• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.40 no.5
• /
• pp.449-456
• /
• 2022

This study proposes a deep learning algorithm to predict crop yield using GIS (Geographic Information System) to extract soil properties from Soilgrids and soil suitability class maps. The proposed model modified the structure of a published CNN-RNN (Convolutional Neural Network-Recurrent Neural Network) based crop yield prediction model suitable for the domestic crop environment. The existing model has two characteristics. The first is that it replaces the original yield with the average yield of the year, and the second is that it trains the data of the predicted year. The new model uses the original field value to ensure accuracy, and the network structure has been improved so that it can train only with data prior to the year to be predicted. The proposed model predicted the yield per unit area of autumn cabbage for kimchi by region based on weather, soil, soil suitability classes, and yield data from 1980 to 2020. As a result of computing and predicting data for each of the four years from 2018 to 2021, the error amount for the test data set was about 10%, enabling accurate yield prediction, especially in regions with a large proportion of total yield. In addition, both the proposed model and the existing model show that the error gradually decreases as the number of years of training data increases, resulting in improved general-purpose performance as the number of training data increases.

• Korean Journal of Agricultural Science
• /
• v.2 no.1
• /
• pp.111-123
• /
• 1975

6 years data on the plant growth and grain yield of 13 rice varieties were investigated to define the relations between meteological conditions and plant growth of rice including grain yield. The results obtained are summarized as follows; 1. Variation of average temperature by years showed great number at early and middle July, middle August and early November, however smallest figure at late June and late July. And that of sunshine hours by years were least at late June and late July, and largest at middle July. 2. Among rice yield components variation of panicle number per hill by years was biggest and that of 1000 grain weight least. Rate of variation of plant growth and rice yield was different by rice varieties. 3. Direct effects on rice yield was greatest at maturing ratio and next at 1000 grain weight. The effects of yield factor on the rice yield wers different by years. 4. Higher temperature and longer sunshine delayed the miaxmum tillering stage but shortened the days from seeding to heading. 5. A significant negative correlation was recognized between the number of panicles per hill and average temperature of 11 to 40 days after transplanting, and number of grains per panicle was correlated with the average temperature of 11 to 70 days after transplanting. High temperature before heading time showed atime decreased maturing ratio. 6. Accumulated temperature was highly correlated with 1000 grain weight in all season. Highest positive correltaion was recognized between grain yield of rice and average temperature of 61 to 70 days after transplanting but correlations between rice yield and average temperature after heading stage were negative. 7. Highly significant correlations were confirmed between maturing ratio and sunshine hours of 31 to 50 days as well as 61 to 70 days after transplanting, and maturing ratio was highly correlated with accumulated hours of sunshine. Correlation between rice yield and hours of sunshine was significant at 51 to 60 days after transplanting.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.64 no.1
• /
• pp.40-47
• /
• 2019

This study was conducted to investigate the effect of planting density on plant growth, yield, and quality in the sweet sorghum cultivar 'Chorong' (Sorghum bicolor (L.) Moench). Plants were cultivated at densities of 16.7, 11.1, 8.3, 6.7, and $5.6plants{\cdot}m^{-2}$. Factors related to yield and yield components were analyzed using correlation and multivariate analyses. There was no significant difference among plant densities in stem length from 20 to 110 days after sowing. But the stem diameter was thin, and a decrease in number of tillers occurred more rapidly as planting density increased. At harvest, juice and sugar yield were higher at densities of 16.7 (42.9, $4.16Mg{\cdot}ha^{-1}$, respectively) and 11.1 (37.1, $3.73Mg{\cdot}ha^{-1}$) $plants{\cdot}m^{-2}$ than at 8.3 (30.5, $2.96Mg{\cdot}ha^{-1}$), 6.7 (26.6, $2.41Mg{\cdot}ha^{-1}$), and 5.6 (24.7, $2.22Mg{\cdot}ha^{-1}$) $plants{\cdot}m^{-2}$. The soluble solids and total sugar contents were not different among treatments, but relatively high values were observed at the density of 11.1 and $8.3plants{\cdot}m^{-2}$. As plant density was increased from 5.6 to $11.1plants{\cdot}m^{-2}$, the lodging index (1 = no, 9 = lodging) increased rapidly from 2.00 to 6.33. To determine the optimal planting density, the number of typhoons and topographical characteristics should be considered. Correlation and principal components analyses revealed that plant density exhibited a positive relationship with fresh stem yield ($r=0.62^{**}$), dry stem yield ($r=0.58^{**}$), juice ($r=0.63^{**}$), and sugar yield ($r=0.66^{**}$), but a negative with stem diameter ($r=-0.65^{**}$). The yield factors were not statistically related to stem height, diameter, and number of nodes.

• Korean Journal of Plant Resources
• /
• v.15 no.2
• /
• pp.144-149
• /
• 2002

This study was carried out to find out the effects of date and number of defoliation by different seeding date on growth and yield of perilla. Saeyuepsildlkkae was seeded from April 30 to June 30 at 20 days interval and defoliated from June 30 to Aug. 15 with different defoliation number. 1. Days to emergence, flowering, and maturing were not significantly influenced by defoliation, but those were shortened as seeding date was delayed. The interaction between defoliation and seeding date showed no significant difference. 2. Stem length, number of nodes on main stem, and weight of dry stem were not significantly affected by defoliation, but those were decreased as seeding date was delayed. 3. Yield of fresh loaves was not significantly influenced by seeding date up to May 20, but markedly decreased by the later seeding date. 4. Cluster length and number of cluster per plant were not affected by seeding date, but number of capsules per cluster was decreased from 32 to 25 and 1,000 grain weight was increased from 3.42 to 4.62g as seeding date was delayed. 5. Grain yield was not significantly affected by defoliation and seeding date up to June 10, but markedly decreased by seeding date June 30.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.29 no.1
• /
• pp.62-66
• /
• 1984

One row of pollen parent to two rows of seed parent (1:2), 2:4 and solid (1:2) planting patterns (PP) were compared in seed parent densities of 3,500, 5,000 and 6,500 plants per 10 ares to determine effects of PP and plant density on growths of seed and pollen plants, and seed yield of seed parent of modified single cross corn hybrid. Planting pattern did not significantly affect agronomic characteristics of seed plant except ear number per 100 plants and seed yield which were greater in solid and 1:2 PP than in 2:4 PP. Significant PP x plant density interaction did not exist for agronomic characteristics of seed parent. In the seed parent, plant height and 100 kernel weight were not affected by plant density, but ear height, ear number per 100 plants, and kernel number per ear were linearly decreased with increased plant densities. Seed yield ranged from 330 to 460 kg per 10 ares and overall yield response to plant density was quadratic. Tassel length and spikelet number per tassel of the pollen parent were significantly affected by PP and plant density. Significant PP x plant density interaction existed for tassel length and spikelet number per tassel. Tassel length and spikelet number per tassel were greater in 1:2 and 2:4 PP compared to solid PP and were greatly reduced with increased plant densities in solid and 1:2 PP. The results indicated that 1:2 or 2:4 PP at around 5,000 plants per 10 ares for seed parent would be suitable for seed production of modified single cross com hybrid.

• Korean Journal of Environmental Agriculture
• /
• v.29 no.1
• /
• pp.1-6
• /
• 2010

Organic farming system in rice paddy is rapidly expanding in Korea. This study was to find out optimum japonica rice cultivars for organic farming. A field research was conducted to evaluate the characteristics of japonica rice cultivars under no-tillage paddy at Doo-ryangmyeon, Sacheon, Gyeongsangnam-do, Korea. The experimental soil was Juggog series (silty clay loam: 56.0% silt, 31.2% clay and 12.8% sand). In experiment, ten lines of Japanese rice cultivars were tested under no-tillage amended with rye (NTR) and no-tillage without cover crop treatment (NTNT). In addition, two Korean japonica rice cultivars as check cultivars were used in this study. The grain yield in NTR was significantly higher in 6.13 Mg $ha^{-1}$ for Kinuhikari, 5.30 Mg $ha^{-1}$ for Komekogane, 5.25 Mg $ha^{-1}$ for Kosihikari, 5.22 Mg $ha^{-1}$ for Mazizbare and 5.12 Mg $ha^{-1}$ for Akitakomachi compared to two Korean rice cultivars (4.57 Mg $ha^{-1}$ for Hwayoungbyeo and 4.00 Mg $ha^{-1}$ for Ilmibyeo) in that order. While, grain yield in NTNT was significantly higher in 4.90 Mg $ha^{-1}$ for Akitakomachi 3.81 Mg $ha^{-1}$ for Hinohikari, 3.74 Mg $ha^{-1}$ for Umezkusi, 3.67 Mg $ha^{-1}$ for Kosihikari and 3.54 Mg $ha^{-1}$ for Dondokuri compared to 3.02 Mg $ha^{-1}$ for Ilmibyeo and 2.36 Mg $ha^{-1}$ for Hwayoungbyeo, respectively. The number of panicle per $m^2$ and grain number per panicle were indispensible for increasing the yield of rice. These results were able to find out optimum japonica rice cultivar Akitakomachi for organic farming in no-tillage paddy.

