포렌식에서 활용 가능한 삼성 스마트폰 백업 파일 분석 기법

이규원,황현욱,김기범,장태주,Lee, Gyuwon,Hwang, Hyunuk,Kim, Kibom,Chang, Taejoo 한국정보처리학회 2013 정보처리학회논문지. 컴퓨터 및 통신시스템 Vol.2 No.8

스마트폰의 다양한 기능들이 사용되면서 사용자의 개인정보를 비롯한 대용량의 데이터들이 스마트폰에 저장되고 있다. 그러나 운영체제와 어플리케이션의 잦은 업데이트는 데이터의 손실을 야기할 수 있으며, 개인의 소중한 데이터를 분실할 위험성을 갖게 한다. 이로 인하여 데이터의 백업에 대한 중요성이 크게 증가하였으며 많은 사용자들이 자신의 데이터를 안전하게 보관하기 위해 백업 기능을 사용하고 있다. 그러나 포렌식 관점에서 이 백업 파일들은 스마트폰의 은닉 및 데이터의 고의 삭제시 중요한 수사 대상이 된다. 따라서, 이 논문에서는 세계에서 스마트폰 점유율이 가장 높은 삼성 스마트폰의 Kies 백업 파일에 대한 구조를 분석하고, 백업 파일을 복원하는 기법을 제안한다. 실험 결과 제안된 기법은 다양한 유형의 파일들을 분석하여 타 도구들 대비 높은 파일 추출 결과를 보였다. As various features of the smartphone have been used, a lot of information have been stored in the smartphone, including the user's personal information. However, a frequent update of the operating system and applications may cause a loss of data and a risk of missing important personal data. Thus, the importance of data backup is significantly increasing. Many users employ the backup feature to store their data securely. However, in the point of forensic view these backup files are considered as important objects for investigation when issued hiding of smartphone or intentional deletion on data of smartphone. Therefore, in this paper we propose a scheme that analyze structure and restore data for Kies backup files of Samsung smartphone which has the highest share of the smartphone in the world. As the experimental results, the suggested scheme shows that the various types of files are analyzed and extracted from those backup files compared to other tools.

[특별기고] 미술 작가의 셀프 브랜딩, 그 효과는?

이규원 (Gyuwon Lee) 한국미술경영학회 2023 한국미술경영학회 논문집 Vol.- No.4

성공한 미술 작가는 어떻게 될 수 있을까? 성공한 작가가 되려면 기본적으로 작품이 좋아야 하는 것은 당연한 이 야기이다. 하지만 어떤 작가는 누가 봐도 훌륭한 작품을 하지만 유명하지 않은 경우를 우리는 주위에서 쉽게 볼 수 있 다.그리고 작품만큼 중요한 것은 작가와 작품의 홍보일 것이다. 예전에는 학연, 지연 등등 물리적인 방법으로 본인을 홍보했다면, 현재는 SNS를 통해서 작가의 일상과 작품을 알리고 있다. 불과 몇 년 전까지만 해도 그 효과에 대해 기성 작가들은 회의적인 반응이었다. 하지만 이제는 미술 작가라면 인스타그램은 당연히 이용해야 하는 홍보 수단이 되었 다. 다니엘 아샴은 비교적 젊은 작가답게 140만 명 이상의 팔로우를 가진 인스타그램을 자신의 홍보와 작품 판매에 활 용하고 있었다. 일본 작가인 이노우에는 전직 그래픽 디자이너답게 인스타그램을 감각적으로 활용해 마치 온라인 갤 러리 같은 효과를 보여주고 있다. 또한 이상원 작가는 인스타그램을 통해서 다시 한번 기회를 잡은 미술 작가가 되었 다. 마지막으로 정디안은 자신의 강점을 인스타그램을 통해 적극적으로 어필하는 젊은 작가였다. 결국, 현재 미술 작 가의 셀프 브랜딩을 위해서 인스타그램은 가장 효과적인 방법이라 할 수 있다. How does one become a successful artist? It goes without saying that to be a successful artist, you basically need to have good artwork However, we can easily see cases where some artists have great artwork but are not famous- And perhaps just as important as the artwork is the promotion of the artist and their work.. In the past, artists promoted themselves through physical means, such as school ties and personal connections. Until just a few years ago, established writers were skeptical about its effectiveness. However, Instagram has now become a valuable public relations tool that artists should utilize. Daniel Asham is a relatively young artist who uses Instagram to promote himself and sell his work, with over 1.4 million followers. Inoue, a former graphic designer, uses Instagram in a clever way to create the effect. Lee Sang-won is a fine artist who took a second chance through Instagram. Finally, Chung Dian is a young artist who actively uses Instagram to showcase her strengths. In the end, we can say that Instagram is the most effective way for artists to self-brand.

