基于注意力机制与多尺度特征融合的行人重识别方法

针对行人重识别中因遮挡、姿态变化使模型特征无法充分表达行人信息的问题,提出了基于注意力机制与多尺度特征融合的行人重识别方法。该方法首先使用改进的骨干网络（R-ResNet50）提取图像特征;其次,抽取网络不同尺度的特征层嵌入注意力机制（DANet）,使模型更关注于重点信息;最后,对提取出的关键特征进行多尺度特征融合,实现特征间的优势互补,并使用联合交叉熵损失、难样本采样三元组损失和中心损失的多损失函数策略对网络模型进行训练。实验结果表明,本文所提方法在Market1501、DukeMTMC-ReID数据集上的首位命中率（Rank-1）和平均精度均值（mAP）分别达到了92.7%、80.4%和86.4%、71.0%,模型提取的特征更具有判别性,识别率更高。     

面向智能船舶的水面小目标检测算法

针对智能船舶中基于视觉传感器的水面小目标识别具有识别区域分辨率低、图像模糊、信噪比低等问题,提出了一种新的基于卷积神经网络的水面小目标检测算法——自注意力特征融合检测算法.首先,为了提高视觉信息处理的效率与准确性,在网络模型中引入了自注意力模块,更多关注小目标的细节信息.其次,在网络模型中采用了结构化的特征融合算法,通过多尺度语义信息融合提升对小目标的检测性能.最后,为了解决目标检测的定位问题,在smooth L1损失函数的基础上设计了一种大梯度定位损失函数.通过与传统的Faster R-CNN目标检测算法在船舶数据库上进行仿真对比,验证了所提算法在解决水面小目标检测问题上的有效性.     

基于多尺度特征融合的单目图像深度估计

为解决从单目图像中很难恢复出准确、有效深度信息的问题,提出一种多尺度特征融合的单目图像深度估计算法.算法采用端对端训练的卷积神经网络(CNN)结构,引入从图像编码器到解码器的跳层连接来实现在不同尺度上特征的提取和表达,设计了一种多尺度的损失函数来提升卷积神经网络的训练效果.通过在NYU Depth V2室内场景深度数据集和KITTI室外场景深度数据集上的训练、验证和测试,实验结果表明:提出的多尺度特征融合方法得到的深度图边缘清晰、层次分明,且在室内场景和室外场景中均能适用,具有较强的泛化性,可以适应多种实际场景的需求.     

基于注意力特征融合的SqueezeNet细粒度图像分类模型

针对现有细粒度图像分类算法普遍存在的模型结构复杂、参数多、分类准确率较低等问题,提出一种注意力特征融合的SqueezeNet细粒度图像分类模型.通过对现有细粒度图像分类算法和轻量级卷积神经网络的分析,首先使用3个典型的预训练轻量级卷积神经网络,对其微调后在公开的细粒度图像数据集上进行验证,经比较后选择了模型性能最佳的SqueezeNet作为图像的特征提取器;然后将两个具有注意力机制的卷积模块嵌入至SqueezeNet网络的每个Fire模块;接着提取出改进后的SqueezeNet的中间层特征进行双线性融合形成新的注意力特征图,与网络的全局特征再融合后分类;最后通过实验对比和可视化分析,网络嵌入Convolution Block Attention Module(CBAM)模块的分类准确率在鸟类、汽车、飞机数据集上依次提高了8.96%、4.89%和5.85%,嵌入Squeeze-and-Excitation(SE)模块的分类准确率依次提高了9.81%、4.52%和2.30%,且新模型在参数量、运行效率等方面比现有算法更具优势.     

基于YOLOv3的改进行人检测算法研究

针对将YOLOv3通用目标检测算法应用于行人检测时的检测精度低、定位不准确的问题,提出了一种基于YOLOv3的适用于行人体态特征的目标检测算法。在预处理生成先验框部分,将MSCOCO通用数据集改进为MSCOCO中的person子集来生成仅针对行人体态特征的锚框,并将生成先验框的K-means算法改进为K-means++算法以弥补K-means算法选择中心点的随意性。针对特征提取,改进了卷积神经网络(CNN) Darknet-53的结构以使其能够提取到更多与行人尺度有关的特征信息。损失函数中的目标定位部分改进为更加符合行人体态特征的损失函数。实验使用MSCOCO训练集中的person子集训练模型,分别使用MSCOCO测试集的person子集和自制的行人数据集作为测试集验证模型。结果显示,对比YOLOv3的通用目标检测算法,该文改进算法更易于提取行人特征并提供有效反馈,提高了检测精度。     

基于多层次深度学习网络的行人重识别

提出了多层级特征融合模型,该模型利用深度学习网络提取行人图像的全局特征和局部特征,并将全局和局部特征联合起来,以生成更具辨识度的描述符.在模型中,基于部分的多层级网络用于提取不同网络深度的局部特征,从而将网络底层到高层中提取的局部特征组合起来.全局—局部网络分支则提取网络深层的局部特征和全局特征,用于识别行人.该模型在三个数据集上进行了实验并得到了更好的结果.     

结合注意力机制与时空特征融合的长时程行为识别方法

为了建立基于视频行为识别的长时程图像序列的时空信息模型,文章提出了一种嵌入注意力的时空特征融合网络(attention-embedded spatial-temporal feature fusion network, ASTFFN)的深度神经网络模型。ASTFFN将一个包含动作的长时程图像序列分割成多个重叠的片段,并使用一个注意力嵌入特征提取网络(attention-embedded feature extraction network, AFEN)从每个片段中的RGB图像或光流图像中提取经过注意力加权的时空特征,进而融合每个片段的时空加权特征,生成行为识别的视频级预测。基于UCF101和HMDB51行为识别数据库进行了大量的实验,验证了方法的有效性。与目前主流的行为识别算法相比,该文方法在识别精度上取得了较好的效果。     

基于双层视觉及多尺度注意力融合的图像去雾

针对现有去雾算法缺乏对雾霾图像不同区域噪音浓度的关注以及远近景特征的区分问题,本文提出了一种新的生成对抗网络模型.模型中通过两个UNet3+网络实现全尺度的跳跃连接和深度监督,使用多尺度融合的方法结合不同尺度特征图中的高低级语义;而深度监督的加入可以更好地学习图像中的远近层次表示.同时在生成器结构中加入融合改进自注意力机制的多尺度金字塔特征融合模块,以便更好地保留特征图的多尺度结构信息,并且提高了对不同雾霾浓度区域的关注度.实验结果显示,在NTIRE 2020、NTIRE 2021、O-Haze数据集和Dense-Haze数据集上, 本文所提出的算法网络相比BPPNET等其他先进算法可以得到更好的视觉效果,在Dense-Haze数据集上,峰值信噪比和结构相似性指数分别达到24.82和0.769.     

基于多模态融合的深度神经网络图像复原方法

针对水下图像成像环境复杂常受偏色等因素干扰而影响后续图像分析的问题,提出一种基于多尺度特征与三重注意力多模态融合的深度卷积神经网络图像复原方法.首先,深度卷积神经网络在抽取图像空间特征的基础上,引入图像多尺度变换特征;其次,通过通道注意力、监督注意力和非局部注意力,挖掘图像特征的尺度间相关性、特征间相关性;最后,通过设计多模态特征融合机制,将上述两类特征有效融合.在公开的水下图像测试集上进行测试并与当前主流方法进行对比的实验结果表明,该方法在峰值信噪比、结构相似性等定量对比以及颜色、细节等定性对比上都优于对比方法.     

