• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.5
• /
• pp.581-587
• /
• 2005

This paper presents a hybrid face recognition method of both the first moment of image and the independent component analysis(ICA) of fixed point(FP) algorithm based on Newton method. First moment is a method for finding centroid of image, which is applied to exclude the needless backgrounds in the face recognitions by shifting to the centroid of face image. FP-ICA is also applied to find a set of independent basis images for the faces, which is a set of statistically independent facial features. The proposed method has been applied to the problem for recognizing the 48 face images(12 persons o 4 scenes) of 64*64 pixels. The 3 distances such as city-block, Euclidean, negative angle are used as measures when match the probe images to the nearest gallery images. The experimental results show that the proposed method has a superior recognition performances(speed, rate) than conventional FP-ICA without preprocessing. The city-block has been relatively achieved more an accurate similarity than Euclidean or negative angle.

• Journal of radiological science and technology
• /
• v.46 no.2
• /
• pp.131-139
• /
• 2023

High-dose I-131 used for the treatment of thyroid cancer causes localized exposure among radiology technologists handling it. There is a delay between the calibration date and when the dose of I-131 is administered to a patient. Therefore, it is necessary to directly measure the radioactivity of the administered dose using a dose calibrator. In this study, we attempted to apply machine learning modeling to measured external dose rates from shielded I-131 in order to predict their radioactivity. External dose rates were measured at 1 m, 0.3 m, and 0.1 m distances from a shielded container with the I-131, with a total of 868 sets of measurements taken. For the modeling process, we utilized the hold-out method to partition the data with a 7:3 ratio (609 for the training set:259 for the test set). For the machine learning algorithms, we chose linear regression, decision tree, random forest and XGBoost. To evaluate the models, we calculated root mean square error (RMSE), mean square error (MSE), and mean absolute error (MAE) to evaluate accuracy and R² to evaluate explanatory power. Evaluation results are as follows. Linear regression (RMSE 268.15, MSE 71901.87, MAE 231.68, R² 0.92), decision tree (RMSE 108.89, MSE 11856.92, MAE 19.24, R² 0.99), random forest (RMSE 8.89, MSE 79.10, MAE 6.55, R² 0.99), XGBoost (RMSE 10.21, MSE 104.22, MAE 7.68, R² 0.99). The random forest model achieved the highest predictive ability. Improving the model's performance in the future is expected to contribute to lowering exposure among radiology technologists.

• Korean Journal of Remote Sensing
• /
• v.17 no.3
• /
• pp.219-230
• /
• 2001

The daytime and nighttime sea surface temperature (SST) differences and their seasonal variabilities in the East Sea were studied using Argos drifters data during 1996~1999. The SST differences for 1,438 data set were derived from 30 Argos drifters related to the NOAA satellite-based location and data collection system. The horizontal variation of SST differences in summer in the East Sea were higher than those in winter. The relationship between the SST differences and the half day moving distances of Argos drifters was studied. Monthly SST difference in the northern and southern part of 38$^{\circ}$N in the East Sea was considered. The SST differences derived from NOAA-14 satellite were compared with those from Argos drifter between daytime and nighttime in the turbulent eddy off Wonsan coast of Korea.

• Land and Housing Review
• /
• v.11 no.1
• /
• pp.109-116
• /
• 2020

This study aims to establish an approach to assess urban flood vulnerability by identifying social characteristics such as the road transportation and the vulnerable groups. Assessment procedures comprise three steps as: (1) composing the assessment criteria to reflect the urban characteristics; (2) calculating the weight; and (3) evaluating the vulnerability. The criteria were adopted by Delphi survey technique. Four criteria as land cover, residents, vulnerable areas, and disaster response were adopted in the current study. To determine the weight set of criteria, subjective and objective methods were combined. The weight set was determined using the combined method which reflects the Delphi method and Entropy analysis. In the process of data-based construction, GIS tools wwere used to extract administrative unit materials such as land cover, road status, and slope. Data on population and other social criteria were collected through the National Statistical Office and the Seoul Metropolitan statistical data. TOPSIS (Technique for Order of Preference by Similarity to Ideal Solution) technique, which uses materials from cell units in order to rank the closest distance to the best case and the farthest distance from the worst case by calculating the distances to the area of assessment, was applied to assess. The study area was the Dorimcheon basin, a flood special treatment area of Seoul city. The results from the current study indicates that the established urban flood vulnerability assessment approach is able to predict the inherent vulnerable factors in urban regions and to propose the area of priority control.

• Journal of Korean Society for Geospatial Information Science
• /
• v.16 no.3
• /
• pp.43-49
• /
• 2008

The Real Time Location System (RTLS) is very important in the ubiquitous society for real time tracking of men, high price assets, and logistics products. In this work, we developed an active RF system for RTLS and tested its performance. The RTLS system developed in this work was constructed of three active readers and one active tag. The small size tag developed in this work operated with a coin type battery. To make the tag smaller, we used an internal PCB antenna and a chip antenna. We tested the performance of the tag. To reduce the manufacturing cost of our RF system, we used low price RF transceiver CC2510 chip-set. The CC2510 chip-set provided RSSI(Received Signal Strength Indicator) signal which could be used to determine the distances between an active tag and three active readers.

