http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
신경망을 이용한 로봇 Endeffector 추적 시스템 설계
朴東宣 全北大學校 1996 論文集 Vol.42 No.-
In this paper, we describe a neural net robot endeffector tracking system. The neural network approach is employed to recognize the robot endeffector covering the situation of three types of motion: translation, scaling and rotation. Features for the neural network to detect the position of the endeffector are extracted from the preprocessed images. Artificial neural networks are used to store models and to match with unknown input features recognizing the position of the robot endeffector. Since a minimal number of samples are used for different directions of the robot endeffector in the system, and artificial neural network with the generalization capability can be utilized for unknown input features. A feedforward neural network trained with the back propagation learning is used to detect the position of the robot endeffector. Another feedforward neural network module is used to estimate the motion from a sequence of images and to control movements of the robot endeffector. Combining the two neural networks for recognizing the robot endeffector and estimating the motion with the preprocessing stage, the whole system keeps tracking of the robot endeffector effectively.
다차원 GIS에서 과거데이터의 무결성을 위한 연산의 설계
박동선,김재홍,배해영,Park, Dong-Seon,Kim, Jae-Hong,Bae, Hae-Young 한국정보처리학회 2000 정보처리논문지 Vol.7 No.6
지난 90년대 초반부터 시간차원을 포함하는 다차원 지리정보시스템에 대한 연구가 활발히 진행되고 있으며, 다수의 시간 지원 공간 연산과 시간지원 공간 길의어에대한 연구 결과가 발표되었다. 그러나 이들 연구들은 대부분 현재시점에 유효한 데이터를 대상으로 하는 연산에 치중하고 있으며, 과거데이터의 오류를 수정하기 위한 연산에 대한 연구는 미비하다. 본 논문에서는 시간을 지원하는 다차원 GIS의 과거 데이터에서 발생할 수 있는 오류의 수정이나, 누락된 정보를 삽입하여 과거데이터이 무결성을 유지할 수 있는 연산들을 설계하며, 설계하는 연산이 효율적으로 수행될 수 있는 새로운 데이터 모델을 개안한다. In the last decade, considerable research effort has been studied ot multi-dimensional GIS and many spatiotemporal operations and query languages have been proposed. Many of them have focused on designing of operation, which process with data that is valid at current point in time. But, research that studied to updata operation that updates or inserts the past data is very rare. This paper designs an update operation and an insertion operation for integrity of past data of multi-dimensional GIS, and propose a new data model for efficient processing of the operations.
A Case of Acute Colonic Diverticulitis as a Complication of Colonoscopy
박동선,박지원,김성열,장길수,홍은영,안정선,김소연,백일현,김종혁,박충기 대한장연구학회 2013 Intestinal Research Vol.11 No.2
Colonoscopy is a good diagnostic tool and facilitates treatment of various colonic diseases. Nevertheless, it can induce many serious complications such as perforation and hemorrhage. Diverticulitis has also been reported as a serious complication of colonoscopy, with an incidence ranging from 0.04% to 0.08%. A 44-year-old male with chronic hepatitis B was presented with general weakness, myalgia, and febrile sensation. After admission for evaluation, pneumonia detected in the left upper and lower lobe and treated. We performed colonoscopy for screening and found multiple colonic diverticula in the right side of the colon. After 48 hours, the patient complained of abdominal pain and febrile sensation. Physical examination revealed tender-ness in the right side of the abdomen. Abdomen-pelvis computed tomography showed bowel wall thickening of the cecum and ascending colon and multiple inflamed diverticula at the cecum with pericolic fat infiltration and fluid collection. We diagnosed the patient with acute diverticulitis after colonoscopy. Thereafter, he was treated with bowel rest and broad-spectrum intrave-nous antibiotics, and recovered. With a review of the relevant literature, we report a case of acute colonic diverticulitis as a com-plication of colonoscopy. (Intest Res 2013;11:146-148)
박동선,Park, Dong-Seon 한국정보처리학회 1997 정보처리학회논문지 Vol.4 No.2
본 논문에서는 클래스내부와 클래스간의를 확정하게 제어할 수 있는 랜덤 프로세스 모델을 제어하는 프리세스 내부의 파라메다들을 변화시키며, 프로세스간의 통계적인 차이와 랜덤 잡음을 변화시켜 학습을 위한 패턴들을 생성한다. 이 랜덤 프로세스 모델에서 생성된 패턴들을 이용하여 역전파알고리즘으로 학습된 다단 신경망의 성능 성능을 평가한다. 평가 실험결과는 패턴 분류문제에서 일반화된 통계적인 거리가 분류문제의 난이도에 대한 좋은 예측기가 되는 것을 보여 준다. 또한 본 논문에서는 다단신경망의 성능과 베이스패턴분류기의 성능을 비교하기 위하여 베이스분류기의 이론적인 성능분석과 모의실험을 통한 평가를 하였다. 다단신경망의 분류성능이 이론적인 성능과 실헝치와 매우 근사하며 그 두 성능 중간에 위치함을 발견하였다. We describe a random prcess model that prvides sets of patterms whth prcisely contrlolled within-class varia-bility and between-class distinctions.We used these pattems in a simulation study wity the back-propagation netwoek to chracterize its perfotmance as we varied the process-controlling parameters,the statistical differences between the processes,and the random noise on the patterns.Our results indicated that grneralized statistical difference between the processes genrating the patterns provided a good predictor of the difficulty of the clssi-fication problem. Also we analyzed the performance of the Bayes classifier whith the maximum-likeihood cri-terion and we compared the performance of the neural network to that of the Bayes classifier.We found that the performance of neural network was intermediate between that of the simulated and theoretical Bayes classifier.