针对目前变分模态分解法在心电信号降噪时存在模态分量难以取舍的问题,提出了一种改进的变分模态分解方法.首先对含噪心电信号进行变分模态分解,通过各模态分量的中心频率和模态分量与原始心电信号的互相关来确定噪声占优的模态分量与信号占优的模态分量.然后选取中心频率处于医学心跳频率范围的模态分量来提取心跳频率对应的采样点数,根据心跳频率对噪声占优的模态分量和信号占优的模态分量分别进行平滑滤波.最后使用处理过的模态分量重构心电信号,完成基线漂移和肌电噪声的去除.实验结果表明该方法的去噪效果优于小波阈值法、变分模态分解法及两者相结合的方法.

Learning from the limited labeled examples, denoted as few-shot, has been successfully applied into the image detection and classification. However, their ability to learn new concepts quickly is limited. Meta-learning is one approach to address this issue, by enabling the network to learn how to learn. However, most of metric-based meta-learning few-shot models low recognition accuracy and due to the large difference in data distribution in few-shot learning, the importance of model feature extraction is more reflected. Especially, some existing methods only utilize high-level semantic feature, thus ignore the underlying information that is essential for few-shot learning. In addition, these methods adopted the fixed distance functions for feature measurement, thus they cannot fully integrate features. To solve these problems, an end-to-end, few-shot classification model (AMBR) with the adaptive multi-branch feature fusion attention is proposed in this paper. The multi-path feature residual module is designed to extract and fuse the features for making full use of the underlying information. Then, while utilizing the metric module to learn the metric between the support set and query set features, the attention module is employed to enhance the learned features. Finally, a non-linear classifier is proposed for image classification. We evaluate our method on two benchmark datasets, namely Omniglot and Mini-imagenet. Experimental results demonstrate our proposed AMBR algorithm achieves the excellent performance in these two datasets through a shallow feature extraction network, and is superior to the state-of-the-art.

Recently, many segmentation methods based on supervised deep learning have been widely used in remote sensing images. However, these approaches often require a large number of labeled samples, which is difficult to obtain them for remote sensing images. Self-supervision is a new learning paradigm, and can solve the problem of lack of labeled samples. In this method, a large number of unlabeled samples are employed for pre-training, and then a few of labeled samples are leveraged for downstream tasks. Contrast learning is a typical self-supervised learning method. Inspired, we propose a Dense Multi-scale Feature Contrastive Learning Network (DMF-CLNet), which is divided into global and local feature extraction parts. Firstly, in the global part, instead of traditional ASPP, DenseASPP can obtain more context information of remote sensing images in a dense way without increasing parameters. Secondly, in the global and local parts, Coordinate Attention (CA) modules are introduced respectively to improve the overall performance of the segmentation model. Thirdly, in the global and local parts, the perceptual loss is calculated to extract deeper features. Two remote sensing image segmentation datasets are evaluated. The experimental results show that our model is superior to the current self-supervised contrastive learning methods and ImageNet pre-training techniques.

主成分分析(PCA)在图像识别及高维数据降维中有着普遍的应用.为提升基于主成分分析的图像重建性能,在经典PCA算法的基础上提出了广义主成分分析(TPCA),并利用该算法进行图像重建.该算法利用图像像素的空间邻域构成固定尺寸的数组并将其作为广义标量进行代数运算,可以方便有效地描述各像素的空间约束,从而提升图像的重建效果,同时也为其他图像分析算法的改进和升级提供了思路.实验分别将经典PCA算法与TPCA算法运用于公有数据进行图像重建,并对重建质量进行定量分析,结果表明运用TPCA算法的图像重建效果明显优于经典PCA算法.

提出一种基于广义奇异值分解的高阶图像低秩近似新方法.在传统矩阵分析的基础上,介绍高阶广义矩阵的生成及定义,得出广义奇异值分解不仅适用于传统的实数矩阵,对高阶广义复数矩阵亦具有重要意义.实验在高阶图像低秩近似的基础上,提出两种改进方案,一是将传统的实数矩阵扩展成为高阶广义复数矩阵,二是在领域选取时,分析比较指数增长和线性增长方式的近似效果.数值实验验证了高阶广义复数矩阵具有更高的低秩近似效果,指数增长方式与线性增长方式相比具有明显的优越性.

