• Journal of Korea Water Resources Association
• /
• v.47 no.12
• /
• pp.1107-1119
• /
• 2014

Interesting in abnormal climate is currently growing because of climate change. With this, an increasing number of people continue to show concern over the negative effects of such changes. In Korea, the annual average rainfall amount increased to about 19% from 1,155 mm in the 1910s to 1,375 mm in the 2000s. By the end of the 21st century, it has been projected that rainfall will further increase to about 17%. In particular, the 10-year frequency of localized heavy rain of more than 100-mm rainfall per day reached 385 days in the last 10 years. As such, it increased 1.7 times from 222 in the 1970s-80s. The extreme events caused by climate change is thus reported as having exacerbated over the years. Gangwon-province will suffer more from climate change than any other region in Korea because of its mostly mountainous terrain. It is a special region with both mountainous and oceanic climates divided alongside the eastern and western regions of the Taebaek Mountain Range. As such, this paper try to quantify using ETCCDI (Expert Team on Climate Change Detection and Indices) the recent climate changes in this region.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.9
• /
• pp.17-24
• /
• 2020

This study examined the spectral information and reflectance of cracks of an embankment with drone-based hyperspectral imagery for crack detection. A Nano-Hyperspec mounted on a drone was used as a sensor, and hyperspectral videos of different intensities of illumination of the cracks on the embankment located in the downstream of Andong-Dam were obtained. An analysis of the data value of the illumination and peak data-value, the coefficients of determination were calculated to be 0.9864 of the uncracked areas and 0.9851 of the cracked area. The reflectance of each area showed a similar value and pattern, regardless of the intensity of illumination. This result may have occurred because the reference values of the white reference as the calculation criteria of reflectance varied according to the intensity of illumination. The reflectance at the cracked area was 5.65% lower in visible light and 4.58% lower in near-infrared light than that at the uncracked area. The detection of cracks may offer more precise results in further studies when the gimbal direction and camera angles of the drone are calibrated. Because hyperspectral imagery enables the detection of crack depths and types of clay minerals, which are difficult to identify in general RGB imagery, it can serve as a preemptive measure for evaluating the embankment stability.

• Korean Journal of Optics and Photonics
• /
• v.28 no.4
• /
• pp.166-171
• /
• 2017

Hydrogen gas is a green energy sources because it features no emission of pollutants during combustion. But hydrogen gas is very dangerous, being flammable and very explosive. Hydrogen gas detection is very important for the safety of a nuclear power plant. Hydrogen gas is generated by oxidation of nuclear fuel cladding during a critical accident, and leads to serious secondary damage in the containment building. This paper discusses the development of a Raman lidar system for remote detection and measurement of hydrogen gas. A small, portable Raman lidar system was designed, and a measurement algorithm was developed to quantitatively measure hydrogen gas concentration. To verify the capability of measuring hydrogen gas with the developed Raman lidar system, experiments were carried out under daytime outdoor conditions by using a gas chamber that can adjust the hydrogen gas density. As results, our Raman lidar system is able to measure a minimum density of 0.67 vol. % hydrogen gas at a distance of 20 m.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.6 no.3
• /
• pp.95-106
• /
• 2003

This study evaluated the architectural density of urban district using airborne laser scanning(ALS) that is a method used in urban planning, water resources and disaster prevention with high interest recently. First, digital elevation model(DEM) and digital surface model(DSM) was constructed from Light detection and ranging(LiDAR). For getting the height of building, ZONALMEAN filter was used in DEM and ZONALMAJORITY filter was used in DSM. This study compared the floor from filtering with the floor from survey and got standard error, which is ${\pm}0.199$ floor. Also, through the overlay and statistical analysis of total-area layer and zone layer, we could present floor area ratio by zone. As a result of comparison with floor area ratio between airborne laser scanning data and survey data, the standard error of floor area ratio shows ${\pm}2.68%$. Therefore, we expect that airborne laser scanning data can be a very efficient source to decision makers who set up landuse plan in near future.

• 재활복지
• /
• v.21 no.3
• /
• pp.129-146
• /
• 2017

In an event of an emergency, such as fire in a building, visually impaired and blind people are prone to exposed to a level of danger that is greater than that of normal people, for they cannot be aware of it quickly. Current fire detection methods such as smoke detector is very slow and unreliable because it usually uses chemical sensor based technology to detect fire particles. But by using vision sensor instead, fire can be proven to be detected much faster as we show in our experiments. Previous studies have applied various image processing and machine learning techniques to detect fire, but they usually don't work very well because these techniques require hand-crafted features that do not generalize well to various scenarios. But with the help of recent advancement in the field of deep learning, this research can be conducted to help solve this problem by using deep learning-based object detector that can detect fire using images from security camera. Deep learning based approach can learn features automatically so they can usually generalize well to various scenes. In order to ensure maximum capacity, we applied the latest technologies in the field of computer vision such as YOLO detector in order to solve this task. Considering the trade-off between recall vs. complexity, we introduced two convolutional neural networks with slightly different model's complexity to detect fire at different recall rate. Both models can detect fire at 99% average precision, but one model has 76% recall at 30 FPS while another has 61% recall at 50 FPS. We also compare our model memory consumption with each other and show our models robustness by testing on various real-world scenarios.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.25 no.6
• /
• pp.555-567
• /
• 2023

