• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.30 no.6_2
• /
• pp.653-663
• /
• 2012

Using soft classification, it is possible to obtain the land cover proportions from the remotely sensed image. These land cover proportions are then used as input data for a procedure called "super-resolution mapping" to produce the predicted hard land cover layers at higher resolution than the original remotely sensed image. Superresolution mapping can be implemented using a number of algorithms in which the Hopfield Neural Network (HNN) has showed some advantages. The HNN has improved the land cover classification through superresolution mapping greatly with the high resolution data. However, the super-resolution mapping is based on the spatial dependence assumption, therefore it is predicted that the accuracy of resulted land cover classes depends on the relative size of spatial features and the spatial resolution of the remotely sensed image. This research is to evaluate the capability of HNN to implement the super-resolution mapping for SPOT image to create higher resolution land cover classes with different zoom factor.

• Korean Journal of Remote Sensing
• /
• v.22 no.3
• /
• pp.199-209
• /
• 2006

The Global Land Cover 2000 (GLC 200) project, as a most recent issue, is to provide for the year 2000 a harmonized land cover database over the whole globe. The classifications were performed according to continental or regional scales by corresponding organization using the data of VEGETATION sensor onboard the SPOT4 Satellite. Even if the global land cover classification for Asia provided by Chiba University showed a good accuracy in whole Asian area, some problems were detected in Korean region. Therefore, the construction of new land cover database over Korea is strongly required using more recent data set. The present study focuses on the development of a new upgraded land cover map at 1 km resolution over Korea considering the widely used K-means clustering, which is one of unsupervised classification technique using distance function for land surface pattern classification, and the principal components transformation. It is based on data sets from the Earth observing system SPOT4/VEGETATION. Newly classified land cover was compared with GLC 2000 for Korean peninsula to access how well classification performed using confusion matrix.

• Journal of Korean Society on Water Environment
• /
• v.24 no.6
• /
• pp.690-700
• /
• 2008

The Soil and Water Assessment Tool (SWAT) model users tend to use the readily available input dataset, such as the Ministry of Environment (MOE) land cover data ignoring temporal and spatial changes in land cover. The SWAT model was calibrated and validated with this land cover data. The EI values were 0.79 and 0.85 for streamflow calibration and validation, respectively. The EI were 0.79 and 0.86 for sediment calibration and validation, respectively. With newly prepared landcover dataset for the Doam-dam watershed, the SWAT model better predicts hydrologic and sediment behaviors. The number of HRUs with new land cover data increased by 70.2% compared with that with the MOE land cover, indicating better representation of small-sized agricultural field boundaries. The SWAT estimated annual average sediment yield with the MOE land cover data was 61.8 ton/ha/year for the Doam-dam watershed, while 36.2 ton/ha/year (70.7% difference) of annual sediment yield with new land cover data. Especially the most significant difference in estimated sediment yield was 548.0% for the subwatershed #2. Therefore it is recommended that one needs to carefully validate land cover for the study watershed for accurate hydrologic and sediment simulation with the SWAT model.

• Proceedings of the KSRS Conference
• /
• 1999.11a
• /
• pp.13-18
• /
• 1999

Land cover map of 30 arc-second grid by NOAA AVHRR data for the whole Asia was produced by the authors as the project of the Asian Association on Remote Sensing(AARS). Land cover change monitoring of continental scale by satellite data needs preprocessing to remove undesirable factors due to noises, atmosphere, or the effect by solar zenith angle. The paper describes the method to remove these factors. The most important thing for better mapping/monitoring in the future is the accumulation of ground truth data by many land cover related researchers. The project of the development of Global Land Cover Ground Truth Database(GLCGT-DB) is proposed.

• Korean Journal of Remote Sensing
• /
• v.37 no.5_1
• /
• pp.871-884
• /
• 2021

The purpose of this study was to determine ways to increase efficiency in constructing and verifying artificial intelligence learning data on land cover using aerial and satellite images, and in applying the data to AI learning algorithms. To this end, multi-resolution datasets of 0.51 m and 10 m each for 8 categories of land cover were constructed using high-resolution aerial images and satellite images obtained from Sentinel-2 satellites. Furthermore, fine data (a total of 17,000 pieces) and coarse data (a total of 33,000 pieces) were simultaneously constructed to achieve the following two goals: precise detection of land cover changes and the establishment of large-scale learning datasets. To secure the accuracy of the learning data, the verification was performed in three steps, which included data refining, annotation, and sampling. The learning data that wasfinally verified was applied to the semantic segmentation algorithms U-Net and DeeplabV3+, and the results were analyzed. Based on the analysis, the average accuracy for land cover based on aerial imagery was 77.8% for U-Net and 76.3% for Deeplab V3+, while for land cover based on satellite imagery it was 91.4% for U-Net and 85.8% for Deeplab V3+. The artificial intelligence learning datasets on land cover constructed using high-resolution aerial and satellite images in this study can be used as reference data to help classify land cover and identify relevant changes. Therefore, it is expected that this study's findings can be used in the future in various fields of artificial intelligence studying land cover in constructing an artificial intelligence learning dataset on land cover of the whole of Korea.

• Atmosphere
• /
• v.21 no.4
• /
• pp.405-414
• /
• 2011

In this study, the influence of a change in land use on the local weather fields is investigated around the Lake Shihwa area using synthetic land cover data and a high-resolution mesoscale model - the Weather Research and Forecasting (WRF). The default land cover data generally used in the WRF is based on the land use category of the United States Geological Survey (USGS), which erroneously presents most land areas of the Korean Peninsula as savannas. To revise such a fault, a multi-temporal land cover data, provided by the Ministry of Environment of Korea, was employed to generate a land cover map of 2005 subject to the land use in Korea at that time. A new land cover map of 1989, before the construction of the Lake Shihwa, was made based on the 2005 map and the Landsat 4-5 TM satellite images of two years. Over the areas where the land use had been changed (e.g., from sea to wetlands, towns, etc.) due to the Lake Shihwa development project, the skin temperature decreased by up to $8^{\circ}C$ in the winter case while increased by as much as $14^{\circ}C$ in the summer case. Changes in the water vapor mixing ratio were mostly affected by advection and topography in both seasons, with considerable increase in the summer case due to continuous sea breeze. Local decrease in water vapor occurred over high land use change areas and/or over downstream of such areas where alteration in wind fields were induced by changes in skin temperature and surface roughness at the areas of land use changes. The albedo increased by about 0.1% in the regions where sea was converted into wetland. In the regions where urban areas were developed, such as Songdo New Town and Incheon International Airport, the albedo increased by up to 0.16%.

• Atmosphere
• /
• v.21 no.4
• /
• pp.373-389
• /
• 2011

This research focuses on the sensitivity of the WRF(Weather Research and Forecasting) Model according to three different land cover data(USGS(United States Geological Survey), MODIS(Moderate Resolution Imaging Spectroradiometer)30s+USGS, and KLC (Korea Land Cover)) for an event of sea breeze, occurred over the Gangwon Yeongdong region on 13 May 2009. Based on the observation, the easterly into Gangneung, due to the sea-breeze circulation, was identified between 1000 LST and 1640 LST. It did not reach beyond the Taebaek Mountain Range and thus the easterly was not observed near Daegwallyeong. On the other hand, the numerical simulations utilizing land cover data of USGS, MODIS30s+USGS, and KLC showed easterlies beyond the Taebaek Mountain Range up to Daegwallyeong. In addition, rather different penetration distances of each easterly, and different timings of beginning and ending of sea breeze were identified among the simulations. The Bias, MAE(Mean Absolute Error) and RMSE(Root Mean Square Error) of the wind from WRF simulation using MODIS30s+USGS land cover data were the least among the simulations particularly over Gangwon Yeongdong coastal area(Sokcho, Gangneung and Donghae), while those of the wind over the Gangwon Mountain area(Daegwallyeong and Jinbu) from the simulation using KLC land cover data were the least among them. The wind field over Gangwon Yeongdong coastal area from the simulation using USGS land cover data was rather poor among them.

• Korean Journal of Remote Sensing
• /
• v.39 no.5_1
• /
• pp.507-519
• /
• 2023

Multi-temporal optical images have been utilized for time-series monitoring of croplands. However, the presence of clouds imposes limitations on image availability, often requiring a cloud removal procedure. This study assesses the applicability of various machine learning algorithms for effective cloud removal in optical imagery. We conducted comparative experiments by focusing on two key variables that significantly influence the predictive performance of machine learning algorithms: (1) land-cover types of training data and (2) temporal variability of land-cover types. Three machine learning algorithms, including Gaussian process regression (GPR), support vector machine (SVM), and random forest (RF), were employed for the experiments using simulated cloudy images in paddy fields of Gunsan. GPR and SVM exhibited superior prediction accuracy when the training data had the same land-cover types as the cloud region, and GPR showed the best stability with respect to sampling fluctuations. In addition, RF was the least affected by the land-cover types and temporal variations of training data. These results indicate that GPR is recommended when the land-cover type and spectral characteristics of the training data are the same as those of the cloud region. On the other hand, RF should be applied when it is difficult to obtain training data with the same land-cover types as the cloud region. Therefore, the land-cover types in cloud areas should be taken into account for extracting informative training data along with selecting the optimal machine learning algorithm.

• Proceedings of the KSRS Conference
• /
• 1999.11a
• /
• pp.439-442
• /
• 1999

It is important to analyze the seasonal change profiles of land cover type in large scale for establishing preservation strategy and environmental monitoring. Because the NOAA-AVHRR data sets provide global data with high temporal resolution, it is suitable for the land cover classification of the large area. The objectives of this study were to classify land cover of Korea, to investigate the phenological profiles of land cover. The NOAA-AVHRR data from Jan. 1998 to Dec. 1998 were received by Korea Ocean Research & Development Institute(KORDI) and were used for this study. The NDVI data were produced from this data. And monthly maximum value composite data were made for reducing cloud effect and temporal classification. And the data were classified using the method of supervised classification. To label the land cover classes, they were classified again using generalized vegetation map and Landsat-TM classified image. And the profiles of each class was analyzed according to each month. Results of this study can be summarized as follows. First, it was verified that the use of vegetation map and TM classified map was available to obtain the temporal class labeling with NOAA-AVHRR. Second, phenological characteristics of plant communities of Korea using NOAA-AVHRR was identified. Third, NDVI of North Korea is lower on Summer than that of South Korea. And finally, Forest cover is higher than another cover types. Broadleaf forest is highest on may. Outline of covertype profiles was investigated.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.7
• /
• pp.1858-1872
• /
• 2023

With the advance of radar technologies, the availability of synthetic aperture radar (SAR) images increases. To improve application of SAR images, a management system for SAR images is proposed in this paper. The system provides trainable land cover classification module and display of SAR images on the map. Users of the system can create their own classifier with their data, and obtain the classified results of newly captured SAR images by applying the classifier to the images. The classifier is based on convolutional neural network structure. Since there are differences among SAR images depending on capturing method and devices, a fixed classifier cannot cover all types of SAR land cover classification problems. Thus, it is adopted to create each user's classifier. In our experiments, it is shown that the module works well with two different SAR datasets. With this system, SAR data and land cover classification results are managed and easily displayed.