手势控制智能搬运小车中手指类型识别
Finger Type Recognition for Gesture Controlled Smart Handling Cart
黄玉银 1李志扬1
作者信息
- 1. 华中师范大学 物理科学与技术学院,湖北 武汉 430000
- 折叠
摘要
采用手势控制的智能搬运小车可大幅度减轻快递货物分拣中的体力劳动强度,提升工作效率,其中手指类型识别对手势识别有重要辅助作用.通过车载摄像头采集手势,经过图像处理获选取基于中心线的角度FBA和垂直长度FPD两个不受图像旋转和平移影响的特征向量,通过多元混合高斯统计模型实现手指类型识别,识别率可达93％.
Abstract
Gesture controlled smart handling carts can greatly reduce labor intensity in sorting express delivery parcels and improve the working efficiency, where finger type recognition plays an important auxiliary role in gesture identification. In this paper, we captured the images of the hand gestures using onboard camera, obtained the two eigenvectors of FBA (the angle from the center line) and FPD (the vertical length) which are not affected by the rotation or translation of the images, and achieved finger type recognition through the multivariate mixed Gaussian statistical model where the recognition rate can be up to 93％.
关键词
手势控制/手指类型识别/智能搬运小车/快递分拣Key words
gesture controlled/finger type recognition/smart handling cart/sorting of express delivery parcels引用本文复制引用
出版年
2018
NETL
NSTL
维普
万方数据