多目标宏形状特征点集快速检测技术研究

提出一种新的图像多目标特征点集检测技术。该技术研究了多目标分割、目标细微形状提取和宏形状提取,进而检测表示目标宏形状特征的特征点集。各目标宏形状特征点集进行识别与跟踪．研究了上述技术的快速实现。实际DSP跟踪系统处理结果表明该技术计算量小,抗噪能力强且准确性高,能准确识别目标,适用于实际图像目标特征点集的检测。     

基于扩展卡尔曼滤波器的矢量跟踪算法研究

该文针对矢量跟踪算法缺乏定量分析的问题,对线性化模型与跟踪误差进行了研究.为了提高矢量跟踪环的动态跟踪性能,修正了已有线性化观测矩阵忽略接收机伪距率对位置的偏导项的问题;提出了新的环路反馈量,简化了运算过程.推导出了矢量跟踪环的热噪声跟踪误差,与标量跟踪环相比跟踪门限明显降低,并且在相同条件下,观测卫星数越多,矢量跟踪环的跟踪门限越小;推导了矢量跟踪环的动态应力跟踪门限,证明矢量跟踪环可以处理高动态微弱信号.通过 Monte Carlo 仿真验证了理论分析结果的正确性.     

Siamese visual tracking with multilayer feature fusion and corner distance IoU loss

The tracker based on the Siamese network regards tracking tasks as solving a similarity problem between the target template and search area. Using shallow networks and offline training, these trackers perform well in simple scenarios. However, due to the lack of semantic information, they have difficulty meeting the accuracy requirements of the task when faced with complex backgrounds and other challenging scenarios. In response to this problem, we propose a new model, which uses the improved ResNet-22 network to extract deep features with more semantic information. Multilayer feature fusion is used to obtain a high-quality score map to reduce the influence of interference factors in the complex background on the tracker. In addition, we propose a more powerful Corner Distance IoU (intersection over union) loss function so that the algorithm can better regression to the bounding box. In the experiments, the tracker was extensively evaluated on the object tracking benchmark data sets, OTB2013 and OTB2015, and the visual object tracking data sets, VOT2016 and VOT2017, and achieved competitive performance, proving the effectiveness of this method.     

A dynamic and weighted spectrum decision mechanism based on SNR Tracking in CRAHNs

The spectrum decision concept in Cognitive Radio Ad-Hoc Networks (CRAHNs) introduces important challenges. These include the time-dependent SNR observations of individual CRAHN users due to the fading and shadowing effects in the licensed channels, the necessity of fusion mechanisms for accurate decisions, and the difficulties depending on multi-hop deployment. Considering these challenges, in this paper, we propose a dynamic, cooperative and distributed spectrum decision mechanism in order to decide the channel usage in CRAHNs accurately. The proposed mechanism introduces the SNR Tracking System which considers the time-varying local SNR observations and decisions of the CRAHN users. The proposed mechanism employs a distributed Weighted Fusion Scheme (WFS), to combine the individual decisions and hence, to obtain the cooperative decision. The proposed spectrum decision mechanism adapts itself dynamically to the multi-hop architecture of the network. The performance of the proposed mechanism is compared to some conventional fusion mechanisms based on the AND, OR and MAJORITY rules, and it is shown that the proposed weighted mechanism gives lower false alarm and higher detection probabilities compared to the conventional fusion mechanisms.     

基于模糊控制的光伏并网逆变器MPPT控制研究

最大功率点跟踪控制的目的是为了将光伏阵列发出的最大能量实时地提供给负载,使光伏发电系统的能量利用率达到最大。在光伏阵列产生电能的应用中,有许多不确定因素,如太阳光照强度、光伏阵列温度的变化、负载的变化、光伏阵列输出特性的非线性,则建立模型分析光伏阵列输出最大功率要考虑很多的因素。从模糊控制技术的分析中知道,模糊控制不需要对被控对象建立精确的数学模型,是一种相对简单的智能控制方法,对处理非线性问题有很好的效果。因此,用模糊控制法来实现MPPT可以得到比较好的效果。本文基于此研究了光伏阵列的非线性功率输出特性,建立了基于Matlab simulink/Power system的光伏阵列仿真模型,对基于模糊控制采用扰动观察法进行光伏发电最大功率点跟踪进行了仿真验证。     

基于单片机的智能小车行驶控制方案设计

以STC89C52单片机作为控制核心,提出了智能小车行驶控制方案,完成了硬件系统的设计,实现了小车的红外循迹和自动避障的功能.设计方案简单低廉、性能稳定、通用性强、控制精度高,具有较高的使用价值.     

一种强跟踪抗噪声自适应滤波算法

通过分析几种典型的LMS算法,结合几种典型算法的优点,在不同时期采用不同的控制步长策略,本文提出基于相关箕舌线的自适应滤波算法(CTCLMS算法),并对其性能指标进行了比较仿真。在同一仿真条件下的仿真结果表明:CTCLMS算法除了具有收敛速度快,稳态误差小,计算简单的特点外,还同时具有了强跟踪能力和强抗噪声能力,解决了系统跟踪能力和抗噪声性能之间的矛盾。     

基于多帧边缘差异的视频运动对象的分割与跟踪算法

从视频场景中分割和跟踪感兴趣的视频对象对于MPEG-4等基于对象的视频编码来说是关键性的技术之一。针对目前大部分视频对象分割和追踪算法相当复杂但仍不能有效地去除背景噪声的问题,该文提出用于分割和跟踪视频运动对象的一种基于多帧边缘差异的算法。该算法利用一组帧的边缘差异来提取运动对象区域,通过聚类方法去除背景像素点,利用形态学算子得到对象分割模板,同时通过建立前帧感兴趣对象与当前帧运动对象的帧间向量跟踪当前帧的感兴趣视频对象。不同标准视频测试序列的测试结果表明,该算法能够实现对感兴趣的视频运动对象更为精确、快速和有效地分割和跟踪。