• Journal of Korea Multimedia Society
• /
• v.9 no.2
• /
• pp.172-188
• /
• 2006

Sensor localization is one of the fundamental problems in wireless sensor networks. Previous localization algorithms can be classified into two categories, the GGB (Global Geometry-Based) approaches and the LGB (Local Geometry-Based). In the GGB approaches, there are a fixed set of reference nodes of which the coordinates are pre-determined. Other nodes determine their positions based on the distances from the fixed reference nodes. In the LGB approaches, meanwhile, the reference node set is not fixed, but grows up dynamically. Most GGB algorithms assume that the nodes are deployed in a convex shape area. They fail if either nodes are in a concave shape area or there are obstacles that block the communications between nodes. Meanwhile, the LGB approach is vulnerable to the errors in the distance estimations. In this paper, we propose new localization algorithms to cope with those two limits. The key technique employed in our algorithms is to determine, in a fully distributed fashion, if a node is in the line-of-sight from another. Based on the technique, we present two localization algorithms, one for anchor-based, another for anchor-free localization, and compare them with the previous algorithms.

• Journal of The Korean Astronomical Society
• /
• v.37 no.5
• /
• pp.605-609
• /
• 2004

We present a set of high-resolution 3D MHD simulations exploring the evolution of light, supersonic jets in cluster environments. We model sets of high- and low-Mach jets entering both uniform surroundings and King-type atmospheres and propagating distances more than 100 times the initial jet radius. Through complimentary analyses of synthetic observations and energy flow, we explore the detailed interactions between these jets and their environments. We find that jet cocoon morphology is strongly influenced by the structure of the ambient medium. Jets moving into uniform atmospheres have more pronounced backflow than their non-uniform counterparts, and this difference is clearly reflected by morphological differences in the synthetic observations. Additionally, synthetic observations illustrate differences in the appearances of terminal hotspots and the x-ray and radio correlations between the high- and low-Mach runs. Exploration of energy flow in these systems illustrates the general conversion of kinetic to thermal and magnetic energy in all of our simulations. Specifically, we examine conversion of energy type and the spatial transport of energy to the ambient medium. Determination of the evolution of the energy distribution in these objects will enhance our understanding of the role of AGN feedback in cluster environments.

• Korean Journal of Agricultural Science
• /
• v.23 no.1
• /
• pp.99-107
• /
• 1996

Eighteen nuclear probes were used to examine RFLP(restriction fragment length polymorphism) between six species of Artemisia spp. of Korea. Total DNA from six different species of Artemisia was separately cut with three restrict enzymes. The PstI enzyme was showed to reduce the variation of polymorphisms than the other two enzymes(EcoRl and BamHI). The genetic variation of polymorphism was similar between the Dhewegiki-ssug and Cham-ssug. RAPD analysis was applied to the same six species of Artemisia spp. in order to assess the degree of DNA polymorphism within the Artemisia genus. Six species of Artemisia were evaluated for variation using a set of 11 random 10-mer primers. Nine out of the eleven primers revealed scorable polymorphisms between six species of Artemisia spp. Genetic distances between each of the species were calculated and cluster analysis was used to generate a dendrogram showing phylogenetic relationships between them This result indicates that molecular markers will be more usable in intraspecific study of Artemisia spp. than isoenzyme markers.

• Asia pacific journal of information systems
• /
• v.29 no.4
• /
• pp.838-855
• /
• 2019

Online product reviews are a vital source for companies in that they contain consumers' opinions of products. The earlier methods of opinion mining, which involve drawing semantic information from text, have been mostly applied in one dimension. This is not sufficient in itself to elicit reviewers' comprehensive views on products. In this paper, we propose a novel approach in opinion mining by projecting online consumers' reviews in a multidimensional framework to improve review interpretation of products. First of all, we set up a new framework consisting of six dimensions based on a marketing management theory. To calculate the distances of review sentences and each dimension, we embed words in reviews utilizing Google's pre-trained word2vector model. We classified each sentence of the reviews into the respective dimensions of our new framework. After the classification, we measured the sentiment degrees for each sentence. The results were plotted using a radar graph in which the axes are the dimensions of the framework. We tested the strategy on Amazon product reviews of the iPhone and Galaxy smartphone series with a total of around 21,000 sentences. The results showed that the radar graphs visually reflected several issues associated with the products. The proposed method is not for specific product categories. It can be generally applied for opinion mining on reviews of any product category.

• Journal of IKEEE
• /
• v.27 no.4
• /
• pp.430-438
• /
• 2023

The Modular Multilevel Converter (MMC) has emerged as a key component in HVDC systems due to its ability to efficiently transmit large amounts of power over long distances. In such systems, accurate estimation of the MMC capacitor voltage is of utmost importance for ensuring optimal system performance, stability, and reliability. Traditional methods for voltage estimation may face limitations in accuracy and robustness, prompting the need for innovative approaches. In this paper, we propose a novel distributed neural network observer specifically designed for MMC capacitor voltage estimation. Our observer harnesses the power of a multi-layer neural network architecture, which enables the observer to learn and adapt to the complex dynamics of the MMC system. By utilizing a distributed approach, we deploy multiple observers, each with its own set of neural network layers, to collectively estimate the capacitor voltage. This distributed configuration enhances the accuracy and robustness of the voltage estimation process. A crucial aspect of our observer's performance lies in the meticulous initialization of random weights within the neural network. This initialization process ensures that the observer starts with a solid foundation for efficient learning and accurate voltage estimation. The observer iteratively updates its weights based on the observed voltage and current values, continuously improving its estimation accuracy over time. The validity of proposed algorithm is verified by the result of estimated voltage at each observer in capacitor of MMC.