식물의 잎의 크기나 면적을 아는 것은 생장을 예측하고 실내농장의 생산성의 향상에 중요한 요소이다. 본 연구에서는 상추 잎 사진을 이용해 엽장과 엽폭을 예측할 수 있는 CNN기반모델을 연...
http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
https://www.riss.kr/link?id=A108818006
2023
Korean
data augmentation ; lettuce imaging ; plant growth ; transfer learning ; vertical farming ; 데이터 증강 ; 상추이미지 ; 수직농장 ; 식물 생장 ; 전이학습
KCI등재
학술저널
434-441(8쪽)
0
상세조회0
다운로드국문 초록 (Abstract)
식물의 잎의 크기나 면적을 아는 것은 생장을 예측하고 실내농장의 생산성의 향상에 중요한 요소이다. 본 연구에서는 상추 잎 사진을 이용해 엽장과 엽폭을 예측할 수 있는 CNN기반모델을 연...
식물의 잎의 크기나 면적을 아는 것은 생장을 예측하고 실내농장의 생산성의 향상에 중요한 요소이다. 본 연구에서는 상추 잎 사진을 이용해 엽장과 엽폭을 예측할 수 있는 CNN기반모델을 연구하였다. 데이터의 한계와 과적합 문제를 극복하기 위해 콜백 함수를 적용하고, 모델의 일반화 능력을 향상시키기 위해 K겹 교차 검증을 사용했다. 또한 데이터 증강을 통한 학습데이터의 다양성을 높이기 위해 image generator를 사용하였다. 모델 성능을 비교하기 위해 VGG16, Resnet152, NASNetMobile 등 사전학습된 모델을 이용하였다. 그 결과너비 예측에서 R2 값 0.9436, RMSE 0.5659를 기록한 NASNetMobile이가장 높은 성능을 보였으며 길이 예측에서는 R2 값이 0.9537, RMSE가 0.8713로 나타났다. 최종 모델에는NASNetMobile 아키텍처, RMSprop 옵티마이저, MSE 손실 함수, ELU 활성화함수가 사용되었다. 모델의 학습 시간은Epoch당 평균 73분이 소요되었으며, 상추 잎 사진 한 장을 처리하는 데 평균 0.29초가 걸렸다. 본 연구는 실내 농장에서 식물의 엽장과 엽폭을 예측하는 CNN 기반 모델을 개발하였고이를 통해 단순한 이미지 촬영만으로도 식물의 생장 상태를신속하고 정확하게 평가할 수 있을 것으로 기대된다. 또한 그결과는 실시간 양액 조절 등의 적절한 농작업 조치를 하는데 활용됨으로써 농장의 생산성 향상과 자원 효율성을 향상시키는데 기여할 것이다.
다국어 초록 (Multilingual Abstract)
Determining the size or area of a plant's leaves is an important factor in predicting plant growth and improving the productivity of indoor farms. In this study, we developed a convolutional neural network (CNN)-based model to accurately predict the l...
Determining the size or area of a plant's leaves is an important factor in predicting plant growth and improving the productivity of indoor farms. In this study, we developed a convolutional neural network (CNN)-based model to accurately predict the length and width of lettuce leaves using photographs of the leaves. A callback function was applied to overcome data limitations and overfitting problems, and K-fold cross-validation was used to improve the generalization ability of the model. In addition, ImageDataGenerator function was used to increase the diversity of training data through data augmentation. To compare model performance, we evaluated pre-trained models such as VGG16, Resnet152, and NASNetMobile. As a result, NASNetMobile showed the highest performance, especially in width prediction, with an R_squared value of 0.9436, and RMSE of 0.5659. In length prediction, the R_squared value was 0.9537, and RMSE of 0.8713. The optimized model adopted the NASNetMobile architecture, the RMSprop optimization tool, the MSE loss functions, and the ELU activation functions. The training time of the model averaged 73 minutes per Epoch, and it took the model an average of 0.29 seconds to process a single lettuce leaf photo. In this study, we developed a CNN-based model to predict the leaf length and leaf width of plants in indoor farms, which is expected to enable rapid and accurate assessment of plant growth status by simply taking images. It is also expected to contribute to increasing the productivity and resource efficiency of farms by taking appropriate agricultural measures such as adjusting nutrient solution in real time.
