北京邮电大学学报

  • EI核心期刊

北京邮电大学学报 ›› 2024, Vol. 47 ›› Issue (4): 83-89.

• 体系化人工智能专题 • 上一篇    下一篇

面向工业膏体浓度监测的体系化智能视觉信息融合模型

王贺正1,2,马博渊1,3,4,李潇睿3,4,郭利杰5,刘光生5   

  1. 1. 北京科技大学 北京材料基因工程高精尖创新中心 2. 北京科技大学 钢铁共性技术协同创新中心 3. 北京科技大学 智能科学与技术学院 4. 北京科技大学 顺德创新学院 5. 矿冶科技集团有限公司
  • 收稿日期:2023-12-31 修回日期:2024-03-17 出版日期:2024-08-28 发布日期:2024-08-26
  • 通讯作者: 马博渊 E-mail:mbytony@ustb.edu.cn
  • 基金资助:
    国家重点研发计划项目; 佛山市科技创新专项资金项目; 中央高校基本科研业务费专项项目

Industrial Paste Concentration Measurement Through Holistic Smart Visual Information Fusion Model

WANG Hezheng1,2, MA Boyuan1,3,4, LI Xiaorui3,4, GUO Lijie5, LIU Guangsheng5   

  • Received:2023-12-31 Revised:2024-03-17 Online:2024-08-28 Published:2024-08-26
  • Contact: Bo-Yuan Ma E-mail:mbytony@ustb.edu.cn

摘要: 针对膏体充填技术中对于膏体浓度监控方法存在的精准度低、仪器使用寿命短、测量耗时长、因安全问题导致使用受限等局限性,提出了一种双流视觉信息融合模型以实现膏体浓度自动化准确估计,以减少膏体浓度监测对人工的需求,增加矿场的自动化程度,促进体系化人工智能在智能采矿领域中的应用。所提模型以卷积神经网络模型为基础,使用双流结构,通过分析膏体视频及相应的光流信息,学习提取其中的空间特征和时间特征,形成对膏体视频的双流特征感知表示;特征融合模块进一步强化对有效特征的识别,使模型能够对膏体浓度进行非接触式的准确监测。在此基础上,采集实际生产环境下膏体视频数据并构建真实的膏体视频 浓度数据集,并以此数据集对所提模型进行验证。实验结果表明,所提模型在真实膏体视频数据集中识别准确率能够达到 94.16% ,在同等条件下准确率较其他深度学习模型高出 3.47% ,实现了在生产环境下对膏体浓度进行实时准确监测的目的。

关键词: 膏体充填, 膏体浓度准确估计 , 视频分类, 多模态

Abstract: To counter existing limitations in paste concentration monitoring of paste backfilling technique, such as lack of accuracy, short life expectancy of related device, prolonged detection time, and limitations due to safety issues, a two-stream visual feature fusion model for automatic soft measuring of paste concentration is proposed, reducing the need for manual participation, increasing automation, and furthering the application of holistic artificial intelligence in the field of smart mining. The model, based on convolutional neural network model, adopts a two-stream architecture, analyzes the paste video and the corresponding optic flow information, extracts spatial and temporal features from the input, and produces two-stream feature representation. The feature fusion module further enhances representation for effective features, enabling the model to accurately measure paste concentration through non-contact method. In addition, video data of paste under production environment is collected to construct a dataset, and evaluated the proposed model with the dataset. The experiment results show that the proposed model can reach an accuracy of 94.16% , surpassing other deep learning methods by 3.47% under the same condition, fulfilling the need to conduct accurate real-time paste concentration detection in production environment.

Key words: paste backfilling , precise paste concentration prediction ,  video classification ,  multi-modal

中图分类号: