• Journal of Information Processing Systems
• /
• v.16 no.3
• /
• pp.612-628
• /
• 2020

This paper presents a new approach for detecting abnormal behaviors in complex surveillance scenes where anomalies are subtle and difficult to distinguish due to the intricate correlations among multiple objects' behaviors. Specifically, a cascaded probabilistic topic model was put forward for learning the spatial context of local behavior and the temporal context of global behavior in two different stages. In the first stage of topic modeling, unlike the existing approaches using either optical flows or complete trajectories, spatio-temporal correlations between the trajectory fragments in video clips were modeled by the latent Dirichlet allocation (LDA) topic model based on Markov random fields to obtain the spatial context of local behavior in each video clip. The local behavior topic categories were then obtained by exploiting the spectral clustering algorithm. Based on the construction of a dictionary through the process of local behavior topic clustering, the second phase of the LDA topic model learns the correlations of global behaviors and temporal context. In particular, an abnormal behavior recognition method was developed based on the learned spatio-temporal context of behaviors. The specific identification method adopts a top-down strategy and consists of two stages: anomaly recognition of video clip and anomalous behavior recognition within each video clip. Evaluation was performed using the validity of spatio-temporal context learning for local behavior topics and abnormal behavior recognition. Furthermore, the performance of the proposed approach in abnormal behavior recognition improved effectively and significantly in complex surveillance scenes.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.341-343
• /
• 2022

온실가스 배출 문제가 세계적인 현안으로 부각되면서 수소를 에너지원으로 사용하는 수소경제가 주목받고 있다. 수소연료전지는 수소경제의 구성요소 중 하나로, 수소를 활용해 열과 전기를 생산하며 에너지 변환 효율이 높이는데 장점이 있다. 본 연구는 세계적인 온라인 커뮤니티인 레딧(Reddit)에서 수집한 수소연료전지와 관련된 소셜 데이터를 텍스트마이닝과 감성분석 기법으로 분석하였다. 분석 결과 9,211건의 댓글을 LDA(Latent Dirichlet Allocation)을 이용해 4개의 토픽 그룹으로 분류할 수 있었다. 이 중 수소연료전지와 관련이 높은 그룹을 선정해 STM(Structural Topic Model) 분석으로 10개 토픽을 추출하였고, 기후 환경, 수소 산업, 수소 차와 관련 있는 토픽 3개를 발견할 수 있었다. 이 연구 결과를 통해 수소연료전지의 세계적으로 실제적인 내용을 빠르고 효과적으로 파악하여 수소연료전지에 대한 예측하고, 우리나라의 수소연료전지 관련 국가R&D의 정책적 방향을 제시하고자 한다.

• Proceedings of the IEEK Conference
• /
• 2006.06a
• /
• pp.797-798
• /
• 2006

In this paper, a codec design method is pro-posed for the face images based on JPEG and its application to face recognition is presented. Quantization table design is dis-cussed using R-D optimization for Yale face data. For the usage in the embedded systems, fast codec design is also considered. The proposed codec has better performance than JPEG codec for face images. Through the recognition experiment using PCA and LDA, it has been shown that the proposed codec has better performance than JPEG codec.

• Journal of Information Technology Services
• /
• v.18 no.2
• /
• pp.123-141
• /
• 2019

The goal of this study is to review the major research trend on the convergence studies of AI and healthcare technologies. For the study, 15,260 English articles on AI and healthcare related topics were collected from Scopus for 55 years from 1963, and text mining techniques were conducted. As a result, seven key research topics were defined : "AI for Clinical Decision Support System (CDSS)", "AI for Medical Image", "Internet of Healthcare Things (IoHT)", "Big Data Analytics in Healthcare", "Medical Robotics", "Blockchain in Healthcare", and "Evidence Based Medicine (EBM)". The result of this study can be utilized to set up and develop the appropriate healthcare R&D strategies for the researchers and government. In this study, text mining techniques such as Text Analysis, Frequency Analysis, Topic Modeling on LDA (Latent Dirichlet Allocation), Word Cloud, and Ego Network Analysis were conducted.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.47 no.6
• /
• pp.12-18
• /
• 2010

