纺织学报 ›› 2017, Vol. 38 ›› Issue (11): 156-161.doi: 10.13475/20161203906

• 管理与信息化 • 上一篇    下一篇

应用机器视觉的鞋面数量统计与尺寸测量

  

  • 收稿日期:2016-12-23 修回日期:2017-08-15 出版日期:2017-11-15 发布日期:2017-11-15

Quantity statistics and size measurement of vamps based on machine vision

  • Received:2016-12-23 Revised:2017-08-15 Online:2017-11-15 Published:2017-11-15

摘要:

针对鞋面数量统计和尺寸测量过程中人工检测方法效率低、成本高的问题。提出利用机器视觉的鞋面数量统计与尺寸测量方法。采用基于边缘方向的模板匹配算法对鞋面进行数量统计。运用最小外接矩形(MER)算法对鞋面进行尺寸测量。为验证方法的可靠性,对3种鞋面分别在光照变化、混乱和遮挡条件下进行了检测实验,并与绝对误差和算法、归一化相关系数法和Hausdorff距离匹配算法的检测结果进行了对比。将MER测量结果和人工测量结果进行对比,结果表明:基于边缘方向的模板匹配算法在复杂环境下的鲁棒性更高,可以完成鞋面数量统计;MER方法和人工测量的相对误差在0.67%以内,满足工业生产中的实际需求。

关键词: 鞋面数量, 鞋面尺寸, 边缘方向, 最小外接矩形算法

Abstract:

Focusing on the problems of low efficiency and high cost of artificial detection in vamps statistics and size measurement, a machine vision method of counting and measuring vamps was presented. Using the template matching algorithm based on the edge direction to complete quantity statistics of vamps. The minimum enclosing rectangle algorithm measured the size of vamps was employed. To verify the reliability of the proposed method, three kinds of vamps were detected in illumination changes, clutter and occlusion, the results were compared with Sum of absolute differences, normalized cross correlation and Hausdorff distance matching algorithm; The measurement results of minimum enclosing rectangle algorithm were compared with manual test. The experimental results demonstrate that template matching algorithm based on the edge direction is more robust in complex environment, and it can finish the quantity statistics of vamps; The relative error between the minimum enclosing rectangle method and the artificial measurement is less than 0.67%, which meets the actual needs of industrial production.

Key words: vamp quantity, vamp size, edge direction, minimum enclosing rectangle algorithm

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!