Machine learning application for predicting the strawberry harvesting time
-
- Korean Journal of Agricultural Science
- /
- v.46 no.2
- /
- pp.381-393
- /
- 2019
A smart farm is a system that combines information and communication technology (ICT), internet of things (IoT), and agricultural technology that enable a farm to operate with minimal labor and to automatically control of a greenhouse environment. Machine learning based on recently data-driven techniques has emerged with big data technologies and high-performance computing to create opportunities to quantify data intensive processes in agricultural operational environments. This paper presents research on the application of machine learning technology to diagnose the growth status of crops and predicting the harvest time of strawberries in a greenhouse according to image processing techniques. To classify the growth stages of the strawberries, we used object inference and detection with machine learning model based on deep learning neural networks and TensorFlow. The classification accuracy was compared based on the training data volume and training epoch. As a result, it was able to classify with an accuracy of over 90% with 200 training images and 8,000 training steps. The detection and classification of the strawberry maturities could be identified with an accuracy of over 90% at the mature and over mature stages of the strawberries. Concurrently, the experimental results are promising, and they show that this approach can be applied to develop a machine learning model for predicting the strawberry harvesting time and can be used to provide key decision support information to both farmers and policy makers about optimal harvest times and harvest planning.
The cultivation of Sparassis crispa in the beginning of the 2000s in South Korea, and the cultivar 'Neowul' bred in the Chonbuk Agricultural Research and Extension Service were registered first in 2016. However, there is no manual for the cultivation of Sparassis crispa, and therefore, there remains a big difference in its harvest rate across farms. Herein, we aimed to study the primordium formation conditions of Sparassis crispa 'Neowul' according to the medium pH, medium moisture content, and inoculum volume of liquid spawn and develop a stable production technology. We found the annual yield per bottle relating to the cultivation period, harvest rate, and the weight of fruiting body to be the highest at 363.6 g in the area cultivated at pH 3.8. However, it is thought that cultivation by adjusting the pH to 3.9±1 would be necessary for stable production, considering that at pH 3.6, the yield sharply reduced to 189.5 g. Moreover, the culture period was shorter at pH 4.0 compared with that at pH3.8, and the cultivation period at pH 4.0 was the same as that at pH 3.8. No significant difference in the weight of the fruiting body at different conditions was recognized. Additionally, it is difficult to regulate the pH precisely in practical applications in the farms. It is thought that 341.8 g Sparassis crispa will be produced per bottle annually if the medium moisture content is adjusted to 65%, liquid spawn inoculum volume is equivalent to 4% of the medium volume, and the humidity in the culture room is set to below 50%.
This paper aimed to suggest decision-making method for forest management planning using integer programming. Thus, the study examined 85 stands consisting of Korean pine, Japanese larch and oak stands-all of which were at the most suitable time for tending, selection thinning, commercial thinning and final cutting-in the experimental forest of Kangwon National University. The forest management model comprised one objective function, maximizing harvest volume in each stand according to tree species and the kinds of practices, and seven constraints: frequency and stands of practices, minimum and maximum yields, even yields, maximum production, and decision-making varialbes. Besides, the entire period intended by the study was 10 years, divided into 5 management periods. In conclusion, the forest management planning model using integer programming proved that among 85 stands, forest practices were conducted over 68 stands (202.8 ha), producing the total harveted volume of
This article investigates the energy harvesting characteristics of a magneto-electro-elastic (MEE) cantilever beam reinforced with carbon nanotubes (CNT) under transverse vibration. To this end, the well-known lumped parameter model is used to represent the coupled multiphysics problem mathematically. The proposed system consists of the MEE-CNT layer on top and an inactive substrate layer at the bottom. The substrate is considered to be made of either an isotropic or composite material. Basic laws such as Gauss's Law, Newton's Law and Faraday's Law are used to arrive at the governing equations. Surface electrodes across the beam are used to harvest the electric potential produced, together with a wound coil, for the generated magnetic potential. The influence of various distributions of the CNT and its volume fraction, substrate material, length-to-thickness ratio, and thickness ratio of substrate to MEE layer on the energy harvesting behaviour is thoroughly discussed. Further, the effect of external resistances and changes in substrate material on the response is analysed and reported. The article aims to explore smart material-based energy harvesting systems, focusing on their behaviour when reinforced with carbon nanotubes. The results of this study may lead to an improved understanding of the design and analysis of CNT-based smart structures.
Agriculture is a primary industry that influenced by the weather or meterological factors more than other industry. Global warming and worldwide climate changes, and unusual weather phenomena are fatal in agricultural industry and human life. Therefore, many previous studies have been made to find the relationship between weather and the productivity of agriculture. Meterological factors also influence on the distribution of agricultural product. For example, price of agricultural product is determined in the market, and also influenced by the weather of the market. However, there is only a few study was made to find this link. The objective of this study is to investigate the effects of meterological factors on the distribution of agricultural products, focusing on the distribution of chinese cabbages. Chinese cabbage is a main ingredient of Kimchi, and basic essential vegetable in Korean dinner table. However, the production of chinese cabbages is influenced by weather and very fluctuating so that the variation of its price is so unstable. Therefore, both consumers and farmers do not feel comfortable at the unstable price of chinese cabbages. In this study, we analyze the real transaction data of chinese cabbage in wholesale markets and meterological factors depending on the variety and geography. We collect and analyze data of meterological factors such as temperatures, humidity, cloudiness, rainfall, snowfall, wind speed, insolation, sunshine duration in producing and consuming region of chinese cabbages. The result of this study shows that the meterological factors such as temperature and humidity significantly influence on the volume and price of chinese cabbage transaction in wholesale market. Especially, the weather of consuming region has greater correlation effects on transaction than that of producing region in all types of chinese cabbages. Among the whole agricultural lifecycle of chinese cabbages, 'seeding - harvest - shipment - wholesale', meterological factors such as temperature and rainfall in shipment and wholesale period are significantly correlated with transaction volume and price of crops. Based on the result of correlation analysis, we make a regression analysis to verify the meterological factors' effects on the volume and price of chines cabbage transaction in wholesale market. The results of stepwise regression analysis are shown in