Automatic Target Detection (ATD) systems that use forward-looking infrared (FLIR) consists of three stages. preprocessing, detection, and clutter rejection. All potential targets are extracted in preprocessing and detection stages. But, this results in a high false alarm rates. To reduce false alarm rates of ATD system, true targets are extracted in the clutter rejection stage. This paper focuses on clutter rejection stage. This paper presents a new clutter rejection technique using PCA features and stochastic features of clutters and targets. PCA features are obtained from Euclidian distances using which potential targets are projected to reduced eigenspace selected from target eigenvectors. CV is used for calculating stochastic features of edges in targets and clutters images. To distinguish between target and clutter, LDA (Linear Discriminant Analysis) is applied. The experimental results show that the proposed algorithm accurately classify clutters with a low false rate compared to PCA method or CV method

• Journal of Biomedical Engineering Research
• /
• v.15 no.4
• /
• pp.489-497
• /
• 1994

The most serious problems related to the cardiovascular prothesis are thrombosis and hemolysis. It is known that the flow pattern of cardiovascular prostheses is highly correlated with thrombosis and hemolysis. Laser Doppler Anemometry (LDA) is a usual method to get flow pattern, which is difficult to operate and has narrow measure region. Particle Image Velocimetry (PIV) can solve these problems. Because the flow speed of valve is too high to catch particles by CCD camera, high-speed camera (Hyspeed : Holland-Photonics) was used. The estimated maximum flow speed was 5m/sec and maximum trackable length is 0.5 cm, so the shutter speed was determined as 1000 frames per sec. Several image processing techniques (blurring, segmentation, morphology, etc) were used for the preprocessing. Particle tracking algorithm and 2-D interpolation technique which were necessary in making gridrized velocity pronto, were applied to this PIV program. By using Single-Pulse Multi-Frame particle tracking algorithm, some problems of PIV can be solved. To eliminate particles which penetrate the sheeted plane and to determine the direction of particle paths are these solving methods. 1-D relaxation fomula is modified to interpolate 2-D field. Parachute artificial heart valve which was developed by Seoul National University and Bjork-Shiely valve was testified. For each valve, different flow pattern, velocity profile, wall shear stress and mean velocity were obtained.

• Journal of Industrial Convergence
• /
• v.19 no.6
• /
• pp.11-19
• /
• 2021

The purpose of this study is to explore trends in research at home and abroad related to "Sport Facilities", and seek the direction of further research. 1,801 abstracts of papers including "Sport Facilities" were collected from the SCOPUS DB from 2016 to 2020. Topic modeling techniques based on Latent Dirichlet Allocation (LDA) algorithm implemented in R language, TD-IDF techniques, and word cluds using Tagxedo was conducted to analyze the data. As a result, 8 topics were optimally determined, and "sports", "facilities", "health", "physical", "data", and "using" were derived as the main keywords for topics. This results indicated that studies on physical activity, health and using facilities regarding sports facilities at home and abroad have been actively carried out in recent years. This indicates that papers in SCOPUS DB are paying attention to the instrumental value of sport facilities, such as health promotion and improving the quality of life. Therefore, various studies that help participants who use sport facilities for a healthy life should be continuously conducted in the future.

• Korean Journal of Remote Sensing
• /
• v.38 no.5_3
• /
• pp.873-885
• /
• 2022

The particulate matter (PM) has emerged as a hot topic around the world as it has been reported that PM is related to an increase in mortality and prevalence rates. In South Korea, the importance of PM has been recognized since the late 1990s, and various studies on PM have been conducted. This study investigated the PM research topics and trends for papers (D=2,764) published in Research Information Sharing Service (RISS) using topic modeling based on Latent Dirichlet Allocation (LDA). As a result, a total of 10 topics were identified in the whole papers, and the PM research topics were classified as 'PM reduction (Topic 1)', 'Government policy and management (Topic 2)', 'Characteristics of PM (Topic 3)', 'PM model (Topic 4)', 'Environmental education (Topic 5)', 'Bio (Topic 6)', 'Traffic (Topic 7)', 'Asian dust (Topic 8)', 'Indoor PM (Topic 9)', 'Human risk (Topic 10)'. In particular, the proportion of papers on topics 'Government policy and management (Topic 2)', 'PM model (Topic 4)', 'Environmental education (Topic 5)', and 'Bio (Topic 6)' to the toal number of papers increased over time (linear slope > 0). The results of this study provide the new literature review methodology related to particulate matter and the history and insight.

• International Journal of Naval Architecture and Ocean Engineering
• /
• v.5 no.1
• /
• pp.81-100
• /
• 2013

During the final quarter of the last century considerable efforts have been spent to reduce the hull pressure fluctuations caused by unsteady propeller cavitation. This has resulted in further changes in propeller design characteristics including increased skew, tip unloading and introduction of "New Blade Sections" (NBS) designed on the basis of the so-called Eppler code. An experimental study was carried out to investigate flow characteristics of alternative two-dimensional (2-D) blade sections of rectangular planform, one of which was the New Blade Section (NBS) developed in Newcastle University and other was based on the well-known National Advisory Committee for Aeronautics (NACA) section. The experiments comprised the cavitation observations and the measurements of the local velocity distribution around the blade sections by using a 2-D Laser Doppler Anemometry (LDA) system. Analysis of the cavitation tests demonstrated that the two blade sections presented very similar bucket shapes with virtually no width at the bottom but relatively favourable buckets arms at the suction and pressure sides for the NACA section. Similarly, pressure analysis of the sections displayed a slightly larger value for the NBS pressure peak. The comparative overall pressure distributions around the sections suggested that the NBS might be more susceptible to cavitation than the NACA section. This can be closely related to the fundamental shape of the NBS with very fine leading edge. Therefore a further investigation into the modification of the leading edge should be considered to improve the cavitation behaviour of the NBS.

• Journal of Embryo Transfer
• /
• v.22 no.3
• /
• pp.167-172
• /
• 2007

Pituitary-specific transcription factor (PIT1) 유전자는 동물의 성장을 조절하고 근육 형성에 관여하는 유전자로서 최근에는 단일염기다형성 변이가 한우를 비롯한 동물에서 관찰되었으며, 한우의 경제 형질과 연관성이 보고되었다. 본 연구는 PIT1 유전자의 단일염기다형성 변이가 한우에서 성장 인자에 미치는 영향과 경제 형질에 대한 유전자형간 육종가와의 상관성에 대해 알아보고자 하였다. 도체 성적을 보유하고 있는 한우 후보종모우 집단 268두를 대상으로 PIT1 유전자 A1256G 다형성을 조사하여 유전자형의 빈도를 분석하였고 각각의 유전형에 따른 기본적인 검정 성적을 바탕으로 경제 형질과의 연관성을 비교 분석하였다. 268두의 한우에서 PIT1 유전자의 A1256G 유전자형 빈도는 MseI 제한 효소를 사용했을 때 A 유전자 빈도(0.37)보다 G 유전자 빈도(0.62)가 높게 나타났다. 통계적 분석을 통하여 각 유전자형에 대한 경제 형질과의 관련성을 분석한 결과, 각 유전자형 간에 12개월령 체중 (body weight 12, BW12)에서 유의한 차이를 보였고, 등지방 두께 육종가 (Backfat thickness-estimated breeding value, BF-EBV)와도 유의한 차이가 있었지만 (p<0.05), marbling score (MS), carcass weight (CW), M. longissimus dorsi area (LDA) 등 다른 경제 형질과는 통계학적으로 유의한 차이가 없었다. PIT1 유전자의 A1256G 다형성은 한우의 성장과 도체체중에 관여하는 인자로 작용하는 것으로 보여진다.