http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
鄭元溶(Wonyong Chung) 경성대학교 인문과학연구소 2003 인문학논총 Vol.8 No.-
This paper purports to investigate the Agentivity of English Middle Verb. It is generally accepted that English Middle Verbs have the following characteristics. First, they are non-eventive. They prefer present tense and denote generic events. As a result, the past tense with definite past moments and the progressive aspects are incompatible with them. Second, middles involve the modal notion of possibility. Third, another characteristic is that they require certain adverbials like easily, well, etc. Fourth, the grammatical subjects of middles are responsible for the action denoted by the verb. Fifth, middle constructions have implicit Agents. Sometimes, it is said they are explicit in the form of for-Phrase. The present paper is especially concerned with the fourth and fifth characteristics. The Agentivity of middle construction poses problems and various solutions. Among others, it is said that there's no such thing as movement from the first and it is also said that there's apparent agentivity and it comes form the inner object of the verb. Some recent researches suggest that it depends on the verbs whether they have agentivity or not. Present paper suggests another approach by investigating the verb "SELL" The suggestion is that middles and ergatives share so many similarities that the agentivity depends on the semantic and pragmatic effects, not by the Middle Verbness.
레이저 텍스처링을 이용한 점과선을 결합한 Pattern의 마찰특성
정우영(WooYoung Chung),김상호(SangHo Kim),한형욱(HyungWook Han),민준원(JoonWon Min),정원용(WonYong Jeung),이명호(MeungHo Rhee) 한국트라이볼로지학회 2010 한국트라이볼로지학회 학술대회 Vol.2010 No.6
Laser Surface Texturing(LST) is typical high speed micro forming process using the pulsed laser to form micro dimple shapes on friction area with high lubrication characteristics like e.g. piston ring. In view of new low friction methods, special surface processes like Laser Texturing gain importance for the associated field. For a better understanding of the working mechanisms and the process for the coupling of line patterning and dimple texturing models, advanced combination rules as well as detailed experimental investigations is required. New results of research concerning correlations among dimple and line depths, and lubricant conditions are presented.
지루성 두피염 진단을 위한 Faster R-CNN과 Atrous 컨볼루션 기반의 두피 각질 검출 기법
조성국(Seongkuk Cho),김형준(Hyungjoon Kim),정원용(Wonyong Chung),황인준(Eenjun Hwang) 한국정보과학회 2021 정보과학회 컴퓨팅의 실제 논문지 Vol.27 No.9
지루성 두피염은 과도한 스트레스나 화학제품의 남용, 불균형한 영양 섭취 등 다양한 요인으로 인해 발생하는 질병이다. 두피염의 대표적인 증상은 두피의 각질과 통증이며, 증상이 심해지면 탈모가 발생할 수 있다. 이에 각질 검출을 통해 두피염을 초기에 진단할 수 있는 스마트폰 기반의 애플리케이션이 다수 개발되었다. 하지만, 대부분은 수치로 된 분석 결과만을 제공할 뿐, 그 수치에 관한 근거를 확인하기 어렵다. 따라서, 본 논문에서는 지루성 두피염 진단을 위한 근거를 시각적으로 보여주기 위해 두피 영상에서 각질의 유무와 위치를 추출하는 기법을 제안한다. 이를 위해, 각질 검출에 Faster R-CNN (Faster Regions with Convolutional Neural Network) 모델을 사용하였으며, 모델의 성능을 더욱 향상하기 위해 Atrous 컨볼루션 연산을 적용하였다. 기존 컨볼루션과 Atrous 컨볼루션을 적용한 Faster R-CNN의 비교를 통해 우리가 제안하는 기법이 효과적으로 두피 각질을 검출할 수 있음을 보인다. Seborrheic scalpitis can be caused by various factors, such as excessive stress, abuse of chemicals, and unbalanced nutrition. Typical symptoms of this disease entity are keratin and pain over the scalp, and when the symptoms become severe, hair loss may occur. Although various smartphone applications have been developed for early detection of scalpitis, it is difficult to ascertain the rationale for the results because they only provide numerical analysis. Therefore, in this paper, we propose a method for detection of keratin from scalp images to visualize the basis for diagnosing seborrheic scalpitis. In particular, we use the Faster Regions with Convolutional Neural Network (Faster R-CNN) model for keratin detection and apply atrous convolution to further improve the detection performance. By comparing our proposed method to the original Faster R-CNN method, we show that our method can effectively diagnose the symptoms of seborrheic scalpitis.