• Journal of the Architectural Institute of Korea Structure & Construction
• /
• v.35 no.1
• /
• pp.11-19
• /
• 2019

Recently, with the development of the fourth industrial revolution that has been achieved through the fusion of information and communication technology (ICT), the technologies of AI, IOT, BIG-DATA, it is increasing utilization rate by industry and research and development of application technologies are being actively carried out. Especially, in the case of unmanned aerial vehicles, the construction market is expected to be one of the most commercialized areas in the world for the next decade. However, research on utilization of unmanned aerial vehicles in the construction field in Korea is insufficient. In this study, We have developed a quantitative building inspection method using the unmanned aerial vehicle and presented the protocol for it. The proposed protocol was verified by applying it to existing old buildings, and defect information could be quantified by calculating length, width, and area for each defect. Through this technical research, the final goal is to contribute to the development of safety diagnosis technology using unmanned aerial vehicle and risk assessment technology of buildings in case of disaster such as earthquake.

• Korean Journal of Remote Sensing
• /
• v.37 no.4
• /
• pp.747-761
• /
• 2021

With the increasing application of UAVs in intelligent transportation systems, vehicle detection for aerial images has become an essential engineering technology and has academic research significance. In this paper, a vehicle detection method for aerial images based on the YOLOv4 deep learning algorithm is presented. At present, the most known datasets are VOC (The PASCAL Visual Object Classes Challenge), ImageNet, and COCO (Microsoft Common Objects in Context), which comply with the vehicle detection from UAV. An integrated dataset not only reflects its quantity and photo quality but also its diversity which affects the detection accuracy. The method integrates three public aerial image datasets VAID, UAVD, DOTA suitable for YOLOv4. The training model presents good test results especially for small objects, rotating objects, as well as compact and dense objects, and meets the real-time detection requirements. For future work, we will integrate one more aerial image dataset acquired by our lab to increase the number and diversity of training samples, at the same time, while meeting the real-time requirements.

• Journal of Korea Multimedia Society
• /
• v.25 no.5
• /
• pp.685-694
• /
• 2022

The goal of few-shot semantic segmentation is to build a network that quickly adapts to novel classes with extreme data shortage regimes. Most existing few-shot segmentation methods leverage single or multiple prototypes from extracted support features. Although there have been promising results for natural images, these methods are not directly applicable to the aerial image domain. A key factor in few-shot segmentation on aerial images is to effectively exploit information that is robust against extreme changes in background and object scales. In this paper, we propose a Mask-Guided Attention module to extract more comprehensive support features for few-shot segmentation in aerial images. Taking advantage of the support ground-truth masks, the area correlated to the foreground object is highlighted and enables the support encoder to extract comprehensive support features with contextual information. To facilitate reproducible studies of the task of few-shot semantic segmentation in aerial images, we further present the few-shot segmentation benchmark iSAID-, which is constructed from a large-scale iSAID dataset. Extensive experimental results including comparisons with the state-of-the-art methods and ablation studies demonstrate the effectiveness of the proposed method.

• Culinary science and hospitality research
• /
• v.15 no.3
• /
• pp.284-293
• /
• 2009

This study investigated chemical composition and antioxidant activity in the aerial parts or roots of Korean and Chinese Cirsium. Water was abundant in their roots than their aerial parts, and its content was similar between Korean Cirsium and Chinese one. Ash was abundant in the aerial parts of Korean one. Protein was abundant in the aerial parts of Korean one and the roots of Chinese one. Lipid content was abundant in the roots of Chinese one. Carbohydrate was abundant in the roots of Korean one while K content was abundant in the roots of Chinese one. Ca and Mg contents were abundant in the aerial parts of Korean one. P and Fe contents were abundant in the aerial parts of Korean one. Cu content was abundant in the aerial parts of Korean and Chinese ones. Vitamin A and vitamin C were not detected, but vitamin E and $\beta$-carotene were contained 5.35 IU/100 g, 1113.18 ${\mu}g$/100 g in the aerial parts of Korean one. Yield, total phenolics and total flavonoid contents were abundant in the roots of Korean one. Silymarin, cynarin, and narirutin were not detected in Korean and Chinese ones, but apigenin was identified in the aerial parts of Korean one. Acacetin was identified in the roots of Chinese one. Antioxidant activity was low in Korean and Chinese ones. These results suggest that nutrition composition, total phenolic content, total flavonoid content and apigenin content in the aerial parts of Korean Cirsium was superior than those in the roots of Korean and Chinese Cirsium.

• Korean Journal of Geomatics
• /
• v.3 no.1
• /
• pp.1-6
• /
• 2003

The purpose of this study is to make out a history of the coastline deformation with annual aerial photos of the target area and to suggest an efficient method for interpreting the coastline deformation. First, there were difficulties in obtaining annual aerial photos of an identical area as well as collecting periodical datum because of too much change of the real area during more than 27 years. Besides the past aerial photo is inferior to the latest one in quality. So there is nothing but to exclude an accuracy evaluation of the ortho photo. Taken into account the extent of the coastline change for 27 years is more than 10M and the all errors of the ortho image is included in this extent, both an accuracy and an error are ignored. The result of this study show that the coastline in the sea area of Namhangjin maintain advanced forward the ocean and keeps on moving with maintaining fairly wider shape of balance beach. Also The coastline deformation off l㎞ from the estuary is greatly irregular, which means the erosion of the sea in this area is in progress. The latest data being the aerial photo in 1996, it is difficult to find out the current conditions of a coastal erosion. However, considered the construction of a breakwater in Anmok Harbor is going on, the beach erosion becomes more accelerative recently. The aerial photos of the present Namhangjin's situation will make it possible to understand the history of the coastline change more accurately.

