• Journal of the Society of Disaster Information
• /
• v.16 no.1
• /
• pp.79-86
• /
• 2020

Purpose: In this study, a model was developed to estimate the storage in Cheonan reservoir using images taken by Sentinel-1 satellite. Method: A total of three reservoirs were studied. All three reservoirs are small reservoirs whose water level is being measured. The preprocessing of Sentinel-1 images was done using SNAP distributed by the European Space Agency(ESA), and the storage was estimated by classifying water surface by the threshold classification method. The estimated reservoir area was compared with satellite and drones images taken on the same day. The correlation was derived by comparing the estimated reservoir area with the actual measurement. Results and Conclusions: The storage values estimated by satellite image analysis showed similar values to the actual measurement data. However, because of the underestimation of the reservoir area due to green algae and Epilithic diatom of summer reservoirs and the low resolution of satellite images, it is dificult to detect reservoir area by satellite images less than 10,000㎡.

• Korean Journal of Remote Sensing
• /
• v.34 no.6_1
• /
• pp.1119-1130
• /
• 2018

Soyang Lake has been contributing to economic growth by preventing flood damage in the metropolitan area, the water level in the middle to upper flow of lake has been greatly decreased due to the drought in 2015. In order to restore the existing flow rate, Bupyungbo has been built in Bupyeong-ri, Shin Nam-myeon, Inje-gun to cause artificial changes on the sedimentary environment of Bupyeong freshwater region. Therefore, this study intends to confirm the changes of sedimentary environment since Bupyeongbo has been utilized. For this study, we used the Sentinel-2 satellite image data periodically to measure the dimension of water according to the volume of water kept near Bupyung district and analyzed the particle size and the percentage of water content of the sediments through field study. The Sentnel-2 satellite images showed us how the water surface has been changed and that during the period from September 2017 to October 2018, the minimum and maximum area of water surface was observed in June 2018 and in January 2018, respectively. In addition, we find that the smaller being the particle size, the higher having the water content and that there is higher the correlation between the water content and the grain size of the sediment layer. Hereafter, if we will acquire the drone images at Bupyung district, we expect that we will be able to measure the distribution of sediments in the same area according to different time periods and observe various kinds of sediment through field work.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.40 no.1
• /
• pp.15-22
• /
• 2022

In order to improve the accuracy of the deep learning object detection technique, the effect of magnification rate conditions and seasonal factors on detection accuracy in aerial photographs and drone images was analyzed through experiments. Among the deep learning object detection techniques, Mask R-CNN, which shows fast learning speed and high accuracy, was used to detect the vehicle to be detected in pixel units. Through Seoul's aerial photo service, learning images were captured at different screen magnifications, and the accuracy was analyzed by learning each. According to the experimental results, the higher the magnification level, the higher the mAP average to 60%, 67%, and 75%. When the magnification rates of train and test data of the data set were alternately arranged, low magnification data was arranged as train data, and high magnification data was arranged as test data, showing a difference of more than 20% compared to the opposite case. And in the case of drone images with a seasonal difference with a time difference of 4 months, the results of learning the image data at the same period showed high accuracy with an average of 93%, confirming that seasonal differences also affect learning.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.38 no.5
• /
• pp.435-441
• /
• 2020

Sentinel-2A satellite imagery provides VNIR (Visible Near InfraRed) and SWIR (ShortWave InfraRed) wavelength bands, and it is known to be effective for land cover classification, cloud detection, and environmental monitoring. Greenhouse is one of the middle classification classes for land cover map provided by the Ministry of Environment of the Republic of Korea. Since greenhouse is a class that has a lot of changes due to natural disasters such as storm and flood damage, there is a limit to updating the greenhouse at a rapid cycle in the land cover map. In the present study, we utilized Sentinel-2A satellite images that provide both VNIR and SWIR bands for the detection of greenhouse. To utilize Sentinel-2A satellite images for the detection of greenhouse, we produced high-resolution SWIR bands applying to the fusion technique performed in two stages and carried out the detection of greenhouse using SVM (Support Vector Machine) supervised classification technique. In order to analyze the applicability of SWIR bands to greenhouse detection, comparative evaluation was performed using the detection results applying only VNIR bands. As a results of quantitative and qualitative evaluation, the result of detection by additionally applying SWIR bands was found to be superior to the result of applying only VNIR bands.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.24 no.3
• /
• pp.1-9
• /
• 2021

In this study, we evaluated the accuracy of deep learning-based tree species classification model trained by using high-resolution images. We selected five species classed, i.e., pine, birch, larch, korean pine, mongolian oak for classification. We created 5,000 datasets using high-resolution orthophoto and forest type map. CNN deep learning model is used to tree species classification. We divided training data, verification data, and test data by a 5:3:2 ratio of the datasets and used it for the learning and evaluation of the model. The overall accuracy of the model was 89%. The accuracy of each species were pine 95%, birch 89%, larch 80%, korean pine 86% and mongolian oak 98%.