针对合成孔径雷达图像易受外界环境干扰导致获取的图像信息准确度降低的问题,提出一种基于广义奇异值分解的高阶图像低秩近似方法.首先,在经典奇异值分解的基础上,利用邻域选取法将经典二维矩阵推广到广义高阶矩阵;其次,利用"t-product"模型将经典矩阵算法推广到广义矩阵的相关算法,得出广义奇异值分解的具体实现过程;最后,在广义奇异值分解和经典奇异值分解的条件下,通过实例分析并利用结构相似度和峰值信噪比比较其低秩近似性能.仿真实验验证了广义奇异值分解技术相比于经典奇异值分解,不仅充分考虑图像像素点之间的相互作用与空间结构,而且随着广义矩阵阶数的扩展,所获得的图像结构相似度越大,峰值信噪比越高,可将其应用于高阶图像的低秩近似、重构.

为提高待识别图像信息提取的准确性,提出了一种基于广义奇异值分解的图像去噪算法.该算法通过邻域选取,将经典的奇异值分解技术扩展为广义奇异值分解技术,达到增强去噪能力和保留原始数据结构的目的.利用Len-na、MonaLisa和House噪声图像对该算法进行性能测试,并与基于奇异值分解的去噪算法进行去噪效果比对.实验结果表明,使用该算法能获得最优的灰度图像去噪结果,在峰值信噪比方面相比于传统算法具有明显优势.

针对电子密码锁安全性及易用性差等问题,设计了一种以STC89C52单片机为核心,通过HC-06蓝牙模块与安卓手机连接,实现随机密码加密的密码锁.利用单片机自身的EEPROM存储数据,通过E0加密算法对数据进行保护,实现了手机对密码锁的控制.实验表明,蓝牙加密密码锁可以实现随机密码功能,具有安全性高、使用方便、价格低廉等特点.

High frequency electronic circuit is one of the important basic courses in electricity specialty .In order to improve the teaching quality and better the teaching effect, in this paper, we discuss some concrete ideas from the teaching content, teach-ing methods and experimental teaching along with the authors ’ teaching practice.Through the course reform, we promote the com-prehensive knowledge level of the students.

[目的]设计开发一种能有效观察留置尿管病人尿量变化的智能监测系统,观察使用效果,同时对尿量进行自动记录,保存相关信息并将信息传输至系统终端。[方法]尿量监测系统主要由特殊精密尿袋、尿液监测装置及系统终端组成。尿袋采用聚苯乙烯、聚氯乙烯、硅胶等材料,尿液监测装置由液位传感器、智能检测系统、显示面板等组成,能显示测量结果,及时发出报警信息;通过某医院135例肾移植术后多尿期病人近1年的实验观察。[结果]该系统能够在集满尿液时及时发出报警信号,漏报率为0.7%,误报率为2.2%。[结论]该尿量监测系统的研制和应用,为临床医护人员及时了解病情及调整治疗方案提供了信息,并大大减轻了医护人员的劳动量,减少了医疗差错发生率。

Aiming at solving the shortcomings and problems of the 3related courses in video technical direction of information class,a series of integration schemes are presented,which include optimizing the course outline of each course in the video technical direction,optimizing the course content,getting through the teaching way between graduate course and undergraduate course to form directional longitudinal teaching hierarchy, designing video technical directional experiment,designing video technical directional course exercise,and constructing an integrated practical teaching system in and out of class.The implementation schemes and specific measures are provided.Preliminary achievements have been achieved in course system,practical teaching content,teaching method and digital video processing multimedia practicing platform building.

This paper introduces an eight-way wireless remote control system for industrial control based on RF technology. The system takes 8051 single chip computer as main MCU, and it can complete data sending and receiving by controlling RF transceiver module RF905, also it can control working condition of equipments by controlling opening and closing of relay finally.

Bistatic inverse synthetic aperture radar (BiISAR) imaging provides complementary information to monostatic ISAR imaging. A suggestion that the conventional bistatic ISAR can be replaced by an equivalent monostatic radar is shown to be inadequate in this paper. We prove that the BiISAR image is a sheared version of the projection of the target on the range-Doppler plane. A bistatic interferometric radar configuration is proposed to correct this distortion and to form 3-D image. Simulation results have validated our analysis and shown the viability of the proposed 3-D imaging algorithm.