• The Journal of the Korea Contents Association
• /
• v.8 no.3
• /
• pp.270-278
• /
• 2008

This study : An analysis of position accuracy of topography according to using LiDAR data, aerial photo and digital map for a study area was conducted. The study area was selected in Hadan area, Sahagu, Busan aerial LiDAR data and aerial photo in the scales of 1:20,000, which are high tech surveying ways were used. The final digital orthoimage according to orientation process for each image and image resampling was producted. Using it, a checkpoint was chosen, information about the checkpoints selected was extracted, a coordinate of Horizontal Position through the screen digitizing was also extracted. Both the coordinates of LiDAR data and aerial photo using digital map in the scales of 1:20,000 announced to the public from NGII(National Geographic Information Institute) were gradually compared and analyzed. Based on the digital map, as a result of being both compared and analyzed, it has shown to us that horizontal position of aerial photo is more accurate than that of aerial LiDAR surveying (RMSE-building x:24cm, y:26cm).

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.64 no.7
• /
• pp.1081-1087
• /
• 2015

Recently, the researches of aerial drones are actively executed in various areas, the researches of surface drones and underwater drones are also executed in marine areas. In case of surface drones, they essentially utilize acoustic information by the sonar and consequently have the local information in the obstacle avoidance as the sonar has the limitations due to the beam width and detection range. In order to overcome this, more global method that utilizes optical images by the camera is required. Related to this, the aerial drone with the camera is desirable as the obstacle detection of the surface drone with the camera is impossible in case of the existence of clutters. However, the dynamic-floating aerial drone is not desirable for the long-term operation as its power consumption is high. To solve this problem, a collaborative obstacle avoidance method based on the acoustic information by the sonar of the surface drone and the optical image by the camera of the static-floating aerial drone is proposed. To verify the performance of the proposed method, the collaborative obstacle avoidances of a MSD(Micro Surface Drone) with an OAS(Obstacle Avoidance Sonar) and a BMAD(Balloon-based Micro Aerial Drone) with a camera are executed. The test results show the possibility of real applications and the need for additional studies.

• Journal of the Korea Society of Computer and Information
• /
• v.20 no.6
• /
• pp.87-97
• /
• 2015

An Aerial video summarization is not only the key to effective browsing video within a limited time, but also an embedded cue to efficiently congregative situation awareness acquired by unmanned aerial vehicle. Different with previous works, we utilize sensor operation mode of unmanned aerial vehicle, which is global, local, and focused surveillance mode in order for accurately summarizing the aerial video considering flight and surveillance/reconnaissance environments. In focused mode, we propose the moving-react tracking method which utilizes the partitioning motion vector and spatiotemporal saliency map to detect and track the interest moving object continuously. In our simulation result, the key frames are correctly detected for aerial video summarization according to the sensor operation mode of aerial vehicle and finally, we verify the efficiency of video summarization using the proposed mothed.

• Journal of the Society of Disaster Information
• /
• v.15 no.3
• /
• pp.427-439
• /
• 2019

Purpose: We generated geospatial information of unmanned aerial vehicle based on various SW and analyzed the location accuracy of orthoimage and DSM and texture mapping of 3D mesh. Method: The same unmanned aerial image data is processed using two different SW, and spatial information is generated. Among the generated spatial information, the orthoimage and DSM were compared with the spatial information generation results of the unmanned aerial photogrammetry SW by performing quantitative analysis by calculating RMSE of horizontal position and vertical position error and performing qualitative analysis. Results: There were no significant differences in the positional accuracy of the orthoimage and DSM generated by each SW, and differences in texture mapping in 3D mesh. The creation of the 3D mesh indicated the impact of the Unmanned Aerial Photogrammetry SW. Conclusion: It is shown that there is no effect of SW on the creation of orthoimage and DSM for geospatial analysis based on unmanned aerial vehicle. However, when 3D visualization is performed, texture mapping results are different depending on SW.

• Journal of Advanced Navigation Technology
• /
• v.25 no.3
• /
• pp.256-266
• /
• 2021

Domestic and foreign studies point to human factors as the main cause of unmanned aerial vehicle accidents, and HFACS is introduced as a technique to effectively analyze these human factors. Until now, domestic and foreign cases of analyzing the human factors of unmanned aerial vehicle accidents using HFACS were mainly targeted by military unmanned aerial vehicles, which can be used as an objective cause identification and similar accident prevention tool. In particular, identifying the focus of HFACS application considering the performance and operation conditions of domestic civilian unmanned aerial vehicles is expected to greatly help identify the cause and prevent recurrence in the event of an accident. Based on HFACS version 7.0, this study analyzed the accident investigation report data conducted by Korea Aviation and Railway Accident Investigation Board to identify the focus of HFACS application that can be used for domestic civilian unmanned aircraft accident investigations.