http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
자생 상록 굴거리나무의 실내조명 하에서 적정광도 분석 및 도입방안
신현철,윤재길,최경옥,Shin, Hyeon-Cheol,Yun, Jae-Gill,Choi, Kyoung-Ok 한국조경학회 2012 한국조경학회지 Vol.40 No.1
This study was performed to determine light-adaptability under indoor lighting for Korean wild Daphniphyllum macropodum, and to provide an introduction plan for indoor spaces of landscape architecture. The experimental plants, Daphniphyllum macropodum were purchased from a farmhouse in Jejudo as anannualplant, and it was acclimatized in the glass-green house of the general farm of Gyeongnam National University of Science and Technology for40 days. The experiment was performed from February 15, 2010 to November 15, 2010. First, in the case of the fluorescent lamp, the growth and development condition was poor because the plants withered, or the leaves fell off under the 100lux to 500lux, but the condition under 1,000lux was good In the aspects of the number of leaves, form of the tree, photosynthesis rate in its body, and the value of sight, the best light intensity for the growth condition was under the 1,000lux. Second, in the case of the LED light, the growth and development condition was poor because the plants withered, or the leaves fell off under 100lux to 1,000lux, but the smooth growth and development was done under a more light intensity. The best intensity for the growth condition was under 2,000lux. Third, in the case of the three-wave light, the withering was serious by 1,000lux, and the growth and development was the worst amongst the four introduced lighting systems, therefore, growth under the three-wave light was incongruous. The best intensity for the growth condition was under 2,000lux. Fourth, in the case of the optical fiver, the withering did not existed under 100lux and growth was possible. The growth and development was the amongst in the four introduced lighting systems. Generally, in the more light intensity, more growth was observed, but the value of sight was higher under 1,000lux than under 2,000lux because the falling rate of leaves and the form of the tree was stable. The most effective light was under 1,000lux due to the high photosynthesis in its body. When Daphniphyllum macropodum is introduced into the indoor landscaping space considering the light, the optical fiber, fluorescent lamp, and LED light are suitable to introduce. The three-wave light is unfit. The most proper light intensity by the light source, in the case of the optical fiber, and fluorescent lamp, is 1,000lux, and, in the case of LED light, is 2,000lux. The wild Daphniphyllum macropodum is the species of tree to substitute the Schefflera actinophylla which is the introduced species, and it is expected to be use as the central tree in indoor spaces.
해상 영상에서 관심영역 추출과 합성곱 신경망을 활용한 선박 분류
신현철(Hyeon-Cheol Shin),이광일(Kwang-il Lee) 한국지능시스템학회 2019 한국지능시스템학회논문지 Vol.29 No.4
최근에 해양에서는 선박 스스로 주변 상황을 인지하고 운항할 수 있는 자율운항 기술개발이 활발하게 이루어지고 있다. 이를 위해, 카메라를 통한 영상정보를 활용하여 인간의 시각 정보를 대신할 수 있는 기술에 대한 중요성이 대두되고 있다. 카메라 영상을 기반으로 한 상황인지 기술을 위해서는 영상에서 존재하는 다양한 객체 정보를 분석하는 객체 분류 기술이 필수적이다. 본 논문에서는 해양 영상에서 관심 영역 추출과 합성곱 신경망을 활용하여 선박 분류 정확도를 향상시킬 수 있는 방법을 제안한다. 본 논문에서 제안된 방식의 성능을 검증하기 위해 공개 데이터 셋을 이용하여 기존의 합성곱 신경망 기반 방법과의 비교 실험을 수행하였으며, 실험을 통해 본 논문에서 제안된 방식이 선박 분류 정확도를 향상시킬 수 있음을 확인하였다. 제안된 방법을 활용하여 자율운항선박에서는 다른 해상 장비와의 센서 퓨전을 통하여 객체 데이터의 신뢰성을 확보할 수 있으며, 이를 통해 충돌 회피 및 안전한 운항이 가능할 것으로 기대된다. Recently, autonomous navigation technology has been actively developed in order to recognize and operate the vessel itself. For this purpose, importance is attached to technologies that can substitute human visual information by utilizing image information through a camera. For the context recognition based on the camera image, object classification technology for analyzing various object information existing in the image is essential. In this paper, we propose a method to improve the accuracy of ship classification by using region of interest and artificial neural network. In order to verify the performance of the propose method in this paper, we performed a comparative experiment with the convolution artificial neural network based on the open data set. Experimental results show that the proposed method improves the accuracy of vessel classification. By using the proposed method, it is possible to secure the reliability of object data through sensor fusion with autonomous vessels, and it is expected that collision avoidance and safe operation will be possible.
종설(綜說) : 재발성(再發性) 방광염(膀胱炎) 및 하부뇨로감염(下部尿路感炎)의 침술(針術) 치료(治療)에 관한 고찰
신현철 ( Hyeon Cheol Shin ) 대구한의대학교 제한동의학술원 2010 東西醫學 Vol.35 No.4
Objectives: Frequently recurrent acute uncomplicated lower urinary tract infection(cystitis), defined as acute cystitis in otherwise healthy adult women is common. Usually even though they use prophylactic antibacterial treatment, long-term antibacterial treatment can lead to unintended increased bacterial resistance, which is a medical problem of increasing concern. The main purpose of this study is to evaluate the effect of acupuncture in the treatment and prevention of recurrent cystitis and lower urinary tract infection. Methods and Materials: Writer ran a computer literature search in Pubmed. The search terms were 「cystitis acupuncture」 and 「lower urinary tract infection acupuncture」, and selected 5 literatures. Results Recurrent cystitis and lower urinary tract infection frequently occurred in the middle generation women in the latter half of their thirties. In Zang-Fu they were connected with Spleen, Kidney and Liver. In the frequency of use, Sanyinjiao and P`angkuangshu in the Jing-Points and Spleen-Channel and Urinary Bladder-Channel in the Meridians were most. The difference of the incidence rate between acupuncture treated group and nontreated group was more than twice. Especially in the nontreated group the residual urine was unchanged, but in the treated group reduced to a half at least. Conclusions: In the treatment and prevention of recurrent cystitis and lower urinary tract infection acupuncture-therapy seems to be effective, but in many ways it is likely that further research is need.