• 한국정보과학회논문지:시스템및이론
• /
• 제35권8호
• /
• pp.407-415
• /
• 2008

본 연구에서는 긴 춤 시퀀스를 춤 세그먼트들로 분절하는 방법과 이 춤 세그먼트들의 자연스러운 연결을 위한 전이 방법을 제안한다. 제안하는 분절화(segmentation) 방법은 라반 운동 분석(LMA)을 기반으로 한다. 춤 세그먼트 안에서 춤추는 캐릭터 외형의 변화가 일정한 패턴을 이루며 발생하는 것에 착안하여, 패턴이 변화되는 구간에서 춤 세그먼트의 경계를 찾는다. 이 방법은 춤 시퀀스에서 얻어낸 LMA의 외형 곡선에서 키 자세(key pose)를 찾고, 나타나는 패턴을 분석하여 수행된다. 제안하는 전이(transition) 방법은 춤과 같이 유사도가 낮은 모션을 대상으로 하고 있다. 이 방법은 선형 보간 방법을 기본으로 이용하고, 발에 발생하는 제약사항을 만족하도록 중간 자세를 생성하고, 전이 구간 중간의 키 프레임으로 사용하여 전이를 완성한다. 우리가 제안하는 방법들은 기존의 춤 모션 데이타를 사용한 새로운 춤 시퀀스 합성에 활용이 가능하다.

• 한국전자통신학회논문지
• /
• 제11권1호
• /
• pp.73-80
• /
• 2016

자동차 번호판 인식은 뺑소니차량의 추적이나 교통량의 측정, 교통사고의 조사 및 차량의 증가에 따른 차량범죄의 추적에 이용되고 있다. 실제 적용되는 교통 환경에서는 눈이나 비 그리고 주야간의 조명 변화에 따라서 입력되는 영상에 외란의 영향을 받기 쉬우며, 또한 영상을 촬영하는 순간의 차량의 직진방향과 카메라가 보는 방향에 따라서 동일한 번호판에 대해서도 기하학적으로 변형된 영상이 입력되게 된다. 본 연구에서는 이러한 카메라를 이용한 번호판 인식 환경의 문제를 해결하는 방법으로 호모그래피를 이용하여 기하학적으로 변형된 영상을 원래의 영상으로 변환하는 방법과 투영 히스토그램을 이용한 문자의 분리 방법을 제안하였다. 분리된 영상은 다중 퍼셉트론방법을 이용하여 문자와 숫자를 인식하였고 특히 최적한 입력, 은닉, 출력 층의 비율을 실험을 통하여 도출 하였다.

• Archives of Plastic Surgery
• /
• 제40권4호
• /
• pp.392-396
• /
• 2013

Background The typical clinical symptoms of glomus tumors are pain, tenderness, and sensitivity to temperature change, and the presence of these clinical findings is helpful in diagnosis. However, the tumors often pose diagnostic difficulty because of variations in presentation and the nonspecific symptoms of glomus tumors. To the best of our knowledge, few studies have reported on the usefulness of magnetic resonance imaging (MRI) in diagnosing glomus tumors in patients with unspecific symptoms. Methods The inclusion criteria of this study were: having undergone surgery for subungual glomus tumor of the hand, histopathologic confirmation of glomus tumor, and having undergone preoperative MRI. Twenty-one patients were enrolled. The characteristics of the tumors and the presenting symptoms including pain, tenderness, and sensitivity to temperature change were retrospectively reviewed. Results Five out of 21 patients (23%) did not show the typical glomus tumor symptom triad because they did not complain of pain provoked by coldness. Nevertheless, preoperative MRI showed well-defined small soft-tissue lesions on T1- and T2-weighted images, which are typical findings of glomus tumors. The tumors were completely resected and confirmed as glomus tumor histopathologically. Conclusions Early occult lesions of glomus tumor in the hand may not be revealed by physical examination because of their barely detectable symptoms. Moreover, subungual lesions may be particularly difficult to evaluate on physical examination. Our cases showed that MRI offers excellent diagnostic information in clinically undiagnosed or misdiagnosed patients. Preoperative MRI can accurately define the character and extent of glomus tumor, even though it is impalpable and invisible.

• 한국정보통신학회논문지
• /
• 제25권10호
• /
• pp.1427-1434
• /
• 2021

본 연구는 사회성 상호작용 훈련자극에 반응하는 아동의 행동 및 정서적 긴장상태를 연산하는 심리반응 상태 추정 프로세스를 제안한다. 행동 중재에 필요한 훈련자극으로는 공동 주의(Joint attention) 사회성 훈련을 채택하고, 훈련프로토콜은 불연속 개별시도 훈련(DTT: Discrete trial training)기법이 적용된다. 공동 주의 훈련에서 사용자의 집중과 긴장 정도를 확인하기 위해 3가지 훈련자극용 콘텐츠를 구성한 후, 캐릭터 형태의 탁상 로봇을 이용하여 사용자에게 훈련자극을 수행하게 된다. 그런 다음, 비전 기반 헤드 포즈 인식기와 기하학 연산모델로 사용자 응시반응을 추정하고, PPG와 GSR의 생체신호를 심박변이도와 히스토그램 기법으로 신경계 반응을 분석한다. 로봇을 활용한 실험에서 공동 주의에 대한 각 콘텐츠 별 훈련에 사용자의 심리반응을 정량화 할 수 있음을 확인하였다.

• 영미문화
• /
• 제16권2호
• /
• pp.89-110
• /
• 2016

Since he was an exile, Henry James himself was well aware of agonies as an outsider in either Europe or America. Such an anguish is deftly depicted in the character of Felix Young with James's unique ironic tone. Unlike James, however, Felix is neither affluent nor distinguished as an artist. Nor is he supported by any patron. Furthermore, at first, he doesn't seem to survive the strict joyless environment in New England, but he possesses his own survival value. His unique esthetic value and his beautiful smile enable him to win Gertrude's heart. His adroit balance between pleasure-seeker and respect for American serious culture without hostility ultimately ends up with his marrying Gertrude. His arrival in Boston might pose a threat as Mr. Wentworth fears. Actually he subverts the traditional idea of an artist. He is armed with amiability and frankness, which are incongruous with a stereotypical idea of an artist: a willful, freakish, and self-righteous person. Felix here suggests to us that a new kind of modern art be possible. Gertrude is also a new woman who opposes to staying put under the patriarchal society. She is always wavering in and out of the house, searching for opportunities to quench her curiosity to see the world by breaking the bond of New England. Her ceaseless quest for independent values results in fortuitous encounter with a new species of artist Felix. Unlike Henry James's other novels, in which male characters assume a role of sophisticated "fortune-hunter," the union of Felix and Gertrude in The Europeans represents the compromise between two different cultures. According to Nietzsche, the birth of superman is possible by the union of Athens and Jerusalem. In other words, the matrimony of Felix and Gertrude means the commingling of his liberal arts and Gertrude's moral seriousness might contribute to the birth of the new culture.

• International journal of advanced smart convergence
• /
• 제10권1호
• /
• pp.1-11
• /
• 2021

Document page segmentation is an important step in building a quality optical character recognition module. The study examined already existing work on the topic of page segmentation and focused on the development of a segmentation model that has greater functional significance for application in an organization, as well as broad capabilities for managing the quality of the model. The main problems of document segmentation were highlighted, which include a complex background of intersecting objects. As classes for detection, not only classic text, table and figure were selected, but also additional types, such as signature, logo and table without borders (or with partially missing borders). This made it possible to pose a non-trivial task of detecting non-standard document elements. The authors compared existing neural network architectures for object detection based on published research data. The most suitable architecture was RetinaNet. To ensure the possibility of quality control of the model, a method based on neural network modeling using the RetinaNet architecture is proposed. During the study, several models were built, the quality of which was assessed on the test sample using the Mean average Precision metric. The best result among the constructed algorithms was shown by a model that includes four neural networks: the focus of the first neural network on detecting tables and tables without borders, the second - seals and signatures, the third - pictures and logos, and the fourth - text. As a result of the analysis, it was revealed that the approach based on four neural networks showed the best results in accordance with the objectives of the study on the test sample in the context of most classes of detection. The method proposed in the article can be used to recognize other objects. A promising direction in which the analysis can be continued is the segmentation of tables; the areas of the table that differ in function will act as classes: heading, cell with a name, cell with data, empty cell.