CNN 기반 토마토의 흰가루병 발병 인식 방법

다국어 초록 (Multilingual Abstract)

The smart greenhouse, which is equipped with an autonomous environmental control system, is attracting attention as an effective alternative to solve problems in the modern agricultural industry. Although the smart greenhouse enables the monitoring of environmental information of the greenhouse, the system to directly capture the state of the crops is required to enhance the productivity of the smart greenhouse. In this paper, we propose an image recognition algorithm based on the convolution neural network (CNN) to detect an outbreak of powdery mildew on tomatoes. We propose a method to artificially generate the powdery mildew images using an image fusion technique to prepare various forms of CNN learning data. The artificial powdery mildew images are produced in three steps: mildew image extraction, transformation, and overlapping. The CNN is learned using these artificial images, and we test the recognition performance of the CNN using real tomato leap images captured in the greenhouse. The experimental results show that the proposed image recognition algorithm presents a recognition rate of 93.02% for 43 test images.

목차 (Table of Contents)

• Abstract
• Ⅰ. 서론
• Ⅱ. 관련연구
• Ⅲ. CNN 기반 흰가루병 영상 인식 방법
• Ⅴ. 흰가루병 영상인식 모듈 성능분석
• Ⅵ. 결론
• REFERENCES

참고문헌 (Reference)

1 전호진, "합성곱 신경망을 이용한 복잡한 형상을 가진 공의 인식" 제어·로봇·시스템학회 23 (23): 354-362, 2017

2 박용규, "컨볼루션 신경망 기반의 차량 전면부 검출 시스템" 제어·로봇·시스템학회 21 (21): 1008-1016, 2015

3 장인훈, "심층인공신경망을 이용한 신발 갑피 패턴인식 자동화 시스템" 제어·로봇·시스템학회 23 (23): 705-710, 2017

4 여욱현, "스마트팜 구현을 위한 연구동향 및 ICT 핵심기술 분석" (사) 한국생물환경조절학회 25 (25): 30-41, 2016

5 M. A. Ebrahimi, "Vision-based pest detection based on SVM classification method" 137 : 52-58, 2017

6 Y. Li, "Vision-based pest detection and automatic spray of greenhouse plant" 2009

7 A. Manickavasagan, "Thermal imaging to detect infestation by cryptolestes ferrugineus inside wheat kernels" 44 (44): 186-192, 2008

8 G. Liu, "Research on prediction about fruit tree diseases and insect pests based on neural network" 731-740, 2005

9 Korea Institute of Science and Technology Evaluation and Planning (KISTEP), "Progress direction and Project of agriculture, fisheries and food R & D to realize smart-agriculture" 3-36, 2015

10 박은수, "MSERs와 CNN 기반의 실시간 교통표지판 인식" 제어·로봇·시스템학회 23 (23): 488-495, 2017

11 M. H. Javed, "K-means based automatic pests detection and classification for pesticides spraying" 8 (8): 236-240, 2017

12 A. Krizhevsky, "Imagenet classification with deep convolutional neural networks" 1097-1105, 2012

13 K. Venugoban, "Image classification of paddy field insect pests using gradient-based features" 4 (4): 1-5, 2014

14 R. K. Priya, "Fuzzy C-Means method for colour image segmentation with L*U*V* colour transformation" 1 (1): 123-127, 2011

15 J. C. Olivo-Marin, "Extraction of spots in biological images using multiscale products" 35 (35): 1989-1996, 2002

16 Y. S. Kim, "Easy-to-understand pests and diseases and physiological disorders" 2015

17 C. B. Singh, "Detection of insect-damaged wheat kernels using near-infrared hyperspectral imaging" 45 (45): 151-158, 2009

18 G. Kandalkar, "Classification of agricultural pests using DWT and back propagation neural networks" 5 (5): 4034-4037, 2014

19 이우영, "CNN 기반 신발갑피 패턴인식 및 점착제 도포를 위한 경로점 생성" 제어·로봇·시스템학회 23 (23): 725-731, 2017

20 S. N. A. Hassan, "Automatic classification of insects using color-based and shape-based descriptors" 2 (2): 2014

21 E. Larios, "Automated insect identification through concatenated histograms of local appearance features" 2007

22 A. Camargo, "An image-processing based algorithm to automatically identify plant disease visual symptoms" 102 (102): 9-21, 2009