煤炭工程 ›› 2016, Vol. 48 ›› Issue (3): 116-118.doi: 10.11799/ce201603036

• 研究探讨 • 上一篇    下一篇

基于角点匹配的强力输送带接头识别算法

佘影   

  1. 中煤科工集团重庆研究院有限公司
  • 收稿日期:2015-12-03 修回日期:2015-12-07 出版日期:2016-03-10 发布日期:2016-04-08
  • 通讯作者: 佘影 E-mail:sysysy.boy@163.com

CONVEYOR BELT JOINT RECOGNITION BASED ON CORNER MATCHING ALGORITHM

  • Received:2015-12-03 Revised:2015-12-07 Online:2016-03-10 Published:2016-04-08

摘要: 为精确定位输送带内部钢丝绳损伤,提出一种基于角点特征向量匹配的接头识别算法。该算法以接头为基准点,根据归一化积相关算法筛选出完整的接头图像,利用Harris角点作为目标特征点计算该特征点的haar小波响应分布,形成接头图像的特征向量并依据特征向量匹配识别出每个接头。实验表明,该算法不仅具有良好的光照、平移鲁棒性和识别率,而且速度快、效率高,可以有效解决接头识别和故障定位问题。

关键词: 损伤定位, Harris角点, 小波响应, 特征向量

Abstract: In order to accurate the positioning of conveyor belt internal wire rope damage, a joint recognition algorithm based on corner feature vectors was proposed. Reference point for the joint, the joint image intact was selected According to the normalized product correlation algorithm, the Haar wavelet response distribution was calculated by using the Harris corner points as feature points, each joint was identified according to the feature vector matching of joint image. Experiments show that, this algorithm not only has good illumination robustness and recognition rate, but also has high speed and high efficiency, which can effectively solve the joint identification and fault location.

中图分类号: