• Journal of the Korea Society of Computer and Information
• /
• v.29 no.11
• /
• pp.113-118
• /
• 2024

This paper proposes an optimized fruit set prediction model for cucumbers using an AI-based automatic growth control system. Based on data collected from experimental farms at Sunchon National University and Suncheon Bay cucumber farms, we constructed and compared the performance of models using three machine learning algorithms: Random Forest, XGBoost, and LightGBM. The models were trained using 19 environmental and growth-related variables, including temperature, humidity, and CO2 concentration. The LightGBM model showed the best performance (RMSE: 1.9803, R-squared: 0.5891). However, all models had R-squared values below 0.6, indicating limitations in capturing data nonlinearity and temporal dependencies. The study identified key factors influencing cucumber fruit set prediction through feature importance analysis. Future research should focus on collecting additional data, applying complex feature engineering, introducing time series analysis techniques, and considering data augmentation and normalization to improve model performance. This study contributes to the practical application of smart farm technology and the development of data-driven agricultural decision support systems.

• Advanced Industrial SCIence
• /
• v.3 no.2
• /
• pp.17-22
• /
• 2024

Recently, as the ICT field has been used in various environments, it has become possible to analyze pests by crops, use robots when harvesting crops, and predict by big data by utilizing ICT technologies in a sustainable agricultural environment. However, in a sustainable agricultural environment, efforts to solve resource depletion, agricultural population decline, poverty increase, and environmental destruction are constantly being demanded. This paper proposes an artificial intelligence-based big data processing analysis method to reduce the production cost and increase the efficiency of crops based on a sustainable agricultural environment. The proposed technique strengthens the security and reliability of data by processing big data of crops combined with AI, and enables better decision-making and business value extraction. It can lead to innovative changes in various industries and fields and promote the development of data-oriented business models. During the experiment, the proposed technique gave an accurate answer to only a small amount of data, and at a farm site where it is difficult to tag the correct answer one by one, the performance similar to that of learning with a large amount of correct answer data (with an error rate within 0.05) was found.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.19 no.4
• /
• pp.771-780
• /
• 2024

In this study, we implemented a real-time pest detection and prediction system for a strawberry farm using a computer vision model based on the YOLOv5 architecture and an Isolation Forest Classifier. The model performance evaluation showed that the YOLOv5 model achieved a mean average precision (mAP 0.5) of 78.7%, an accuracy of 92.8%, a recall of 90.0%, and an F1-score of 76%, indicating high predictive performance. This system was designed to be applicable not only to strawberry farms but also to other crops and various environments. Based on data collected from a tomato farm, a new AI model was trained, resulting in a prediction accuracy of over 85% for major diseases such as late blight and yellow leaf curl virus. Compared to the previous model, this represented an improvement of more than 10% in prediction accuracy.

• Korean Journal of Environmental Agriculture
• /
• v.1 no.2
• /
• pp.89-92
• /
• 1982

Persistence of cyanofenphos on Chinese cabbage under the different climate conditions was studied by spraying the insecticide at the rate of 0.5 and 0.75 ㎏ AI/ha at 22 and 36 days after transplanting and monitoring its residues upto 35 days after the final spray. At both spraying rates the degradation patterns of the insecticide, regardless of climate condition, showed similar trends; cyanofenphos residues on Chinese cabbage declined rapidly upto 14 days after the final spray but more slowly thereafter. Half-life for cyanofenphos on Chinese cabbage was $6{\sim}7$ days. The half-life was little affected by the spraying rate and time. Based on the FAO/WHO maximum residue limit of cyanofenphos on common cabbage (2 ppm), it is recommended that the pre-harvest intervels of the insecticide on Chinese cabbage could be 16 and 19 days for 0.5 and 0.75 ㎏ AI/ha, respectively.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.18 no.3
• /
• pp.441-448
• /
• 2023

The quality of the casting process generates the largest source of defects in the manufacturing process, so its management is a key factor in productivity and quality evaluation. Based on the results of factor analysis, correlation analysis, and regression analysis with process data, this study aims to optimize the machine learning model to reduce the defect rate and verify the data suitability for smart factories.

• Asian-Australasian Journal of Animal Sciences
• /
• v.18 no.6
• /
• pp.787-794
• /
• 2005

The aim of this study was 1) to confirm the practical efficiency of a routine milk P4 monitoring system for postpartum reproductive management of a dairy herd, and 2) to evaluate the relationship between the blood metabolic profiles, milk quality and body weight of individual cows in the farm records, which may reflect the postpartum nutritional condition, and the time of postpartum resumption of ovarian activity of dairy cows. A total of 116 Holstein cows was used in the present study. First, during the period of Experiment 1, postpartum reproductive management based on weekly measured milk P4 concentration from individual cows was conducted. Compared with the reproductive records of the past two years without P4 monitoring, although the day from calving to first AI did not change, both the number of AI until pregnant (with P4; 1.9 times vs. without P4; 2.9 times) and the days open (with P4; 95.1 days vs. without P4; 135.8 days and 133.8 days) were significantly decreased. In Experiment 2, the measurement of blood constituents such as albumin, blood urea nitrogen, packed cell volume, ammonia, glucose, total cholesterol, non-esterified, AST and $\gamma$-GTP was performed on the blood samples taken once approximately 14 days postpartum, to monitor both health and nutritional conditions. The milk constituent parameters, such as milk protein (MP), milk fat (MF), SNF and lactose, collected from the monthly progeny test of individual cows, were used to monitor the postpartum nutritional status. Furthermore, the data obtained from the routine measurements of body weight were used to calculate the rate of peripartum body weight loss. The resumption day of the postpartum estrous cycle was assumed from the milk P4 profiles of individual cows. There was no clear relationship between each parameter from blood examination and those from resumption time. However, the cows had low values of MP, and SNF, which significantly affected the resumption of the postpartum estrous cycle. Similarly, a higher rate of body weight loss indicated a significant delay (more than 1 month) in the resumption of the postpartum estrous cycle, compared with the groups that had a medium or lower rate of body weight loss. The results of the present study demonstrated that the implementation of routine milk P4 monitoring-based postpartum reproductive management, together with milk quality parameters and routine BW data available in field conditions may be utilized as a practical approach for increasing the postpartum reproductive efficiency of a high yielding dairy herd.