• Korean Journal of Remote Sensing
• /
• v.36 no.1
• /
• pp.29-39
• /
• 2020

Forest fire is a critical disaster that causes massive destruction of forest ecosystem and economic loss. Hence, accurate estimation of the burned area is important for evaluation of the degree of damage and for preparing baseline data for recovery. Since most of the area size damaged by wildfires in Korea is less than 1 ha, it is necessary to use satellite or drone images with a resolution of less than 10m for detecting the damage area. This paper aims to detect wildfire-damaged area from a Kompsat-3 image using the indices such as NDVI (normalized difference vegetation index) and FBI (fire burn index) and to examine the classification characteristics according to the methods such as Otsu thresholding and ISODATA(iterative self-organizing data analysis technique). To mitigate the salt-and-pepper phenomenon of the pixel-based classification, a gaussian filter was applied to the images of NDVI and FBI. Otsu thresholding and ISODATA could distinguish the burned forest from normal forest appropriately, and the salt-and-pepper phenomenon at the boundaries of burned forest was reduced by the gaussian filter. The result from ISODATA with gaussian filter using NDVI was closest to the official record of damage area (56.9 ha) published by the Korea Forest Service. Unlike Otsu thresholding for binary classification,since the ISODATA categorizes the images into multiple classes such as(1)severely burned area, (2) moderately burned area, (3) mixture of burned and unburned areas, and (4) unburned area, the characteristics of the boundaries consisting of burned and normal forests can be better expressed. It is expected that our approach can be utilized for the high-resolution images obtained from other satellites and drones.

• Korean Journal of Remote Sensing
• /
• v.38 no.1
• /
• pp.83-101
• /
• 2022

Accurate atmospheric correction is essential for the analysis of land surface and environmental monitoring. Aerosol optical depth (AOD) information is particularly important in atmospheric correction because the radiation attenuation by Mie scattering makes the differences between the radiation calculated at the satellite sensor and the radiation measured at the land surface. Thus, it is necessary to use high-quality AOD data for an appropriate atmospheric correction of high-resolution satellite images. In this study, we examined the Second Simulation of a Satellite Signal in the Solar Spectrum (6S)-based atmospheric correction results for the Sentinel-2 images in South Korea using raster AOD (MODIS) and single-point AOD (AERONET). The 6S result was overall agreed with the Sentinel-2 level 2 data. Moreover, using raster AOD showed better performance than using single-point AOD. The atmospheric correction using the single-point AOD yielded some inappropriate values for forest and water pixels, where as the atmospheric correction using raster AOD produced stable and natural patterns in accordance with the land cover map. Also, the Sentinel-2 normalized difference vegetation index (NDVI) after the 6S correction had similar patterns to the up scaled drone NDVI, although Sentinel-2 NDVI had relatively low values. Also, the spatial distribution of both images seemed very similar for growing and harvest seasons. Future work will be necessary to make efforts for the gap-filling of AOD data and an accurate bi-directional reflectance distribution function (BRDF) model for high-resolution atmospheric correction. These methods can help improve the land surface monitoring using the future Compact Advanced Satellite 500 in South Korea.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.8 no.4
• /
• pp.205-212
• /
• 2013

In recent, the demand of Unmanned Aerial Vehicles (UAVs) that can autonomous navigation and remote control has been increased in military, civil and commercial field. Particularly, existing researches focused on autonomous navigation method based on vanish point and remote control method based on event processing in indoor environments. However, the existing methods have some problems. For instance, a detected vanish point in intersection point has too much detection errors. In addition, the delay is increased in existing remote control system for processing images in real time. Thus, we propose improved vanish point algorithm by removing detection errors in intersection point. We also develop a remote control system with android platform by separating flying control and image process. Finally, we compare the proposed methods with existing methods to show the improvement of our approaches.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2024.05a
• /
• pp.229-230
• /
• 2024

최근 드론으로 극한 환경에서 범죄 수배자 및 실종자를 탐색하는 시도가 활발하다. 이때 생체 인증 기술인 얼굴 인증 기술을 사용하면 탐색 효율이 높아지지만, 암호화되지 않은 인증 프로토콜 적용 시 생체 정보 유출의 위험이 있다. 본 논문에서는 드론이 수집한 얼굴 이미지 템플릿을 암호화하여 안전하게 인증할 수 있는 효율적인 생체 인증 프레임워크인 DF-PPHDM(Privacy-Preserving Hamming Distance biometric Matching for Drone-collected Facial images)을 제안한다. 수집된 얼굴 이미지는 암호문 형태로 서버에 전달되며 서버는 기존 등록된 암호화된 템플릿과의 Hamming distance 분석을 통해 검증한다. 제안한 DF-PPHDM을 RaspberryPI 4B 환경에서 직접 실험하여 분석한 결과, 한정된 리소스를 소유한 드론에서 효율적인 구현이 가능하며, 인증 단계에서 7.83~155.03 ㎲ (microseconds)가 소요된다는 것을 입증하였다. 더불어 서버는 드론이 전송한 암호문으로부터 생체 정보를 복구할 수 없으므로 프라이버시 침해 문제를 예방할 수 있다. 향후 DF-PPHDM에 AI(Artificial Intelligence)를 결합하여 자동화 기능을 추가하고 코드 최적화를 통해 성능을 향상시킬 예정이다.

• Journal of Multimedia Information System
• /
• v.4 no.4
• /
• pp.195-204
• /
• 2017

Mobile robots are used for years as a valuable research and educational tool in form of available open-platform designs and Do-It-Yourself kits. Rapid development and costs reduction of Unmanned Air Vehicles (UAV) and ground based mobile robots in recent years allowed researchers to utilize them as an affordable research platform. Despite of recent developments in the area of ground and airborne robotics, only few examples of Unmanned Surface Vehicle (USV) platforms targeted for research purposes can be found. Aim of this paper is to present the development of open-design USV drone with integrated multi-level control hardware architecture. Proposed catamaran - type water surface drone enables direct control over wireless radio link, separate development of algorithms for optimal propulsion control, navigation and communication with the ground-based control station. Whole design is highly modular, where each component can be replaced or modified according to desired task, payload or environmental conditions. Developed USV is planned to be utilized as a part of the system for detection and identification of marine and lake waste. Cameras mounted to the USV would record sea or lake surfaces, and recorded video sequences and images would be processed by state-of-the-art computer vision and machine learning algorithms in order to identify and classify marine and lake waste.