基于Transformer改进的两分支行人重识别算法

针对基于卷积神经网络的行人重识别算法全局信息建模不足的问题，分析了卷积操作的局限性，提出一种基于Transformer改进的全局-局部两分支行人重识别算法.首先利用相对位置编码改进多头自注意力机制，并将其嵌入到Resnet50骨干网络中.之后在全局分支中对图像进行空间几何划分并利用Transformer的全局感受野增强抽象特征的提取能力；在局部分支中对Layer＿3输出进行降维监督，利用多尺度池化获得更丰富的局部特征.实验结果表明，该算法在公开数据集Market-1501和DukeMTMC-reID上的mAP/Rank-1分别达到了93.45%/95.61%和88.79%/90.35%,相对于单纯基于卷积神经网络的算法，本文算法达到更高的精度.     

An Index mechanism for multi-scale view in spatial databases

There are numerous geometric objects stored in the spatial databases. An importance function in a spatial database is that users can browse the geometric objects as a map efficiently. Thus the spatial database should display the geometric objects users concern about swiftly onto the display window. This process includes two operations:retrieve data from database and then draw them onto screen. Accordingly, to improve the efficiency, we should try to reduce time of both retrieving object and displaying them. The former can be achieved with the aid of spatial index such as R-tree, the latter require to simplify the objects. Simplification means that objects are shown with sufficient but not with unnecessary detail which depend on the scale of browse. So the major problem is how to retrieve data at different detail level efficiently. This paper introduces the implementation of a multi-scale index in the spatial database SISP (Spatial Information Shared Platform) which is generalized from R-tree. The difference between the generalization and the R-tree lies on two facets: One is that every node and geometric object in the generalization is assigned with a importance value which denote the importance of them, and every vertex in the objects are assigned with a importance value,too. The importance value can be use to decide which data should be retrieve from disk in a query. The other difference is that geometric objects in the generalization are divided into one or more sub-blocks, and vertexes are total ordered by their importance value. With the help of the generalized R-tree, one can easily retrieve data at different detail levels.Some experiments are performed on real-life data to evaluate the performance of solutions that separately use normal spatial index and multi-scale spatial index. The results show that the solution using multi-scale index in SISP is satisfying.     

Characterization of Ordovician carbonate reservoirs, southeastern Saskatchewan, Canada

The discovery of the prolific Ordovician Red River reservoirs in 1995 in southeastern Saskatchewan was the catalyst for extensive exploration activity which resulted in the discovery of more than 15 new Red River pools. The best yields of Red River production to date have been from dolomite reservoirs. Understanding the processes of dolomitization is, therefore, crucial for the prediction of the connectivity, spatial distribution and heterogeneity of dolomite reservoirs.The Red River reservoirs in the Midale area consist of 3～4 thin dolomitized zones, with a total thickness of about 20 m, which occur at the top of the Yeoman Formation. Two types of replacement dolomite were recognized in the Red River reservoir: dolomitized burrow infills and dolomitized host matrix. The spatial distribution of dolomite suggests that burrowing organisms played an important role in facilitating the fluid flow in the backfilled sediments. This resulted in penecontemporaneous dolomitization of burrow infills by normal seawater. The dolomite in the host matrix is interpreted as having occurred at shallow burial by evaporitic seawater during precipitation of Lake Almar anhydrite that immediately overlies the Yeoman Formation. However, the low δ18O values of dolomited burrow infills (-5.9‰～ -7.8‰, PDB) and matrix dolomites (-6.6‰～ -8.1‰, avg. -7.4‰ PDB) compared to the estimated values for the late Ordovician marine dolomite could be attributed to modification and alteration of dolomite at higher temperatures during deeper burial, which could also be responsible for its 87Sr/86Sr ratios (0.7084～0.7088) that are higher than suggested for the late Ordovician seawaters (0.7078～0.7080). The trace amounts of saddle dolomite cement in the Red River carbonates are probably related to "cannibalization" of earlier replacement dolomite during the chemical compaction.     

