The Journal of Bigdata (한국빅데이터학회지)

Korea Bigdata Society (한국빅데이터학회)
권호 표지
DOI QR코드

DOI QR Code

A Study on GPR Image Classification by Semi-supervised Learning with CNN

CNN 기반의 준지도학습을 활용한 GPR 이미지 분류

  • 김혜미 (부산대학교 산업공학과 산업데이터공학융합전공) ;
  • 배혜림 (부산대학교 산업공학과 산업데이터공학융합전공)
  • Received : 2021.07.23
  • Accepted : 2021.08.23
  • Published : 2021.08.31
Download PDF
⟨ Previous Next ⟩

Abstract

GPR data is used for underground exploration. The data gathered are interpreted by experts based on experience as the underground facilities often reflect GPR. In addition, GPR data are different in the noise and characteristics of the data depending on the equipment, environment, etc. This often results in insufficient data with accurate labels. Generally, a large amount of training data have to be obtained to apply CNN models that exhibit high performance in image classification problems. However, due to the characteristics of GPR data, it makes difficult to obtain sufficient data. Finally, this makes neural networks unable to learn based on general supervised learning methods. This paper proposes an image classification method considering data characteristics to ensure that the accuracy of each label is similar. The proposed method is based on semi-supervised learning, and the image is classified using clustering techniques after extracting the feature values of the image from the neural network. This method can be utilized not only when the amount of the labeled data is insufficient, but also when labels that depend on the data are not highly reliable.

GPR(Ground Penetrating Radar)에서 수집된 데이터는 지하 탐사를 위해 사용된다. 이 때, 지반 아래의 시설물들이 GPR을 반사하는 경우가 종종 발생하여 수집된 데이터는 전문가에 경험에 의존하여 해석된다. 또한, GPR 데이터는 수집 장비, 환경 등에 따라 데이터의 노이즈, 특성 등이 다르게 나타난다. 이로 인해 정확한 레이블을 가지는 데이터가 충분히 확보되지 못하는 경우가 많다. 일반적으로 이미지 분류 문제에서 높은 성능을 보이는 인공신경망 모델을 적용하기 위해서는 많은 양의 학습 데이터가 확보되어야 한다. 그러나 GPR 데이터의 특성 상 데이터에 정확한 레이블을 붙이는 것은 많은 비용을 필요로 하여 충분한 데이터를 확보하기가 어렵다. 이는 결국 일반적으로 활용되는 지도학습 방법을 기반으로 인공신경망을 적절히 학습시킬 수 없게 한다. 본 논문에서는 각 레이블의 정확도가 유사한 수준을 갖도록 하는 것을 목표로 데이터 특성을 바탕으로 하는 이미지 분류 방법을 제안한다. 제안 방법은 준지도학습을 기반으로 하고 있으며, 인공신경망으로부터 이미지의 특징값을 추출한 후 클러스터링 기법을 활용하여 이미지를 분류한다. 이 방법은 라벨링 된 데이터가 충분하지 않은 경우 라벨링할 때 뿐 만 아니라 데이터에 달린 레이블의 신뢰도가 높지 않은 경우에도 활용할 수 있다.

Keywords

Acknowledgement

이 논문은 2020년도 정부(과학기술정보통신부)의 재원으로 한국연구재단의 지원을 받아 수행된 연구임 (No. 2020R1A2C110229411).

