• Journal of Korean Society for Atmospheric Environment
• /
• v.22 no.5
• /
• pp.661-678
• /
• 2006

Products developed in this research is a software which can transfer the type of shape(.shp) into the type of ascii using the land cover data and the topography data in the metropolitan area of Seoul. In addition, it can calculate the $CO_2$ flux according to distribution of plants within the land cover data. The $CO_2$ flux is calculated by the experimental equation which is compose of the meteorological parameters such as the solar radiation and the air temperature. The net flux was shown in about $-19ton/km^2$ by removing $CO_2$ through the photosynthesis during daytime, and in 2 ton/km2 by producing it through the respiration during nighttime on 10 August 2004, the maximum day of air temperature during the period of 3yr(2001 to 2004), in the metropolitan area of Seoul. Spatial distribution of the air temperature and the wind field is simulated by substituting the middle classification of the land cover map data, upgraded by the Korean Ministry of Environment(KME), for the land-use data of the United States Geological Survey(USGS) within the Meteorological Mesoscale Model Version 5(MM5) on 10 August 2006 in the metropolitan area of Seoul. Difference of the air temperature between both data was shown in the maximum range of $-2^{\circ}C\;to\;2.9^{\circ}C$, and the air temperature due to the land use data of KME was higher than that of USGS in average $0.4^{\circ}C$. Also, those of wind vectors were meanly lower than that of USGS in daytime and nighttime. Furthermore, the hourly time series of Volatile Organic Components(VOCs) is calculated by using the Biosphere Emission and Interaction System Version 2(BEIS2) including the new land cover data and the meteorological parameters such as the air temperature and so]ar insolation. It is possible to calculate the concentration of ozone due to the biogenic emission of VOCs.

• Korean Journal of Remote Sensing
• /
• v.22 no.2
• /
• pp.101-110
• /
• 2006

The accuracy of rural land cover using airborne multispectral images and LEAR (Light Detection And Ranging) data was analyzed. Multispectral image consists of three bands in green, red and near infrared. Intensity image was derived from the first returns of LIDAR, and vegetation height image was calculated by difference between elevation of the first returns and DEM (Digital Elevation Model) derived from the last returns of LIDAR. Using maximum likelihood classification method, three bands of multispectral images, LIDAR vegetation height image, and intensity image were employed for land cover classification. Overall accuracy of classification using all the five images was improved to 85.6% about 10% higher than that using only the three bands of multispectral images. The classification accuracy of rural land cover map using multispectral images and LIDAR images, was improved with clear difference between heights of different crops and between heights of crop and tree by LIDAR data and use of LIDAR intensity for land cover classification.

• Korean Journal of Remote Sensing
• /
• v.36 no.6_2
• /
• pp.1591-1604
• /
• 2020

This study is a study on classifying land cover by applying high-resolution satellite images to deep learning algorithms and verifying the performance of algorithms for each spatial object. For this, the Fully Convolutional Network-based algorithm was selected, and a dataset was constructed using Kompasat-3 satellite images, land cover maps, and forest maps. By applying the constructed data set to the algorithm, each optimal hyperparameter was calculated. Final classification was performed after hyperparameter optimization, and the overall accuracy of DeeplabV3+ was calculated the highest at 81.7%. However, when looking at the accuracy of each category, SegNet showed the best performance in roads and buildings, and U-Net showed the highest accuracy in hardwood trees and discussion items. In the case of Deeplab V3+, it performed better than the other two models in fields, facility cultivation, and grassland. Through the results, the limitations of applying one algorithm for land cover classification were confirmed, and if an appropriate algorithm for each spatial object is applied in the future, it is expected that high quality land cover classification results can be produced.

• Journal of Korean Society for Geospatial Information Science
• /
• v.22 no.4
• /
• pp.53-61
• /
• 2014

Landsat OLI satellite imagery can be applied to various remote sensing applications, such as generation of land cover map, urban area analysis, extraction of vegetation index and change detection, because it includes various multispectral bands. In addition, land cover map is an important information to monitor and analyze land cover using GIS. In this paper, land cover map is generated by using Landsat OLI and existing land cover map. First, training dataset is obtained using correlation between existing land cover map and unsupervised classification result by K-means, automatically. And then, spectral signatures corresponding to each class are determined based on training data. Finally, abundance map and land cover map are generated by using iterative spectral mixture model. The experiment is accomplished by Landsat OLI of Cheongju area. It shows that result by our method can produce land cover map without manual training dataset, compared to existing land cover map and result by supervised classification result by SVM, quantitatively and visually.

• Korean Journal of Remote Sensing
• /
• v.22 no.3
• /
• pp.211-219
• /
• 2006

A random forests classifier is applied to multi-sensor data fusion for supervised land-cover classification in order to account for the importance of variable. The random forests approach is a non-parametric ensemble classifier based on CART-like trees. The distinguished feature is that the importance of variable can be estimated by randomly permuting the variable of interest in all the out-of-bag samples for each classifier. Two different multi-sensor data sets for supervised classification were used to illustrate the applicability of random forests: one with optical and polarimetric SAR data and the other with multi-temporal Radarsat-l and ENVISAT ASAR data sets. From the experimental results, the random forests approach could extract important variables or bands for land-cover discrimination and showed reasonably good performance in terms of classification accuracy.