레이더와 지상관측소 강우자료를 이용한 시공간 강우 조정 모형

신태성,이규원,김용구,Shin, Tae Sung,Lee, Gyuwon,Kim, Yongku 한국데이터정보과학회 2017 한국데이터정보과학회지 Vol.28 No.1

현재 국내에는 무인 관측소인 AWS를 포함하여 많은 지상 강우 관측소를 통해 강우 데이터를 생성하고 있으며, 최근 국토해양부에서는 국내 최초로 대구 비슬산에 이중편파 강우레이더를 설치하여 운영 중에 있으며 이를 통해 전국적으로 강우강도와 강우의 분포를 추정하고 있다. 일반적으로 AWS의 경우 실제 지면에 내린 강우량을 직접 측정하므로 실제 강우량과 근사한 값을 갖지만 AWS의 개수가 적어 강우의 공간분포를 확인하기에는 불충분하다. 반면에 레이더의 경우 광범위하게 측정할 수 있으나 강우 강도와 반사율의 관계식을 통해 강우량을 측정하기 때문에 실제 지면에 내린 강우량과 는 다소 차이가 있을 수 있다. 본 논문에서는 시간적 공간적 강우강도 및 분포를 확인하는데 이점이 있는 레이더 강우 자료와 실제 강우량과 근접한 지상 관측소에서 측정한 강우자료를 사용하여 두 자료에 대한 선형모형을 적용하고, 이 선형모형으로 설명되지 않는 오차에 대해서 공간구조를 가정하여 AWS의 강우자료가 없는 지역의 지상강우량을 추정하여 지상 강우 필드를 생성하였다. Precipitation is an important component for hydrological and water control study. In general, AWS data provides more accurate but low dense information for precipitation while radar data gives less accurate but high dense information. The objective of this study is to construct adjusted precipitation field based on hierarchical spatial model combining radar data and AWS data. Here, we consider a Bayesian hierarchical model with spatial structure for hourly accumulated precipitation. In addition, we also consider a redistribution of hourly precipitation to 2.5 minute precipitation. Through real data analysis, it has been shown that the proposed approach provides more reasonable precipitation field.

소형 이중편파 레이더를 이용한 서울지역 시공간 고분해능 강우 추정 및 예측기술 개발

이대형(Lee, Dae-Hyung),방원배(Bang, Wonbae),류근수(Lyu, Geunsu),이규원(Lee, GyuWon),민기홍(Min, Ki-Hong) 대한토목학회 2018 대한토목학회 학술대회 Vol.2018 No.10

Ka-밴드 구름레이더 자료품질 및 구름통계 기초연구

예보영(Bo-Young Ye),이규원(GyuWon Lee),권수현(Soohyun Kwon),이호우(Ho-Woo Lee),하종철(Jong-Chul Ha),김연희(Yeon-Hee Kim) 한국기상학회 2015 대기 Vol.25 No.1

The Ka-band cloud radar (KCR) has been operated by the National Institute of Meteorological Research (NIMR) of Korea Meteorological Administration (KMA) at Boseong National Center for Intensive Observation of severe weather since 2013. Evaluation of data quality is an essential process to further analyze cloud information. In this study, we estimate the measurement error and the sampling uncertainty to evaluate data quality. By using vertically pointing data, the statistical uncertainty is obtained by calculating the standard deviation of each radar parameter. The statistical uncertainties decrease as functions of sampling number. The statistical uncertainties of horizontal and vertical reflectivities are identical (0.28 dB). On the other hand, the statistical uncertainties of Doppler velocity (spectrum width) are 2.2 times (1.6 times) larger at the vertical channel. The reflectivity calibration of KCR is also performed using X-band vertically pointing radar (VertiX) and 2-dimensional video disdrometer (2DVD). Since the monitoring of calibration values is useful to evaluate radar condition, the variation of calibration is monitored for five rain events. The average of calibration bias is 10.77 dBZ and standard deviation is 3.69 dB. Finally, the statistical characteristics of cloud properties have been investigated during two months in autumn using calibrated reflectivity. The percentage of clouds is about 26% and 16% on September to October. However, further analyses are required to derive general characteristics of autumn cloud in Korea.

3차원 레이더 반사도를 이용한 대류세포 판별과 추적 알고리즘의 개발

정성화(Sung-Hwa Jung),이규원(GyuWon Lee),김형우(Hyung-Woo Kim),국봉재(BongJae Kuk) 한국기상학회 2011 대기 Vol.21 No.3

This paper presents the development of new algorithm for identifying and tracking the convective cells in three dimensional reflectivity fields in Cartesian coordinates. First, the radar volume data in spherical coordinate system has been converted into Cartesian coordinate system by the bilinear interpolation. The three-dimensional convective cell has then been identified as a group of spatially consecutive grid points using reflectivity and volume thresholds. The tracking algorithm utilizes a fuzzy logic with four membership functions and their weights. The four fuzzy parameters of speed, area change ratio, reflectivity change ratio, and axis transformation ratio have been newly defined. In order to make their membership functions, the normalized frequency distributions are calculated using the pairs of manually matched cells in the consecutive radar reflectivity fields. The algorithms have been verified for two convective events in summer season. Results show that the algorithms have properly identified storm cells and tracked the same cells successively. The developed algorithms may provide useful short-term forecasting or nowcasting capability of convective storm cells and provide the statistical characteristics of severe weather.

변분에코추적법을 이용한 제주도 지역 여름철 강수계의 이동 특성 분석

김권일(Kwonil Kim),이호우(Ho-Woo Lee),정성화(Sung-Hwa Jung),류근수(Geunsu Lyu),이규원(GyuWon Lee) 한국기상학회 2018 대기 Vol.28 No.4

Nowcasting algorithms using weather radar data are mostly based on extrapolating the radar echoes. We estimate the echo motion vectors that are used to extrapolate the echo properly. Therefore, understanding the general characteristics of these motion vectors is important to improve the performance of nowcasting. General characteristics of radar-based motions are analyzed for warm season precipitation over Jeju region. Three-year summer season data (June~August, 2011~2013) from two radars (GSN, SSP) in Jeju are used to obtain echo motion vectors that are retrieved by Variational Echo Tracking (VET) method which is widely used in nowcasting. The highest frequency occurs in precipitation motion toward east-northeast with the speed of 15~16 m s<SUP>−1</SUP> during the warm season. Precipitation system moves faster and eastward in June-July while it moves slower and northeastward in August. The maximum frequency of speed appears in 10~20 m s<SUP>−1</SUP> and 5~10 m s<SUP>−1</SUP> in June~July and August respectively while average speed is about 14~15 m s<SUP>−1</SUP> in June~July and 8 m s<SUP>−1</SUP> in August. In addition, the direction of precipitation motion is highly variable in time in August. The speed of motion in Lee side of the island is smaller than that of the windward side.

겨울철 영동지역 눈 결정 습성과 성상 변화 에피소드 분석

최영길,김병곤,김지윤,김태연,한진헌,이규원,김권일,김기훈,임병환,Young-Gil Choi,Byung-Gon Kim,Ji-Yun Kim,Tae-Yeon Kim,Jin-Heon Han,GyuWon Lee,Kwonil Kim,Ki-Hoon Kim,Byung-Hwan Lim 한국기상학회 2024 대기 Vol.34 No.2

The Yeongdong region has suffered from severe snowstorms and the relevant damage such as traffic accidents on slippery roads, and the collapse of greenhouses and temporary buildings. While a lot of research on snowfall has been conducted, the detailed study of snow crystals' phase and habit through intensive observations and the relevant microphysical analysis is still lacking. Therefore, a snowflake camera, PARSIVEL, and intensive radiosonde soundings were utilized to investigate phase and habit changes in solid precipitation. Two remarkable episodes of phase and habit changes were selected such as 19 March 2022 and 15 February 2023. Both events occurred in the synoptic condition of the High in the north and the Low passing by the south, which was accompanied by rapid temperature cooling below 2.5 km. During the events of a short period between 3 to 6 hours, the temperature at 850 hPa decreased by about 4 to 6℃. This cooling led to a change in the main habit of snow particles from riming to aggregate, identified with both MASC and PARSIVEL. Meanwhile, the LDAPS model analyses do not successively represent the rapid cooling and short-term variations of solid precipitation, probably by virtue of overestimating low-level equivalent potential temperature during these periods. The underlying causes of these the low-level temperature variations within 6 hours, still remain unclear. It might be associated with mesoscale orographic phenomenon due to the mountains and East Sea effects, which certainly needs an intensive and comprehensive observation campaign.

2단 변속기를 적용한 전기 스포츠카의 성능 및 효율 특성 분석

김시원(Siwon Kim),김영환(Younghwan Kim),이규원(Gyuwon Lee),최길수(Gilsu Choi) 대한전기학회 2021 대한전기학회 학술대회 논문집 Vol.2021 No.7

LSTM 오토인코더를 이용한 이상 탐지의 임계치 결정 방법

전승현(Seunghyeon Jeon),박채린(Chaelyn Park),이규원(Gyuwon Lee),김성종(Seongjong Kim),구본근(Bongen Gu) 한국정보기술학회 2023 한국정보기술학회논문지 Vol.21 No.4

The threshold is an important factor that affects the performance of anomaly detection using LSTM autoencoder for the purpose of reducing direct and indirect losses due to equipment failure. The LSTM autoencoder trained on normal data set has the characteristics that the reconstruction error of abnormal data is greater than that of normal data. In this paper, we propose a method of determining the threshold by using the ratio between the maximum reconstruction error of normal data and the minimum reconstruction error of abnormal data. To show the validity of the proposed threshold determination method, we implement an LSTM autoencoder model for detecting anomalies in pump data that is publicly available through Kaggle and determine the threshold using the threshold determination data set. The experimental results using the determined threshold on the test data set showed an improvement of 5% in accuracy and 25% in recall compared to the existing precision-recall curve method.