• Journal of KIISE:Software and Applications
• /
• v.29 no.1_2
• /
• pp.80-90
• /
• 2002

Clustering is a useful technique for grouping data points such that points within a single group/cluster have similar characteristics. Many clustering algorithms have been developed and used in engineering applications including pattern recognition and image processing etc. Recently, it has drawn increasing attention as one of important techniques in data mining. However, clustering algorithms such as K-means and Fuzzy C-means suffer from difficulties. Those are the needs to determine the number of clusters apriori and the clustering results depending on the initial set of clusters which fails to gain desirable results. In this paper, we propose a new clustering algorithm, which solves mentioned problems. In our method we use evolutionary algorithm to solve the local optima problem that clustering converges to an undesirable state starting with an inappropriate set of clusters. We also adopt a new measure that represents how well data are clustered. The measure is determined in terms of both intra-cluster dispersion and inter-cluster separability. Using the measure, in our method the number of clusters is automatically determined as the result of optimization process. And also, we combine heuristic that is problem-specific knowledge with a evolutionary algorithm to speed evolutionary algorithm search. We have experimented our algorithm with several sets of multi-dimensional data and it has been shown that one algorithm outperforms the existing algorithms.

• The KIPS Transactions:PartB
• /
• v.15B no.1
• /
• pp.25-36
• /
• 2008

In this paper, we present a robust hand recognition approach to sudden illumination changes. The proposed approach constructs a background model with respect to hue and hue gradient in HSI color space and extracts a foreground hand region from an input image using the background subtraction method. Eighteen features are defined for a hand pose and multi-class SVM(Support Vector Machine) approach is applied to learn and classify hand poses based on eighteen features. The proposed approach robustly extracts the contour of a hand with variations in illumination by applying the hue gradient into the background subtraction. A hand pose is defined by two Eigen values which are normalized by the size of OBB(Object-Oriented Bounding Box), and sixteen feature values which represent the number of hand contour points included in each subrange of OBB. We compared the RGB-based background subtraction, hue-based background subtraction and the proposed approach with sudden illumination changes and proved the robustness of the proposed approach. In the experiment, we built a hand pose training model from 2,700 sample hand images of six subjects which represent nine numerical numbers from one to nine. Our implementation result shows 92.6% of successful recognition rate for 1,620 hand images with various lighting condition using the training model.

• Korean Journal of Remote Sensing
• /
• v.35 no.6_2
• /
• pp.1107-1115
• /
• 2019

Satellite data play a major role in supporting knowledge about forest fire by delivering rapid information to map areas damaged. This study, we used 7 Sentinel-2A images to detect change area in forests of Sokcho on April 4, 2019. The process of classify forest fire severity used 7 levels from Sentinel-2A dNBR(differenced Normalized Burn Ratio). In the process of classifying forest fire damage areas, the study selected three areas with high regrowth of vegetation level and conducted a detailed spatial analysis of the areas concerned. The results of dNBR analysis, regrowth of coniferous forest was greater than broad-leaf forest, but NDVI showed the lowest level of vegetation. This is the error of dNBR classification of dNBR. The results of dNBR time series, an area of forest fire damage decreased to a large extent between April 20th and May 3rd. This is an example of the regrowth by developing rare-plants and recovering broad-leaf plants vegetation. The results showed that change area was detected through the change detection of danage area by forest category and the classification errors of the coniferous forest were reached through the comparison of NDVI and dNBR. Therefore, the need to improve the precision Korean forest fire damage rating table accompanied by field investigations was suggested during the image classification process through dNBR.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.10
• /
• pp.284-295
• /
• 2019

Celebrities form a dominant culture of one era and are cultivating influence to lead new fashion. Domestic women celebrities attending Seoul Fashion Week build their image as a fashion leader by exposing their own fashion aesthetics to the public. This promotes PR and sales of the brand of fashion designers in the collection. This study considers the fashion trends of women celebrities by analyzing the fashion style of women celebrities in Seoul Collection over the recent five years. The results of this study on contents analysis are as follows. First, straight silhouettes, achromatic colors, and an absence of a pattern or decoration comprised a high proportion. Celebrities preferred a modern and minimal style. Second, they preferred a feminine style that shows traditional feminine beauty through slim and princess silhouettes, one-piece dress or skirts, soft materials, and decorations with ruffles. In addition, as a matching jacket on one-piece dresses also appeared, it showed that celebrities attempted to change the chic feminine style with masculine beauty. Third, boxy silhouettes, multi-color mixed with colorful colors, graphic or lettering patterns, glittering materials and lots of details comprised a high proportion. This means celebrities pursue a maximal style that reveals their strong presence as fashion leaders.