This paper introduces a wireless medical liquid-level monitoring and controlling system based on RF technology, in which liquid levels beyond limits are monitored by using the principle of different refractive indexes of light rays in air and in liquid and adopting the non-contact photoelectric monitoring method, and 8051 single chip computer were used as main MCU, through controlling RF module RF905 to transmit and receive alarm signals, also control the nixie tubes and buzzer to give sound and light alarms.

The invention discloses a frequency and phase difference precision measurement method based on frequency and phase relationship auxiliary processing. The frequency and phase difference precision measurement method comprises the following steps of: allowing a 10-MHz frequency scale output by a caesium clock to pass through a shaping circuit and an adjustable pulse generation circuit to generate a pulse signal; automatically synthesizing frequency f0 by using a digital display scope (DDS), wherein the value of the f0 depends on fx roughly measured by a singlechip microcomputer, so that the integer multiple of the group period of the fx and the f0 is equal to the time value of a gate and the value of the group phase quantum of the fx and the f0 is equal to the resolution ratio of a group phase coincidence detection circuit; transmitting the f0 and the fx to a pilot frequency phase coincidence detection circuit to generate the actual measurement gate; controlling work of a counter; calculating the value of the fx according to the counting result by using a microprogrammed control unit (MCU); and finally displaying output by using a liquid crystal display (LCD). According to the frequency and phase difference precision measurement method disclosed by the invention, one DDS is introduced to determine the frequency relationship between the fx and the f0, and a good measuring effect of any frequency signals within the required gate time is achieved.

Median filter has good capabilities for reducing a variety kind of random noise, and causes less ambiguity than linear smoothing filters under same processing size. In order to suppress the impulse noise of digital video signal and meet the system's needs of real-time, it is of great significance to do fast filtering of image based on hardware. By analyzing the common 3×3 filtering window's mathematical model, this paper proposes a fast median filtering algorithm based on field programmable gate array (FPGA) and the scheme design. According to the characteristics of parallel structures and its suitable for pipeline design of FPGA. VHDL and schematic design are used in this paper to design the implement circuit. Quartus II is used for timing simulation. The results show that it can filter the impulse noise in real time and improves the quality of image.

Pseudo random codes are easily constructed and can be used in multiple-input multiple-output (MIMO) radar to provide for signal separation. However it has the drawbacks of high aperiodic cross-correlation and auto-correlation sidelobes. Fortunately, in some applications, codes with low correlations are needed only over a narrow window. In this paper, we introduce the Zero Correlation Zone Codes (ZCZC) which have zero correlations in a narrow zone. In other applications, the targets are stationary, which allows for using a long code sequence. In this case we propose an Extended Zero Correlation Zone Codes (EZCZC) where the zero correlation zone is extended to cover the whole sidelobes area. Design examples are given.

A kind of graduation design(paper)on-line management system is introduced based on the network,through analysis of the problems in the link of graduation design(paper).The necessity of construction graduation design(paper) on-line management system,is proposed.The graduation design(paper) on-line management system is designed by adopting B/S development model.All the management work from the declaration of graduation design(paper) to submission can be realized in network.The convert from the previous management by objectives to the present quality monitoring of the whole process is implemented.A new model for the dynamic management of graduation design(paper) is provided.

Ad-hoc Network is an autonomous communication system composed of a group of mobile terminals with radio transceiver.Ad-hoc network is a technology applicable for monitoring and controlling system,data collecting system or emergency communication system for underground coal mine.According to its routing method,Ad-hoc network can fall into two categories: proactive and reactive.AODV is a typical reactive routing method.This paper proposes an AODV-based routing recovery mechanism using secondary path in Ad-hoc network for underground coal mine,and presents some principles for evaluating the performance of the AODV-based route recovery mechanism using secondary path.

介绍了家居安防系统中各种防盗探测器的设计。采用门磁、热释红外、主动红外、振动等传感器,以新型的安防集成电路和高性能收发组件组成外围信号采集和传送电路,硬件实现稳定可靠,有较高的市场推广价值。