• Proceedings of the Korean Society for Agricultural Machinery Conference
• /
• 1996.06c
• /
• pp.478-487
• /
• 1996

A " plucking rolls device" was developed in this study to improve the quality of harvested tea leaves. In this report, the outline of the system and the results of performance experiments in our laboratory are discussed. Tow kinds of performance experiments were carried out. The first experiment checked harvesting accuracy by using a plucking unit that was developed for harvesting machine installation. The second experiment was a harvesting experiment which utilized a fron bar in order to prevent cutting of the tea buds which had been a problem in precious experiments . As a results of the first experiments , it was confirmed that selective harvesting obtained high quality tea leaves. but a cutting problem that, when the harvesting seed was faster than the working speed, which was non-selective harvesting , was also seen. In the second experiment, the cutting rate decreased to a maximum of 50% level, when tea buds most bent ahead by the front bar. The effect was seen that cutt ng problem was alleviated from this.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.11 no.2
• /
• pp.79-86
• /
• 2016

In wireless sensor networks, increasing the sensing rate of each node to improve the data accuracy usually incurs a decrease of network lifetime. In this study, an energy-adaptive data compression scheme is proposed to efficiently control the sensing rate in an energy-harvesting wireless sensor network (WSN). In the proposed scheme, by utilizing the surplus energy effectively for the data compression, each node can increase the sensing rate without any rise of blackout time. Simulation result verifies that the proposed scheme gathers more amount of sensory data per unit time with lower number of blackout nodes than the other compression schemes for WSN.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.2
• /
• pp.421-441
• /
• 2021

Fruit detection in orchards is one of the most crucial tasks for designing the visual system of an automated harvesting robot. It is the first and foremost tool employed for tasks such as sorting, grading, harvesting, disease control, and yield estimation, etc. Efficient visual systems are crucial for designing an automated robot. However, conventional fruit detection methods always a trade-off with accuracy, real-time response, and extensibility. Therefore, an improved method is proposed based on coarse-to-fine multitask cascaded convolutional networks (MTCNN) with three aspects to enable the practical application. First, the architecture of Fruit-MTCNN was improved to increase its power to discriminate between objects and their backgrounds. Then, with a few manual labels and operations, synthetic images and labels were generated to increase the diversity and the number of image samples. Further, through the online hard example mining (OHEM) strategy during training, the detector retrained hard examples. Finally, the improved detector was tested for its performance that proved superior in predicted accuracy and retaining good performances on portability with the low time cost. Based on performance, it was concluded that the detector could be applied practically in the actual orchard environment.

• Journal of Korean Society on Water Environment
• /
• v.24 no.2
• /
• pp.221-229
• /
• 2008

This study revised a model for hydrologically analyzing rainwater harvesting facilities considering their rainfall-runoff properties and the data available. This model has only a few parameters, which can be estimated with rather poor measurements available. The model has a non-linear module for rainfall loss, and the remaining rainfall excess (effective rainfall) is assumed to be inflow to the storage tank. This model has been applied for the rainwater harvesting facilities in Seoul National University, Korea Institute of Construction Technology, and the Daejon World Cup Stadium. As a result, the runoff coefficients estimated were about 0.9 for the building roof as a rainwater collecting surface and about 0.18 for the playground. This result is coincident with that for designing the rainwater harvesting facilities to show the accuracy of model and the simulation results.

• Journal of Advanced Navigation Technology
• /
• v.21 no.1
• /
• pp.99-106
• /
• 2017

In this paper, an energy harvesting and profiling system is designed for smart video devices in internet of things environments without dedicated power source. The energy harvesting module provides the harvested energy from solar panel to the smart video device. The energy profiling module measures the battery outflow current and the battery voltage of the smart video device and the consumed energy of processes, and calculate the harvested energy from the energy harvesting module to the smart video device and the total energy consumption of the smart video device. The accuracy of the harvested energy measured by the device energy profiling module is validated by comparing with the calculated energy using the regional solar radiation provided by Korea Meteorological Administration. Energy harvesting data from the designed energy harvesting and profiling system can be used to design the perpetual operation of smart video devices or Internet of Things sensors.

