DOI QR코드

DOI QR Code

Object Segmentation for Detection of Moths in the Pheromone Trap Images

페로몬 트랩 영상에서 해충 검출을 위한 객체 분할

  • Kim, Tae-Woo (Division of Electrical, Electronic and Communication Engineering, Hanyang Cyber University) ;
  • Cho, Tae-Kyung (Dept. of Information Security Engineering, Sangmyung University)
  • 김태우 (한양사이버대학교 전기전자통신공학부) ;
  • 조태경 (상명대학교 정보보안공학과)
  • Received : 2017.11.01
  • Accepted : 2017.12.08
  • Published : 2017.12.31

Abstract

The object segmentation approach has the merit of reducing the processing cost required to detect moths of interest, because it applies a moth detection algorithm to the segmented objects after segmenting the objects individually in the moth image. In this paper, an object segmentation method for moth detection in pheromone trap images is proposed. Our method consists of preprocessing, thresholding, morphological filtering, and object labeling processes. Thresholding in the process is a critical step significantly influencing the performance of object segmentation. The proposed method can threshold very elaborately by reflecting the local properties of the moth images. We performed thresholding using global and local versions of Ostu's method and, used the proposed method for the moth images of Carposina sasakii acquired on a pheromone trap placed in an orchard. It was demonstrated that the proposed method could reflect the properties of light and background on the moth images. Also, we performed object segmentation and moth classification for Carposina sasakii images, where the latter process used an SVM classifier with training and classification steps. In the experiments, the proposed method performed the detection of Carposina sasakii for 10 moth images and achieved an average detection rate of 95% of them. Therefore, it was shown that the proposed technique is an effective monitoring method of Carposina sasakii in an orchard.

객체 분할 방식은 객체를 먼저 분할한 후, 검출된 객체에 대해 해충 검출 알고리즘을 적용하므로 해충 개체를 검출하는 데 필요한 처리 비용이 줄어드는 장점이 있다. 본 논문에서는 페로몬 트랩 영상에서 해충 검출을 위한 객체 분할 방법을 제안한다. 제안한 방법은 전처리, 문턱치 처리, 형태학적 필터링, 레이블링 처리로 구성된다. 이들 과정 중 문턱치 처리는 객체 분할의 성능을 좌우하는 매우 중요한 처리 과정이다. 제안한 방법은 문턱치 처리 과정에서 해충 영상의 국소적 특성을 반영하므로 매우 정교한 문턱치 처리를 할 수 있다. 과수원에 설치된 페로몬 트랩에서 수집된 복숭아심식나방 영상에 대해 Otsu의 방법의 전역적 방식과 국소적 방식, 그리고 제안한 방법으로 처리한 결과, 제안한 방법이 조명과 배경의 특성을 잘 반영함을 알 수 있었다. 페로몬 트랩에 수집된 복숭아심식나방 영상에 대해 객체 분할과 개체 분류를 수행하였다. 개체 분류는 SVM 분류기로 학습하여 사용하였다. 실험에서 제안한 방법으로 10개의 해충 영상에 대해 복숭아심식나방 검출 결과 95%의 평균 검출율을 보임으로써 과수원의 복숭아심식나방의 개체 모니터링 방법으로서 효과적임을 보였다.

Keywords

References

  1. Sun-Young Lee, Kyung-Hee Choi, Yun-Su Do, Soon-Won Lee, Changmann Yoon and Gil-Hah Kim, "Management of Grapholita molesta and Carposina sasakii Using Mating Disruption in Non-Chemical or Organic Apple Orchards", Korean J. Appl. Entomol., 53(2), pp. 103-110, 2014. DOI: https://doi.org/10.5656/KSAE.2013.11.0.056
  2. Jeong, Sun-A, Lalit Prasad Sah, Jeong Joon Ahn, Young-il Kim, and Chuleui Jung, "Occurrence Patterns of Three Major Fruit Moths, Grapholita molesta, Grapholita dimorpha and Carposina sasakii, Monitored by Sex Pheromone in Plum Orchards", Korean J. Appl. Entomol., 51(4), pp. 449-459, 2012. DOI: https://doi.org/10.5656/KSAE.2012.10.0.061
  3. Adriano Guarnieri, Stefano Maini, Giovanni Molari, Valda Rondelli, "Automatic trap for moth detection in integrated pest management", Bulletin of Insectology, 64(2), pp. 247-251, 2011.
  4. Young-Seuk Park, Man-Wi Han, Hwang-Yong Kim, Ki-Baik Uhm, Chang-Gyu Park, JangMyung Lee, and Tae-Soo Chon, "Density Estimation of Rise Planthoppers Using Digital Image Processing Algorithm", Korean J. Appl. Entomol., 42(1), pp.57-63, 2003.
  5. Chang Bae Moon, Byeong Man Kim, Jong Yeol Yi, Jae Wook Hyun, and Pyoung Ho Yi, "Detection of Candidate Areas for Automatic Identification of Scirtothrips Dorsalis", J Korea Industr Inf Syst Res, Volume 17, Number 6, pp.51-58, 2012. DOI: https://doi.org/10.9723/jksiis.2012.17.6.051
  6. Chenglu Wen, Daniel E. Guyer, Wei Li, "Local feature-based identification and classification for orchard insects", Biosystems Engineering, 104, pp.299-307, 2009. DOI: https://doi.org/10.1016/j.biosystemseng.2009.07.002
  7. Weiguang Diang, Graham Taylor, "Automatic moth detection from trap images for pest management", Computers and Electronics in Agriculture, 123, pp.17-28, 2016. DOI: https://doi.org/10.1016/j.compag.2016.02.003
  8. Chenglu Wen, Daniel Guyer, "Image-based orchard insect automated identification and classification method", Computers and Electronics in Agriculture, 89, pp.110-115, 2012. DOI: https://doi.org/10.1016/j.compag.2012.08.008
  9. Hyeon-Joong Yoo, Tae-Woo Kim, and Chun-Seok Oh, Digital Image Processing, Firstbook, 2013.
  10. Nobuyuki Otsu, "A Threshold Selection Method from Gray-Level Histogram", IEEE Transactions on Systems, Man, and Cybernetics, 9(1), pp.62-66, 1979. DOI: https://doi.org/10.1109/TSMC.1979.4310076