毕业设计（论文）-基于视频的运动物体的识别与跟踪精选.docxVIP

编号毕业设计题 目基于视频的物体识别与跟踪学生姓名学 号学 院专 业班 级指导教师二〇一七年六月本科毕业设计（论文）诚信承诺书本人郑重声明：所呈交的毕业设计（论文）（题目：基于视频的物体识别与跟踪）是本人在导师的指导下独立进行研究所取得的成果。尽本人所知，除了毕业设计（论文）中特别加以标注引用的内容外，本毕业设计（论文）不包含任何其他个人或集体已经发表或撰写的成果作品。作者签名：年月日（学号）：基于视频的物体识别与跟踪摘 要本文以高速路上的运动车辆为例，研究基于视频的物体的识别与跟踪。本文所使用算法均以C++语言在visual studio2015平台上实现，并以开源OpenCV为依托，简化了程序的代码书写。本文首先使用Gloyer方法完成了背景建模，并在此基础上使用光流法进行运动车辆的识别与跟踪。具体地，使用Shi-Tomasi 算法进行特征点检测，并使用金字塔LK算法实现特征点的跟踪。对于Shi-Tomasi算法和金字塔LK算法中的具体参数选择，本文采用了理论分析和实验检测相结合的方法，获得了适用于本文的各参数值。最后，本文对多个视频帧中的信息进行统计分析，获得了由多个特征点对应到同一车辆的各阈值的具体数值。经实验，这些阈值数值适应于本文视频，并且获得了较好的识别跟踪效果，可以以较高的正确率返回视频中所有车辆的实时运动速度和位置信息。关键词：Shi-Tomasi 算法，金字塔LK算法，多特征，阈值选取Video - based object recognition and trackingAbstractIn this paper, I use the moving vehicles on the highway as an example and study the recognition and tracking of moving objects based on videos. The algorithms used in this paper are based on C ++ language in the visual studio 2015 platform with the basement of open source codes OpenCV in order to simplify the program code writing.In this paper, firstly, I use the Gloyer method to complete the background modeling. Then I make the use of optical flow method for the identification and tracking of moving vehicles. Specifically, the Shi-Tomasi algorithm is used to detect the feature points while the pyramid LK algorithm is used to track the feature points. For the choice of the parameters of Shi-Tomasi algorithm and the pyramid LK algorithm, this paper uses a combination of theoretical analysis and experimental testing. Finally, this paper analyzes the information in multiple video frames and successfully achieves the specific values of the thresholds which are used to distinguish the multiple features points that belongs to the same car. Experiments show that these values of the thresholds are adapted to the video used in this paper and have a good recognition and tracking effect. They can return the real-time velocity and position information of all the vehicles showed in the video with high matching r