• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.16 no.4
• /
• pp.97-108
• /
• 2016

In this paper, we consider a cooperative monitoring network, which consists of a large number of UAVs, in order to promptly detect event in a disaster area. A command center may not be able to control each UAV individually due to resource constraints. Therefore, UAVs need to autonomously construct a mobile monitoring network in order to maximize monitoring coverage and to adapt the network formation according to environment changes in the disaster area. To that end, we propose multiple UAVs-based cooperative monitoring schemes that uses virtual forces. In this monitoring scheme, an effective monitoring is enabled by extending monitoring coverage using each UAV's circle movements. The UAVs-based monitoring network can also be splitted or merged in order to increase the monitoring effectiveness. Through simulations, we show that the proposed scheme can effectively monitor a large area and achieve a high event detection ratio.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.37 no.2
• /
• pp.459-466
• /
• 2017

Several reservoirs that were managed by local governments and the Korea Rural Community Corporation have recently collapsed. One of them is the Goeyeon reservoir in Yeongcheon-si, Gyeongsangbuk-do that collapsed mainly around the spillway due to heavy rain at 9 O'clock, on 21 August 2014. The Goeyeon reservoir was an aging agricultural reservoir over 70 years since it was built. In this study, the collapse situation of the reservoir was reproduced through the DAMBRK model. Flood inundation maps were reconstructed for the breach outflow of the dam analyzed by the DAMBRK model. We estimated the breach duration and outflow of the reservoir as compared with the inundation image taken by the Unmanned Aerial Vehicle (UAV) at the time when the Goeyeon reservoir collapsed. The results of this study are expected to be useful for predicting damage in the downstream inundation area when a reservoir collapses.

• Journal of Advanced Navigation Technology
• /
• v.21 no.6
• /
• pp.529-536
• /
• 2017

The general ground control system has control and information display functions for the operation of a single unmanned aerial vehicle. Recently, the function of the single ground control system extends to the operation of multiple UAVs. As a result, operators have been exposed to more diverse tasks and are subject to task overload due to various factors during their mission. This study proposes an adaptive ground control system that reflects the operator's condition through the task overload measurement of multiple UAV operators. For this, the ground control software is developed to control multiple UAVs at the same time, and the simulator with six degree-of-freedom aircraft dynamics is constructed for realistic human-machine-interface experiments by the operators.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.42 no.1
• /
• pp.47-58
• /
• 2014

In this research, the initial sizing of a HALE(High Altitude Long Endurance) UAV which uses solar power and hydrogen fuel cell as an alternative energy was performed. Instead of a wing box type, a glider type was chosen since it is relatively easy to get a data thanks to many researches abroad. Maximum takeoff weight is around 150Kg and the propulsion system is composed of motor, propeller, solar cell, and hydrogen fuel cell which can be recharged through electrolysis. Maximum takeoff weight was estimated as aspect ratio, wing span, wing area change while considering energy balance of required energy which is necessary for flight during the entire day and available energy which can be taken from the solar cell.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.10 no.6
• /
• pp.626-632
• /
• 2017

Since natural disaster occurs increasingly and becomes complicated, it causes deaths, disappearances, and damage to property. As a result, there is a growing interest in the development of ICT-based natural disaster response technology which can minimize economic and social losses. In this letter, we introduce the main functions of the forest fire management platform by using images from an UAV. In addition, we propose a disaster image analysis technology based on the deep learning which is a key element technology for disaster detection. The proposed deep learning based disaster image analysis learns repeatedly generated images from the past, then it is possible to detect the disaster situation of forest-fire similar to a person. The validity of the proposed method is verified through the experimental performance of the proposed disaster image analysis technique.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.64 no.1
• /
• pp.15-26
• /
• 2022

3D open-field farm model developed from UAV (Unmanned Aerial Vehicle) data could make crop monitoring easier, also could be an important dataset for various fields like remote sensing or precision agriculture. It is essential to separate crops from the non-crop area because labeling in a manual way is extremely laborious and not appropriate for continuous monitoring. We, therefore, made a 3D open-field farm model based on UAV images and developed a crop segmentation model using a supervised machine learning algorithm. We compared performances from various models using different data features like color or geographic coordinates, and two supervised learning algorithms which are SVM (Support Vector Machine) and KNN (K-Nearest Neighbors). The best approach was trained with 2-dimensional data, ExGR (Excess of Green minus Excess of Red) and z coordinate value, using KNN algorithm, whose accuracy, precision, recall, F1 score was 97.85, 96.51, 88.54, 92.35% respectively. Also, we compared our model performance with similar previous work. Our approach showed slightly better accuracy, and it detected the actual crop better than the previous approach, while it also classified actual non-crop points (e.g. weeds) as crops.