참고문헌 (Reference)
1 Simonyan K., "Very deep convolutional networks for large-scale image recognition"
2 Commercialization Promotion Agency for R&D Outcome, "S&T Market Report Vol. 69" COMPA 2019
3 Mack L., "Nondestructive leaf area estimation for chia" 109 : 1960-1969, 2017
4 Peksen E, "Non-destructive leaf area estimation model for faba bean (Vicia faba L.)" 113 : 322-328, 2007
5 Souza M. C., "Non-destructive equations to estimate the leaf area of Styrax pohlii and Styrax ferrugineus" 74 : 222-225, 2014
6 Zoph B., "Neural architecture search with reinforcement learning"
7 Zoph B., "Learning transferable architectures for scalable image recognition" 8697-8710, 2018
8 Fakir M. S. A., "Leaf area estimation by linear regression models in pigeonpea(Cajanus cajan(L. )Millsp. )" 11 : 312-316, 2013
9 Kim S. K., "Horticultural crop growth models for smart farms: utilization of descriptive, explanatory, and structural growth models" 59 : 28-37, 2017
10 Zhang L., "Growth monitoring of greenhouse lettuce based on a convolutional neural network" 7 : 124-, 2020
1 Simonyan K., "Very deep convolutional networks for large-scale image recognition"
2 Commercialization Promotion Agency for R&D Outcome, "S&T Market Report Vol. 69" COMPA 2019
3 Mack L., "Nondestructive leaf area estimation for chia" 109 : 1960-1969, 2017
4 Peksen E, "Non-destructive leaf area estimation model for faba bean (Vicia faba L.)" 113 : 322-328, 2007
5 Souza M. C., "Non-destructive equations to estimate the leaf area of Styrax pohlii and Styrax ferrugineus" 74 : 222-225, 2014
6 Zoph B., "Neural architecture search with reinforcement learning"
7 Zoph B., "Learning transferable architectures for scalable image recognition" 8697-8710, 2018
8 Fakir M. S. A., "Leaf area estimation by linear regression models in pigeonpea(Cajanus cajan(L. )Millsp. )" 11 : 312-316, 2013
9 Kim S. K., "Horticultural crop growth models for smart farms: utilization of descriptive, explanatory, and structural growth models" 59 : 28-37, 2017
10 Zhang L., "Growth monitoring of greenhouse lettuce based on a convolutional neural network" 7 : 124-, 2020
11 NICE Information Service Co. Ltd, "GREENPLUS Technical Analysis Report" NICE 2020
12 Gang M. S., "Estimation of greenhouse lettuce growth indices based on a two-stage CNN using RGB-D images" 22 : 5499-, 2022
13 De Lucena L. R. R., "Estimation of cladode area of Nopalea cochenillifera using digital images" 21 : 32-42, 2019
14 Deng Y., "Estimation of Pinus massoniana leaf area using terrestrial laser scanning" 10 : 660-, 2019
15 Zhang W, "Digital image processing method for estimating leaf length and width tested using kiwifruit leaves (Actinidia chinensis Planch)" 15 : e0235499-, 2020
16 He K., "Deep residual learning for image recognition" 770-778, 2016
17 National Assembly Budget Office, "Current status and improvement tasks of the smart agriculture fostering project" NABO 2022
18 Nasiri A., "Automated grapevine cultivar identification via leaf imaging and deep convolutional neural networks : a proof-of-concept study employing primary iranian varieties" 10 : 1628-, 2021
19 Korea Rural Economic Institute, "Agriculture and Rural Economy Trends Spring 2006" KREI 2006
20 Launay M., "Ability for a model to predict crop production variability at the regional scale : an evaluation for sugar beet" 23 : 135-146, 2003
21 National Information Society Agency, "AI INsight Report Vol. 01" NIA 2019
22 Hajjdiab H., "A vision-based approach for nondestructive leaf area estimation" 53-56, 2010
23 Korea Rural Economic Institute, "A study on analyzing the realities of smart farm operations and researching development direction" KREI 2016
24 Boyacı S., "A research on nondestructive leaf area estimation modeling for some apple cultivars" 64 : 1-7, 2022
1-Methylcyclopropene(1-MCP)와 MAP 처리가 저온 저장 중 복숭아 ‘대홍’의 품질 변화에 미치는 영향
밀폐형 식물생산시스템 내 CO2와 광도에 따른 오이 및 토마토 묘의 생육
상추 재배 식물공장의 환경변화에 따른 연중 에너지 요구량 분석
전산유체역학을 이용한 도로 식재 배치 유형에 따른 미세먼지 저감 분석