• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.18 no.6
• /
• pp.155-163
• /
• 2019

This paper proposes a new methodology to recognize cracks on asphalt road surfaces using the image data obtained with drones. The target section was Yuseong-daero, the main highway of Daejeon. Furthermore, two object detection algorithms, such as Tiny-YOLO-V2 and Faster-RCNN, were used to recognize cracks on road surfaces, classify the crack types, and compare the experimental results. As a result, mean average precision of Faster-RCNN and Tiny-YOLO-V2 was 71% and 33%, respectively. The Faster-RCNN algorithm, 2Stage Detection, showed better performance in identifying and separating road surface cracks than the Yolo algorithm, 1Stage Detection. In the future, it will be possible to prepare a plan for building an infrastructure asset-management system using drones and AI crack detection systems. An efficient and economical road-maintenance decision-support system will be established and an operating environment will be produced.

• Journal of Korea Planning Association
• /
• v.53 no.7
• /
• pp.109-126
• /
• 2018

The purpose of this study is to identify the spatial types for thermal environment improvement considering heat flux and its spatial context through empirical orthodox formulas. First, k-means clustering was used to classify values of three kinds of heat flux - latent, sensible and storage heat. Next, from the k-means clustering, we defined a type of thermal environment (type LHL) where improvement is needed for more comfortable and pleasant thermal environment in the city, among the eight types. Lastly, we compared and analyzed the characteristics of each classified thermal environmental types based on land cover types. From the study, we found that the ratio of impervious surfaces, roads, and buildings of the type LHL is higher than those of the type HLH (relatively thermal comfort environment). In order to improve the thermal environment, the following contents are proposed to urban planners and designers depending on the results of the study. a) Increase the green zone rate by 10% to reduce sensible heat; b) Reduce the percentage of impermeable surfaces and roads by 10% ; c) Latent heat increases when water and green spaces are expanded. This study will help to establish a minimum criterion for a land cover rate for the improvement of the urban thermal environment and a standard index for the thermal environmental improvement can be derived.

• Proceedings of the KSRS Conference
• /
• 2007.10a
• /
• pp.587-590
• /
• 2007

The demand for more accurate and realistic 3D urban models has been increasing more and more. Many studies have been conducted to extract 3D features from remote sensing data such as satellite images, aerial photos, and airborne laser scanning data. In this paper a technique is presented to extract and reconstruct 3D buildings in urban areas using airborne laser scanning data. Firstly all points in a building were divided into some groups by height difference. From segmented laser scanning data of irregularly distributed points we generalized and regularized building boundaries which better approximate the real boundaries. Then the roof points which are subject to the same groups were classified using pre-defined models by least squares fitting. Finally all parameters of the roof surfaces were determined and 3D building models were constructed. Some buildings with complex shapes were selected to test our presented algorithms. The results showed that proposed approach has good potential for reconstructing complex buildings in detail using only airborne laser scanning data.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.27 no.1
• /
• pp.39-53
• /
• 1999

This study quantified carbon uptake and emissions in urban landscape, and the role of urban greenspace in atmospheric carbon reduction for several cities of Chuncheon and Kangleung in Kangwon province. Mean carbon storage by trees and shrubs was 26.0 t (mertric tons)/ha in Chuncheon and 46.7 t/ha in Kangleung for natural lands, and ranged from 4.7 to 6.3 t/ha for urban lands (all land use types except natural and agricultural lands) in both cities. Mean annual carbon uptake by trees and shrubs ranged from 1.60 to 1.71 t/ha/yr for natural lands, and from 0.56 to 0.71 t/ha/yr for urban lands. There was no significant difference (95% confidence level) between the two cities in the carbon storage and annual carbon uptake per ha, except the carbon storage for natural lands. Organic carbon storage in soils (to a depth of 60 cm) of Chuncheon average 24.8 t/ha for urban lands and 31.6 t/ha for natural lands, 1.3 times greater than for urban lands. Annual carbon accumulation in soils was 1.3 t/hr/yr for natural lands of the study cities. Annual per capita carbon emissions from fossil fuel consumption were 1.3 t/yr in Chunceon and 1.8 t/yr in Kangleung. The principal carbon release in urban landscapes was from transport and industry. Total carbon storage by urban greenspace (trees, shrubs, and soils) equaled 66% of total carbon emissions in Chuncheon and 101% in Kangleung. Carbon uptake by urban greenspace annually offset total carbon emissions by approximately 4% in the study cities. Thus, urban greenspace played a partial important role in reducing atmospheric $CO_2$ concentrations. To increase $CO_2$ uptake and storage by urban greenspace, suggested are conservation of natural lands, minimization of hard surfaces and more plantings, selection of tree species with high growth rate, and proper management for longer healthy tree growth.

• KIEAE Journal
• /
• v.15 no.1
• /
• pp.117-127
• /
• 2015

The cool surface ratio could be used as a proxy of the overall thermal environment contributing to heat islands in urban area. This research proposes a comparative evaluation framework in an objective and quantitative way for measuring cool surface ratios. Two university campuses (Kyungpook National University: KNU, South Korea and UC Davis: University of California, Davis, USA) were selected as case study sites in order to monitor cool surface condition. Google Earth combined with digital maps realistically identified the major type of cool surfaces such as cool roofs and water bodies in the study area. Cool surfaces were sparsely identified over the KNU campus while the UC Davis campus was heavily covered by cool surfaces such as cool roofs and water bodies, resulting in almost four times more first-grade cool surfaces, as compared to KNU. It is confirmed that standard remote sensing technology can offer the viable method of measuring and comparing the campus-wide cool surface condition. It is anticipated that this research output could be used as a valuable reference to initiate a nation-wide cool surface strategy since objective evidence has been provided based on area-wide measurement for the cool surface in the two university context.