• Journal of Biosystems Engineering
• /
• v.40 no.1
• /
• pp.28-34
• /
• 2015

Purpose: This study aimed to install an RTK-GPS (Real Time Kinematic-Global Positioning System) and IMU (Inertial Measurement Unit) on a tractor used in a farm to measure positions, pasture topography, posture angles, and vibration accelerations, translate the information into maps using the GIS, analyze the characteristics of grass harvesting work, and establish new technologies and construction standards for pasture infrastructure improvement based on the analyzed data. Method: Tractor's roll, pitch, and yaw angles and vibration accelerations along the three axes during grass harvesting were measured and a GIS map prepared from the data. A VRS/RTK-GPS (MS750, Trimble, USA) tractor position measuring system and an IMU (JCS-7401A, JAE, JAPAN) tractor vibration acceleration measuring systems were mounted on top of a tractor and below the operator's seat to obtain acceleration in the direction of progression, transverse acceleration, and vertical acceleration at 10Hz. In addition, information on regions with bad workability was obtained from an operator performing grass harvesting and compared with information on changes in tractor posture angles and vibration acceleration. Results: Roll and pitch angles based on the y-axis, the direction of forward movements of tractor coordinate systems, changed by at least $9-13^{\circ}$ and $8-11^{\circ}$ respectively, leading to changes in working postures in the central and northern parts of the pasture that were designated as regions with bad workability during grass harvesting. These changes were larger than those in other regions. The synthesized vectors of the vibration accelerations along the y-axis, the x-axis (transverse direction), and the z-axis (vertical direction) were higher in the central and northwestern parts of the pasture at 3.0-4.5 m/s2 compared with other regions. Conclusions: The GIS map developed using information on posture angles and vibration accelerations by position in the pasture is considered sufficiently utilizable as data for selection of construction locations for pasture infrastructure improvement.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.10
• /
• pp.3321-3341
• /
• 2014

This paper investigates the outage performance and optimization for the four-phase two-way transmission network with an energy harvesting (EH) relay. To enable the simultaneous information processing and energy harvesting at the relay, we firstly propose a power splitting-based two-way relaying protocol (PSTWR). Then, we discuss its outage performance theoretically and derive an explicit expression for the system outage probability. In order to find the optimal system configuration parameters such as the optimal power splitting ratio and the optimal transmit power redistribution factor, we formulate an outage-minimized optimization problem. As the problem is difficult to solve, we design a genetic algorithm (GA) based algorithm for it. Besides, we also investigate the effects of the power splitting ratio, the power redistribution factor at the relay, and the source to relay distance on the system outage performance. Finally, extensive simulation results are provided to demonstrate the accuracy of the analytical results and the effectiveness of the GA-based algorithm. Moreover, it is also shown that, the relay position greatly affects the system performance, where relatively worse outage performance is achieved when the EH relay is placed in the middle of the two sources.

• Korean Journal of Agricultural Science
• /
• v.50 no.3
• /
• pp.357-364
• /
• 2023

In this study, we propose a ripe tomato bunch localization method based on convolutional neural networks, to be applied in robotic harvesting systems. Tomato images were obtained from a smart greenhouse at the Rural Development Administration (RDA). The sample images for training were extracted based on tomato maturity and resized to 128 × 128 pixels for use in the classification model. The model was constructed based on four-layer convolutional neural networks, and the classes were determined based on stage of maturity, using a Softmax classifier. The localization of the ripe tomato bunch region was indicated on a class activation map. The class activation map could show the approximate location of the tomato bunch but tends to present a local part or a large part of the ripe tomato bunch region, which could lead to poor performance. Therefore, we suggest a recursive method to improve the performance of the model. The classification results indicated that the accuracy, precision, recall, and F1-score were 0.98, 0.87, 0.98, and 0.92, respectively. The localization performance was 0.52, estimated by the Intersection over Union (IoU), and through input recursion, the IoU was improved by 13%. Based on the results, the proposed localization of the ripe tomato bunch area can be incorporated in robotic harvesting systems to establish the optimal harvesting paths.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.3
• /
• pp.506-512
• /
• 2016

Recently, energy harvesting technology is considered as a tool to improve the lifetime of sensor networks by mitigating the battery capacity limitation problem. However, the previous work on energy harvesting has failed to provide practical information since it has assumed an ideal channel knowledge model with perfect channel state information at transmitter (CSIT). This paper proposes an energy efficient resource allocation scheme that takes account of the channel estimation process and the corresponding estimation error. Based on the optimization tools, we provide information on efficient scheduling and power allocation as the functions of channel estimation accuracy, harvested energy, and data rate. The simulation results confirm that the proposed scheme outperforms the conventional energy harvesting networks without considering channel estimation error in terms of energy efficiency. Furthermore, with taking account of channel estimation error, the results provides a new way for allocating resources and scheduling devices.

• Journal of Biosystems Engineering
• /
• v.14 no.2
• /
• pp.104-114
• /
• 1989

The most important subject in developing agricultural robots for fruit harvesting is to detect accurately the location of a fruit about the given coordinate system. This study was carried out to develop an image processing algorithm which enables finding out the three dimensional locations of a fruit. The digital image processing device consisted of an optosensor (Closed-circuit TV camera), image processing interface board (Digitizer) and microcomputer (IBM PC/AT). A stereo-image processing method using the two cameras attached to the manipulator was evaluated through experiment with apples. The accuracy and quickness of detecting the location of apples by this method was not satisfactory. The maximum errors of the detected locations by the stereo-image processing method in x-, Y-, and z- directions were 3, 4 and 4 cm, respectively. The maximum time required to get the rectangular coordinate data of a fruit was about 2 minutes.