Intelligent Damage Detection for Aircraft Engine with Context Encoder Neural Networks
- 1 November 2021
- journal article
- research article
- Published by American Institute of Aeronautics and Astronautics (AIAA) in Journal of Aerospace Information Systems
- Vol. 18 (11), 803-812
- https://doi.org/10.2514/1.i010960
Abstract
It is vital and essential to accurately and timely detect various damages of aircraft engines in civil aviation. Currently, aircraft engines are manually inspected via borescope images by aircraft maintenance technicians. This process is time-consuming and prone to error due to human factors. The aim of this paper is to automate the aircraft engine inspection, and this work presents a deep learning framework with a context encoder neural network structure such that the damaged structures can be accurately segmented from borescope images. Moreover, the proposed network structure is further optimized through an orthogonal-array-based method. With the real borescope images collected from a commercial airline company, the proposed framework is compared with existing deep-learning-based methods from various aspects. The experimental results validate that various damages can be automatically detected and recognized with high accuracy and efficiency by the proposed solution.Funding Information
- National Nature Science Foundation of China (61802176)
This publication has 14 references indexed in Scilit:
- CE-Net: Context Encoder Network for 2D Medical Image SegmentationIEEE Transactions on Medical Imaging, 2019
- UAV-based Inspection of Airplane Exterior Screws with Computer VisionPublished by INSTICC ,2019
- NB-CNN: Deep Learning-Based Crack Detection Using Convolutional Neural Network and Naïve Bayes Data FusionIEEE Transactions on Industrial Electronics, 2017
- Generalised Dice Overlap as a Deep Learning Loss Function for Highly Unbalanced SegmentationsLecture Notes in Computer Science, 2017
- DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFsIEEE Transactions on Pattern Analysis and Machine Intelligence, 2017
- Deep Learning‐Based Crack Damage Detection Using Convolutional Neural NetworksComputer-Aided Civil and Infrastructure Engineering, 2017
- Context Encoders: Feature Learning by InpaintingPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2016
- Deep Residual Learning for Image RecognitionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2016
- A review of human error in aviation maintenance and inspectionInternational Journal of Industrial Ergonomics, 2000
- Ridge Regression: Biased Estimation for Nonorthogonal ProblemsTechnometrics, 2000