방송공학회논문지 (Journal of Broadcast Engineering)

  • 제25권2호
  • /
  • Pages.208-217
  • /
  • 2020
  • /
  • 1226-7953(pISSN)
  • /
  • 2287-9137(eISSN)
한국방송∙미디어공학회 (The Korean Institute of Broadcast and Media Engineers)
권호 표지
DOI QR코드

DOI QR Code

합성곱 신경망을 이용하는 수퍼픽셀 기반 사과잎 병충해의 분류

Superpixel-based Apple Leaf Disease Classification using Convolutional Neural Network

  • 김만배 (강원대학교 컴퓨터정보통신공학과) ;
  • 최창열 (강원대학교 컴퓨터정보통신공학과)
  • Kim, Manbae (Dept. of Computer and Communications Engineering, Kangwon National University) ;
  • Choi, Changyeol (Dept. of Computer and Communications Engineering, Kangwon National University)
  • 투고 : 2020.02.25
  • 심사 : 2020.03.23
  • 발행 : 2020.03.30
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

원예작물을 카메라로 촬영하여 병해충의 종류를 판단하려는 연구가 오랫동안 있어왔다. 일반적으로 영역분할로 병해충 영역을 추출하고, 통계적 특징을 추출한 후 다양한 기계학습 기법으로 병해충 종류를 판단한다. 최근에는 딥러닝의 종단간 학습으로 병해충을 판별하는 연구가 많이 진행되고 있다. 영역분할은 조명 등의 주변 환경 변화에 따라 만족스러운 성능이 어렵고, 전체 잎 영상을 사용하는 종단간 신경망은 학습 영상과 실제 영상과의 차이 때문에 실제 적용이 어려운 문제가 있다. 이를 해결하기 위해서 본 논문에서는 수퍼픽셀 및 합성곱신경망을 이용하는 병해충 분류 방법을 제안한다. 실험에서는 PlantVilllage의 사과 병충해 영상들을 이용하여 실험한 결과, 분류정확도는 전체영상과 수퍼픽셀이 각각 (98.29, 92.43)%이고, 다변량 F1-score는 각각 (0.98. 0.93)이다. 제안하는 수퍼픽셀 기법은 성능 측면에서 약간 저하되지만, 현실적으로 실제 환경에서 적용 가능함을 확인하였다.

The classification of plant diseases by images captured by a camera sensor has been studied over past decades. A method that has gained much interest is to use image segmentation, from which statistical features are derived and analyzed by machine learning. Recently, deep learning has been adopted in this area. However, image segmentation is still a difficult task to achieve stable performance due to a variety of environmental variations. The end-to-end learning in neural network has a demerit that train images may be different from real images acquired in outdoor fields. To solve these problems, we propose superpixel-based disease classification method using end-to-end CNN (convolutional neural network) learning. Based on experiments performed on PlantVillage apple images, the classification accuracy is 98.29% and 92.43% for full-image and superpixel. As well, the multivariate F1-score is (0.98, 0.93). Therefore we validate that the method of using superpixel is comparable to that of full-image.

키워드

참고문헌

  1. S. Chouhan, A. Kaul, U. Singh, and S. Jaini, "Bacterial foraging optimization based radial basis function neural network (BRBFNN) for identification and classification of plant leaf diseases: an automatic approach towards plant pathology", IEEE Access, Digital Object Identifier 10.1109/ACCESS.2018.280068, Feb. 2018.
  2. M. Khan, I. Lali, M. Sharif, K. Javed, K. Aurangzeb, S. Haider, A. Altmarah, and T. Akram, "An optimized method for segmentation and classification of apple diseases based on strong correlation and genetic algorithm based feature selection", IEEE Access, Digital Object Identifier 10.1109/ACCESS.2019.2908040, March 28, 2019.
  3. K. Jagan. M. Balasubramanian. "Recognition of paddy plant diseases based on histogram oriented gradient features", International Journal of Advanced Research in Computer and Communication Engineering, Vol. 5, Issue 3, March 2016.
  4. S. Yun, W. Xiangfeng, Z. Shanwen, and Z. Chuanlei. "PNN based crop disease recognition with leaf image features and meteorological data", Int. J. Agric & Biol. Eng., Vol. 8, No. 4, Aug. 2015.
  5. G. Wang, Y. Sun, and J. Wang, "Automatic image-based plant disease severity estimation using deep learning", Hindawi Computational Intelligence and Neuroscience, Vol. 2017, pp. 1-8, Article ID 2917536.
  6. S. P. Mohanty, D. P. Hughes, and M. Salathe, "Using deep learning for image-based plant disease detection," Frontiers in Plant Science, Vol. 7, Article 1419, Sep. 2016.
  7. H. Lim, Y. Lee, M. Ji, H. Kim, W. Kim, "Efficient inference of image objects using semantic segmentation", Journal of Broadcast Engineering, Vol. 24, No. 1, pp. 67-76, Jan. 2019. https://doi.org/10.5909/JBE.2019.24.1.67
  8. J. Pujari, R. Yakkundimath and A. Byadgi, "Classification of fungal disease symptoms affected on cereals using color texture features.", International Journal of Signal Processing, Vol. 6, No. 6, pp. 321-330, June 2013.
  9. R. Achanta, A. Shaji, K. Smith, A. Lucchi, P. Fua and S. Susstrunk, "SLIC superpixels compared to state-of-the-art superpixel methods", IEEE Trans. on Pattern Analysis and Machine Intelligence, Vol. 34, No. 11, pp. 2274-2281, Nov. 2012. https://doi.org/10.1109/TPAMI.2012.120
  10. S. Lee, C. Choi, and M. Kim, "CNN-based people recognition for vision occupancy sensors", Journal of Broadcast Engineering, Vol. 23, No. 2, March 2018, pp. 274-282. https://doi.org/10.5909/JBE.2018.23.2.274
  11. E. Kim and W. Kim, "Face anti-spoofing based on combination of luminance and chrominance with convolutional neural networks", Journal of Broadcast Engineering, Vol. 24, No. 6, pp. 1113-1121, Nov. 2018. https://doi.org/10.5909/JBE.2019.24.6.1113
  12. J. Yuna, H. Nagaharab, and I. Park, "Classification and restoration of compositely degraded Images using deep learning", Journal of Broadcast Engineering, Vol. 24, No. 3, May 2019.
  13. D. Kingma and J. Ba, "ADAM: A Method for stochastic optimization", Int' Conf. Learning Representations (ICLR), May 2015.
  14. D. Hughes and M. Salathe, "An open access repository of images on plant health to enable the development of mobile disease diagnostics through machine learning and crowd sourcing," arXiv:1511.08060 [cs.CY], Nov. 2015. [Online] Available: http://arxiv.org/abs/1511.08060