YOLOv5+deepsort算法实现门店客流量统计，源码加模型及测试视频。也可以自己训练模型，应用于车流量或者其他目标计数。 如果下载后，使用有什么问题可以留言或者私聊！

是利用离子滤波在MATLAB下所实现的目标跟踪

This lecture series gives comprehensive overview of the broad field of advanced radar systems, signal and data processing. The series starts with a lecture by U. Nickel in which the basic and fundamental of signal processing for phased array radar and their problems with grating lobes, ambiguities, and angle estimation for instance. The lecture “Advanced target tracking techniques” by W. Koch gives a short introduction to the principle of target tracking and several approaches are discussed for sequential track extraction and for phased-array radars. In the third lecture P. Berens gives an introduction to the synthetic aperture radar (SAR). T. Johnsen provides an overview of bi- and multistatic radar and their associated problems like synchronization, timing, and signal processing. The second lecture of U. Nickel focuses on the problem of adaptive array signal processing and provides the fundamental understanding for the next two lectures. The focus of these lectures, presented by W. Bürger, is on space-time adaptive processing. In his second lecture P. Berens continues with the topic of the synthetic aperture radar and expands the presented techniques to wideband SAR and multichannel SAR/MTI systems. W. Koch’s second paper focuses on sensor data and information fusion, which is essential to extract key-information for the final judgement using several sensors. In summery, this Lecture Series presents a unique overview of the state of the art of advanced radar and the associated signal and data processing research. It offers a variety of material for all those being involved in this scientific area, e.g. students, university teachers, researchers, industrial system designers, and military users.

提出融合遮挡感知的在线Boosting跟踪算法，该算法对跟踪结果实时进行遮挡检测，根据检测结果自适应调整分类器更新策略。该方式能够有效维护分类器特征池的纯净，提高算法在遮挡环境下的顽健性。实验结果表明，与传统的在线Boosting跟踪算法相比，改进的算法能有效解决目标遮挡问题。

运动目标跟踪系统c++源码

visdrone数据集场景下YOLOv5-deepsort视觉检测和跟踪代码，包括用Visdrone数据集训练好的YOLOv5s-visdrone.pt和YOLOv5m-visdrone.pt两个模型，并附上了训练曲线图；代码都已配置好，安装好环境后可以直接使用，包含有相关场景下的测试视屏和跟踪结果，并保存了目标运动的质心坐标和可以绘制出目标运动轨迹

vot-toolkit, 可视化目标跟踪( VOT ) 挑战评估工具 VOT评估套件这个库包含了可视对象跟踪( VOT ) 挑战插件的官方评估工具包。 存储库包含实际的评估工具包作为一组 Matlab ( Octave 兼容) 脚本，一个文档和一组不同编程语言集成示例。有关详细信息，请参阅源文档中的文档

卡尔曼滤波器 C ++中基于卡尔曼滤波的多目标跟踪的实现 安装 编译并生成重构文件mkdir build cd build cmake .. # 生成供make编译的规则文件makefile make # 编译源文件并生成可执行文件multiple_target_tracking 运行 启动多目标跟踪模拟程序./multiple_target_tracking

 采用霍夫变换法对雷达目标进行起始，解决了机动目标的非线性强的问题，得到精确的航迹起始初值信息，并将初值信息作为无迹卡尔曼滤波目标跟踪的初始输入，实现对机动目标的跟踪。较其它的算法，霍夫-无迹卡尔曼滤波具有更高的精度。实验仿真，证明了其有效性。

【滤波跟踪】基于Kalman滤波的视频目标跟踪matlab 源码.md