DOI QR코드

DOI QR Code

A Smart Farm Environment Optimization and Yield Prediction Platform based on IoT and Deep Learning

IoT 및 딥 러닝 기반 스마트 팜 환경 최적화 및 수확량 예측 플랫폼

  • Received : 2019.11.16
  • Accepted : 2019.12.23
  • Published : 2019.12.31

Abstract

This paper proposes "A Smart Farm Environment Optimization and Yield Prediction Platform based on IoT and Deep Learning" which gathers bio-sensor data from farms, diagnoses the diseases of growing crops, and predicts the year's harvest. The platform collects all the information currently available such as weather and soil microbes, optimizes the farm environment so that the crops can grow well, diagnoses the crop's diseases by using the leaves of the crops being grown on the farm, and predicts this year's harvest by using all the information on the farm. The result shows that the average accuracy of the AEOM is about 15% higher than that of the RF and about 8% higher than the GBD. Although data increases, the accuracy is reduced less than that of the RF or GBD. The linear regression shows that the slope of accuracy is -3.641E-4 for the ReLU, -4.0710E-4 for the Sigmoid, and -7.4534E-4 for the step function. Therefore, as the amount of test data increases, the ReLU is more accurate than the other two activation functions. This paper is a platform for managing the entire farm and, if introduced to actual farms, will greatly contribute to the development of smart farms in Korea.

본 논문은 농장의 바이오 센서 데이터를 수집해서 농장에서 재배중인 농작물의 질병을 진단하고, 그 해 수확량을 예측하는 IoT 및 딥 러닝 기반 스마트 팜 환경 최적화 및 수확량 예측 플랫폼을 제안한다. 이 플랫폼은 현재 날씨, 토양 미생물 등 수집 가능한 모든 정보를 수집하여 작물이 잘 성장할 수 있도록 농장 환경을 최적화하고, 농장에서 재배중인 작물의 잎을 이용하여 작물의 질병을 진단하고, 그리고, 농장의 모든 정보를 사용하여 올해 수확량을 예측한다. 실험 결과 AEOM(Agricultural Environment Optimization Module)의 평균 정확도는 RF(Random Forest)보다 약 15%, GBD(Gradient Boosting Tree)보다 약 8% 높고, 데이터가 증가해도 RF나 GBD에 비해 정확도가 덜 감소한다. 선형 회귀에 따르면 정확도의 기울기는 ReLU의 경우 -3.641E-4, Sigmoid의 경우 -4.0710E-4, 계단함수의 경우 -7.4534E-4이다. 따라서 ReLU 사용시 정확도 기울기가 가장 낮으므로 테스트 데이터의 양이 증가함에 따라 ReLU는 다른 두 가지 활성화 기능보다 더 정확하다. 본 논문에서 제안한 EOYPP는 농장 전체를 관리하는 플랫폼으로 실제 농장에 도입된다면 국내 스마트 팜의 발전에 크게 이바지할 것이다.

Keywords

References

  1. Choi, Hwi-Min, Kim, Joo-Man, "Anomaly Detection System of Smart Farm ICT Device", The Institute of Internet, Broadcasting and Communication, Vol. 19, No. 2, pp. 169-174, 2019.
  2. Sung-Il Hwang, Jong-Moon Joo, Seong-Yong Joo "ICT-Based Smart Farm Factory Systems through the Case of Hydroponic Ginseng Plant Factory", J The Journal of Korean Institute of Communications and Information Sciences, Vol 40, No. 4, pp. 780-790, 2015. https://doi.org/10.7840/kics.2015.40.4.780
  3. Sihyun Lee, "Fundamental Functionality Design of a Smart Farm Using an Embedded Computing Platform", Journal of the Institute of Electronics Engineers of Korea, Vol 55, No. 4, pp. 151-157, 2018.
  4. Wonyoung An, YunHi Chang, "A Study on the Livestock Feed Measuring Sensor and Supply Management System Implementation based on the IoT.", kiiect, Vol. 10, No. 5, pp. 442-454, 2017.
  5. Yongju Park, Jun Moon, "Smart Dairy Management System Development Using Biometric/Environmental Sensors and Farm Control Gateway.", IEMEK Journal of Embedded Systems and Applications, Vol. 11, No. 1, pp. 15-20, 2016. https://doi.org/10.14372/IEMEK.2016.11.1.15
  6. Kim Young-hwan, Kim Tae-wook, Won Hyun-kyu, Lee Kyung-hak, Shin Man-yong, "Estimation of Timber Production by Thinning Scenarios Using a Forest Stand Yield Model.", Korean Society Of Forest Science, Vol. 101, No. 4, pp. 592-598, 2012.
  7. Jae Hun Sung, Dong Hun Lee, "Rice Yield Prediction Based on the Soil Chemical Properties Using Neural Network Model.", Korean Society for Agricultural Machin, Vol. 30, No. 6, pp. 360-365, 2005.
  8. Ju-Hee Kim, Seong-Soo Cheong, Ju-Rak Yim, Ki-Kwon Lee, "Yield Loss Assessment and Determination of Control Thresholds for Powdery Mildew of Chili pepper (Capsicum annuum L).", The Korean society of pesticide science, Vol. 19, No. 2, pp. 113-118, 2015. https://doi.org/10.7585/kjps.2015.19.2.113
  9. Jung-won Oh, Haeng-gon Kim, Il-tae Kim, "Design and implementation of fruit harvest time predicting system based on machine learning.", Smart media journal, Vol. 8, No. 1, pp. 74-71, 2019.
  10. Yeong Mo Son, Jin Taek Kang, Jeong Sun Hwang, Park Hyun, Kang Su Lee, "Assessment and Prediction of Stand Yield in Cryptomeria japonica Stands.", Korean Society Of Forest Science, Vol. 104, No. 3, pp. 421-426, 2015. https://doi.org/10.14578/jkfs.2015.104.3.421