• Smart Media Journal
• /
• v.8 no.1
• /
• pp.74-81
• /
• 2019

Recently, machine learning technology has had a significant impact on society, particularly in the medical, manufacturing, marketing, finance, broadcasting, and agricultural aspects of human lives. In this paper, we study how to apply machine learning techniques to foods, which have the greatest influence on the human survival. In the field of Smart Farm, which integrates the Internet of Things (IoT) technology into agriculture, we focus on optimizing the crop growth environment by monitoring the growth environment in real time. KT Smart Farm Solution 2.0 has adopted machine learning to optimize temperature and humidity in the greenhouse. Most existing smart farm businesses mainly focus on controlling the growth environment and improving productivity. On the other hand, in this study, we are studying how to apply machine learning with respect to harvest time so that we will be able to harvest fruits of the highest quality and ship them at an excellent cost. In order to apply machine learning techniques to the field of smart farms, it is important to acquire abundant voluminous data. Therefore, to apply accurate machine learning technology, it is necessary to continuously collect large data. Therefore, the color, value, internal temperature, and moisture of greenhouse-grown fruits are collected and secured in real time using color, weight, and temperature/humidity sensors. The proposed FPSML provides an architecture that can be used repeatedly for a similar fruit crop. It allows for a more accurate harvest time as massive data is accumulated continuously.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.22 no.4
• /
• pp.239-249
• /
• 2020

Rice flour varieties have been developed to replace wheat, and consumption of rice flour has been encouraged. damage related to pre-harvest sprouting was occurring due to a weather disaster during the ripening period. Thus, it is necessary to develop pre-harvest sprouting rate prediction system to minimize damage for pre-harvest sprouting. Rice cultivation experiments from 20 17 to 20 19 were conducted with three rice flour varieties at six regions in Gangwon-do, Chungcheongbuk-do, and Gyeongsangbuk-do. Survey components were the heading date and pre-harvest sprouting at the harvest date. The weather data were collected daily mean temperature, relative humidity, and rainfall using Automated Synoptic Observing System (ASOS) with the same region name. Gradient Boosting Machine (GBM) which is a machine learning model, was used to predict the pre-harvest sprouting rate, and the training input variables were mean temperature, relative humidity, and total rainfall. Also, the experiment for the period from days after the heading date (DAH) to the subsequent period (DA2H) was conducted to establish the period related to pre-harvest sprouting. The data were divided into training-set and vali-set for calibration of period related to pre-harvest sprouting, and test-set for validation. The result for training-set and vali-set showed the highest score for a period of 22 DAH and 24 DA2H. The result for test-set tended to overpredict pre-harvest sprouting rate on a section smaller than 3.0 %. However, the result showed a high prediction performance (R2=0.76). Therefore, it is expected that the pre-harvest sprouting rate could be able to easily predict with weather components for a specific period using machine learning.

• Journal of Biosystems Engineering
• /
• v.41 no.3
• /
• pp.177-183
• /
• 2016

Purpose: Mee (Madhuca longifolia) is an economically important tree growing throughout Sri Lanka. Its importance is mainly attributed to its oil with high nutritional and medicinal values. However, an inefficient extraction method limits its use. This study revealed the possibility of extracting oil from mee seeds by using a screw-type oil expeller. Methods: A popular screw-type oil expeller was used in the experiment. Extract bar clearance and speeds of the main spiral shaft were altered to increase the oil expelling efficiency of the machine. The quality of refined oil at the optimum oil yield was determined by measuring the refractive index, saponification value, iodine value, unsaponifiable matter, free fatty acid, and specific gravity. Results: An optimum yield of 35% oil was obtained when the machine capacity was 30 kg/h and energy consumption was 0.13 kWh/kg. This optimum machine condition was observed at an extract bar clearance of 0.5 mm and a main spiral shaft speed of 90 rpm. The refractive index, saponification value, iodine value, unsaponifiable matter, free fatty acid, and specific gravity of the oil were 1.4, 203, 59, 3.5%, 0.2%, and 0.907 g/cm3 respectively. Color of the mee oil was closer to yellow, which is revealed by the lightness value (L) of 24.93 and positive value (b) of 11.81. Conclusion: The screw-type oil expeller can be used for economically extracting mee oil on a commercial scale.

