湖库疏浚技术及淤积物资源化利用技术进展

文章介绍了我国部分河湖水库的淤积现状,评述了当前国内外对河湖疏浚设备和淤泥资源化利用的几种主要途径,包括物理利用、生物处理以及化学处理方法等,同时对新型的干馏法进行了介绍,指出了未来淤泥资源化利用的主要发展方向.     

Construction and verification of retinal vessel segmentation algorithm for color fundus image under BP neural network model

To improve the accuracy of retinal vessel segmentation, a retinal vessel segmentation algorithm for color fundus images based on back-propagation (BP) neural network is proposed according to the characteristics of retinal blood vessels. Four kinds of green channel image enhancement results of adaptive histogram equalization, morphological processing, Gaussian matched filtering, and Hessian matrix filtering are used to form feature vectors. The BP neural network is input to segment blood vessels. Experiments on the color fundus image libraries DRIVE and STARE show that this algorithm can obtain complete retinal blood vessel segmentation as well as connected vessel stems and terminals. When segmenting most small blood vessels, the average accuracy on the DRIVE library reaches 0.9477, and the average accuracy on the STARE library reaches 0.9498, which has a good segmentation effect. Through verification, the algorithm is feasible and effective for blood vessel segmentation of color fundus images and can detect more capillaries.

     

众筹项目推荐:面向隐式反馈的深度学习协同过滤

目前,在推荐系统研究中,用户的隐式反馈,以及极度稀疏的数据,已成为影响协同过滤推荐效果的主要问题.针对这一现象,本文提出了深度学习协同过滤算法,先利用卷积神经网络,对用户-项目矩阵的隐层特征进行学习,再结合协同过滤,对用户-项目的交互信息进行建模,并将两种特征融合预测推荐列表.以众筹平台的数据为实验对象,比较模型中各参数对推荐效果的影响,并设计与基线方法的对比实验.实验结果表明:均匀采集负反馈,并在一定卷积层数的网络中,数据稀疏度越高,效果越好;对比基线方法,本文提出的算法在公开数据集(Yahoo!Movie)上取得了最好的推荐结果.本文提出的算法有助于提高众筹平台的融资成功率,同时也丰富了推荐系统的研究体系.     

Bead-based polymerase chain reaction on a microchip

We present a bead-based approach to microfluidic polymerase chain reaction (PCR), enabling fluorescent detection and sample conditioning in a single microchamber. Bead-based PCR, while not extensively investigated in microchip format, has been used in a variety of bioanalytical applications in recent years. We leverage the ability of bead-based PCR to accumulate fluorescent labels following DNA amplification to explore a novel DNA detection scheme on a microchip. The microchip uses an integrated microheater and temperature sensor for rapid control of thermal cycling temperatures, while the sample is held in a microchamber fabricated from (poly)dimethylsiloxane and coated with Parylene. The effects of key bead-based PCR parameters, including annealing temperature and concentration of microbeads in the reaction mixture, are studied to achieve optimized device sensitivity and detection time. The device is capable of detecting a synthetically prepared section of the Bordetella pertussis genome in as few as 10 temperature cycles with times as short as 15?min. We then demonstrate the use of the procedure in an integrated device; capturing, amplifying, detecting, and purifying template DNA in a single microfluidic chamber. These results show that this method is an effective method of DNA detection which is easily integrated in a microfluidic device to perform additional steps such as sample pre-conditioning.     

基于光通信的数据单向导入系统的设计与实现

随着信息网络的日趋成熟,网络信息安全尤其是内网信息安全,越来越受到人们的关注,成为一时研究的热点。其中如何防止在移动存储设备与内网计算机信息交流过程中内网信息流失的问题,显得尤为迫切。为此,本文提出一种基于光通信的数据单向导入系统,为用户提供一种安全地将信息或者数据单向导入到内网的解决方案。利用光的单向性,从传输通道上卡断了反向的可能,有效地保护内网的机密信息不会流失到外部,同时做到良好的人机交互,用户可以任意选择存储设备中的文件发送。     

独立分量分析理论在海底管道漏磁检测中的应用

应用独立分量分析(ICA)方法对海底管道缺陷漏磁法检测中电磁干扰的消除进行了研究，阐述了漏磁检测和负熵法ICA快速算法的原理，应用相应的分析工具研究了试验数据。结果表明，在试验室条件下该方法能很好地分离出缺陷信号和电磁干扰信号。     

磁耦合谐振式无线能量传输的阻抗匹配特性

针对无线电能传输过程中传输效率随匹配阻抗大小变化的特点,提出两种阻抗匹配的方法:一种是基于步进电动机通过检测网络反馈信号,控制网络通过GA神经网络算法控制步进电动机,改变电容实现阻抗匹配;另一种是基于直直变换器中Cuk电路的占空比大小来匹配阻抗的方法.在负载和线圈之间嵌入单相桥式整流电路和Cuk升降压变压器实现匹配.根据最大传输功率理论,定量分析了改变电路的阻抗值大小,使系统的传输功率及效率最大.在MATLAB/Simulink平台验证了两种阻抗匹配设计方案的有效性.结果 表明,利用基于步进电动机和Cuk电路的阻抗匹配方法可以实现最佳阻抗值匹配,使无线电能传输效率最大化.     

带特征指标约束描述的设计模式分类挖掘

为解决设计模式挖掘正确率较低的问题, 提出一种带特征指标约束描述的设计模式分类挖掘方法。归纳了47种特征指标, 给出了设计模式约束定义, 对设计模式特征进行了描述, 以Adapter模式、Command模式、Factory Method模式为例, 分结构型、行为型与创建型3类挖掘设计模式, 设计了3个基准系统与4个经典系统的设计模式挖掘试验。试验结果表明, 本研究方法对基准系统的Adapter模式、Command模式、Factory Method模式挖掘正确率为96.13%、91.67%、72.23%, 对经典系统挖掘正确率分别为84.3%、81.26%、73.17%, 与传统方法比较, 本研究方法取得了较好的效果。