References

  1. B. Park, et al. "Underground object classification for urban roads using instantaneous phase analysis of ground-penetrating radar (GPR) data." Remote Sensing Vol.10, No.9, pp.1417, 2018. https://doi.org/10.3390/rs10091417
  2. C. Tan, et al. "A survey on deep transfer learning." International conference on artificial neural networks. Springer, Cham, 2018.
  3. E. Pasolli, M. Farid, and D. Massimo, "Automatic analysis of GPR images: A pattern-recognition approach." IEEE Transactions on Geoscience and Remote Sensing Vol.47, No.7, pp.2206-2217, 2009. https://doi.org/10.1109/TGRS.2009.2012701
  4. El-Mahallawy, S. Mohamed, and H. Mazlan. "Material classification of underground utilities from GPR images using DCT-based SVM approach." IEEE Geoscience and Remote Sensing Letters Vol.10, No.6, pp.1542-1546, 2013. https://doi.org/10.1109/LGRS.2013.2261796
  5. E. Pasolli, et al. "Automatic detection and classification of buried objects in GPR images using genetic algorithms and support vector machines." IGARSS 2008-2008 IEEE International Geoscience and Remote Sensing Symposium. Vol. 2, 2008.
  6. R. N. Nagashree, et al. "Detection and classification of ground penetrating radar image using textrual features." 2014 International Conference on Advances in Electronics Computers and Communications. IEEE, 2014.
  7. Ozkaya, Umut, and Levent Seyfi. "Deep dictionary learning application in GPR B-scan images." Signal, Image and Video Processing Vol.12, No.8, pp.1567-1575, 2018. https://doi.org/10.1007/s11760-018-1313-x
  8. Lameri, Silvia, et al. "Landmine detection from GPR data using convolutional neural networks." 2017 25th European Signal Processing Conference (EUSIPCO). IEEE, 2017.
  9. Elsaadouny, Mostafa, J. Barowski, and I. Rolfes. "ConvNet transfer learning for GPR images classification." 2020 German Microwave Conference (GeMiC). IEEE, 2020.
  10. Z. Gong, Zhimin, and H. Zhang. "Research on GPR image recognition based on deep learning." MATEC Web of Conferences. Vol. 309. EDP Sciences, 2020.
  11. H. Harkat, et al. "GPR target detection using a neural network classifier designed by a multi-objective genetic algorithm." Applied Soft Computing Vol.79, pp.310-325, 2019. https://doi.org/10.1016/j.asoc.2019.03.030
  12. U. Ozkaya, et al. "GPR B scan image analysis with deep learning methods." Measurement Vol.165, pp.107770, 2020. https://doi.org/10.1016/j.measurement.2020.107770
  13. A. Bhuiyan, and B. Nath. "Anti-personnel mine detection and classification using GPR image." 18th International Conference on Pattern Recognition (ICPR'06). Vol. 2. IEEE, 2006.
  14. X. Zhu, and A. B. Goldberg. "Introduction to semi-supervised learning." Synthesis lectures on artificial intelligence and machine learning Vol.3, No.1 pp.1-130, 2009.
  15. Chang, Jianlong, et al. "Deep adaptive image clustering." Proceedings of the IEEE international conference on computer vision. 2017.
  16. Coleman, Guy Barrett, and Harry C. Andrews. "Image segmentation by clustering." Proceedings of the IEEE Vol.67, No.5, pp.773-785, 1979. https://doi.org/10.1109/PROC.1979.11327
  17. Vaswani, Ashish, et al. "Attention is all you need." Advances in neural information processing systems. 2017.
  18. C. Tian, et al. "Attention-guided CNN for image denoising." Neural Networks Vol.124, pp.117-129, 2020. https://doi.org/10.1016/j.neunet.2019.12.024
  19. Perkins, N. David, and G. Salomon. "Transfer of learning." International encyclopedia of education 2, pp.6452-6457, 1992.
  20. F. Zhuang, et al. "A comprehensive survey on transfer learning." Proceedings of the IEEE Vol.109, No.1, pp.43-76, 2020. https://doi.org/10.1109/JPROC.2020.3004555
  21. Hussain, Mahbub, Jordan J. Bird, and R. Faria. "A study on cnn transfer learning for image classification." UK Workshop on computational Intelligence. Springer, Cham, 2018.
  22. I. Goodfellow, Y. Bengio, and A. Courville. Deep learning. MIT press, 2016.
  23. Hartigan, A. John, A. and W. A. Manchek, "Algorithm AS 136: A k-means clustering algorithm." Journal of the royal statistical society. series c (applied statistics) Vol.28, No.1, pp.100-108, 1979.
  24. https://www.tensorflow.org/tutorials/images/classification?hl=ko