Improved mean shift target tracking based on self-organizing maps

Thanks to its simplicity and real-time processing possibility, mean shift has been widely used for video tracking. However, it often fails when the background is similar to the intended object or when the object is partially or completely occluded. To address these two problems, in this paper we propose a novel algorithm based on mean shift by exploring simultaneously the temporal and spatial information of the tracked object. A cascade classification method based on nearest neighbor and self-organizing maps is employed as a confirmation step to eliminate spurious objects through the structure information of the object. The forward and backward tracking results are further combined to improve the localization accuracy and tolerate at the same time scale variation. Experiments have shown clearly the superior performance of the proposed system in terms of accuracy, stability and robustness.     

A protection scheme for microgrid with multiple distributed generations using superimposed reactive energy

With the growing integration of distributed generation, distribution networks have evolved toward the concept of microgrids. Microgrids can be operated in either the grid-connected mode to achieve peak shaving and power loss reduction or the islanded mode to increase the reliability and continuity of supply. These two modes of operation cause a challenge in microgrid protection, because the magnitude of fault current decreases significantly during the transition of a microgrid from the grid-connected mode to the islanded mode. This paper proposes a protection scheme for the microgrid based on superimposed reactive energy. The proposed scheme uses the Hilbert transform to calculate the superimposed reactive energy (SRE). The sequence components of superimposed current are adopted to detect fault incidents in the microgrid. The faulty phase and section are recognised by using the directional characteristics of SRE along with a threshold value. Moreover, a relay structure, which enables the proposed protection scheme, is designed. The significant feature of the proposed protection scheme is that it has the ability to protect the looped and radial microgrids against solid and high-impedance faults. To verify the efficacy of the proposed approach, extensive simulations have been carried out using the MATLAB/SIMULINK software package. The results show that the proposed scheme successfully identifies and isolates various types of fault in a microgrid and performs well with different fault resistances and fault locations.     

电子鼻技术在果蔬检测中的应用

电子鼻技术作为一种新兴的智能仿生技术,在工业和农业方面都得到了广泛的研究与应用,具有样品处理简单、检测速度快、识别效果好、实时、无损的优点。由于电子鼻具有满足农产品品质无损检测要求等优点,电子鼻系统正在被越来越多地运用到农业研究之中。本文介绍了电子鼻的发展史和一些常见的电子鼻型号,其中典型的有PEN3型和FOX4000等。分析了电子鼻系统的组成构造,介绍了电子鼻系统的工作原理。其中重点阐述了电子鼻技术在水果与蔬菜中的研究与应用。在农业生产过程中,电子鼻在对区分果蔬的不同类型与品种、合理地制定与调整果蔬的贮藏时间和条件以及鉴别与评价果蔬品质等方面有着很大的帮助。最后,主要针对电子鼻技术在农业中的应用,从多个方面对电子鼻技术存在的问题及其发展趋势进行了探讨。     

Chaos and rigorous verification of horseshoes in a class of Hopfield neural networks

In this paper, chaos in a new class of three-dimensional continuous time Hopfield neural networks is investigated. Numerical experiments show that this class of Hopfield neural networks can have chaotic attractors and limit cycles for different parameter configurations. By virtue of horseshoes theory in dynamic systems, rigorous computer-assisted verifications are done for their chaotic behavior. In terms of topological entropy, quantitative interpretations of these neural networks’ complexity are given. A brief analysis is also presented about their robustness.     

Decoupled optimal design for power electronic circuits with adaptive migration in coevolutionary environment

This paper presents a genetic based decoupled optimal design method for power electronics circuit design using an adaptive collaboration approach in a cooperative coevolutionary environment. The circuit parameters of the power conversion stage and the feedback network of a buck regulator are optimized through two parallel coadaptive genetic based optimization processes. The best candidate of the tunable parameters in one evolutionary process for the design of the power conversion stage is merged to the other evolutionary process for the design of the feedback network as untunable factors through a collaboration controller in which the collaboration strategy is adaptively controlled by a first-order projection of the maximum and minimum bounds of the fitness value of the genes representing the circuit design parameters in each generation. The proposed design methodology is suitable for parallel computation resulting in considerable improvement in searching efficiency. Simulated results of the design of a buck regulator with the proposed approach were verified with experimental results from the actual hardware implementation. It showed that the design with the proposed scheme was compatible with the design specification.     

引入“西蒙学习法”提高计算机专业学生的实践能力

针对学校计算机科学与技术专业学生普遍存在学习专业知识范围广泛,难以熟练掌握一门开发技术的现状,文章以实践教学中引入"西蒙学习法"为立足点,在分析计算机专业Java知识学习的基础上,介绍如何使用西蒙学习法,结合实践教学的实际,提高学生的实践动手能力,培养出面向市场需求的本科计算机人才。     

U-Net与自适应阈值脉冲耦合神经网络相结合的眼底血管分割方法

由于眼底血管结构复杂多变,且图像中血管与背景对比度低,眼底血管分割存在巨大困难,尤其是微小型血管难以分割.基于深层全卷积神经网络的U-Net能够有效提取血管图像全局及局部信息,但由于其输出为灰度图像,并采用硬阈值实现二值化,这会导致血管区域丢失、血管过细等问题.针对这些问题,提出一种结合U-Net与脉冲耦合神经网络(P...     

宽带 U 型槽矩形贴片微带天线设计

提出并设计了一种宽带广角覆盖的波束扫描折叠式反射阵列（folded reflect array,FRA）天线. 为了在宽带范围内以及较宽的波束覆盖范围（beam coverage range,BCR）内获得较低的扫描损耗,提出一种考虑照射权重因子、截获功率权重因子以及频率权重因子的多波束优化方法;为了降低FRA天线的剖面,采用小尺寸贴片天线作为馈源,将其与单层反射板集成在一起. 设计了一个尺寸为264 mm×264 mm、工作在Ku波段的11波束扫描的FRA天线,仿真验证本文所提设计方法的有效性,结果表明：该设计在13.5~16.5 GHz的宽带范围内,实现了3 dB波束可覆盖−28°~28°范围,中心频率处最大增益为26.6 dBi,口径效率范围为14%~20.9%,所设计的共口径FRA天线可以获得相当平坦的增益,扫描损耗都在3 dB以内. 该天线具有覆盖范围广、覆盖带宽宽、集成度高、增益高、口径效率高等优点,在无线通信、雷达成像和目标识别等领域具有广阔的应用前景.