• The Korean Journal of Pesticide Science
• /
• v.14 no.2
• /
• pp.133-137
• /
• 2010

A field investigation was carried out for two years to analyze yield loss due to soybean anthracnose caused by Colletotrichum truncatum and to determine its economic threshold limit. Anthracnose severity in terms of % diseased pods was negatively correlated with yield, number of normal seeds per plant and number of pods per plant, and positively correlated with % abnormal seeds with correlation coefficients of -0.85, -0.78, -0.64, and 0.80, respectively. A simple linear regression model was obtained as Y=-1.7781X+164.22 with $R^2$=0.8092, when the soybean yields (Y) were predicted using anthracnose severity (X) as an independent variable. The yield levels could be predicted as high as 80.92%. Based on this equation, spray threshold without economic considerations was estimated as 6.9 in % pods infected with anthracnose. Economic threshold limit and economic spray threshold able to compensate the costs of fungicide sprays were determined as 11.9% and 9.5%, respectively.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.42 no.1
• /
• pp.89-94
• /
• 1997

The present study was conducted to know the growth and yield of in response to the drought stress at the different soybean growth stage. Drought stress was given to the soybean plants on early vegetative growth at fourth-node stage(FNS), mid-growth at beginning pod stage(BPS) and late growth at beginning seed stage(BSS) for 30 days, which are high availability in soil water stress on climate condition of Korea. Dry weight was decreased severely by water stress at FNS, and BPS and BSS has no difference compared to control. Chlorophyll content of leaf severely decreased at the end of water stress of FNS and BPS, but was recovered at the harvest stage. Drought-stressed root distributed mainly near the soil surface and number and dry weight of root nodule were decreased severely by drought stress at BPS. Number of pod, seed weight and yield were decreased by drought stress and showed the highest yield loss at BPS.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.45 no.3
• /
• pp.163-166
• /
• 2000

Cassia mimosoides var. nomame was grown at five phosphate rates (0, 5, 15, 25 and 35 kg/10a) at Cheju in 1998 to determine the influence of phosphate rate on growth, forage yield and quality. As phosphate rate increased from 0 to 35kg/10a, the plant growth was increased such as days to flowering, plant height, the number of branches and green leaves per plant, stem diameter and SPAD (Soil Plant Analysis Development) reading values of leaves but the number of withering leaves per plant decreased. Fresh forage yield was 3, 291kg/10a at 0 kg/10a of phosphate rate and increased 5, 200 and 5, 230 kg/10a at 25 and 35 kg/10a of phosphate level, respectively. Dry matter, crude protein and total digestible nutrients (TDN) showed the same tendency of changes as the fresh forage yield. Crude protein, crude fat, nitrogen free extract and TDN content increased but crude ash and crude fiber content decreased as the increasing of phosphate rate increased.

• Journal of Sericultural and Entomological Science
• /
• v.39 no.2
• /
• pp.205-209
• /
• 1997

In a line X tester crossing programme (24 lines 2 testers) the general combining ability (GCA) and specific combining ability (SCA) effects were analyzed for five economic characters in the bivoltine silkworm, Bombyx mori L. The results showed desired GCA effects in 934D1 (9500), 934B (9789) and 934A1 (9855) for cocoon yield per 10,000 larvae brushed by number. Likewise, the lines found to be superior based on GCA effects for other characters were as follows; 931D (14.040 Kgs), 935E (17.023 Kga.), 934D1 (15.643 Kgs.) and 934B (15.687 Kgs.) for cocoon yield by weight: 931D (1.717 g) and 930E (1.796 g) for single cocoon weight; 932B (0.330 g) for single shell weight: 931D (1.717 g) and 930e (1.796 g) for single cocoon weight; 932B (0.330 g) for single shell weight; 932B (18.7%), 933A (18.86%) and 935A (19.89%) for shell ratio. SCA effects showed the superiority of 932D$\times$KA (9822 cocoon yield per 10,000 larvae brushed by number); 932A$\times$NB4D2 (16.933 Kgs. cocoon yield per 10,000 larvae brushed by weight); 931C$\times$KA (1.911 g single cocoon weight); 934$\times$NB4D2 (0.371 g single shell weight and 21.0% shell ratio). The analysis indicated non-additive gene action for all the five characters.