计算机应用 ›› 2021, Vol. 41 ›› Issue (1): 270-274.DOI: 10.11772/j.issn.1001-9081.2020060964

所属专题: 第八届中国数据挖掘会议(CCDM 2020)

• 第八届中国数据挖掘会议(CCDM 2020) • 上一篇    下一篇

复杂环境下的冰箱金属表面缺陷检测

袁野1,2,3, 谭晓阳1,2,3   

  1. 1. 南京航空航天大学 计算机科学与技术学院, 南京 211106;
    2. 模式分析与机器智能工业和信息化部重点实验室(南京航空航天大学), 南京 211106;
    3. 软件新技术与产业化协同创新中心, 南京 211106
  • 收稿日期:2020-05-31 修回日期:2020-08-07 出版日期:2021-01-10 发布日期:2020-11-12
  • 通讯作者: 袁野
  • 作者简介:袁野(1996-),男,湖南岳阳人,硕士研究生,主要研究方向:机器视觉、目标检测;谭晓阳(1971-),男,重庆人,教授,博士,CCF会员,主要研究方向:人脸识别、模式识别、计算机视觉、机器学习。
  • 基金资助:
    国家自然科学基金资助项目(61732006,61672280,61976115);全军共用信息系统装备预研项目(315025305)。

Defect detection of refrigerator metal surface in complex environment

YUAN Ye1,2,3, TAN Xiaoyang1,2,3   

  1. 1. College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing Jiangsu 211106, China;
    2. MIIT Key Laboratory of Pattern Analysis and Machine Intelligence(Nanjing University of Aeronautics and Astronautics), Nanjing Jiangsu 211106, China;
    3. Collaborative Innovation Center of Novel Software Technology and Industrialization, Nanjing Jiangsu 211106, China
  • Received:2020-05-31 Revised:2020-08-07 Online:2021-01-10 Published:2020-11-12
  • Supported by:
    This work is partially supported by the National Natural Science Foundation of China (61732006, 61672280, 61976115), the Project of Pre-research on Military Shared Information System Equipment (315025305).

摘要: 为了提升冰箱金属表面的缺陷检测效率,从而应对复杂的生产情况,提出了Metal-YOLOv3模型。使用随机参数变换,将缺陷数据进行了数百倍的扩充,改变原有YOLOv3模型的损失函数,引入了基于完整交并比(CIoU)所设计的CIoU损失函数,用缺陷的分布特性来降低非极大值抑制算法的阈值,并基于K均值聚类算法计算出更适合数据特点的先验框(anchors)值以提升检测精度。在一系列的实验后,发现Metal-YOLOv3模型在检测速度上远胜于主流的区域卷积神经网络(R-CNN)模型,每秒传输帧数(FPS)达到7.59,是Faster R-CNN的14倍,而且平均精确度(AP)也达到了88.96%,比Faster R-CNN高11.33个百分点,说明所提模型同时具备良好的鲁棒性与泛化性能。可见该方法具备有效性,能实际应用于金属制品的生产。

关键词: 金属表面, 缺陷, 冰箱, 损失函数, YOLOv3, 完整交并比

Abstract: In order to improve the efficiency of detecting defects on the metal surface of refrigerators and to deal with complex production situations, the Metal-YOLOv3 model was proposed. Using random parameter transformation, the defect data was expanded hundreds of times; the loss function of the original YOLOv3 (You Only Look Once version 3) model was changed, and the Complete Intersection-over-Union (CIoU) loss function based on CIoU was introduced; the threshold of non-maximum suppression algorithm was reduced by using the distribution characteristics of defects; and the anchor value that is more suitable for the data characteristics was calculated based on K-means clustering algorithm, so as to improve the detection accuracy. After a series of experiments, it is found that the Metal-YOLOv3 model is far better than the mainstream Regional Convolutional Neural Network (R-CNN) model in term of detection speed with the Frames Per Second (FPS) reached 7.59, which is 14 times faster than that of Faster R-CNN, and has the Average Precision (AP) reached 88.96%, which is 11.33 percentage points higher than Faster R-CNN, showing the good robustness and generalization performance of the proposed model. It can be seen that this method is effective and can be practically applied to the production of metal products.

Key words: metal surface, defect, refrigerator, loss function, YOLOv3 (You Only Look Once version 3), Complete Intersection-over-Union (CIoU)

中图分类号: