现代电子技术2017，Vol.40Issue(17)：56-59,63,5.DOI:10.16652/j.issn.1004-373x.2017.17.014
基于图像识别的钻头自动跟踪算法
Drill automatic tracking algorithm based on image recognition
摘要
Abstract
The current methods are difficult to describe the drill target template accurately and have complex operation and big tracking error in the drill automatic tracking process,therefore a drill automatic tracking method based on image recognition is proposed. The double convolution theory is used to segment the threshold of the drill original image to extract the drill moving image target. The extracted drill moving image target is filtered to get the normalized central moment of the drill image. The Gauss model is used to model the drill moving image target and background. On this basis,the thoughts of Kalman filtering esti-mation and mean shift are integrated to apply to the tracking process of drill moving image target,so as to track the drill auto-matically. The experimental results show that the method can improve the detection accuracy of drill moving image target effec-tively,and has high stability of target tracking.关键词
图像识别/钻头/自动跟踪/卡尔曼滤波Key words
image recognition/drill/automatic tracking/Kalman filtering分类
信息技术与安全科学引用本文复制引用
杨怀林..基于图像识别的钻头自动跟踪算法[J].现代电子技术,2017,40(17):56-59,63,5.基金项目
江苏省淮安市科技项目:新型医用缝合针打孔智能化装置研发(HAG2014040) （HAG2014040）
