http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
김병천(B C. Kim),최형일(H.I. Choi) 한국정보과학회 1990 한국정보과학회 학술발표논문집 Vol.17 No.2
본 논문에서는 프레임 시스템을 기초로 한 얼굴 영상 분석 시스템을 제안한다. 제안된 시스템은 영역 처리기, 제어기, 지식베이스로 구성되어 있으며, 문제 영역에 관한 지식을 프레임으로 표현하여 영역 분할과 해석 단계에서 통합적으로 이용한다. 이를 위한 제어구조는 통합 제어구조로서, 문제 영역에 관한 지식표현 구조가 프레임 형태이므로 기본적으로는 목표 지향적인 하향식 제어구조이다. 그러나 하위레벨의 지식이 예시화 되어 상위레벨로 전파되고, 전파된 지식을 상위레벨에서 검증하여 다른 하위레벨의 지식이나 인접한 다른 지식에 영향을 주는 귀환제어의 특성이 있다.
김창덕 ( Chang Duk Kim ),김기훈 ( Gi Hun Kim ),박재홍 ( Jae Hong Park ),김병천 ( Byung Chun Kim ),이규석 ( Kyu Suk Lee ) 대한피부과학회 2003 대한피부과학회지 Vol.41 No.4
Congenital self-healing reticulohistiocytosis(CSHRH) consists of multiple disseminated papulonodular skin lesions of varying sizes and regress spontaneously in several weeks to months with little or no scarring, and until recently it has been regarded as a benign cutaneous disease without systemic involvement. However, involvement of organ systems other than the skin has been described occasionally and recurrence of disease at sites distant from the skin has been documented. We present a case of CSHRH in a full termed, normal spontaneous vaginal delivered male neonate. He had reddish colored nodules on the whole body without systemic symptoms. Four weeks later, the skin lesions cleared concurrently with bony involvement of the skull. Because multiple organ systems can be involved and recurrences are possible, long-term follow-up of these patients is indicated. (Korean J Dermatol 2003;41(4) : 466~469)
전기로 제강 더스트 배출 스크류 컨베이어의 냉각성능 수치해석
김병천(Byeongcheon Kim),이상욱(Sang-Wook Lee) 대한기계학회 2018 대한기계학회 춘추학술대회 Vol.2018 No.12
Numerical simulations are conducted to analyze the conjugate heat transfer of turbulent flow in conveying screw for a furnace using Ansys FLUENT. The coolant flow in a concentric tube contacts with solid wall and prevents corrosion by making solid wall temperature down. The helical fins are attached in the flow path in order to create helical flow. In this study, the effect of helical fin along flow path on cooling performance is investigated. The result of full height fin model showed that rotational flow driven by helical fin increases the cooling performance of conveying screw.
평활근 증식과 점액성 기질 변화를 보인 사구맥관근종 1 예
김창덕(Chang Duk Kim),박재홍(Jae Hong Park),류영욱(Young Wook Ryoo),김병천(Byung Chun Kim),이규석(Kyu Suk Lee) 대한피부과학회 2001 대한피부과학회지 Vol.39 No.10
Glomus tumors are benign neoplasms that are derived from modified smooth muscle cells known as glomus cells. Histologically, it can be subdivided as glomus tumor proper, glomangioma, and glomangiomyoma according to relative proportions of components. Glomangiomyomas are the least frequent type and their overall architectural pattern may resemble glomus tumor proper or glomangioma, but there is a gradual transition from glomus cells to elongated mature smooth muscle cells. This transition is most obvious in the region surrounding large vessels. We present a case of glomangiomyoma of the left upper arm and the left fourth finger, in which ten-year history of two painful, bluish-colored subcutaneous nodules. On histologic examination, this case showed marked smooth muscle cell proliferation around large vessles and mucinous stromal change. (Korean J Dermatol 2001;39(10) 1147-1151)
김창덕 ( Chang Duk Kim ),김창욱 ( Chang Ook Kim ),김병천 ( Byung Chun Kim ),이규석 ( Kyu Suk Lee ) 대한피부과학회 2003 大韓皮膚科學會誌 Vol.41 No.7
N/A Background : Many studies focused on the distribution of skin tumors in Korea, but none of them included whole data refered from deparments other than dermatology and primary physicians. Objective : This study was to investigate the incidence and distribution of skin tumors among Korean people. Methods : A study was made on 1448 cases of skin tumors which were obtained mainly from surgical pathology specimen during the period of 10 years from 1992 to 2001 at Department of Dermatology in Keimyung University Dongsan Medical center. Results : 1. Of the 1448 cases of skin tumors, 1171 cases(80.9%) were benign tumor, 277 cases (19.1%) were malignant tumor and male to female ratio was 1:1.1. 2. Among the 1171 cases of benign tumors, 174 cases of melanocytic nevi, 173 cases of epidermal cyst, 109 cases of seborrheic keratosis, 62 cases of granuloma pyogenicum were noted. 3. Among 277 cases of malignant tumors, 107 cases of basal cell carcinoma, 58 cases of squamous cell carcinoma, 27 cases of metastatic cutaneous carcinoma, 24 cases of Bowen disease were noted. 4. The predilection sites of Intradermal nevi were face(38.4%), trunk(17.4%), scalp(16.3%), and those of epidermal cyst were face(41.0%), trunk(29.5%), neck(8.1%), The predilection site of basal cell carcinoma, squamous cell carcinoma was face(80.4%, 58.6%). Conclusion : The 1448 cases of skin tumors, which were confirmed histologically from surgical pathology specimen at Department of Dermatology in the Keimyung University Dongsan Medical Center during the period of 10 years from 1992 to 2001, were reviewed.
김병천(Kim Byung Cheon),윤병주(Yoon Byung Joo) 한국정보처리학회 1999 정보처리학회논문지 Vol.6 No.3
Reinforcement learning performs learning through interacting with trial-and-error in dynamic environment. Therefore, in dynamic environment, reinforcement learning method like Q-learning and TD(Temporal Difference)-learning are faster in learning than the conventional stochastic learning method. However, because many of the proposed reinforcement learning algorithms are given the reinforcement value only when the learning agent has reached its goals state, most of the reinforcement algorithms are given the reinforcement value only when the learning agent has reached its goal state, most of the reinforcement algorithms converge to the optimal solution too slowly. In this paper, we present COMREL(COMpressed REinforcement Learning) algorithm for finding the shortest path fast in a maze environment. COMREL compress the given maze environment, select the candidate states that can guide the shortest path in compressed maze environment, and learn only the candidate states to find the shortest path. After comparing COMREL algorithm with the already existing Q-learning and Priortized Sweeping algorithm, we could see that the learning time shortened very much.
김병천(Kim Byung Cheon),윤병주(Yoon Byung Joo) 한국정보처리학회 1999 정보처리학회논문지 Vol.6 No.5
In order to learn in dynamic environments, reinforcement learning algorithms like Q-learning, TD(0)-learning, TD(%u03BB)-learning have been proposed, However, most of them have a drawback of very slow learning because the reinforcement value is given when they reach their goal state. In this thesis, we have proposed a reinforcement learning method that can approximate fast to the goal state in maze environments. The proposed reinforcement learning method is separated into global learning and local learning, and then it executes learning. Global learning is a learning that uses the replacing eligibility trace method to search the goal state. In local learning, it propoagates the goal state value that has been searched through global learning to neighboring state, and then searches goal state in neighboring states. We can show through experiments that the reinforcement learning method proposed in this thesis can find out an optimal solution faster than other reinforcement learning methods like Q-learning, TD(0)-learning, TD(%u03BB)-learning.