Besed On Visual Target Detection And Tracking Technology Research

Visual movement target detection and tracking technology has always been a popular research topic in the field of machine vision and is challenging.In research for the target detection involves a lot of technology,those includes the classification of object detection,image pattern recognition,image processing technology and computer application,etc.With the development of science and technology,the field of machine vision has also boomed.Improvements in camera technology,hardware system cost reduction,also increased computing speed and storage technology have contributed to its development.Many excellent theories have been proposed and applied in recent decades.But there is still reasonable algorithms for different practical scenarios.This article is aimed at solving practical problems,on the basis of the traditional motion target detection technology,combined with RobotMaster competition car motion detection identification as the background.A specific visual algorithm and the solution are proposed for single camera,single target motion object detection and tracking.First,the technique of image pretreatment was studied to prepare for follow-up work,Second,the use of the interframe differential method is applied to the detection of this low-velocity moving object.When the object does not satisfy the interframe difference method,the algorithm of template matching is introduced,and the traditional template matching algorithm has been designed and improved to identify and track moving objects.The stability of the algorithm is also improved.And get the center coordinate data and the curve of the moving object.Again,because of the different colors of the sports car,the new Camcum5 sensor is used to identify the object's color.and combining the STM32 MCU with the improved PID algorithm to control the cloud platform and make up the cloud tracking system.At the same time,the paper expounds the communication mode and the identification tracking based on color recognition.Finally,the proposed algorithm and solution are combined to form a visual movingobject detection and tracking system based on this specific application and can obtain a better identification and tracking effect.This paper presents a method of motion detection and tracking based on visual algorithm,and puts forward a solution to the application.