The cracks in the tunnel are currently determined through visual inspections conducted by inspectors based on images acquired using tunnel imaging acquisition systems. This labor-intensive approach, relying on inspectors, has inherent limitations as it is subject to their subjective judgments. Recently research efforts have actively explored the use of deep learning to automatically detect tunnel cracks. However, most studies utilize public datasets or lack sufficient objectivity in the analysis process, making it challenging to apply them effectively in practical operations. In this study, we selected test datasets consisting of images in the same format as those obtained from the actual inspection system to perform an objective evaluation of deep learning models. Additionally, we introduced ensemble techniques to complement the strengths and weaknesses of the deep learning models, thereby improving the accuracy of crack detection. As a result, we achieved high recall rates of 80%, 88%, and 89% for cracks with sizes of 0.2 mm, 0.3 mm, and 0.5 mm, respectively, in the test images. In addition, the crack detection result of deep learning included numerous cracks that the inspector could not find. if cracks are detected with sufficient accuracy in a more objective evaluation by selecting images from other tunnels that were not used in this study, it is judged that deep learning will be able to be introduced to facility safety inspection.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.28 no.4
• /
• pp.55-61
• /
• 2024

Recently, the number of aging concrete structures is steadily increasing. This is because many of these structures are reaching their expected lifespan. Such structures require accurate inspections and persistent maintenance. Otherwise, their original functions and performance may degrade, potentially leading to safety accidents. Therefore, research on objective inspection technologies using deep learning and computer vision is actively being conducted. High-resolution images can accurately observe not only micro cracks but also spalling and exposed rebar, and deep learning enables automated detection. High detection performance in deep learning is only guaranteed with diverse and numerous training datasets. However, surface damage to concrete is not commonly captured in images, resulting in a lack of training data. To overcome this limitation, this study proposed a method for generating concrete surface damage images, including cracks, spalling, and exposed rebar, using stable diffusion. This method synthesizes new damage images by paired text and image data. For this purpose, a training dataset of 678 images was secured, and fine-tuning was performed through low-rank adaptation. The quality of the generated images was compared according to three base models of stable diffusion. As a result, a method to synthesize the most diverse and high-quality concrete damage images was developed. This research is expected to address the issue of data scarcity and contribute to improving the accuracy of deep learning-based damage detection algorithms in the future.

• Environmental Engineering Research
• /
• v.12 no.5
• /
• pp.203-210
• /
• 2007

Since the emissions composition from the household products have potentially been associated with health risks for building occupants, the chemical composition emitted from the products should be surveyed. The current study identified the emission composition for 42 liquid household products, using a purge-and-trap method. This evaluation was done by classifying the household products into five product classes (deodorizers, household cleaners, color removers, pesticides, and polishes). Nineteen compounds were chosen on the basis of selection criteria. The quality control program for purge-and-trap and analytical systems included tests of laboratory blank Tenax traps and blank water samples, and the determination of calibration equation, measurement precision, method detection limit (MDL), and recovery. The number of chemicals varied according to the product categories, ranging from 4 for the product category of bleaches to 12 for the product categories of air fresheners and nail color removers. For all product categories, the emission composition and concentrations varied broadly according to product. It is noteworthy that most household products emit limonene: 19 of 25 cleaning products; 5 of 6 deodorizers; 1 of 3 pesticides; 3 of 3 color removers; and 4 of 5 polishes. It was suggested that the use of household products sold in Korea could elevate the formation of secondary toxic pollutants in indoor environments, by the reaction of limonene with ozone, which entered indoor environments or might be generated by indoor sources such as electronic air cleaning devices and copying machines.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.28 no.1
• /
• pp.83-92
• /
• 2010

Emergence of high-resolution digital aerial cameras and airborne laser scanners have made innovative progress in photogrammetry and spatial information technology. The purpose of this study is to generate true orthoimage by recovering occlusion areas. The orthoimages were generated patch-based transformation. The occlusion areas were mutually corrected by using multiple aerial images. This study proposed a novel method of building roof based orthoimage generation and an effective method of occlusion area detection and recovery. The proposed methods could be efficient to generate true orthoimages in urban areas where occlusion areas are problematic.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.9 no.5
• /
• pp.109-114
• /
• 2009

At the present time, buildings are designed higher and more complex than ever before. Therefore the potential disasters are happened such as fire, power outage, earthquake, flood, hurricanes. Their disasters require people inside buildings to be evacuated as soon as possible. This paper presents a new disaster evacuation guidance concept of inner buildings, whiche aims at integrated the constructing of a sensor network and a guidance light networks in order to provide a quick detection of disasters and accurate evacuation guidance based on indoor geo-information, and sends these instructions to people. In this paper, we present the integrated logical model based on sensor and guidance light networks for the fire disaster management in inner building using our concept. And we verify proposed logical model according to experiments with visualization and operations on map.