| YOLOv4目标检测算法在煤矿工人口罩佩戴监测工作中的应用研究* |
| |
| 引用本文： | 崔铁军,王凌霄.YOLOv4目标检测算法在煤矿工人口罩佩戴监测工作中的应用研究*[J].中国安全生产科学技术,2021,17(10):66-71. |
| |
| 作者姓名： | 崔铁军 王凌霄 |
| |
| 作者单位： | （辽宁工程技术大学 安全科学与工程学院，辽宁 葫芦岛 125105） |
| |
| 基金项目： | * 基金项目: 国家自然科学基金项目（52004120）;国家重点研发计划项目(2017YFC1503102);辽宁省教育厅项目(LJ2020QNL018);辽宁工程技术大学学科创新团队项目(LNTU20TD-31) |
| |
| 摘 要： | 为防止煤矿工人吸入过量粉尘而导致职业性尘肺病，基于Keras框架利用YOLOv4 (you only look once)目标检测算法对井下人员佩戴防尘口罩情况进行高精度且快速的检测与识别，并与MTCNN(Multi-task convolutional neural network)和FaceNet构成的人脸识别算法相结合，进行煤矿工人口罩佩戴监测的研究。结果表明:模型对井下人员口罩佩戴有较高的检测精度，识别已佩戴口罩的矿井下作业人员的平均精度达到92.78%，识别未佩戴防尘口罩检测的平均精度为91.63%，与其他主流算法相比算法具有更好的鲁棒性和检测效果。研究结果为预防煤矿工人职业性尘肺病提供1种有效的技术手段。
|
| 关 键 词： | 目标检测 口罩检测 深度学习 人脸识别 职业安全卫生 |
Research on application of YOLOv4 object detection algorithm in monitoring on masks wearing of coal miners |
| |
| CUI Tiejun,WANG Lingxiao.Research on application of YOLOv4 object detection algorithm in monitoring on masks wearing of coal miners[J].Journal of Safety Science and Technology,2021,17(10):66-71. |
| |
| Authors: | CUI Tiejun WANG Lingxiao |
| |
| Affiliation: | (College of Safety Science and Engineering,Liaoning Technical University,Huludao Liaoning 125105,China) |
| |
| Abstract: | In order to prevent the coal miners from inhaling excessive dust and causing occupational pneumoconiosis,based on the Keras framework,the YOLOv4 (you only look once) object detection algorithm was used to detect and identify the situation of underground workers wearing dust-proof masks accurately and rapidly.Combined with the face recognition algorithm composed of MTCNN (Multi-task convolutional neural network) and FaceNet,the research on the monitoring of coal miners’ mask wearing was carried out.The results showed that the model had a high detection accuracy for the masks wearing of underground workers.The average accuracy for the recognition on the masks wearing of underground miners was 92.78%,and the average accuracy for the recognition on without wearing dust-proof masks was 91.63%,so the algorithm had better robustness and detection effect compared with other mainstream algorithms.The model provides a way and method for the coal miners to prevent the occupational pneumoconiosis,and its effectiveness was verified by experiments. |
| |
| Keywords: | object detection mask detection deep learning face recognition occupational safety and health |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《中国安全生产科学技术》浏览原始摘要信息 |
|
点击此处可从《中国安全生产科学技术》下载全文 |