• Korean Journal of Soil Science and Fertilizer
• /
• v.44 no.6
• /
• pp.1035-1041
• /
• 2011

Salinity of soil under the plastic film houses in Korea is known as a significant factor to lower the crop production and to hamper the sustainable agricultural land management. In this study we propose a field monitoring technique to examine the methods applied to minimize the adverse effect of salts in soil based on the relationship between soil electrical characteristics and soil properties. Field experiments for 4 different treatments (water only, fertilizer only, DTPA only, and DTPA and fertilizer together) were conducted on soils at the plastic film house built for cultivating a cucumber plant located at Chunan-si, Chungchungnam-do in Korea. The electrical resistivity was measured by both a dipole-dipole and wenner multi-electrodes array method. After the electrical resistivity measurement we also measured the soil water content, temperature, and electrical conductivity on surface soil. The resulted image of the interpreted resistivity by the inversion technique presented a unique spatial distribution depending on the treatment, implying the effect of the different chemical components. It was also highly suspected that resistivity response changed with the nutrients level, suggesting that our proposed technique could be the effective tool for the monitoring soil water as well as nutrient during the cropping period. Especially, subsoils under DTPA treatment at 40 to 60 cm depth typically presented lower soil water accumulation comparing to subsoils under non-DTPA treatment. It is considered that DTPA resulted in increase of a root water uptake. However, our demonstrated results were mainly based on qualitative comparison. Further experiments need to be conducted to monitor temporal changes of electrical resistivity using time lapse analysis, providing that a plant root activity difference based on changes of soil water and nutrients level in time.

• Korean Journal of Remote Sensing
• /
• v.37 no.6_3
• /
• pp.2027-2034
• /
• 2021

Using multi-temporal KOMPSAT-3/3A high-resolution satellite images, the Normalized Difference Vegetation Index (NDVI) for the area around the Fukushima daiichi nuclear power plant was determined, and the pattern of vegetation changes was analyzed. To calculate the NDVI, surface reflectance from the KOMPSAT-3/3A satellite image was used. Satellite images from four years were used, and the zones where the images overlap was designated as the area of interest (AOI) for the study, and by setting a profile passing through highly vegetated area as a data analysis method, the changes by year were examined. In addition, random points were extracted within the AOI and displayed as a box plot to quantitatively indicate change of NDVI distribution pattern. The main results of this study showed that the NDVI in 2014 was low within AOI in the vicinity of the nuclear power plant, but vegetated area continued to expand until 2021. These results were also confirmed in the change monitoring results shown in a profile or box plot. In disaster areas where access is restricted, such as the Fukushima nuclear power plant area, where it is difficult to collect field data, obtaining land cover classification products with high accuracy using satellite images is challenging, so it is appropriate to analyze them using primary outputs such as vegetation indices obtained from high-resolution satellite imagery. It is necessary to establish an international cooperation system for jointly utilizing satellite images. Meanwhile, to periodically monitor environmental changes in neighboring countries that may affect the Korean peninsula, it is necessary to establish utilization models and systems using high-resolution satellite images.

