计算机应用 ›› 2021, Vol. 41 ›› Issue (8): 2352-2357.DOI: 10.11772/j.issn.1001-9081.2020101575

所属专题: 前沿与综合应用

• 前沿与综合应用 • 上一篇    下一篇

航空发动机损伤图像的二分类到多分类递进式检测网络

樊玮1, 李晨炫1, 邢艳1, 黄睿1, 彭洪健2   

  1. 1. 中国民航大学 计算机科学与技术学院, 天津 300300;
    2. 厦门航空有限公司 维修工程部, 福建 厦门 361006
  • 收稿日期:2020-10-12 修回日期:2021-01-05 出版日期:2021-08-10 发布日期:2021-01-27
  • 通讯作者: 黄睿
  • 作者简介:樊玮(1968-),男,陕西咸阳人,教授,博士,CCF会员,主要研究方向:机器学习、收益管理;李晨炫(1994-),男,山东济宁人,硕士研究生,主要研究方向:深度学习、计算机视觉;邢艳(1987-),女,河北沧州人,讲师,博士,CCF会员,主要研究方向:数据挖掘;黄睿(1987-),男,宁夏中卫人,讲师,博士,CCF会员,主要研究方向:计算机视觉、图像处理、机器学习;彭洪健(1988-),男,福建龙岩人,工程师,主要研究方向:航空发动机孔探及叶片润滑。
  • 基金资助:
    中央高校基本科研业务费项目中国民航大学专项(3122018C020);天津市教委科研计划项目(2019KJ126)。

Binary classification to multiple classification progressive detection network for aero-engine damage images

FAN Wei1, LI Chenxuan1, XING Yan1, HUANG Rui1, PENG Hongjian2   

  1. 1. College of Computer Science and Technology, Civil Aviation University of China, Tianjin 300300, China;
    2. Maintenance and Engineering Department, Xiamen Airlines, Xiamen Fujian 361006, China
  • Received:2020-10-12 Revised:2021-01-05 Online:2021-08-10 Published:2021-01-27
  • Supported by:
    This work is partially supported by the Special Program for CAUC of the Fundamental Research Funds for the Central Universities (3122018C020), the Scientific Research Program of Tianjin Municipal Education Commission (2019KJ126).

摘要: 航空发动机损伤是影响飞行安全的重要因素。当前基于计算机视觉的发动机孔探图像损伤检测存在两个主要问题:一是孔探图像背景复杂,使得模型对损伤的检测精度较低;二是孔探图像数据来源受限,导致模型可检测类别较少。为解决这两个问题,提出了基于Mask R-CNN的二分类到多分类递进式航空发动机损伤图像检测网络。通过在Mask R-CNN中增加二分类检测分支,首先对图像中的损伤进行二分类检测并对定位坐标进行回归优化;其次使用原始检测分支递进地进行多分类检测,以进一步回归优化损伤的检测结果并确定损伤类型;最后根据多分类检测的结果,通过Mask分支对对损伤进行实例分割。为了增加模型检测类别及验证方法的有效性,构建了包含八种损伤类型,共1 315张孔探图像的数据集。在该集合上进行的训练和测试结果表明,多分类检测的平均精度(AP)和AP75与Mask R-CNN相比分别提高3.34%、9.71%,可见所提方法能够有效提高对孔探图像中的损伤的多分类检测精度。

关键词: 孔探图像, 目标检测, 实例分割, 航空发动机探伤, 递进式检测

Abstract: Aero-engine damage is an important factor affecting flight safety. There are two main problems in the current computer vision-based damage detection of engine borescope image:one is that the complex background of borescope image makes the model detect the damage with low accuracy; the other one is that the data source of borescope image is limited, which leads to fewer detectable classes for the model. In order to solve these two problems, a Mask R-CNN (Mask Region-based Convolutional Neural Network) based progressive detection network from binary classification to multiple classification was proposed for aero-engine damage images. By adding a binary classification detection branch to the Mask R-CNN, firstly, the damage in the image was detected in binary way and regression optimization was performed to the localization coordinates. Secondly, the original detection branch was used to progressively perform multiple classification detection, so as to further optimize the damage detection results by regression and determine the damage class. Finally, instance segmentation was performed to the damage through the Mask branch according to the results of multiple classification detection. In order to increase the detection classes of the model and verify the effectiveness of the method, a dataset of 1 315 borescope images with 8 damage classes was constructed. The training and testing results on this set show that the Average Precision (AP) and AP75 (Average Precision under IoU (Intersection over Union) of 75%) of multiple classification detection are improved by 3.34% and 9.71% respectively, compared with those of Mask R-CNN. It can be seen that the proposed method can effectively improve the multiple classification detection accuracy for damages in borescope images.

Key words: borescope image, object detection, instance segmentation, aero-engine damage inspection, progressive detection

中图分类号: