• Korean Journal of Remote Sensing
• /
• v.25 no.3
• /
• pp.233-242
• /
• 2009

This paper presents two different types of supervised classifiers such as support vector machine (SVM) and spectral angle mapper (SAM). The Compact Airborne Spectrographic Imager (CASI) high resolution aerial image was classified with the above two classifier. The image was classified into eight land use /land cover classes. Accuracy assessment and Kappa statistics were estimated for SVM and SAM separately. The overall classification accuracy and Kappa statistics value of the SAM were 69.0% and 0.62 respectively, which were higher than those of SVM (62.5%, 0.54).

• Korean Journal of Remote Sensing
• /
• v.28 no.5
• /
• pp.489-499
• /
• 2012

In this paper, a classifier ensemble framework for remote sensing data classification is presented that combines classification results generated from both different training sets and different classifiers. A core part of the presented framework is to increase a diversity between classification results by using both different training sets and classifiers to improve classification accuracy. First, different training sets that have different sampling densities are generated and used as inputs for supervised classification using different classifiers that show different discrimination capabilities. Then several preliminary classification results are combined via a majority voting scheme to generate a final classification result. A case study of land-cover classification using multi-temporal ENVISAT ASAR data sets is carried out to illustrate the potential of the presented classification framework. In the case study, nine classification results were combined that were generated by using three different training sets and three different classifiers including maximum likelihood classifier, multi-layer perceptron classifier, and support vector machine. The case study results showed that complementary information on the discrimination of land-cover classes of interest would be extracted within the proposed framework and the best classification accuracy was obtained. When comparing different combinations, to combine any classification results where the diversity of the classifiers is not great didn't show an improvement of classification accuracy. Thus, it is recommended to ensure the greater diversity between classifiers in the design of multiple classifier systems.

• Journal of Soil and Groundwater Environment
• /
• v.20 no.7
• /
• pp.53-60
• /
• 2015

This study suggested the new site classification system according to land use, type of contamination and contaminants. Because the present site classification system can not cover all the areas, we changed the concept of land use to more detail one and enlarged the concept of other areas to cover all the areas not defined as certain land use. In case of the present industrial area, it was merged as other areas to avoid the confusion with oil and toxic material storage tank farm area. Accident area was separated from other areas and defined as only accident area caused by the mobile storage facility. In addition to classify the sites according to the basic land use, we classify the sites again in lower level according to the type of contamination and contaminants. With this classification system, we proposed different soil sampling strategy with the consideration of the origin of contamination and the interactions between soil and contaminants. We removed the surface soil sample (0~15 cm depth) around above storage tank because it was not a effective sample to assess whether that area contaminated or not. We also proposed to take the deeper soil samples at minimum three sampling points to confirm the depth of contamination in exploratory soil survey. We also proposed to remove the one point of 15 m depth sampling because it is not effective to confirm contaminated soil depth and needs the exhausted labor and cost. Instead of doing this, we added the continuous sampling to uncontaminated subsoil. Soil sampling points and depth in detailed soil survey is determined based on the results of exploratory soil survey. Therefore, effectiveness and reinforcements of exploratory soil survey would play an important role in improving the reliability of detailed soil survey.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.38 no.1
• /
• pp.23-33
• /
• 2020

Recently, high-resolution images can be easily acquired using UAV (Unmanned Aerial Vehicle), so that it is possible to produce small area observation and spatial information at low cost. In particular, research on the generation of cover maps in crop production areas is being actively conducted for monitoring the agricultural environment. As a result of comparing classification performance by applying RF(Random Forest), SVM(Support Vector Machine) and CNN(Convolutional Neural Network), deep learning classification method has many advantages in image classification. In particular, land cover classification using satellite images has the advantage of accuracy and time of classification using satellite image data set and pre-trained parameters. However, UAV images have different characteristics such as satellite images and spatial resolution, which makes it difficult to apply them. In order to solve this problem, we conducted a study on the application of deep learning algorithms that can be used for analyzing agricultural lands where UAV data sets and small-scale composite cover exist in Korea. In this study, we applied DeepLab V3 +, FC-DenseNet (Fully Convolutional DenseNets) and FRRN-B (Full-Resolution Residual Networks), the semantic image classification of the state-of-art algorithm, to UAV data set. As a result, DeepLab V3 + and FC-DenseNet have an overall accuracy of 97% and a Kappa coefficient of 0.92, which is higher than the conventional classification. The applicability of the cover classification using UAV images of small areas is shown.

• Journal of Forest and Environmental Science
• /
• v.28 no.3
• /
• pp.152-157
• /
• 2012

Sarawak is the largest state in Malaysia that covers 37.5% of the total land area. Multitemporal satellite images of Landsat and SPOT were used to examine deforestation and forest fragmentation in Sarawak between 1990 and 2009. Supervised classification with maximum likelihood classifier was used to classify the land cover types in Sarawak. The overall accuracies of all classifications were more than 80%. Our results showed that forests were reduced at 0.62% annually during the two decades. The peat swamp forest suffered a tremendous loss of almost 50% between 1990 and 2009 especially at coastal divisions due to intensified oil palm plantation development. Fragmentation analysis revealed the loss of about 65% of the core area of intact forest during the change period. The core area of peat swamp forest had almost completely disappeared during the two decades.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.712-714
• /
• 2003

The modifications taking place within the alluvial plains impart a larger extent of disturbances to hydrologic systems. The objective of the present investigation is to detect the sensitivity of multi-temporal image data from Landsat TM (Thematic Mapper) for finding out the land-cover/land-use changes associated with alluvial low land. The eastern coast of Chiba Prefecture, Japan, forms a very important geographic unit owing to the existence of a unique alluvial landform. The alluvial plain occupied in the study area is widely known as 'Kujukuri Plain'. The TM images have been classified by means of maximum likelihood supervised classifier and the extent of changes has been estimated.

• Proceedings of the KIEE Conference
• /
• 1992.07a
• /
• pp.449-452
• /
• 1992

Before performing an adaptive stereo matching using satellite images, classification is needed as a preprocessing step. This paper describes that classification of three land cover types : river, mountain, and agricultural fields. We proposed that classification algorithm using pyramid structure and texture features. Results of applying the proposed algorithm to satellite image improved classification accuracy.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.9 no.4
• /
• pp.113-124
• /
• 1989

In this study, we tried to extract the land-use change information of Seoul city using the multiple date images of the same geographic area. Multiple date image set is MSS('72, '79, '81, '93) and TM('85), and we carried out geometric correction, digitizing(due to the administrative boundary) in pre-processing process. In addition, we performed land-use classification with MLC(Maximum Likelihood Classifier) after improving the predictive accuracy of classification by filtering technique. At the stage of classification, ground truth data, topographic maps, aerial photographs were used to select the training field and statistical data of that time were compared with the classification result to prove the accuracy. As a result, urban area in Seoul has been increased('72 : 25.8 %${\rightarrow}$'81 : 43.0 %${\rightarrow}$'85 : 51.9 %) and Forest area decreased ('72 : 39.0 %${\rightarrow}$'85 : 28.4 %) as we estimated. Finally, it is concluded that the utilzation of satellite imagery is very effective, economical and helpful in the urban land-use/land-cover monitoring.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.8 no.4
• /
• pp.43-51
• /
• 2005

Due to rapid and reckless economic development, natural resources of the national land have been damaged and polluted. Accordingly, the necessity for environment-friendly development has been on the rise and many have made efforts to assess the environmental value of the national land. This study divides the conservation areas by means of using relative elevation, slope, and development of housing land based on environmental evaluation of the national land. The relative elevation is obtained by the difference of altitude at the edge of the forest patch extracted from the land cover classification map based on the ridgeline, and the slope is obtained by environment-oriented land suitability analysis. The development of housing land is classified in accordance with the progress of each project. Twenty-six evaluation criteria are divided into five different grades using the minimal indicator approach and then sub-divided into ten grades by means of using the following two scenarios. The first one uses the weight of input materials while the second one relies on the size of patches that are emphasized in landscape ecology. Consequently, such a study demonstrated the following results. The method relying on the weight of input materials revealed the limitation of emphasizing input materials excessively, whereas the method of considering the size of patches resulted in the division of conservation areas that embody regional characteristics. This study is meaningful in that it classifies the conservation areas by reflecting the regional characteristics and the ecological values of animals and plants.

• Proceedings of the KSRS Conference
• /
• 2008.10a
• /
• pp.169-172
• /
• 2008

Land-use or land-cover classification of satellite images is one of the important tasks in remote sensing application and many researchers have been tried to enhance classification accuracy. Previous studies show that the classification technique based on wavelet transform is more effective than that of traditional techniques based on original pixel values, especially in complicated imagery. Various wavelets can be used in wavelet transform. Wavelets are used as basis functions in representing other functions, like sinusoidal function in Fourier analysis. In these days, some basis functions such as Haar, Daubechies, Coiflets and Symlets are mainly used in 2D image processing. Selecting adequate wavelet is very important because different results could be obtained according to the type of basis function in classification. However, it is not easy to choose the basis function which is effective to improve classification accuracy. In this study, we computed the wavelet coefficients of satellite image using 10 different basis functions, and then classified test image. After evaluating classification results, we tried to ascertain which basis function is the most effective for image classification. We also tried to see if the optimum basis function is decided by energy parameter before classifying the image using all basis function. The energy parameter of signal is the sum of the squares of wavelet coefficients. The energy parameter is calculated by sub-bands after the wavelet decomposition and the energy parameter of each sub-band can be a favorable feature of texture. The decision of optimum basis function using energy parameter in the wavelet based image classification is expected to be helpful for saving time and improving classification accuracy effectively.