A computer generator for randomly layered structures

AcomputergeneratorforrandomlylayeredstructuresYUJia shun1,2,HEZhen hua2(1.TheInstituteofGeologicalandNuclearSciences,NewZealand;2.StateKeyLaboratoryofOilandGasReservoirGeologyandExploitation,ChengduUniversityofTechnology,China)Abstract:Analgorithmisintrod…     

Perspherical Approaches for Electrons in Atoms and Charged Excitons in Quantum Wells

Instead of following Fock’s expansion,we solve the Schrodinger equation for some quantum mechanical manybody systems such as electrons in atoms and charged excitons in quantum wells in a similar way in hyperspherical coordinates by expanding the wave functions into orthonormal complete basis sets of the hyperspherical hannonics(HHs)of hyperangles and generalized Laguerre polynomials(GLPs)of the hyperradius.This leads the equation to     

Advances in broadband wireless communications under high-mobility scenarios

Future mobile communication systems aim at providing very high data transmission rates, even in high-mobility scenarios such as high-speed wheel-track trains, maglev trains, highway vehicles, airplanes, guided missiles or spacecraft. A particularly important commercial application is the strong and increasing worldwide demand for high- speed broadband wireless communications （up to 574.8 km/ h test speeds or 380 km/h commercial speeds） in railways, providing data, voice and video services for applications such as onboard entertainment services to passengers, train control, train dispatch, train sensor status handling and sur- veillance. In such high-mobility scenarios, there are a number of communication challenges, including fast hand- over, location updating, high-speed channel modeling, estimation and equalization, anti-Doppler spreading tech- niques, fast power control, and dedicated network architec- ture. Because signal transmission in very high-speed scenarios will inevitably experience serious deterioration, it is imperative to develop key broadband mobile communi- cation techniques for such high-speed vehicles.     

Wuhan University Journal of Natural Sciences TOTAL CONTENTS

<正>~~     

Scale-based spatial data model for GIS

Being the primary media of geographical information and the elementary objects manipulated, almost all of maps adopt the layer-based model to represent geographic information in the existent GIS. However, it is difficult to extend the map represented in layer-based model. Furthermore, in Web-Based GIS, It is slow to transmit the spatial data for map viewing. In this paper, for solving the questions above, we have proposed a new method for representing the spatial data. That is scale-based model. In this model we represent maps in three levels: scale-view, block, and spatial object, and organize the maps in a set of map layers, named Scale-View, which associates some given scales.Lastly, a prototype Web-Based GIS using the proposed spatial data representation is described briefly.     

海南岛地体及其毗邻陆缘晚中生代—新生代古地磁研究和构造演化

本文叙述了对海南岛及其毗邻大陆边缘白垩纪到第四纪地层岩石进行古地磁研究的全部工作过程。通过分析岩石中剩余磁矢量的磁偏角及磁倾角的变化,提出海南岛白垩纪以来经历的构造演化模式如下:早期伴随顺时针旋转而向南迁移,后期伴随逆时针转动并向北运移。联系该地区及邻区的地质、地球物理资料,对海南岛上述的构造地体运动提出以下认识:北部湾内早期有一拉张作用,主要是该作用使湾内地壳显著伸长减薄,形成北部湾盆地。从而导致了海南岛的早期构造运动,而海南岛后期的构造运动则主要是受南海海底扩张的影响。海南地体运动规律的阐明对于了解北部湾油气盆地的形成演化有重要的理论和实际意义。     

Exciton Seebeck in Molecular Systems

Various applications relevant to the exciton dynamics,such as the organic solar cell,the large-area organic light-emitting diodes and the thermoelectricity,are operating under temperature gradient.The potential abnormal behavior of the exicton dynamics driven by the temperature difference may affect the efficiency and performance of the corresponding devices.In the above situations,the exciton dynamics under temperature difference is mixed with     

Achieving Unusual States of Matter Under High Pressure

The elongation method,originally proposed by Imamura was further developed for many years in our group.As a method towards O(N)with high efficiency and high accuracy for any dimensional systems.This treatment designed for one-dimensional(ID)polymers is now available for three-dimensional(3D)systems,but geometry optimization is now possible only for 1D-systems.As an approach toward post-Hartree-Fock,it was also extended to