• Korean Journal of Remote Sensing
• /
• v.27 no.3
• /
• pp.359-367
• /
• 2011

A suitable pan-sharpening method has to be chosen with respect to the used spectral characteristic of the multispectral bands and the intended application. The research on pan-sharpening algorithm in improving the accuracy of image classification has been reported. For a classification, preserving the spectral information is important. Other applications such as road detection depend on a sharp and detailed display of the scene. Various criteria applied to scenes with different characteristics should be used to compare the pan-sharpening methods. The pan-sharpening methods in our research comprise rather common techniques like Brovey, IHS(Intensity Hue Saturation) transform, and PCA(Principal Component Analysis), and more complex approaches, including wavelet transformation. The extraction of matching pairs was performed through SIFT descriptor and Canny edge detector. The experiments showed that pan-sharpening techniques for spatial enhancement were effective for extracting point and linear features. As a result of the validation it clearly emphasized that a suitable pan-sharpening method has to be chosen with respect to the used spectral characteristic of the multispectral bands and the intended application. In future it is necessary to design hybrid pan-sharpening for the updating of features and land-use class of a map.

• Korean Journal of Remote Sensing
• /
• v.21 no.5
• /
• pp.383-392
• /
• 2005

Vegetation classes, especially grass and tree classes, are often confused in classification when conventional spectral pattern recognition techniques are used to classify urban areas. This paper reports on a study to improve the classification results by using an automated process of considering height information in separating urban vegetation classes, specifically tree and grass, using three-band, high-spatial resolution, digital aerial imagery. Height information was derived photogrammetrically from stereo pair imagery using cross correlation image matching to estimate differential parallax for vegetation pixels. A threshold value of differential parallax was used to assess whether the original class was correct. The average increase in overall accuracy for three test stereo pairs was $7.8\%$, and detailed examination showed that pixels reclassified as grass improved the overall accuracy more than pixels reclassified as tree. Visual examination and statistical accuracy assessment of four test areas showed improvement in vegetation classification with the increase in accuracy ranging from $3.7\%\;to\;18.1\%$. Vegetation classification can, in fact, be improved by adding height information to the classification procedure.

• Korean Journal of Remote Sensing
• /
• v.24 no.1
• /
• pp.57-64
• /
• 2008

In the previous study, the Haar wavelet was used as the sole basis function for the 3D discrete wavelet transform because the number of bands is too small to decompose a remotely sensed image in band direction with other basis functions. However, it is possible to use other basis functions for wavelet decomposition in horizontal and vertical directions because wavelet decomposition is independently performed in each direction. This study aims to classify a high spatial resolution image with the six types of basis function including the Haar function and to compare those results. The other wavelets are more helpful to classify high resolution imagery than the Haar wavelet. In overall accuracy, the Coif4 wavelet has the best result. The improvement of classification accuracy is different depending on the type of class and the type of wavelet. Using the basis functions with long length could be effective for improving accuracy in classification, especially for the classes of small area. This study is expected to be used as fundamental information for selecting optimal basis function according to the data properties in the 3D DWT based image classification.

• Korean Journal of Remote Sensing
• /
• v.37 no.4
• /
• pp.763-776
• /
• 2021

This paper introduces a method of selecting priority update areas for subdivided land cover maps by training orthoimages and serial cadastral maps in a deep learning model. For the experiment, orthoimages and serial cadastral maps were obtained from the National Spatial Data Infrastructure Portal. Based on the VGG-16 model, 51,470 images were trained on 33 subdivided classifications within the experimental area and an accuracy evaluation was conducted. The overall accuracy was 61.42%. In addition, using the differences in the classification prediction probability of the misclassified polygon and the cosine similarity that numerically expresses the similarity of the land category features with the original subdivided land cover class, the cases were classified and the areas in which the boundary setting was incorrect and in which the image itself was determined to have a problem were identified as the priority update polygons that should be checked by operators.

• Journal of Engineering Education Research
• /
• v.25 no.3
• /
• pp.26-34
• /
• 2022

Students' questions are essential and important for learning, but previous research and experience of instructors shows that there is a lack of interaction between instructors and students in online classes. This research studies how learners can effectively ask questions in online classes at engineering colleges. Based on two axes of anonymity and openness, the four different types of questioning methods were suggested as 'onymous/public', 'onymous/private', 'anonymous/public' and 'anonymous/private.' In this study, seven communication channels were applied to check their effectiveness in an online class. The results showed that learners' satisfaction with learning outcomes increased compared to previous offline classes, while satisfaction with teaching methods was similar. Additionally, among the four types of questioning methods, the preference and effectiveness of 'anonymous/public' was highest, followed by 'onymous/private'. This study suggests several implications of educational approaches to online education in engineering colleges.

• Journal of the Society of Disaster Information
• /
• v.19 no.4
• /
• pp.924-935
• /
• 2023

Purpose: The central aim of this study is to leverage machine learning techniques for the classification of Intrusion Detection System (IDS) data, with a specific focus on identifying the variables responsible for enhancing overall performance. Method: First, we classified 'R2L(Remote to Local)' and 'U2R (User to Root)' attacks in the NSL-KDD dataset, which are difficult to detect due to class imbalance, using seven machine learning models, including Logistic Regression (LR) and K-Nearest Neighbor (KNN). Next, we use the SHapley Additive exPlanation (SHAP) for two classification models that showed high performance, Random Forest (RF) and Light Gradient-Boosting Machine (LGBM), to check the importance of variables that affect classification for each model. Result: In the case of RF, the 'service' variable and in the case of LGBM, the 'dst_host_srv_count' variable were confirmed to be the most important variables. These pivotal variables serve as key factors capable of enhancing performance in the context of classification for each respective model. Conclusion: In conclusion, this paper successfully identifies the optimal models, RF and LGBM, for classifying 'R2L' and 'U2R' attacks, while elucidating the crucial variables associated with each selected model.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_4
• /
• pp.1925-1934
• /
• 2022

Monitoring nuclear activity for inaccessible areas using remote sensing technology is essential for nuclear non-proliferation. In recent years, deep learning has been actively used to detect nuclear-activity-related small objects. However, high-resolution satellite imagery containing small objects can result in class imbalance. As a result, there is a performance degradation problem in detecting small objects. Therefore, this study aims to improve detection accuracy by analyzing the effect of the ratio of small objects related to nuclear activity in the input data for the performance of the deep learning model. To this end, six case datasets with different ratios of small object pixels were generated and a U-Net model was trained for each case. Following that, each trained model was evaluated quantitatively and qualitatively using a test dataset containing various types of small object classes. The results of this study confirm that when the ratio of object pixels in the input image is adjusted, small objects related to nuclear activity can be detected efficiently. This study suggests that the performance of deep learning can be improved by adjusting the object pixel ratio of input data in the training dataset.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.15 no.6
• /
• pp.1209-1216
• /
• 2020

In this study, we propose a method for detecting coastal surface wastes using an UAV(Unmanned Aerial Vehicle) remote sensing method and an object detection algorithm based on deep learning. An object detection algorithm based on deep neural networks was proposed to detect coastal debris in aerial images. A deep neural network model was trained with image datasets of three classes: PET, Styrofoam, and plastics. And the detection accuracy of each class was compared with Darknet-53. Through this, it was possible to monitor the wastes landing on the shore by type through unmanned aerial vehicles. In the future, if the method proposed in this study is applied, a complete enumeration of the whole beach will be possible. It is believed that it can contribute to increase the efficiency of the marine environment monitoring field.

• Korean Journal of Remote Sensing
• /
• v.25 no.1
• /
• pp.71-83
• /
• 2009

This paper is written to analyze possible problems encountered with the existing data for the monitoring of land use and land cover change over the Korean peninsula and, further, to provide technical alternatives for the future land monitoring over the area. The oldest type of non-spatial data related to the land use change are cadastral statistics obtained since 1911. Annual statistics of cadastral data in early years (before 1942) can be used to assess land use change over the area. However, the cadastral statistics after the Korean War are not very appropriate for land use monitoring since the land class in cadastral data does not always correspond with actual land cover status. Majority of spatial data available for land monitoring over the area are land cover maps classified from satellite imagery since early 1970's. To analyze the suitability of land cover maps that were produced by two separate institutes with about 10 years interval, we conducted simple change detection analysis using these maps. These maps were not quite ready to be compared each other, in which they did not have the same class definition, classification method, and geometric registration. To achieve continuous and effective monitoring of land use and land cover change, particularly over North Korea, we should have a standard scheme in type and season of satellite imagery, image classification procedure, and class definition, which also should correspond to international standards.

• 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.