• Korean Journal of Remote Sensing
• /
• v.37 no.6_1
• /
• pp.1669-1683
• /
• 2021

AI intelligent agriculture and digital agriculture are important for the science of agriculture. Leaf chlorophyll contents(LCC) are one of the most important indicators to determine the growth status of vegetable crops. In this study, a support vector machine (SVM) regression model was produced using an unmanned aerial vehicle-based RGB camera and a multispectral (MSP) sensor for onions and garlic, and the LCC estimation applicability of the RGB camera was reviewed by comparing it with the MSP sensor. As a result of this study, the RGB-based LCC model showed lower results than the MSP-based LCC model with an average R² of 0.09, RMSE 18.66, and nRMSE 3.46%. However, the difference in accuracy between the two sensors was not large, and the accuracy did not drop significantly when compared with previous studies using various sensors and algorithms. In addition, the RGB-based LCC model reflects the field LCC trend well when compared with the actual measured value, but it tends to be underestimated at high chlorophyll concentrations. It was possible to confirm the applicability of the LCC estimation with RGB considering the economic feasibility and versatility of the RGB camera. The results obtained from this study are expected to be usefully utilized in digital agriculture as AI intelligent agriculture technology that applies artificial intelligence and big data convergence technology.

• The Journal of Bigdata
• /
• v.6 no.1
• /
• pp.37-49
• /
• 2021

The number of dementia patients in Korea is estimated to be over 800,000, and the severity of dementia is becoming a social problem. However, no treatment or drug has yet been developed to cure dementia worldwide. The number of dementia patients is expected to increase further due to the rapid aging of the population. Currently, early detection of dementia and delaying the course of dementia symptoms is the best alternative. This study presented a methodology for early diagnosis of dementia by measuring and analyzing amyloid plaques. This vital protein can most clearly and early diagnose dementia in the retina through AI-based image analysis. We performed binary classification and multi-classification learning based on CNN on retina data. We also developed a deep learning algorithm that can diagnose dementia early based on pre-processed retinal data. Accuracy and recall of the deep learning model were verified, and as a result of the verification, and derived results that satisfy both recall and accuracy. In the future, we plan to continue the study based on clinical data of actual dementia patients, and the results of this study are expected to solve the dementia problem.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.22 no.1
• /
• pp.62-77
• /
• 2019

Open Data Cube(ODC) has been emerging and developing as the open source platform in the Committee on Earth Observation Satellites(CEOS) for the Global Earth Observation System of Systems(GEOSS) deployed by the Group on Earth Observations (GEO), ODC can be applied to the deployment of scalable and large amounts of free and open satellite images in a cloud computing environment, and ODC-based country or regional application services have been provided for public users on the high performance. This study first summarizes the status of ODC, and then presents concepts and some considering points for linking this platform with Korea Multi-Purpose Satellite (KOMPSAT) images. For the reference, the main contents of ODC with the Google Earth Engine(GEE) were compared. Application procedures of KOMPSAT satellite image to implement ODC service were explained, and an intermediate process related to data ingestion using actual data was demonstrated. As well, it suggested some practical schemes to utilize KOMPSAT satellite images for the ODC application service from the perspective of open data licensing. Policy and technical products for KOMPSAT images to ODC are expected to provide important references for GEOSS in GEO to apply new satellite images of other countries and organizations in the future.

• Korean Journal of Heritage: History & Science
• /
• v.46 no.3
• /
• pp.78-93
• /
• 2013

The purpose of this study is to analyze college students' perception of national natural heritages, including natural monumentsand scenic sites, in comparison with the similar concepts of national parks and eco-landscape preservation areas, to suggest managerial implications to enhance competitiveness. The first objective of this study is to measure their attitude toward traveling each type of natural heritage and to rate the relative importance of the four types of heritages in terms of preservation. Natural monuments were perceived as the most strongly preserved type, while scenic sites were perceived as the least strongly preserved type. The second objective is to conduct a correspondence analysis to match the concepts of four types of natural heritages with 10 tourist attractions representing the types. It was found that college students have difficulty appropriately identifying the types of 10 tourist attractions. In addition, multi-dimensional scaling (MDS) was implemented to identify the similarities and differences of the four types of natural heritages and to produce a positioning map matching the four types of natural heritages with the six attributes representing each type as a tourist attraction. Natural monuments were perceived very differently from scenic sites and not very differently from eco-landscape preservation areas. There were a few associations between the four types of natural heritages and the six tourism attributes, implying that little effort is made to promote and position each type of natural heritage in terms of tourism. In conclusion, a public awareness program is required to enhance the brand image of natural monuments and scenic sites in comparison to national parks and eco-landscape preservation areas. In addition, local residents who live near the natural heritages should be bolstered to play a managerial role as supporters and contents providers for sustainability.