《计算机应用》唯一官方网站 ›› 2024, Vol. 44 ›› Issue (1): 167-174.DOI: 10.11772/j.issn.1001-9081.2023010037

• 人工智能 • 上一篇    

不完整实例引导的航空发动机叶片实例分割

黄睿1, 张超群1, 成旭毅1, 邢艳1(), 张宝2   

  1. 1.中国民航大学 计算机科学与技术学院, 天津 300300
    2.天津大学 智能与计算学部, 天津 300000
  • 收稿日期:2023-01-15 修回日期:2023-04-13 接受日期:2023-04-14 发布日期:2023-06-06 出版日期:2024-01-10
  • 通讯作者: 邢艳
  • 作者简介: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;
    张超群(2000—),男,山东聊城人,硕士研究生,主要研究方向:深度学习、计算机视觉;
    成旭毅(1997—),男,山西晋中人,硕士研究生,主要研究方向:深度学习、计算机视觉;
    张宝(1989—),男,安徽灵璧人,工程师,硕士,主要研究方向:计算机视觉。
    第一联系人: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;
  • 基金资助:
    中国民航大学科研启动项目(2017QD15X)

Incomplete instance guided aeroengine blade instance segmentation

Rui HUANG1, Chaoqun ZHANG1, Xuyi CHENG1, Yan XING1(), Bao ZHANG2   

  1. 1.College of Computer Science and Technology,Civil Aviation University of China,Tianjin 300300,China
    2.College of Intelligence and Computing,Tianjin University,Tianjin 300000,China
  • Received:2023-01-15 Revised:2023-04-13 Accepted:2023-04-14 Online:2023-06-06 Published:2024-01-10
  • Contact: Yan XING
  • About author:HUANG Rui, born in 1987, Ph. D., lecturer. His research interests include computer vision, image processing, machine learning.
    ZHANG Chaoqun, born in 2000, M. S. candidate. His research interests include deep learning, computer vision.
    CHENG Xuyi, born in 1997, M. S. candidate. His research interests include deep learning, computer vision.
    ZHANG Bao, born in 1989, M. S., engineer. His research interests include computer vision.
  • Supported by:
    Scientific Research Startup Foundation of Civil Aviation University of China(2017QD15X)

摘要:

当前基于深度学习的实例检测方法在进行发动机叶片分割时,由于缺少带标注的发动机叶片数据,导致无法充分训练网络模型,仅得到次优的分割结果。为了提升航空发动机叶片实例分割精度,提出了不完整实例引导的航空发动机叶片实例分割方法,通过结合已有的实例分割方法和交互式分割方法,可得到较好的发动机叶片分割结果。首先,使用少量标注数据训练实例分割网络,得到发动机叶片的初步分割结果;其次,将检测到的单个叶片分为前景和背景两部分,通过选择前景种子点和背景种子点,利用交互式分割方法的思想,产生完整的单个叶片的分割结果;依次处理完所有的叶片后,将结果合并得到最终的发动机叶片实例分割结果。使用72张图像训练基于稀疏实例激活图的实时实例分割方法(SparseInst)产生初始的实例分割结果,在56张图像上进行测试。所提方法的全类平均准确率(mAP)比SparseInst的全类平均准确率高5.1个百分点;且它的mAP结果均优于当前流行的实例分割方法MASK R-CNN(MASK Region based Convolutional Neural Network)、YOLACT (You Only Look At CoefficienTs)、BMASK-RCNN (Boundary-preserving MASK R-CNN)。

关键词: 航空发动机, 实例分割, 发动机叶片, 损伤检测, 交互式分割

Abstract:

The current deep learning based instance segmentation methods cannot fully train the network model and result in sub-optimal segmentation results due to the lack of labeled engine blade data. To improve the precision of aeroengine blade instance segmentation, an aeroengine blade instance segmentation method based on incomplete instance guidance was proposed. Combining with an existing instance segmentation method and an interactive segmentation method, promising aeroengine blade instance segmentation results were obtained. First, a small amount of labeled data was used to train the instance segmentation network, which generated initial instance segmentation results of aeroengine blades. Secondly, the detected single blade instance was divided into foreground and background. By selecting foreground seed points and background seed points, the interactive segmentation method was used to generate complete segmentation results of the blade. After all the blade instances were processed in turn, the final segmentation result of engine blade instance was obtained by merging the results. All the 72 images were used to train the Sparse Instance activation map for real-time instance segmentation (SparseInst), to produce the initial instance segmentation results. The testing dataset contained 56 images. The mean Average Precision (mAP) of the proposed method is higher than that of SparseInst by 5.1 percentage points. The mAP results of the proposed method are better than those of the state-of-the-art instance segmentation methods, e.g., MASK R-CNN (Mask Region based Convolutional Neural Network), YOLACT (You Only Look At CoefficienTs), BMASK-RCNN (Boundary-preserving MASK R-CNN).

Key words: aeroengine, instance segmentation, engine blade, damage detection, interactive segmentation

中图分类号: