• Journal of the Korean association of regional geographers
• /
• v.7 no.1
• /
• pp.35-50
• /
• 2001

Since remotely sensed images of coastal wetlands are very sensitive to spatial resolution, it is very important to select an optimum resolution for particular geographic phenomena needed to be represented. Scale is one of the most important factors in spatial analysis techniques, which is defined as a spatial and temporal interval for a measurement or observation and is determined by the spatial extent of study area or the measurement unit. In order to acquire the optimum scale for a particular subject (i.e., coastal wetlands), measuring and representing the characteristics of attribute information extracted from the remotely sensed images are required. This study aims to explore and analyze the scale effects of attribute information extracted from remotely sensed coastal wetlands images. Specifically, it is focused on identifying the effects of scale in response to spatial resolution changes and suggesting a methodology for exploring the optimum spatial resolution. The LANDSAT TM image of Sunchon Bay was classified by a supervised classification method, Six land cover types were classified and the Kappa index for this classification was 84.6%. In order to explore the effects of scale in the classification procedure, a set of images that have different spatial resolutions were created by a aggregation method. Coarser images were created with the original image by averaging the DN values of neighboring pixels. Sixteen images whose resolution range from 30 m to 480 m were generated and classified to obtain land cover information using the same training set applied to the initial classification. The values of Kappa index show a distinctive pattern according to the spatial resolution change. Up to 120m, the values of Kappa index changed little, but Kappa index decreased dramatically at the 150m. However, at the resolution of 240 m and 270m, the classification accuracy was increased. From this observation, the optimum resolution for the study area would be either at 240m or 270m with respect to the classification accuracy and the best quality of attribute information can be obtained from these resolutions. Procedures and methodologies developed from this study would be applied to similar kinds and be used as a methodology of identifying and defining an optimum spatial resolution for a given problem.

• Proceedings of the KSRS Conference
• /
• 2001.03a
• /
• pp.42-48
• /
• 2001

오늘날 원격탐사와 GIS를 이용한 시·공간적 분석은 인간활동에서부터 자연환경에 이르기까지 다양한 정보를 추출하기 위한 기법으로 자주 사용되고 있다. 본 연구는 위성원격 탐사자료와 GIS를 활용하여 시기별 도시지역에서의 열 분포 특성을 추출하여 토지피복과의 상관관계를 시·공간적으로 해석하였다. 이를 위하여 세 시기간 도시 열 분포의 특성을 도시성장과 함께 해석함과 동시에 보다 명확 하게 규명하기 위하여 Landsat TM band 6의 DN value를 이용한 지표온도 추출에 있어서 NASA 모델을 활용하여 대구시 주변지역 8개 지점의 AWS 실측 값과 서로 상관 분석한 결 과 평균 0.85의 상관정도를 얻었다. 또한 토지피복분류를 통하여 도시성장에 따른 열 분포 및 식생지수의 변화를 시·공간적으로 해석하기 위하여 1,000지점에서 sample 자료를 추출 하여 지형특성별 열 분포의 패턴을 분석하였다. 이와 같은 결과는 향후 도시환경 특성을 고 려한 환경 친화적인 도시계획수립에 있어서 중요한 인자로 작용할 것으로 사료된다.

• Proceedings of the KGS Conference
• /
• 1999.05a
• /
• pp.47-52
• /
• 1999

• Korean Journal of Remote Sensing
• /
• v.31 no.3
• /
• pp.205-214
• /
• 2015

In this study, a novel land-cover classification framework for multi-temporal SAR data is presented that can combine multiple features extracted through data transforms and multiple classifiers. At first, data transforms using principle component analysis (PCA) and 3D wavelet transform are applied to multi-temporal SAR dataset for extracting new features which were different from original dataset. Then, three different classifiers including maximum likelihood classifier (MLC), neural network (NN) and support vector machine (SVM) are applied to three different dataset including data transform based features and original backscattering coefficients, and as a result, the diverse preliminary classification results are generated. These results are combined via a majority voting rule to generate a final classification result. From an experiment with a multi-temporal ENVISAT ASAR dataset, every preliminary classification result showed very different classification accuracy according to the used feature and classifier. The final classification result combining nine preliminary classification results showed the best classification accuracy because each preliminary classification result provided complementary information on land-covers. The improvement of classification accuracy in this study was mainly attributed to the diversity from combining not only different features based on data transforms, but also different classifiers. Therefore, the land-cover classification framework presented in this study would be effectively applied to the classification of multi-temporal SAR data and also be extended to multi-sensor remote sensing data fusion.

• Korean Journal of Remote Sensing
• /
• v.29 no.5
• /
• pp.569-579
• /
• 2013

In this study, land-cover classification was implemented using features extracted from multi-temporal SAR data through 3D wavelet transform and the applicability of the 3D wavelet transform as a feature extraction approach was evaluated. The feature extraction stage based on 3D wavelet transform was first carried out before the classification and the extracted features were used as input for land-cover classification. For a comparison purpose, original image data without the feature extraction stage and Principal Component Analysis (PCA) based features were also classified. Multi-temporal Radarsat-1 data acquired at Dangjin, Korea was used for this experiment and five land-cover classes including paddy fields, dry fields, forest, water, and built up areas were considered for classification. According to the discrimination capability analysis, the characteristics of dry field and forest were similar, so it was very difficult to distinguish these two classes. When using wavelet-based features, classification accuracy was generally improved except built-up class. Especially the improvement of accuracy for dry field and forest classes was achieved. This improvement may be attributed to the wavelet transform procedure decomposing multi-temporal data not only temporally but also spatially. This experiment result shows that 3D wavelet transform would be an effective tool for feature extraction from multi-temporal data although this procedure should be tested to other sensors or other areas through extensive experiments.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.19 no.2
• /
• pp.163-172
• /
• 2001

This study present the quadratic programming as the classification method of remote sensed data applying to the extraction of landcover and examine it's applicable capability by comparing the classification accuracy of quadratic programming with that of neural network and maximum likelihood method which are used in the extraction of thematic layer. As the results, as drawing the more improved classification results by 6% than maximum likelihood method, we could discern that the method of quadratic programming is appliable to classifying the remote sensed data. Also, in the classification of quadratic programming method, we could definitely indicate the results which was ignored in the previous extreme(binary) classification method by affecting the class decision with the class composition proportion.

• 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.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_2
• /
• pp.1663-1676
• /
• 2022

Recently, a number of deep-learning based land cover segmentation studies have been introduced. Some studies denoted that the performance of land cover segmentation deteriorated due to insufficient training data. In this study, we verified the improvement of land cover segmentation performance through data augmentation. U-Net was implemented for the segmentation model. And 2020 satellite-derived landcover dataset was utilized for the study data. The pixel accuracies were 0.905 and 0.923 for U-Net trained by original and augmented data respectively. And the mean F1 scores of those models were 0.720 and 0.775 respectively, indicating the better performance of data augmentation. In addition, F1 scores for building, road, paddy field, upland field, forest, and unclassified area class were 0.770, 0.568, 0.433, 0.455, 0.964, and 0.830 for the U-Net trained by original data. It is verified that data augmentation is effective in that the F1 scores of every class were improved to 0.838, 0.660, 0.791, 0.530, 0.969, and 0.860 respectively. Although, we applied data augmentation without considering class balances, we find that data augmentation can mitigate biased segmentation performance caused by data imbalance problems from the comparisons between the performances of two models. It is expected that this study would help to prove the importance and effectiveness of data augmentation in various image processing fields.

• Korean Journal of Remote Sensing
• /
• v.29 no.5
• /
• pp.537-544
• /
• 2013

Baton Peninsula, where Sejong station is located, mainly covered with snow and vegetation. Because this area is sensitive to climate change, monitoring of surface variation is important to understand climate change on the polar region. Due to the inaccessibility, the remote sensing is useful to continuously monitor the area. The objectives of this research are 1) map classification of land-cover types in the Barton Peninsular around King Sejong station and 2) grasp distribution of vegetation species in classified area. A KOMPSAT-2 multispectral satellite image was used to classify land-cover types and vegetation species. We performed classification with hierarchical procedure using KOMPSAT-2 satellite image and ground reference data, and the result is evaluated for accuracy as well. As the results, vegetation and non-vegetation were clearly classified although species shown lower accuracies within vegetation class.

• Journal of Korean Society for Geospatial Information Science
• /
• v.8 no.1 s.15
• /
• pp.121-130
• /
• 2000

For the land-cover classification we have usually used imagery of the optical sensors only. But currently a number of the satellite with various sensors are operating and the availability of using the data acquired from them are increasing. SAR sensors, in particular, can produce additional informations on the land-cover which has not been available from optical sensors. On this study, I have applied the SAR Image to the SPOT XS image in the classification procedures, and analysed the classified results. In this procedure I have extracted texture informations from SAR intensity images, then applied both intensity and texture informations. From the accuracy analysis, overall accuracy are increased slightly when the SAR texture was applied. In case of the Built-up class the results showed higher accuracy than those of when only the SPOT XS image was used. From this result I can show that overall accuracy was increased slightly but the spatial distribution of classes was visibly improved.