• 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.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 2003.04a
• /
• pp.45-49
• /
• 2003

According to the rising of national economic level, domestic consumption of vegetables having high additive values is increased continuously due to increased consumption of meat in last decade. These vegetables are produced almost in this country and are limited to import from neighbor countries in due of high transportation expenses for storing in refrigerated container. It is very important to mechanize the harvest work, forming more than 30% for their production cost, in order to cultivate variable vegetables at the same time according to their harvesting seasons. In this state its former harvest methods, with using of human power or semi-automatic harvest, caused to increase their production cost due to high labor cost and low working efficiency.

• Journal of Biosystems Engineering
• /
• v.20 no.1
• /
• pp.22-35
• /
• 1995

Pepper is the most important horticultural plant in Korean farm. Pepper harvesting has been known to be the most difficult process in pepper cultivation so that demand for mechanization is strong. In a research to develop a pepper harvesting machine performance and capacity of the harvester should be determined based on both economical feasibility and machine design concept. In order to accomplish an economical analysis of the pepper harvester, a mathematical model for comparing manual harvesting cost to machine harvest cost was developed. Validity of the model depends on the data used in the model. Economical information for the model variables was acquired from the result of farm survey on pepper cultivation technique and economics of pepper farmer. Technical information on pepper harvester were also collected through literature review and analyzed. Based on the economical analysis and synthesis of the technical information on pepper harvesters, its performance and capacity were determined. The operating performances of the harvester such as cutting, conveying, flipping, pepper removing and post-processing (sorting) were determined. Daisy capacity of the machine was determined to be 0.41 ha. A pepper harvester with the suggested capacity was economically feasible if the price of pepper harvester, pepper recovery ratio and service life of harvester were about 6 million won, 80%, and 4 years, respectively.

• Journal of Sericultural and Entomological Science
• /
• v.40 no.1
• /
• pp.8-12
• /
• 1998

The study aimed at development of a riding-type mulberry harvester for mechanical harvest. A riding-type mulberry harvester has been developed to harvest on sloped land with a higher efficiency. It has been implemented over a period of 2 years from 1996 to 1997. The result is as follows. It moves on carterpillar with a level adjusting system. It reduced only from 14.6 hrs to 0.9hrs/10a for cutting in a range of 25 to 80 cm high and possibly used for both spring and autumn. It reduced only the labor requirements of mulberry harvesting by 94 percent, as compared to that of the manual harvest. All related processes, cutting, binding and loading are simultaneously done by this harvester and totally it can reduce 96 percent of the labor requirements, as compared to 20.4 hrs/10a of the manual harvest. The machine compared to improved mulberry harvest efficiency with 11.11a per hour by about 23 times as compared to 0.49a per hour manpower. Cost analysis indicated that the riding-type mulberry harvester saved overall cost by 66 percent from 980,000 won per ha manpower to 330,000 won per ha.

• Journal of the Korean Society of International Agriculture
• /
• v.30 no.4
• /
• pp.370-375
• /
• 2018