• Journal of Environmental Science International
• /
• v.19 no.9
• /
• pp.1103-1108
• /
• 2010

This study was conducted to be used as basic data of environmental friendly construction planning by comparing and analyzing thermal environment, find particles and biotope area rate according to land cover types of outside space of schools located in Chung-ju. When meteorological factors were analyzed according to land cover types, for temperature planting area and paved area showed low-and high-temperature ranges, respectively, and relative humidity was negatively related with temperature as low-and high-temperature ranges corresponded to high-and low-humidity ranges, respectively. For Wet Bulb Globe Temperature Index (WBGT) by land cover types, it was observed to be artificial grass> bare land> natural grass. Find particles were different according to land cover types of playground with being bare land> artificial grass> natural grass in the order. Bare land playground, where there were artificial factors and no absorption of fine particles through stomata of leaves as a function of natural circulation, recorded the highest level of $39.8\;{\mu}g/m^3$ and the level was relatively higher compared to the levels by season in Chung-ju. Biotope area rate showed the order of M elementary school> K elementary school> C commercial high school. That was considered to be caused by the difference of land cover type of school playground accounting for a large part of a school.

• Atmosphere
• /
• v.15 no.2
• /
• pp.75-90
• /
• 2005

This study examines the impacts of land cover changes on the East Asia summer monsoon with the National Center for Atmospheric Research Regional Climate Model (NCAR RegCM2), coupled with Biosphere Atmosphere Transfer Scheme (BATS). To assess the goals, two types of land cover maps were used in the simulation of summer climate. One type was NCAR land cover map (CTL) and the other was current land cover map derived from satellite data (land cover: LCV). Warm and cold surface temperature biases of $1-3^{\circ}C$ occurred over central China and Mongolia in CTL. The model produced excessive precipitation over northern land area but less over southern ocean of the model domain. Changes of biophysical parameters, such as albedo, minimum stomatal resistance and roughness length, due to the land cover changes resulted in the alteration of land-atmosphere interactions. Latent heat flux and wind speed in LCV increased noticeably over central China where deciduous broad leaf trees have been replaced by mixed farm and irrigated crop. As a result, the systematic warm biases over central China were greatly reduced in LCV. Strong cooling of central China decreased pressure gradient between East Asian continent and Pacific Ocean. The decreased pressure gradient suppressed the northward transport of moisture from south China and South China Sea. These changes reduced not only the excessive precipitation over north China and Mongolia but also less precipitation over south China. However, the land cover changes increased the precipitation over the Korean Peninsula and the Japan Islands, especially in July and August.

• Korean Journal of Remote Sensing
• /
• v.27 no.4
• /
• pp.435-444
• /
• 2011

The land cover map derived from spectral features of high resolution optical images has low spectral resolution and heterogeneity in the same land cover class. For this reason, despite the same land cover class, the land cover can be classified into various land cover classes especially in vegetation area. In order to overcome these problems, detailed vegetation classification is applied to optical satellite image and SAR(Synthetic Aperture Radar) integrated data in vegetation area which is the result of pre-classification from optical image. The pre-classification and vegetation classification were performed with MLC(Maximum Likelihood Classification) method. The hierarchical land cover classification was proposed from fusion of detailed vegetation classes and non-vegetation classes of pre-classification. We can verify the facts that the proposed method has higher accuracy than not only general SAR data and GLCM(Gray Level Co-occurrence Matrix) texture integrated methods but also hierarchical GLCM integrated method. Especially the proposed method has high accuracy with respect to both vegetation and non-vegetation classification.

• Korean Journal of Remote Sensing
• /
• v.22 no.6
• /
• pp.565-574
• /
• 2006

Satellite images have been used to obtain land cover information that is one of important factors for hydrological analysis over a large area. In urban area, more detailed land cover data are often required for hydrological analysis because of the relatively complex land cover types. The number of land cover classes that can be classified with traditional multispectral data is usually less than the ones required by most hydrological uses. In this study, we present the capabilities of hyperspectral data (Hyperion) for the classification of hydrological land cover types in urban area. To obtain 17 classes of urban land cover defined by the USDA SCS, spectral mixture analysis was applied using eight endmembers representing both impervious and pervious surfaces. Fractional values from the spectral mixture analysis were then reclassified into 17 cover types according to the ratio of impervious and pervious materials. The classification accuracy was then assessed by aerial photo interpretation over 10 sample plots.

• Korean Journal of Remote Sensing
• /
• v.28 no.2
• /
• pp.203-214
• /
• 2012

The physical environment of urban areas covered mostly by concrete and asphalt is the main cause of the urban heat island effect, primarily becoming apparent through increased land surface temperature. This study examined the effect of different urban land cover types on the land surface temperature using MODIS, Landsat ETM+ and RapidEye satellite data. As a result, the remote sensing based land surface temperature showed a marked difference according to the land use pattern in the case study of Ilsan new city. The high-rise apartment residential districts with less building-to-land ratio and higher green area ratio revealed lower land surface temperature than the low-story single-family housing districts characterized by relatively high building-to-land ratio and low green area ratio. From the view of climate zone and land cover types, there is a strong linear correlation between the impervious land cover ratio and the land surface temperature; the land surface temperature increases as the impervious built-up areas expand. In contrast, vegetation;water and shadow areas affect the decrease of land surface temperature. There is also a negative (-) correlation between NDVI and land surface temperature but the seasonal variation of NDVI can be hardly corrected.