节点文献

用图象识别法在线判断烧结矿FeO含量等级的研究

STUDY ON ONLINE RECOGNITION OF FeO IN SINTERBY MEANS OF IMAGE RECOGNITION METHOD

  • 推荐 CAJ下载
  • PDF下载
  • 不支持迅雷等下载工具,请取消加速工具后下载。

【作者】 张石郑春清张宏勋李万新

【Author】 ZHANG Shi ZHENG Chunqing ZHANG Hongxun(Northeastern University)LI Wanxin(Benxi Iron & Steel(Group)Co.)

【机构】 东北大学本溪钢铁(集团)公司

【摘要】 介绍了烧结矿FeO含量的测量方法,提出了利用图象识别方法在线判断烧结矿FeO含量等级的方法。根据烧结生产实际,将烧结矿中FeO含量分成低、适中、高、过高四个等级。采集断面图象训练样本,判断其对应的FeO含量,进而计算其对每个FeO含量等级的隶属度。对在线采集的未知FeO含量等级的图象样本,用隶属度加权的平均距离法判断其FeO含量等级。

【Abstract】 In this paper,the methods of measuring ferrous oxide(FeO)in the sinter are introduced and then a new online image recognition method is proposed.According to the practice of sintering production,the concentration of FeO in sinter is classified as four classes:low,medium,high and extrahigh.The crosssection image samples have been collected,the concentration of FeO of every image has been determined by the watcher and the membership function to each class has been calculated.The membership function weighted mean distance classifyng method is used to determine the FeO concentration degree of online collected image. 

  • 【分类号】TF046.4,
  • 【被引频次】16
  • 【下载频次】67
节点文献中: 

本文链接的文献网络图示:

本文的引文网络