In order to promote the mechanized cultivation of perilla for seed, which has been increasing in cultivation area and production recently as demand increases according to the health-functional effects, we carried out this experiment to determine the optimum sowing time of perilla to minimize the seed loss at harvest and increase the yield. We used two different types of perilla varieties, 'Sodam(small-branch)' and 'Deulsaem(multi-branch)', and the sowing time was June 15, June 30, July 15 and August 1. As the sowing time is late, days of growth from sowing to flowering were shortened, and they were shortened from 14, 26 and 31~32 days on June 30, July 15 and August 1 as compared with June 15, respectively. And, the stem length and culm diameter were shortened or tapered and the number of nodes tended to decrease. The number of effective branch was 82%, 61% and 56% on June 30, July 15 and August 1 as compared with June 15, respectively. Accordingly, it seems to make against in securing the yield from July 15. And, the lowest cluster height was generally shorter as the sowing time is late, and the height was below 15cm on July 15 and August 1. It seems that this may work against the machine harvest. There was a high degree of significance between the sowing time and the yield. Although, the total yield was not statistically significant among June 15, June 30 and July 15, the ratio of shattering seed at harvest was in order of July 15, August 1(30.3%)> June 15(15.3%)> June 30(13.5%). Therefore, the net yield except for shattered seed was higher in order of June 30${\geq}$ June 15> July 15> August 1. This tendency was characteristic regardless of variety and sowing method. And, the protein content in perilla seed increased as the sowing time was delayed, and the content was the highest on August 1. The content of crude fat was relatively high on June 15 and July 15 in 'Sodam', and June 30 and July 15 in 'Deulsaem', respectively. And, the content of linolenic acid was found to be the highest on August 1. As a result, the optimal sowing time for machine harvest of perilla for seed is about June 30. At this time, it is determined that the sowing time is the most suitable to be advantageous in increasing the yield of perilla seed, while minimizing the seed loss due to the shattering at harvest.

• Journal of Drive and Control
• /
• v.16 no.4
• /
• pp.74-79
• /
• 2019

In this study, an edible fresh corn harvest testing machine was designed and manufactured. And harvesting performance was analyzed through the field test. The testing machine is of the tractor attached type. It is connected to the tractor PTO shaft to transfer power to the each part of the harvesting machine. And it harvests fresh corn by one row through the processes of cutting, stem crushing, detaching, and collecting. The performance test was performed at PTO speed (540, 750, 1050 rpm, respectively), working speed (0.1, 0.15, 0.2 m/s, respectively), and cropping cultivation (row spacing·hill spacing 70·25 cm, 70·40 cm, 90·30 cm, respectively). The performance test was repeated three times in the 15 m section. The detachment loss ratio, uncollected crop ratio, damage ratio, and harvest ratio were analyzed. As a result of the performance test, it was analyzed that the PTO speed 540 rpm, running speed of 0.1 m/s, and row spacing·hill spacing 70·40 cm were the optimal condition.

• Korean Journal of Plant Resources
• /
• v.9 no.1
• /
• pp.63-69
• /
• 1996

The experiment was conducted to reduce the labor and production cost with the labor save of harvest in cultivating the onion using the machine and the results are as follows. On labor saving effect in transparent vinyl mulching, digging working hours per 10a in the case of using tractor are 55 min., fixing + turning time is 11 min.,the time of harvest is 66 min, digging working hours using cultivator are 90 min. and fixing + turning time is 9 min., but the time of hand harvesting is 693 min and 41 sec. and in digging labor saving effect, tractor shows 90% in the harvesting period and harvest by cultivator 86%. On nonmulching cases, the harvest by tractor takes 44 min. and that by cultivator does 75 min, and digging labor saving effect shows 93.6% in the tractor harvest and 89% in the cultivator harvest. Therefore, on the operation efficiency per hour, in the case of tractor with digger vinyl mulching and nonmulching show $0.091\sim0.136ha$ and in the case of cultivator with digger-both show $0.061\sim0.08ha$, but in the case of hand harvest, vinyl mulching and nonmulching are $0.008\sim0.009ha$, so in the mechanized harvest of onion, the harvest by tractor with digger is the best. On the cost and labor save for harvesting the onion with labor saving effect, tractor shows 19 hours and 26min./10a in vinyl mulching and 18 hours and 54min./10a in nonmulching, so it shows the short hours for harvesting. And labor saving effect shows 37% in comparison with 29 hours and 49 min. $\sim30$ hours and 38 min.($110,587\sim113,925won$) of the hand harvest, so the cost was reduced to $69,525\sim72,225won$. On the cultivator with diggers, vinyl mulching takes 19 hours and 49 min and nonmulching 20 hours and 2 min., so the labor saving effect and cost were reduced to $32\sim36%$($73,087\sim75,075$ won) in comparison with the hand harvest.