Research On Moving Human Tracking Method Of Golf Robotic Caddie

Moving target tracking has been one of the topics at the forefront of visual areas ofconcern in the mobile robot localization, navigation and intelligent monitoring andother aspects are very important applications. Through the tireless efforts of scholars,moving target detection and tracking technology has made great progress. However,because of the visual tracking system application environment complexity and diversityof the target itself, to target detection, tracking technology has brought greatdifficulties. Paper boy robot golf background research on the golf course moving humandetection and tracking method.Thesis detailed analysis of the current detection areas of three kinds of commonlyused algorithms: the background difference method, the frame difference method,optical flow method. When walking speed is quick slow for a human movement targetcharacteristics of the multiple frame difference method and gray projection methodcombined with detection method, solve the problem of not complete detection to extracthuman body target; On the golf course with a background color is a single, such astrees, or flag on the benchmarking the characteristics of distractors, using human bodyexternal torque length-width ratio and detection method of human skin colorcombination, effectively eliminate the human body target; Paper studied the trackingmethod based on Kalman motion estimation and Camshift tracking method, in view ofthe human body Similar color, block and other issues, put forward the Camshift andKalman filtering method, combining the prediction of improve the effectiveness andstability of the track; In order to make the human body target to keep a certain distancewith the golf caddie robot, the paper adopts human target external torque borders anddistance from the edge of the video sequence method, to determine target body stay inthe middle of the field of vision. Through the body’s external torque size and theproportion of the size of the video sequence to determine the distance between the bodyand golf caddie.Paper detection and tracking algorithm is applied to the golf course will humanmovement on the video, simulation studies are carried out using MATLAB simulationplatform, the results prove the validity of the method of the thesis.