• KCID journal
• /
• v.18 no.2
• /
• pp.3-14
• /
• 2011

Increased impervious surfaces alter stream hydrology resulting in lower flows during droughts and higher peak flows during floods. Not only urban area but also rural area has been expanded impervious surfaces because of increasing of greenhouses. The main objective of this study was to evaluate the performance of the AnnAGNPS (Annualized Non-Point Source Pollution Model) on the surface runoff characteristics of the Daejeon-Stream watershed, and to predict the hydrological effects due to increasing of impervious surfaces. The model parameters were obtained from the geographical information system (GIS) databases, and additional parameters calibrated with the observed data. The model was calibrated by using 2004 of the runoff data and validated by using 2002 data obtained from WAMIS (Water Management Information System) to compare the simulated results for the study watershed. R2 values and efficiency index (EI) between observed and simulated runoff were 0.78 and 0.80, respectively at the calibration period. In this study, expanding of impervious surfaces such as greenhouses caused increasing of surface runoff, but caused decreasing of total nitrogen and total phosphorus loads.

• International Journal of High-Rise Buildings
• /
• v.9 no.4
• /
• pp.351-360
• /
• 2020

Monitoring civil structures periodically is necessary for ensuring the fitness of the structures. Cracks on inner and outer surfaces of the building plays a vital role in indicating the health of the building. Conventionally, human visual inspection techniques were carried up to human reachable altitudes. Monitoring of high rise infrastructures cannot be done using this primitive method. Also, there is a necessity for more accurate prediction of cracks on building surfaces for ensuring the health and safety of the building. The proposed research focused on developing an efficient crack classification model using Transfer Learning enabled EfficientNet (TL-EN) architecture. Though many other pre-trained models were available for crack classification, they rely on more number of training parameters for better accuracy. The TL-EN model attained an accuracy of 0.99 with less number of parameters on large dataset. A bench marked METU dataset with 40000 images were used to test and validate the proposed model. The surfaces of high rise buildings were investigated using vision enabled Unmanned Arial Vehicles (UAV). These UAV is fabricated with TL-EN model schema for capturing and analyzing the real time streaming video of building surfaces.

• Proceedings of the KSRS Conference
• /
• 2004.10a
• /
• pp.10-13
• /
• 2004

Hyperspectral data has a great advantage to classify various surface materials that are spectrally similar. In this study, we attempted to classify man-made materials in urban area using Hyperion data. Hyperion imagery of Seoul was initially processed to minimize radiometric distortions caused by sensor and atmosphere. Using color aerial photographs. we defined seven man-made surfaces (concrete, asphalt road. railroad, buildings, roof, soil, shadow) for the classification in Seoul. The hyperspectral data showed the potential to identify those manmade materials that were difficult to be classified by multispectral data. However. the classification of road and buildings was not quite satisfactory due to the relatively low spatial resolution of Hyperion image. Further, the low radiometric quality of Hyperion sensor was another limitation for the application in urban area.

• Journal of Environmental Science International
• /
• v.24 no.9
• /
• pp.1211-1220
• /
• 2015

Soil is the most important factor in natural environment for bio-diversity. Urbanization and development of city devastate urban soil by the fraternization of green network and run off pollution. In these facts, preservation of soil is the main issue in maintain of quality urban environment. In order to handle this issue, the gold network that link fragment soil patches is considered in maintain quality soil. This study researched the infiltration Treebox design technique based on the Low Impact Development. This technique suggest reduction of impervious area of the soil due to urbanization. The main concept of this study is encourage more permeable surfaces in urban area by using a infiltration planter. The function of the planter is hold run off as much as possible from intensive rainfall, and utilizes it in drought season. Also, this planter provides fertile soil for organism habitat by keeping appropriate moisture supplying.

• Journal of Environmental Science International
• /
• v.11 no.10
• /
• pp.1045-1054
• /
• 2002

Urban atmospheric conditions are usually settled as warmer, drier and dirtier than those of rural counterpart owing to reduction of green space and water space area heat retention in surfaces such as concrete and asphalt, and abundant fuel consumption. The characteristics of urban climate has become generally known as urban heat island. The purpose of this study is to investigate the temporal and spatial distribution of the heat emission from human activity, which is a main factor causing urban heat island. In this study, the anthropogenic heat fluxes emitted from vehicles and constructions are estimated by computational grid mesh which is divided by 1km $\times$ 1km. The anthropogenic heat flux by grid mesh can be applied to a numerical simulation model of the local circulation model. The constructions are classified into 9 energy-consumption types - hospital, hotel, office, department store, commercial store, school, factory, detached house and flat. The vehicles classified into 4 energy-consumption types - car, taxi, truck and bus. The seasonal mean of anthropogenic heat flux around central Daegu exceeded $50 W/m^2$ in winter. The annual mean anthropogenic heat flux exceeded $20 W/m^2$. The values are nearly equivalent to the anthropogenic heat flux in the suburbs of Tokyo, Japan.