• Asian-Australasian Journal of Animal Sciences
• /
• v.33 no.12
• /
• pp.1922-1929
• /
• 2020

Objective: The present study aimed to survey seasonal changes in reproductive performance of local cows receiving artificial insemination (AI) in the Pursat province of Cambodia, a tropical country, to investigate if ambient conditions affect the reproductive performance of cows as to better understand the major problems regarding cattle production. Methods: The number of cows receiving AI, resultant number of calving, and calving rate were analyzed for those receiving the first AI from 2016 to 2017. The year was divided into three seasons: cool/dry (from November to February), hot/dry (from March to June), and wet (from July to October), based on the maximal temperature and rainfall in Pursat, to analyze the relationship between ambient conditions and the reproductive performance of cows. Body condition scores (BCS) and feeding schemes were also analyzed in these seasons. Results: The number of cows receiving AI was significantly higher in the cool/dry season than the wet season. The number of calving and calving rate were significantly higher in cows receiving AI in the cool/dry season compared with the hot/dry and wet seasons. The cows showed higher BCSs in the cool/dry season compared to the hot/dry and wet seasons probably due to the seasonal changes in the feeding schemes: these cows grazed on wild grasses in the cool/dry season but fed with a limited amount of grasses and straw in the hot/dry and wet seasons. Conclusion: The present study suggests that the low number of cows receiving AI, low number of calving, and low calving rate could be mainly due to poor body condition as a result of the poor feeding schemes during the hot/dry and wet seasons. The improvement of body condition by the refinement of feeding schemes may contribute to an increase in the reproductive performance in cows during the hot/dry and wet seasons in Cambodia.

• Korean Journal of Agricultural Science
• /
• v.47 no.4
• /
• pp.979-985
• /
• 2020

A wide range of techniques have been developed to separate X or Y- chromosome-bearing sperm. In particular, bovine semen sex-sorted by using flow cytometry based on differences in the amount of DNA between X and Y chromosome bearing sperm is used in dairy farms. The first piglets were produced using sex-sorted sperm 30 years ago. However, sexed sperm have not been commercially available in pigs because the flow cytometry technique is not capable of sorting the high number of sperm required for porcine artificial insemination (AI), and the prolonged exposure to an electrical filed might damage to the DNA in sperm. The purpose of this study was to evaluate a boar sperm sorting method based on magnetic nanoparticles. A flow cytometer assay verified the efficacy of the magnetic nanoparticles (> 90% of sex-sorted sperm). In addition, a duplex polymerase chain reaction (PCR) assay using sex chromosome specific genes including SRY (sex-determining region Y; male), ZFY (zinc finger protein Y-linked; male), and ZFX (zinc finger protein X-linked; female) showed that in vitro fertilized porcine embryos by X and Y-chromosome bearing sperm were 100% female (40/40) and 72% female (35/48), respectively, at 8-cell or morula stages, suggesting that the sex-sorted sperm were fertile. In conclusion, our findings suggest that the sex-sorted method based on magnetic nanoparticles can be utilized for porcine sex-sorted AI.

• Annual Conference of KIPS
• /
• 2024.10a
• /
• pp.821-822
• /
• 2024

This paper describes an automated fruit sorting system using Raspberry Pi and Arduino to classify apples and oranges by freshness, using EfficientNet-B0 for detection. It offers expandability and addresses labor shortages in agriculture through automation.

• Journal of Information Processing Systems
• /
• v.16 no.1
• /
• pp.155-170
• /
• 2020

Passive ranging is a critical part of machine vision measurement. Most of passive ranging methods based on machine vision use binocular technology which need strict hardware conditions and lack of universality. To measure the distance of an object placed on horizontal plane, we present a passive ranging method based on monocular vision system by smartphone. Experimental results show that given the same abscissas, the ordinatesis of the image points linearly related to their actual imaging angles. According to this principle, we first establish a depth extraction model by assuming a linear function and substituting the actual imaging angles and ordinates of the special conjugate points into the linear function. The vertical distance of the target object to the optical axis is then calculated according to imaging principle of camera, and the passive ranging can be derived by depth and vertical distance to the optical axis of target object. Experimental results show that ranging by this method has a higher accuracy compare with others based on binocular vision system. The mean relative error of the depth measurement is 0.937% when the distance is within 3 m. When it is 3-10 m, the mean relative error is 1.71%. Compared with other methods based on monocular vision system, the method does not need to calibrate before ranging and avoids the error caused by data fitting.