Govt.of India,New Delhi,for awarding JRF(NTA Ref.No.:191620024300)University Grants Commission(UGC),New Delhi,India,for awarding National Fellowship for Scheduled Caste Students(UGC-Ref.No.:201610123801)+1 种基金PKS acknowledges the National Board for Higher Mathematics(NBHM)under the Department of Atomic Energy(DAE),Govt.of India,for financial support to carry out the Research project No.:02011/3/2022 NBHM(R.P.)/R&D II/2152 Dt.14.02.2022IUCAA,Pune,India for providing support through the visiting Associateship program.
文摘In the background of f(R,L_(m))gravity,this work investigates three distinct dark matter halo profiles to test the possibility of generalised wormhole geometry within the galactic halo regions.The current study aims to accomplish these goals by examining various dark matter profiles including universal rotation curves(URC),Navarro-Frenk-White(NFW)model-Ⅰ,and NFW model-Ⅱinside two distinct f(R,L_(m))gravity models.According to the f(R,L_(m))=R/2+L^(a)_(m)model,the dark matter(DM)halo density profiles produce suitable shape functions that meet all the necessary requirements for exhibiting the wormhole geometries with appropriate choice of free parameters.In addition,to examine DM profiles under the f(R,L_(m))=R/2+(1+λR)L_(m) model,we consider a specific shape function.Further,we observed that the derived solution from both two models violates the null energy constraints,confirming that the DM supports wormholes to maintain in the galactic halo.
