• Journal of the Korean Association of Geographic Information Studies
• /
• v.18 no.3
• /
• pp.76-88
• /
• 2015

This paper presents a classification method based on the combination of active learning with temporal contextual information extracted from past land-cover maps for the classification of crop cultivation areas. Iterative classification based on active learning is designed to extract reliable training data and cultivation rules from past land-cover maps are quantified as temporal contextual information to be used for not only assignment of training data but also relaxation of spectral ambiguity. To evaluate the applicability of the classification method proposed in this paper, a case study with MODIS time-series vegetation index data sets and past cropland data layers(CDLs) is carried out for the classification of corn and soybean in Illinois state, USA. Iterative classification based on active learning could reduce misclassification both between corn and soybean and between other crops and non crops. The combination of temporal contextual information also reduced the over-estimation results in major crops and led to the best classification accuracy. Thus, these case study results confirm that the proposed classification method can be effectively applied for crop cultivation areas where it is not easy to collect the sufficient number of reliable training data.

• Journal of Korean Society for Geospatial Information Science
• /
• v.24 no.1
• /
• pp.121-128
• /
• 2016

Many studies are focused on image data and classifier for comparison or improvement of classification accuracy. Therefore studies are needed aspect of the training samples on supervised classification which depend on reference data or skill of analyst. This study tries to assess usability of field spectra as training samples on supervised classification. Classification accuracies of hyperspectral and multispectral images were assessed using training samples from image itself and field spectra, respectively. The results shown about 90% accuracy with training sample collected from image. Using field spectra as training sample, accuracy was decreased 10%p for hyperspectral image, and 20%p for multispectral image. Especially, some classes shown very low accuracies due to similar spectral characteristics on multispectral image. Therefore, field spectra might be used as training samples on classification of hyperspectral image, although it has limitation for multispectral image.

• Korean Journal of Environment and Ecology
• /
• v.24 no.6
• /
• pp.751-762
• /
• 2010

In order to understand the influence that the plant cover condition of the high-rise building outer space causes to the temperature change, we selected 12 high-rise building constructed in Seoul City. The land cover type of the outside was classified into six type(outer road, paved surface, shrub/grassland, single-layer tree planting-site, multi-layer planting-site, and waterscape facilities) and the temperature was measured at the representative point for each type in order to analyze the land cover temperature differential for each type of the high-rise building outer space. The study area showing the temperature tendency to be similar based upon one way analysis of variance after selecting the central part of the outer road for a control and measuring a temperature in order to consider the neighboring environmental difference of the dozen building was classified into 4 groups. As to the one-way layout result of variance analysis with the land cover type of the classified group and outer space temperature, the single-layer tree planting-site, waterscape facilities, and multi-layer planting-site belonged mainly to the low temperature section. The shrub/grassland, paved surface, and outer road belonged to the high temperature region. The temperature difference between low temperature region and high temperature region is about $1.06{\sim}6.17^{\circ}C$. However, the temperature in the Outer Space of the Super-High-Rise Building was variously appeared by the influence such as the cramped of the created planting-site and waterscape facilities area, the increase of amount of solar radiation and the reduction of reflection amount of light due to building etc.. Thus, the composition all produced the area of the green quantity required for each space and water space in advance. It was determined that there were the minimum area displaying an effect and the necessity to it secures the green quantity.

• Korean Journal of Remote Sensing
• /
• v.37 no.6_3
• /
• pp.1975-1984
• /
• 2021

Urban expansion results in raising the temperature in the city, which can cause social, economic and physical damage. In order to prevent the urban heat island and reduce the urban land surface temperature, it is important to quantify the cooling effect of the features of the urban space. Therefore, in order to understand the relationship between each object of land cover and the land surface temperature in Seoul, the land cover map was classified into 6 classes. And the correlation and multiple regression analysis between land surface temperature and the area of objects, perimeter/area, and normalized difference vegetation index was analyzed. As a result of the analysis, the normalized difference vegetation index showed a high correlation with the land surface temperature. Also, in multiple regression analysis, the normalized difference vegetation index exerted a higher influence on the land surface temperature prediction than other coefficients. However, the explanatory power of the derived models as a result of multiple regression analysis was low. In the future, if continuous monitoring is performed using high-resolution MIR Image from KOMPSAT-3A, it will be possible to improve the explanatory power of the model. By utilizing the relationship between such various land cover types considering vegetation vitality of green areas with that of land surface temperature within urban spaces for urban planning, it is expected to contribute in reducing the land surface temperature in urban spaces.

• Journal of Wetlands Research
• /
• v.24 no.2
• /
• pp.93-101
• /
• 2022

This study was conducted to analyze the distribution characteristics of abandoned paddy wetlands according to topography and land cover. In Seosan-si, Dangjin-si, Boryeong-si, and Taean-gun, Chungcheongnam-do, abandoned rice wetlands were found through GIS and field surveys, and the distribution status according to slope, elevation and land cover type was analyzed. As a result of the study, a total of 106 abandoned paddy wetlands were identified, and the average elevation of each abandoned paddy wetlands was 38.85m (S.D.32.76), the average slope was 6.27˚ (S.D.5.39), and the total area was 24,200km². 90 sites (84.9%) of abandoned paddy wetlands were distributed on flat land with less than 5˚ slope, 63 sites (12,121.07km²), and 27 sites(9,524.15km²) at 5-10˚ (9,524.15km²) on flat land with less than 10˚. The area is 21,645.22km²(89.5%) of the total area of abandoned paddy wetlands. 48 sites(12,326km²) in the lowlands with an altitude of less than 25 m, 29 sites(4,909.4km²) below 50m. It accounts for 71.2% of the total area of abandoned paddy wetlands. Among environmental factors of abandoned paddy wetlands, there was no statistically significant correlation between slope and altitude. According to the land cover classification, it was widely distributed in artificial grasslands (38), paddy fields (33), and fields (22).

• Journal of the Korean Association of Geographic Information Studies
• /
• v.8 no.2
• /
• pp.21-30
• /
• 2005

The primary purpose of this paper is to prove a clear relationship among land use type, vegetation level and surface temperature. For this purpose, this paper presents series of spatial distribution maps of the three features obtained through the visual interpretation of digital images. The result of study tells us that the spatial distribution of the vegetation level is very similar with that of surface temperature. By analyzing the relationship between surface temperature and land use types, this study categorizes the eighteen urban land uses into 7-8 groups according to their average surface temperature. The Duncan test was conducted to categorize the land uses. The surface temperature of manufacturing related land use is the highest, semi-residential use is the next, non-residential land use is the next to the lowest, and the agricultural and forest land use is the lowest. This paper provides another strong evidence of the relationship by showing the regression result.

• Journal of Korean Society for Geospatial Information Science
• /
• v.8 no.2 s.16
• /
• pp.47-60
• /
• 2000

On this study, Among the classification methods of land cover using satellite imagery, we compared the classification accuracy of Neural Network Classifier and that of Maximum Likelihood Classifier which has the characteristics of parametric and non-parametric classification method. In the assessment of classification accuracy, we analyzed the classification accuracy about testing area as well as training area that many analysts use generally when assess the classification accuracy. As a result, Neural Network Classifier is superior to Maximum Likelihood Classifier as much as 3% in the classification of training data. When ground reference data is used, we could get poor result from both of classification methods, but we could reach conclusion that the classification result of Neural Network Classifier is superior to the classification result of Maximum Likelihood Classifier as much as 10%.

• Korean Journal of Remote Sensing
• /
• v.36 no.6_2
• /
• pp.1567-1577
• /
• 2020

Urban land cover classification is role in urban planning and management. So, it's important to improve classification accuracy on urban location. In this paper, machine learning model, Support Vector Machine (SVM) and Artificial Neural Network (ANN) are proposed for urban land cover classification based on high resolution satellite imagery (KOMPSAT-3A). Satellite image was trained based on 25 m rectangle grid to create training data, and training models used for classifying test area. During the validation process, we presented confusion matrix for each result with 250 Ground Truth Points (GTP). Of the four SVM kernels and the two activation functions ANN, the SVM Polynomial kernel model had the highest accuracy of 86%. In the process of comparing the SVM and ANN using GTP, the SVM model was more effective than the ANN model for KOMPSAT-3A classification. Among the four classes (building, road, vegetation, and bare-soil), building class showed the lowest classification accuracy due to the shadow caused by the high rise building.

• Ecology and Resilient Infrastructure
• /
• v.8 no.4
• /
• pp.179-193
• /
• 2021

This study examines the use of satellite images for river classification and determination of stream reach, which is the first priority in the river environment assessment system. In the river environment assessment system used in South Korea, it is proposed to set a stream reach by using 10 or 25 times the width of the river based on the result of river classification. First, river classification for the main stream section of Cheongmi stream was performed using various river-related data. The maximum likelihood method was applied for land cover classification. In this study, Sentinel-2 satellite imagery, which is an open data technology with a resolution of 10 m, was used. A total of four satellite images from 2018 was used to consider various flow conditions: February 2 (daily discharge = 2.39 m³/s), May 23 (daily discharge = 15.51 m³/s), June 2 (daily discharge = 3.88 m³/s), and July 7 (daily discharge = 33.61 m³/s). The river widths were estimated from the result of land cover classification to determine stream reach. The results of the assessment reach classification were evaluated using indicators of stream physical environments, including pool diversity, channel sinuosity, and river crossing shape and structure. It is concluded that appropriate flow conditions need to be considered when using satellite images to set up assessment segments for the river environment assessment system.

• Journal of Korea Water Resources Association
• /
• v.49 no.4
• /
• pp.305-314
• /
• 2016

In this study, a quantitative assessment was carried out in order to identify the agricultural drought in time and space using the Terra MODIS remote sensing data for the agricultural drought. The Normalized Difference Vegetation Index (NDVI) and Enhanced Vegetation Index (EVI) were selected by MOD13A3 image which shows the changes in vegetation conditions. The land cover classification was made to show only vegetation excluding water and urbanized areas in order to collect the land information efficiently by Type1 of MCD12Q1 images. NDVI and EVI index calculated using land cover classification indicates the strong seasonal tendency. Therefore, standardized Vegetation Stress Index Anomaly (VSIA) of EVI were used to estimated the medium-scale regions in Korea during the extreme drought year 2001. In addition, the agricultural drought damages were investigated in the country's past, and it was calculated based on the Standardized Precipitation Index (SPI) using the data of the ground stations. The VSIA were compared with SPI based on historical drought in Korea and application for drought assessment was made by temporal and spatial correlation analysis to diagnose the properties of agricultural droughts in Korea.