回声信号处理综合实验设计及实现

针对学生对卷积和相关概念不清的问题,结合生活现象,本文设计了回声信号处理综合实验。实验内容包括回声产生、回声特性分析、回声消除,它将难以理解的卷积、自相关、互相关概念有机地串在一起,帮助学生理解和应用。实践表明,理论推导和实验编程相结合有助于学生将所学的知识融会贯通,且学生在做与生活密切联系的综合实验时,积极性高,收获大。     

一种实时多尺度目标检测识别算法

无人机目标检测与识别任务中,目标随着飞行高度的改变尺寸发生显著变化。常规目标检测模型中,获取的小目标细节信息有限,检测精度较低;而适用于小目标的实时检测模型往往容易丢失大目标的背景信息,降低大目标的检测精度。针对以上多尺度目标检测识别任务难点,提出一种基于改进特征金字塔网络(Feature Pyramid Network, FPN)结构的实时多尺度目标检测识别模型。该模型通过增加特征金字塔层级覆盖更广的目标尺度,获取更为丰富的目标信息;同时,利用跨连接增加不同尺度特征融合的多样性,降低特征传导距离,保留更加完整的尺度特征来提高模型检测识别多尺度目标的性能。通过实验发现,相比于原始网络结构和相同特征层级的四层特征金字塔结构,加入改进特征金字塔结构的多尺度目标检测模型识别性能得到了提升。     

复杂场景下多尺度船舶实时检测方法

船舶检测在军事侦察、海上目标跟踪、海上交通管制等任务中发挥着重要作用。然而,受船舶外形尺度多变和复杂海面背景的影响,在复杂海面上检测多尺度船舶仍然是一个挑战。针对此难题,提出了一种基于多层信息交互融合和注意力机制的 YOLOv4 改进方法。该方法主要通过多层信息交互融合（multi-layer information interactive fusion,MLIF）模块和多注意感受野（multi-attention receptive field,MARF）模块构建一个双向细粒度特征金字塔。其中,MLIF模块用于融合不同尺度的特征,不仅能将深层的高级语义特征串联在一起,而且将较浅层的丰富特征进行重塑;MARF由感受野模块（receptive field block,RFB）与注意力机制模块组成,能有效地强调重要特征并抑制冗余特征。此外,为了进一步评估提出方法的性能,在新加坡海事数据集（Singapore maritime dataset,SMD）上进行了实验。实验结果表明,所提方法能有效地解决复杂海洋环境下多尺度船舶检测的难题,且同时满足了实时需求。     

基于改进的SSD监理目标检测研究

针对装饰装修工程中由人工验收带来的诸多问题,文中提出了一种改进的SSD算法并将其应用于监理工作来代替人工验收,推动智能监理的实现。由于SSD算法存在对同一目标复检以及小目标检测效果欠佳等问题,故文中利用DPN网络替换基础特征提取网络VGG16。DPN结合了Resnet和Densenet的优点,具有更好的特征提取能力。通过加权FPN融合特征图,突出不同层特征图的贡献,丰富用于预测的特征图语义。利用深度可分离卷积降低模型的参数量,提高算法的推理速度。实验对比发现,改进后模型的平均精度提升了3.47%,对小数目检测平均精度的提升可达15%,证明新模型在监理目标检测任务中效果良好。     

Single image shadow detection via uncertainty analysis and GCN-based refinement strategy

Learning-based shadow detection methods have achieved an impressive performance, while these works still struggle on complex scenes, especially ambiguous soft shadows. To tackle this issue, this work proposes an efficient shadow detection network (ESDNet) and then applies uncertainty analysis and graph convolutional networks for detection refinement. Specifically, we first aggregate global information from high-level features and harvest shadow details in low-level features for obtaining an initial prediction. Secondly, we analyze the uncertainty of our ESDNet for an input shadow image and then take its intensity, expectation, and entropy into account to formulate a semi-supervised graph learning problem. Finally, we solve this problem by training a graph convolution network to obtain the refined detection result for every training image. To evaluate our method, we conduct extensive experiments on several benchmark datasets, i.e., SBU, UCF, ISTD, and even on soft shadow scenes. Experimental results demonstrate that our strategy can improve shadow detection performance by suppressing the uncertainties of false positive and false negative regions, achieving state-of-the-art results.     

A Procedure for Transforming Impedance Spectra for the Determination of Mechanism of Electrochemical Reactions

Transformation of impedance spectra into relaxation time spectra (RTS) is used for determining contributions of individual processes of the oxygen electroreduction reaction (OER) to the polarization resistance of the electrochemical cell. The transformation technique involves the solution of the convolution equation found with the aid of a modified Van Cittert iteration algorithm checked on model impedance spectra. The technique, when used to analyze impedance spectra of electrochemical cells air|Pt|YSZ|YSZ + Pt|air, shows that the conversion of a globular structure of the YSZ + Pt cermet layer to a columnar one is accompanied by a change of peak amplitudes in RTS. The revealed RTS dynamics when heated to 750°C is compared with peculiarities of individual processes in OER.     

用重复卷积法求解物理宽化函数

用重复卷积法求解物理宽化函数吴万国,林嘉　（福州大学中心实验室,福州３５０００２）（中国（福建）对外贸易中心集团,福州３５０００１）关键词Ｘ射线衍射,晶粒度,晶格畸变,重复卷积粉末试样的Ｘ射线衍射谱线一般具有一定的宽度,引起话线宽化的主要原因有仪器致...     

光周期敏感不育水稻中激素的二阶导数卷积伏安法研究

二阶导数卷积伏安法应用于水稻中痛量雌雄激素的检测，具有快速，灵敏，干扰少的特点。本文以光周期敏感核不育水稻农垦５８Ｓ为材料，以农垦５８作对照，了叶焉和穗中雌，雄性激素的含量，研究了不同光周期条件下两类激素的变化。     

关于CBP算法的一种新型滤波函数和它的性质

用卷积反投影(CBP)算法作CT重建,滤波函数是关键.本文建议一种新型滤波函数,给出了用它作CT重建的误差估计,分析了该滤波函数的时频特性,并用来作局部重建.模拟和实测数据的数字实验表明其在保证空间分辨率的同时能较好地克服Gibbs效应.     

Accuracy and CPU time of JE-TLM,PLRC-TLM and CRC-TLM methods for plasma medium

The current density (JE), the piecewise linear recursive convolution (PLRC) and the constant convolution recursive (CRC) techniques are developed and integrated into the transmission line matrix (TLM) algorithm. A comparison of these three schemes according to the criteria of accuracy and CPU time is presented. Numerical experiments show that JE provides the most accurate solution and requires the least CPU time; PLRC is nearly as accurate but consumes more CPU time than CRC.