DOI QR코드

DOI QR Code

Image Classification of Damaged Bolts using Convolution Neural Networks

합성곱 신경망을 이용한 손상된 볼트의 이미지 분류

  • Lee, Soo-Byoung (School of Mechanical Engineering, Gyeongsang National University) ;
  • Lee, Seok-Soon (School of Mechanical Engineering, Gyeongsang National University)
  • Received : 2022.03.28
  • Accepted : 2022.07.05
  • Published : 2022.08.31

Abstract

The CNN (Convolution Neural Network) algorithm which combines a deep learning technique, and a computer vision technology, makes image classification feasible with the high-performance computing system. In this thesis, the CNN algorithm is applied to the classification problem, by using a typical deep learning framework of TensorFlow and machine learning techniques. The data set required for supervised learning is generated with the same type of bolts. some of which have undamaged threads, but others have damaged threads. The learning model with less quantity data showed good classification performance on detecting damage in a bolt image. Additionally, the model performance is reviewed by altering the quantity of convolution layers, or applying selectively the over and under fitting alleviation algorithm.

딥러닝 기법과 컴퓨터 비전 기술을 융합한 합성곱 신경망 알고리즘은 고성능 컴퓨팅 시스템을 기반으로 이미지 데이터의 분류를 가용하게 한다. 본 논문에서는 합성곱 신경망 알고리즘을 대표적인 딥러닝 프레임워크인 텐서플로와 학습 기법을 이용하여 구현하고 이미지 분류 문제에 적용한다. 모델의 지도학습에 필요한 데이터는 동일 종류의 볼트를 이용하여 나사산이 정상인 볼트와 나사산이 손상된 볼트로 구분하여 이미지를 생성하였다. 소량의 이미지 데이터를 이용한 학습 모델은 좋은 성능으로 볼트의 손상을 탐지하였다. 그리고 모델의 내부 구성에 따른 학습 성능을 비교하기 위해 합성곱 신경망 내 컨볼루션 레이어의 개수를 변경하고 과적합 회피기법을 선택 적용하여 이미지 분류 성능을 확인하였다.

Keywords

References

  1. Patterson, Josh, and Adam Gibson, Deep learning: A practitioner's approach, O'Reilly Media, 2017.
  2. LeCun, Yann, et al, "Gradient-based learning applied to document recognition," Proceedings of the IEEE 86.11, pp. 2278-2324, 1998. https://doi.org/10.1109/5.726791
  3. Sagar Sharma, 2017, "Activation Functions in Neural Networks." Towards Data Science, February 5, 2022. https://towardsdatascience.com/activation-functions-neural-networks-1cbd9f8d91d6.
  4. Srivastava, Nitish, et al, "Dropout: a simple way to prevent neural networks from overfitting," The journal of machine learning research, 15.1, pp. 1929-1958, 2014.
  5. TensorFlow (2022) Classification [Source code]. https://www.tensorflow.org/tutorials/images/classification.
  6. Sreenivas B., (2020) Classification [Source code]. https://github.com/bnsreenu/python_for_microscopists/blob/master/158_classification_CNN_RF.py.