基于自适应对比度增强的红外小目标检测网络

红外探测系统需要尽早发现目标以便及时拦截,但是红外图像上的小目标检测是一个挑战十足的任务。为了提高检测准确率,提出一种基于自适应对比度增强的红外小目标检测方法。为了利用自注意力机制和卷积各自的优势,设计了一个高效的特征提取网络和一个面向小目标的检测头。同时为了解决实际应用中出现的弱目标,在检测子网络前添加了一个图像预处理子网络,该模块可以自适应地调节图像对比度。在红外空中小目标数据集上的实验表明,提出的方法能达到93.76%的检测精度,与经典的检测方法相比,能够更好地平衡检测精度和召回率,证明了方法的巨大应用潜力。     

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.     

跨谱段SAR散射中心多维参数解耦和估计方法

微波光子雷达发射大带宽跨谱段的信号,为目标的精细电磁特性描述和准确识别提供基础的同时,也亟需与之相应的大带宽大转角情况下的电磁模型参数提取方法.相比窄带条件,跨谱段信号数据量大,所含物理量信息维度高且复杂,大转角情况下距离和方位向耦合.该文提出跨谱段SAR散射中心多维参数解耦和估计方法,首先结合极坐标格式算法(PFA)和属性散射中心模型构造2维解耦波数域散射中心模型,再结合坐标下降法(CDA)将复杂的高维耦合参数估计方法简化为循环迭代的1维参数估计方法,有效降低字典维度和估计复杂度,并引入Hooke-Jeeves算法提高估计精度.最后根据各个散射中心的参数估计结果对它们的结构和位置进行识别,对仿真数据的处理实验验证了该文方法的有效性.     

一种密码专用可编程逻辑阵列的分组密码能效模型及其映射算法

密码专用可编程逻辑阵列(CSPLA)是一种数据流驱动的密码处理结构,该文针对不同规模的阵列结构和密码算法映射实现能效关系的问题,首先以CSPLA的特定硬件结构为基础,以分组密码的高能效实现为切入点,建立基于该结构的分组密码算法映射能效模型并分析影响能效的相关因素,然后进一步根据阵列结构上算法映射的基本过程提出映射算法,最后选取几种典型的分组密码算法分别在不同规模的阵列进行映射实验.结果表明越大的规模并不一定能够带来越高的能效,为取得映射的最佳能效,阵列的规模参数应当与具体的硬件资源限制和密码算法运算需求相匹配,CSPLA规模为4×4～4×6时映射取得最优能效,AES算法最优能效为33.68 Mbps/mW,对比其它密码处理结构,CSPLA具有较优的能效特性.     

Research and Implementation of Parallel Artificial Fish Swarm Algorithm Based on Ternary Optical Computer

Mobile Networks and Applications - Artificial fish swarm algorithm (AFS) is used in the field of function optimization problems widely. The traditional AFS algorithm has some problems such as long...     

First Total Synthesis of A New 5-Deoxyflavone

A novel 5-deoxyflavone compound, 7, 8-dimethoxy-3′, 4′-methylene-dioxyflavone (1), was in 2002 isolated from the root bark of Albizia Odoratissima. [1] The compound 1 was preparaed from pyrogallic acid (2) and piperonal(5) in four steps (Scheme 1). [2] The spectroscopic data (1H NMR, 13C NMR, MS and IR) of 1 were in good agreement with those of natural product. [ 1]     

THE STRUCTURAL BASIS OF THE POOR FIBRIN SPECIFICITY OF UROKINASE(Ⅰ)——KNOWLEDGE-BASED PREDICTION OF KRINGLE STRUCTURES OF UROKINASE AND ITS RELATED PROTEINS

The Kringle-1 structure of plasminogen (PGK-1), the Kringle-2 structure of tissue plasminogen activator (PAK-2) and the Kringle structure of prourokinase (UKK) has been modeled on the basis of the three-dimensional structure of Kringle-1 of prothrombin (PTK-1) at 2.8 resolution. The predicted three-dimensional structure of these Kringles shows that the binding site of PGK-1 is characterized by an apparent dipolar site, the polar parts of which are separated by a hydrophobic region. PAK-2 possesses the anionic center but has not a cationic binding center which might be provided by a guanidinium group from Arg-69 located adjacent to the Arg-71 position. UKK possesses neither the anionic binding center nor the cationic center which are probably the main reason for the poor fibrin specificity of urokinase.     

两种航空电子光纤统一网络标准的比较及结合

根据美军最新一代战斗机研制计划提出的"航空电子光纤统一网络"的概念,从拓扑结构、可扩展性和机载应用等方面对FC和SCI两种航空电子光纤统一网络标准进行了比较.根据比较的结果,探索了综合使用两者的模型.     

电爆炸箔断路开关的理论和实验研究

 采用金属箔电爆炸过程的二维数值计算模型，对含电爆炸金属箔断路开关的电感储能脉冲功率调节系统进行了数值计算。以此为基础，研制出了一种低电感型的电爆炸箔断路开关，并以4μF/75kV脉冲电容器组作为初级能源，13Ω电阻作为负载进行实验研究。研究结果表明：在负载上可获得约250kV，脉宽大于400ns的脉冲电压。     

一种改进的基于110GHz在片S参数测试的HEMT器件寄生电阻提取方法

提出了一种改进的高电子迁移率晶体管寄生电阻提取方法,该方法利用了特殊偏置点 （V_gs > V_th, V_ds = 0 V）的等效电路模型, 推导了寄生电阻的表达式,采用半分析法对寄生电阻进行了优化。1 ~110 GHz S参数实测结果和仿真的S参数一致,证明该方法是有效的。