가천대학교 문화기술연구실 영상처리 및 헬스케어 정보 기술 연구

황보택근 한국공학교육학회 2015 Ingenium(人材니움) Vol.22 No.3

공학교육의 현장 Part 3 우수 연구실 소개 : 가천대학교 문화기술연구실 영상처리 및 헬스케어 정보 기술 연구

황보택근 한국공학교육학회(구 한국공학교육기술학회) 2015 공학교육동향 Vol.22 No.3

View Variations and Recognition of 2-D Objects

황보택근,Whangbo, Taeg-Keun Korea Information Processing Society 1997 정보처리논문지 Vol.4 No.11

Recognition of 3D objects using computer vision is complicated by the fact that geometric features vary with view orientation. An important factor in designing recognition algorithms in such situations is understanding the variation of certain critical features. The features selected in this paper are the angles between landmarks in a scene. In a class of polyhedral objects the angles at certain vertices may form a distinct and characteristic alignment of faces. For many other classes of objects it may be possible to identify distinctive spacial arrangements of some readily identifiable landmarks. In this paper given an isotropic view orientation and an orthographic projection the two dimensional joint density function of two angles in a scene is derived. Also the joint density of all defining angles of a polygon in an image is derived. The analytic expressions for the densities are useful in determining statistical decision rules to recognize surfaces and objects. Experiments to evaluate the usefulness of the proposed methods are reported. Results indicate that the method is useful and powerful. 컴퓨터 비전을 이용한 3차원 물체 인식은 카메라의 위치에 따라 화상에 투영되는 물체의 형상이 변하기 때문에 매우 복잡하고 어렵다. 따라서 컴퓨터 비전을 이용한 효과적인 인식 시스템을 구축하기 위해서는 각 3차원 물체에 있어서 유일하고 중요한 특성이 보는 각도에 따라 어떻게 변화하는가를 분석하고 이해하는 것이 매우 중요한 요소이다. 본 연구에서는 특징점들(landmarks)간에 이루어지는 각도 또는 3차원 다각형의 모서리(edge) 사이의 각도를 중요한 특성으로 선택하였고, orthographic 투영과 isotropic view orientation 아래에서 그러한 각도들의 보는 방향에 따른 화상에서의 변화를 2차원 결합 밀도 함수로 유도하였다. 본 논문에서 구한 수리적인 결합 밀도 함수는 통계적인 판단 규칙을 적용하여 효과적으로 물체 인식에 적용될 수 있다. 제안된 방법의 타당성 검토를 위하여 간단한 실험을 수행하였으며, 실험결과 본 방법 이 매우 효과적인 것으로 나타났다.

실시간 지능형 원격교육에 관한 연구

황보택근,홍기천 한국컴퓨터산업학회 2002 컴퓨터産業敎育學會論文誌 Vol.3 No.9

현재의 원격 교육 및 가상 교육은 일반적으로 학습자의 이해 정도에 관계없이 같은 수준의 교육 내용이 제공되며, 학습 진도 또한 학습자에 의해 임의로 조정할 수 있다. 본 논문에서는 현재의 원격교육 및 가상교육의 문제점을 개선하여 보다 효과적인 실시간 지능형 원격교육을 제안하기 위한 것으로, 먼저 학습자 수준을 파악하여 수준에 맞는 강의 내용을 제공하고, 학습자의 학습의 정도와 이해를 분석 판단하여 그 결과를 교육자에게 전달한다. 진단 결과에 따라 교육자는 학습자와 화상통신을 이용하여 질의문답을 할 수 있으며 학습 자료를 공유하고, 칠판을 제공하여 학습자의 취약 부분을 강조하고 이해시켜 학습의 효율을 높이고, 보다 질 높은 교육이 될 수 있도록 한다. Current distance loaming or cyber teaming offers learner the same level of contents regardless of learner's degree of understanding and the progress of learning also may be controlled by learner freely. In this paper we improve the several problems of distance teaming and cyber learning, and propose new real-time intelligent distance learning. First, we provide learners with the contents of lectures that are appropriate to learner's level. After that we analyze md judge the degree of understanding of learners, the results are informed to instructors for further uses. Due to the diagnosis result instructors know the frail part of learners and let them understand that, and highlight the efficiency of learning. Session with instructor and learner is performed with multimedia communication using Internet, and all the educational materials are shared by instructor and learner at the same time.

Multimedia Streaming

황보택근,홍기천 한국멀티미디어학회 2000 한국멀티미디어학회지 Vol.4 No.1

학술 정보기술 : 첨단 디지털 디자인 정보 도서관

황보택근 한국디지틀도서관포럼 1998 디지틀 도서관 Vol.9 No.-

용철에서의 가탄에 관한 연구 (2) - 가탄에 미치는 탄소재료의 입도와 점결제의 영향 -

황보택근,이종남 ( Taeg Keun Whangbo,Jong Nam Lee ) 한국주조공학회 1985 한국주조공학회지 Vol.5 No.1

In order to investigate the effects of binder and particle size of carbon bearing material on carbon pick-up in molten iron, sodium silicate (sodium silicate bonded carburizer) and coal pitch (coal pitch bonded carburizer) were used as the binders. The sodium silicate content was varied as 5wt%, 10wt%, 20wt%, and the particle size of carbon bearing material was classified by the sieving as follows; 50-100 mesh, 100-140 mesh, 140-200 mesh, under 200 mesh. The forme dcarburizers were heattreated at 1300℃ and added to the induction furnace melt. Experimental results obtained from this study were summarized as Follows. 1. The carbon content in molten iron was increased with coarsening of particle size of carbon bearing material. 2. In the case of the sodium silicate bonded carburizer the carbon content in molten iron was higher than that of the coal pitch bonded one. The carbon content in molten iron was decreased with the increase of sodium silicate content. 3. The silicon content in molten iron was increased with the increase of sodium silicate content. 4. The dissolving time was shortened with the increase of particle size of carbon bearing material and sodium silicate content.

Personalized Urination Activity Recognition Based on a Recurrent Neural Network Using Smart Band

황보택근,은성종,정은영,박동균,김수진,김창희,정경진,김계환 대한배뇨장애요실금학회 2018 International Neurourology Journal Vol.22 No.S2

Purpose: Though it is very important obtaining exact data about patients’ voiding patterns for managing voiding dysfunction, actual practice is very difficult and cumbersome. In this study, data about urination time and interval measured by smart band device on patients’ wrist were collected and analyzed to resolve the clinical arguments about the efficacy of voiding diary. By developing a smart band based algorithm for recognition of complex and serial pattern of motion, this study aimed to explore the feasibility of measurement the urination time and intervals for voiding dysfunction management. Methods: We designed a device capable of recognizing urination time and intervals based on specific postures of the patient and consistent changes in posture. These motion data were obtained by a smart band worn on the wrist. An algorithm that recognizes the repetitive and common 3-step behavior for urination (forward movement, urination, backward movement) was devised based on the movement and tilt angle data collected from a 3-axis accelerometer. The sequence of body movements during voiding has consistent temporal characteristics, so we used a recurrent neural network and long short-term memory based framework to analyze the sequential data and to recognize urination time. Real-time data were acquired from the smart band, and for data corresponding to a certain duration, the value of the signals was calculated and then compared with the set analysis model to calculate the time of urination. A comparative study was conducted between real voiding and device-detected voiding to assess the performance of the proposed recognition technology. Results: The accuracy of the algorithm was calculated based on clinical guidelines established by urologists. The accuracy of this detecting device was high (up to 94.2%), proving the robustness of the proposed algorithm. Conclusions: This urination behavior recognition technology showed high accuracy and could be applied in clinical settings to characterize patients’ voiding patterns. As wearable devices are developed and generalized, algorithms detecting consistent sequential body movement patterns reflecting specific physiologic behavior might be a new methodology for studying human physiologic behavior.

인터넷 쇼핑몰에서 효율적인 상품 검색을 위한 가시화방법

황보택근 ( Whangbo Taeg Keun ),이수진 ( Lee Soo Jin ) 한국인터넷정보학회 2005 인터넷정보학회논문지 Vol.6 No.5

전자상거래의 규모가 증가함에 따라 제공되는 상품의 정보와 서비스의 내용도 점차 다양해지고 있지만, 기존 쇼핑몰은 사용자에게 상품 정보를 일괄적인 나열방식으로 제공하기 때문에 사용자가 원하는 상품을 선택하기 위해서는 많은 노력과 시간이 소요된다. 본 논문에서는 속성 탐색기(feature inquirer)와 평행좌표(parallel coordinates)를 이용한 정보 가시화 기법을 적용하여 사용자에게 편리한 인터페이스를 제공함으로써 원하는 상품을 쉽게 검색할 수 있도록 하고, 상품의 효율적인 비교 검색을 통하여 정확한 상품선택이 이루어지도록 하는 방법을 제안하였다. 제안하는 검색 기법의 효율성을 검증하기 위해 피험자를 대상으로 검색 소요시간과 마우스 조작 횟수를 비교하여 타당성을 검증하였다. As the scale of e-commerce has been increasing, information about products and service become diversified. Most current internet shopping mall, however, provides product information to consumers in collective arrangement, thus they devote a lot of efforts and time to select the product they want. In this paper, as providing consumers convenient interfaces adapting feature inquirer and parallel coordinate, a method for selecting a right product through searching the wanted products easily and comparing their features efficiently has been proposed. To evaluate the efficiency of the method proposed in this paper, several experiments to measure the time spent and the number of clicks of mouse taken in finding the product they want have been conducted. The results verify the effectiveness of the proposed method.

화상에서의 각도 변화를 이용한 3차원 물체 인식

황보택근(Whang bo Taeg Keun) 한국정보처리학회 1997 정보처리학회논문지 Vol.4 No.11

Recognition of 3D objects using computer vision is complicated by the fact geometric features vary with view orientation. An important factor in designing recognition algorithms in such situations is understanding the variation of certain critical features. The features selected in this paper are the angles between landmarks in a scene. In a class of polyhedral objects the angles at certain vertices may form a distinct and characteristic alignment of faces. For many other classes of objects it may be possible to identify distinctive spactial arrangements of some readily identifiable landmarks. In this paper given an isotropic view orientation and an orthographic projection the two dimensional joint density function of two angles in a scene is derived. Also the joint density of all defining angles of a polygon in an image is derived. The analytic expressions for the densities are useful in determining statistical decision rules to recognize surfaces and objects. Experiments to evaluate the usefulness of the propesed methods are reported. Results indicate that the method is useful and powerful.