《计算机应用》唯一官方网站 ›› 2023, Vol. 43 ›› Issue (7): 2295-2302.DOI: 10.11772/j.issn.1001-9081.2022060857

• 多媒体计算与计算机仿真 • 上一篇    

改进注意力机制的电梯场景下危险品检测方法

郭奕裕, 周箩鱼(), 刘新瑜, 李尧   

  1. 长江大学 电子信息学院,湖北 荆州 434023
  • 收稿日期:2022-06-13 修回日期:2022-09-06 接受日期:2022-09-08 发布日期:2022-10-08 出版日期:2023-07-10
  • 通讯作者: 周箩鱼
  • 作者简介:郭奕裕(1997—),男,广东潮州人,硕士研究生,主要研究方向:图像处理、机器学习;
    周箩鱼(1985—),男,湖南邵阳人,副教授,博士,主要研究方向:计算机视觉、人工智能;
    刘新瑜(2000—),女,湖北十堰人,硕士研究生,主要研究方向:机器学习、模板匹配;
    李尧(1998—),男,山东济南人,硕士研究生,主要研究方向:机器学习、人工智能。
  • 基金资助:
    国家自然科学基金资助项目(61901059)

Dangerous goods detection method in elevator scene based on improved attention mechanism

Yiyu GUO, Luoyu ZHOU(), Xinyu LIU, Yao LI   

  1. Electronics and Information School,Yangtze University,Jingzhou Hubei 434023,China
  • Received:2022-06-13 Revised:2022-09-06 Accepted:2022-09-08 Online:2022-10-08 Published:2023-07-10
  • Contact: Luoyu ZHOU
  • About author:GUO Yiyu, born in 1997, M. S. candidate. His research interests include image processing, machine learning.
    ZHOU Luoyu, born in 1985, Ph. D., associate professor. His research interests include computer vision, artificial intelligence.
    LIU Xinyu, born in 2000, M. S. candidate. Her research interests include machine learning, template matching.
    LI Yao, born in 1998, M. S. candidate. His research interests include machine learning, artificial intelligence.
  • Supported by:
    National Natural Science Foundation of China(61901059)

摘要:

针对电动自行车和煤气罐搭乘电梯引起的火灾隐患,提出一种改进注意力机制的电梯场景下危险品检测方法。以YOLOX-s为基线模型,首先在加强特征提取网络中引入深度可分离卷积替换标准卷积,提升模型的推理速度。然后提出一种基于混合域的高效卷积块注意力模块(ECBAM)并嵌入主干特征提取网络中。在ECBAM模块的通道注意力部分,使用一维卷积替换两个全连接层,既降低了卷积块注意力模块(CBAM)的复杂度又提高了检测精度。最后提出一种多帧协同算法,通过结合多张图片的危险品检测结果以减少危险品入侵电梯的误报警。实验结果表明:改进后模型比YOLOX-s的平均精度均值(mAP)提升了1.05个百分点,浮点计算量降低了34.1%,模型体积减小了42.8%。可见改进后模型降低了实际应用中的误报警,且满足电梯场景下危险品检测的精度和速度要求。

关键词: 危险品检测, 电梯, YOLOX-s, 深度可分离卷积, 高效卷积块注意力模块, 一维卷积, 多帧协同算法

Abstract:

Aiming at the hidden danger of fire caused by electric bicycles and gas tanks taken into elevators, an improved attention mechanism was proposed to detect dangerous goods in elevator scene, and a method based on the mechanism was proposed. With YOLOX-s as the baseline model, firstly, a depthwise separable convolution was introduced in the enhanced feature extraction network to replace the standard convolution, which improved the reasoning speed of the model. Secondly, an Efficient Convolutional Block Attention Module (ECBAM) based on mixed-domain was proposed and embedded into the backbone feature extraction network. In the channel attention part of ECBAM, two fully connected layers were replaced by a one-dimensional convolution, which not only reduced the complexity of Convolutional Block Attention Module (CBAM) but also improved the detection precision. Finally, a multi-frame collaboration algorithm was proposed to reduce the false alarms of dangerous goods’ intrusion into the elevator by combining the dangerous goods detection results of multiple images. Experimental results show that compared with YOLOX-s, the improved model can increase the mean Average Precision (mAP) by 1.05 percentage points, reduce the floating point computational cost by 34.1% and reduce the model size by 42.8%. The improved model reduces false alarms in practical applications and meets the precision and speed requirements of dangerous goods detection in elevator scene.

Key words: dangerous goods detection, elevator, YOLOX-s (You Only Look Once version X-s), depthwise separable convolution, Efficient Convolutional Block Attention Module (ECBAM), one-dimensional convolution, multi-frame collaboration algorithm

中图分类号: