Aircraft Fuselage Defect Detection using Deep Neural Networks

To ensure flight safety of aircraft structures, it is necessary to have regular maintenance using visual and nondestructive inspection (NDI) methods. In this paper, we propose an automatic image-based aircraft defect detection using Deep Neural Networks (DNNs). To the best of our knowledge, this is the first work for aircraft defect detection using DNNs. We perform a comprehensive evaluation of state-of-the-art feature descriptors and show that the best performance is achieved by vgg-f DNN as feature extractor with a linear SVM classifier. To reduce the processing time, we propose to apply SURF key point detector to identify defect patch candidates. Our experiment results suggest that we can achieve over 96% accuracy at around 15s processing time for a high-resolution (20-megapixel) image on a laptop.

[1]  Andrea Vedaldi,et al.  MatConvNet: Convolutional Neural Networks for MATLAB , 2014, ACM Multimedia.

[2]  Ngai-Man Cheung,et al.  Smartphone and Mobile Image Processing for Assisted Living: Health-monitoring apps powered by advanced mobile imaging algorithms , 2016, IEEE Signal Processing Magazine.

[3]  Colin G. Drury,et al.  Computer-Simulated Aircraft Inspection Tasks for Off-Line Experimentation , 1992 .

[4]  Ngai-Man Cheung,et al.  On classification of distorted images with deep convolutional neural networks , 2017, 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[5]  Yoshua Bengio,et al.  How transferable are features in deep neural networks? , 2014, NIPS.

[6]  Stefan Carlsson,et al.  CNN Features Off-the-Shelf: An Astounding Baseline for Recognition , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops.

[7]  Colin G. Drury,et al.  Task Analysis of Aircraft Inspection Activities: Methods and Findings , 1990 .

[8]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[9]  Peter Cawley,et al.  The Potential of Guided Waves for Monitoring Large Areas of Metallic Aircraft Fuselage Structure , 2001 .

[10]  Jürgen Schmidhuber,et al.  Multi-column deep neural networks for image classification , 2012, 2012 IEEE Conference on Computer Vision and Pattern Recognition.

[11]  Ngai-Man Cheung,et al.  Deepmole: Deep neural networks for skin mole lesion classification , 2016, 2016 IEEE International Conference on Image Processing (ICIP).

[12]  Domingo Mery,et al.  Automatic Defect Recognition in X-Ray Testing Using Computer Vision , 2017, 2017 IEEE Winter Conference on Applications of Computer Vision (WACV).

[13]  Bart De Schutter,et al.  Deep convolutional neural networks for detection of rail surface defects , 2016, 2016 International Joint Conference on Neural Networks (IJCNN).

[14]  Andrew Zisserman,et al.  Return of the Devil in the Details: Delving Deep into Convolutional Nets , 2014, BMVC.

[15]  Ngai-Man Cheung,et al.  Image-based vehicle analysis using deep neural network: A systematic study , 2016, 2016 IEEE International Conference on Digital Signal Processing (DSP).

[16]  Zhenhua Guo,et al.  A Completed Modeling of Local Binary Pattern Operator for Texture Classification , 2010, IEEE Transactions on Image Processing.

[17]  Luc Van Gool,et al.  SURF: Speeded Up Robust Features , 2006, ECCV.

[18]  Prasad V. Prabhu,et al.  A review of human error in aviation maintenance and inspection , 2000 .

[19]  Ngai-Man Cheung,et al.  Deep neural networks on graph signals for brain imaging analysis , 2017, 2017 IEEE International Conference on Image Processing (ICIP).

[20]  Roderick K. Stanley,et al.  Nondestructive Evaluation: A Tool in Design, Manufacturing and Service , 2018 .

[21]  Bartosz Krawczyk,et al.  Learning from imbalanced data: open challenges and future directions , 2016, Progress in Artificial Intelligence.