• Journal of Korean Society for Geospatial Information Science
• /
• v.24 no.4
• /
• pp.39-45
• /
• 2016

In this study, with arbitrary reference points on the water, photogrammetry accuracy analysis was conducted using unmanned aerial vehicle(UAV). A small reservoir is a research area, and twenty buoys were used as arbitrary reference points. Errors of location coordinate were identified with control of amounts of used reference points. cases are categorized by index scores per photos. Accuracy of X is 0.141m~0.166m and accuracy of Y is 0.136m~0.241m. Considering that allowable error for the maritime boundary survey is ${\pm}2m$, it is possible to get the accuracy data available for the photogrammetry of UAV using an reference point. In addition, the coefficient of correlation between the number of reference points per unit and number of buoys used as reference point and the ratio of the reference point per square measure, and percentage of buoys used as reference point and the coefficient of x and y were performed. Each element, x, and y showed a strong correlation and the coefficient of number of buoys used as reference point was irrelevant. The results of this correlation analysis can be analyzed that the number of reference points used in each picture is greater than the actual number of reference points used in location accuracy.

• Journal of Convergence for Information Technology
• /
• v.11 no.1
• /
• pp.136-142
• /
• 2021

Open-pit mine development requires continuous management because of topographical changes and there is a risk of accidents if the current status survey is performed directly in the process of calculating the earthwork. In this study, the application of UAV photogrammetry, which can acquire spatial information without direct human access, was applied to open-pit mines development area and analyzed the accuracy, earthwork, and mountain restoration plan to determine its applicability. As a result of accuracy analysis at checkpoint using ortho image and Digital Surface Model(DSM) by UAV photogrammetry, Root Mean Square Error(RMSE) is 0.120 m in horizontal and 0.150 m in vertical coordinates. This satisfied the tolerance range of 1:1,000 digital map. As a result of the comparison of the earthwork, UAV photogrammetry yielded 11.7% more earthwork than the conventional survey method. It is because UAV photogrammetry shows more detailed topography. And result of monitoring mountain restoration showed possible to determine existence of rockfall prevention nets and vegetation. If the terrain changes are monitored by acquiring images periodically, the utility of UAV photogrammetry will be further useful to open-pit mine development.

• Smart Media Journal
• /
• v.13 no.2
• /
• pp.52-61
• /
• 2024

In the shipyard, aerial images are acquired at regular intervals using Unmanned Aerial Vehicles (UAVs) for the management of external storage yards. These images are then investigated by humans to manage the status of the storage yards. This method requires a significant amount of time and manpower especially for large areas. In this paper, we propose an automated management technology based on a semantic segmentation foundation model to address these challenges and accurately assess the status of external storage yards. In addition, as there is insufficient publicly available dataset for external storage yards, we collected a small-scale dataset for external storage yards objects and equipment. Using this dataset, we fine-tune an object detector and extract initial object candidates. They are utilized as prompts for the Segment Anything Model(SAM) to obtain precise semantic segmentation results. Furthermore, to facilitate continuous storage yards dataset collection, we propose a training data generation pipeline using SAM. Our proposed method has achieved 4.00%p higher performance compared to those of previous semantic segmentation methods on average. Specifically, our method has achieved 5.08% higher performance than that of SegFormer.

• Korean Journal of Remote Sensing
• /
• v.34 no.2_2
• /
• pp.393-406
• /
• 2018

Every year in the ocean, various accidents occur frequently and illegal fishing is rampant. Moreover, their size and frequency are also increasing. In order to reduce losses of life or property caused by these, it is necessary to have a means to perform remote monitoring quickly. As an effective platform of such monitoring means, an Unmanned Aerial Vehicle (UAV) is receiving the spotlight. In these situations where marine accidents or illegal fishing occur, main targets of monitoring are ships. In this study, we propose a UAV based ship monitoring system and suggest a method of determining ship positions using UAV multi-sensory data. In the proposed method, firstly, the position and attitude of individual images are determined by using the pre-performed system calibration results and GPS/INS data obtained at the time when images were acquired. In addition, after the ship being detected automatically or semi-automatically from the individual images, the absolute coordinates of the detected ships are determined. The proposed method was applied to actual data measured at 200 m, 350 m, and 500 m altitude, the ship position can be determined with accuracy of 4.068 m, 8.916 m, and 13.734 m, respectively. According to the minimum standard of a hydrographical survey, the ship positioning results of 200 m and 350 m data satisfy grade S and the results of 500 m data do grade 1a, where the accuracy is required for positioning the coastline and topography less significant to navigation order. Therefore, it is expected that the proposed method can be effectively used for various purposes of marine monitoring or surveying.