Blood cell counting based on U-Net++ and YOLOv5

Clinical information about a variety of disorders is available through blood cell counting, which is usually done by manual methods. However, manual methods are complex, time-consuming and susceptible to the subjective experience of inspectors. Although many efforts have been made to develop automated blood cell counting algorithms, the complexity of blood cell distribution and the highly overlapping nature of some red blood cells (RBCs) remain significant challenges that limit the improvement of analytical accuracy. Here, we proposed an end-to-end method for blood cell counting based on deep learning. Firstly, U-Net++ was used to segment the whole blood cell image into several regions of interest (ROI), and each ROI contains only one single cell or multiple overlapping cells. Subsequently, YOLOv5 was used to detect blood cells in each ROI. Specifically, we proposed several strategies, including fine classification of RBCs, adaptive adjustment for non-maximal suppression (NMS) threshold and blood cell morphology constraints to improve the accuracy of detection. Finally, the detection outcomes for each ROI were combined and superimposed. The results show that our method can effectively address the issue of high overlap and precisely segment and detect blood cells, with a 98.18% accuracy rate for blood cell counting.     

Surface Oxygen Coordination of Hydrogen Bond Chemistry for Aqueous Ammonium Ion Hybrid Supercapacitor

Aqueous ammonium ion hybrid supercapacitor (A-HSC) combines the charge storage mechanisms of surface adsorption and bulk intercalation, making it a low-cost, safe, and sustainable energy storage candidate. However, its development is hindered by the low capacity and unclear charge storage fundamentals. Here, the strategy of phosphate ion-assisted surface functionalization is used to increase the ammonium ion storage capacity of an α-MoO₃ electrode. Moreover, the understanding of charge storage mechanisms via structural characterization, electrochemical analysis, and theoretical calculation is advanced. It is shown that NH₄⁺ intercalation into layered α-MoO₃ is not dominant in the A-HSC system; rather, the charge storage mainly depends on the adsorption energy of surface “O” to NH₄⁺. It is further revealed that the hydrogen bond chemistry of the coordination between “O” of surface phosphate ion and NH₄⁺ is the reason for the capacity increase of MoO₃. This study not only advances the basic understanding of rechargeable aqueous A-HSC but also demonstrates the promising future of surface engineering strategies for energy storage devices.     

小型化低功耗机载显示器图形系统设计与实现

为了满足机载显示器图形生成系统的小尺寸、低功耗需求,提出了一种基于龙芯2K1000的全国产图形生成与处理方案。该方案以2K1000为核心构建硬件平台,使用2K1000内部集成的中央处理器、图形处理器、显示控制器,以及外部的内存实现图形的计算生成和双缓存显示,配合国产可编程逻辑器件实现图形与外视频的叠加显示及机载通信总线扩展。软件上采用航空专用的国产天脉操作系统,基于天脉操作系统设计了关键的图形显示驱动、帧存驱动、显示控制器驱动。实验结果表明,在输出1 024 pixel×768 pixel分辨率显示时,典型机载图形画面帧率达到35 frame/s,整体功耗约10 W。该方案扩展性强、功耗低,满足实时显示需求,在机载显示器领域有着广泛的应用空间。     

电磁散射特征提取与成像识别算法综述

合成孔径雷达(SAR)图像的自动化解译是合成孔径雷达技术应用的重要发展方向之一。电磁散射特征与目标结构具有稳健的关联性,是SAR图像解译的关键支撑。近年来,如何准确地实现电磁特征提取以及利用电磁特征反演目标特性受到了广泛的重视。该文讨论了电磁特征提取和基于电磁散射特征识别方法的研究成果,总结归纳了其中的关键要素和主要思路,详述了电磁散射机理在成像和识别领域的扩展应用,展望了未来电磁散射特征提取与应用的研究发展趋势。      

Upconversion fluorescence modulation of CaTiO3:Yb3+/ Er3+ nanocubes via Zn2+ introduction

rials, CaTiO3:Yb3+/Er3+/Zn2+ nanocubes with uniform size were prepared by solvothermal method in this paper. They were respectively characterized by X-ray diffractometer (XRD), scanning electron microscope (SEM), photoluminescence (PL) spectroscopy and their fluorescence lifetimes. The results show that the average size of the nanocubes is about 550 nm×650 nm×850 nm with good upconversion luminescence (UCL) properties. Under 980 nm laser excitation, the effects of the ratio between activator Er3+ and sensitizer Yb3+ and Zn2+ doping on the upconversion fluorescence properties were investigated, and the optimal ion ratio was obtained. The results of steady-state spectra show that the strongest fluorescence intensity of CaTiO3:Yb3+/Er3+ was obtained with the addition of 10 mol% Zn2+ at a Yb3+/Er3+ molar doping ratio of 3:0.3, which was attributed to the crystal field asymmetry generated by the introduction of Zn2+ ions. The energy transfer and upconversion mechanism between Yb3+ and Er3+ ions in CaTiO3 nanocubes were investigated by analyzing the upconversion fluorescence kinetics.     

基于激光自混合干涉调频信号的位移测量实验

激光自混合干涉在高精度测量领域发挥着越来越重要的作用,微位移测量已应用于大型土木结构、航空航天、健康监测等。传统测量以调幅信号为主,但随着噪声加强,测量误差越来越大,甚至难以测量。为此,采用马赫-曾德尔干涉仪,实现了调幅/调频信号转换。针对大噪声环境下的调频信号,利用多次希尔伯特变换进行相位解卷,实现了原始信号的重构。数值仿真结果证明了该方法的有效性,在10 dB白噪声环境中,系统的信噪比为10.0614 dB,估计误差为0.0614 dB。实验结果表明,微位移重构的误差在100 nm以内,且调频信号的信噪比远高于调幅信号。该方法适用于超精密测量,且在大噪声环境下仍具有较高的信噪比。     

基于涡流脉冲热成像的角焊缝表面裂纹检测北大核心CSCD

角焊缝裂纹是焊接结构中常见的缺陷之一,由于其所处位置特殊而常出现漏检、错检等情况。本文采用涡流脉冲热成像技术对T形角焊缝裂纹进行了研究,并提出了一种包含表面裂纹热图像前、后处理及定量检测的检验方法。首先,采用L形铁氧体磁芯对角焊缝缺陷试件进行检测,然后基于独立成分分析算法(ICA)对整个图像序列进行前处理,并提出了一种改进区域生长算法用于提取缺陷特征,最后利用所提出的检测方法对缺陷进行了定量检测和误差分析。实验结果表明,所提出的检测方法可以实现对焊缝表面裂纹的检测,抑制不均匀加热和边缘效应,改善图像分割中的欠分割和过分割现象,且检测出的缺陷长度和宽度相对误差仅为0.4%和14%。     

“高电压工程”课程思政教改探索与实践

针对“高电压工程”课程思政教学改革的要求,分别从教学目标、教学内容、教学方法和考核手段四个方面,挖掘课程中蕴含的思政元素,通过教学创新,有机融入思政内容,使学生不仅学到专业知识,更体会到专业内容蕴含的社会主义核心价值观。最后,通过课堂教学对“高电压工程”课程思政的改革效果进行了实践和检验,为其他电气类专业课的课程思政教学改革提供参考。