摘要
在服装吊挂系统中,由于衣架指夹难以精确定位,使得挂片机器人不能准确将衣服裁片夹持在衣架指夹上.针对此问题,采用尺度不变特征变换(SIFT,scale-invariant feature transform)算法,给出一种求取匹配点集最小外接矩形的方法来定位衣架.该方法利用SIFT算法对衣架图像进行识别,用Graham算法得到匹配点集的凸包,求取点集最小外接矩形,最后根据衣架自身比例关系确定衣架指夹的图像坐标.实验结果表明,该方法能够有效定位吊挂衣架指夹,对研制挂片机器人有一定的借鉴意义.
In the garment hanger system,because the position of finger clip of coat hanger is not accurate,the hanging robot can not accurately clamp the garment pieces.Thus a method of finding the minimum bounding rectangle of the matching point set is given by using the scale-invariant feature transform(SIFT)algorithm to locate the coat hanger.In this method,the SIFT algorithm is used to recognize the clothes rack image,the Graham algorithm is used to get the convex hull of the matching point set,and the minimum bounding rectangle is obtained.Finally,the image coordinates of the hangers are determined according to the proportion of the hangers.The results show that this method can effectively locate the finger clip of coat hanger on the garment hanger system,and it has certain reference significance for the development of the hanging robot.
作者
董煜文
王晓华
张森宇
李琪
叶浩劼
DONG Yuwen;WANG Xiaohua;ZHANG Senyu;LI Qi;YE Haojie(School of Electronics and Information,Xi′an Polytechnic University,Xi′an 710048,China)
出处
《西安工程大学学报》
CAS
2018年第1期77-82,共6页
Journal of Xi’an Polytechnic University
基金
国家自然科学基金资助项目(51607133)
陕西省科技厅工业攻关项目(2016GY-136)
国家级大学生创新训练项目(20170709047)
