http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Park, Hanwool,Yoo, Yechan,Park, Yoonjin,Lee, Changdae,Lee, Hakkyung,Kim, Injung,Yi, Kang Korean Institute of Information Scientists and Eng 2018 Journal of Computing Science and Engineering Vol.12 No.1
Deep convolutional neural network (DCNN) is an advanced technology in image recognition. Because of extreme computing resource requirements, DCNN implementation with software alone cannot achieve real-time requirement. Therefore, the need to implement DCNN accelerator hardware is increasing. In this paper, we present a field programmable gate array (FPGA)-based hardware accelerator design of DCNN targeting handwritten Hangul character recognition application. Also, we present design optimization techniques in SDAccel environments for searching the optimal FPGA design space. The techniques we used include memory access optimization and computing unit parallelism, and data conversion. We achieved about 11.19 ms recognition time per character with Xilinx FPGA accelerator. Our design optimization was performed with Xilinx HLS and SDAccel environment targeting Kintex XCKU115 FPGA from Xilinx. Our design outperforms CPU in terms of energy efficiency (the number of samples per unit energy) by 5.88 times, and GPGPU in terms of energy efficiency by 5 times. We expect the research results will be an alternative to GPGPU solution for real-time applications, especially in data centers or server farms where energy consumption is a critical problem.
Hanwool Park,Yechan Yoo,Yoonjin Park,Changdae Lee,Hakkyung Lee,Injung Kim,Kang Yi 한국정보과학회 2018 Journal of Computing Science and Engineering Vol.12 No.1
Deep convolutional neural network (DCNN) is an advanced technology in image recognition. Because of extreme computing resource requirements, DCNN implementation with software alone cannot achieve real-time requirement. Therefore, the need to implement DCNN accelerator hardware is increasing. In this paper, we present a field programmable gate array (FPGA)-based hardware accelerator design of DCNN targeting handwritten Hangul character recognition application. Also, we present design optimization techniques in SDAccel environments for searching the optimal FPGA design space. The techniques we used include memory access optimization and computing unit parallelism, and data conversion. We achieved about 11.19 ms recognition time per character with Xilinx FPGA accelerator. Our design optimization was performed with Xilinx HLS and SDAccel environment targeting Kintex XCKU115 FPGA from Xilinx. Our design outperforms CPU in terms of energy efficiency (the number of samples per unit energy) by 5.88 times, and GPGPU in terms of energy efficiency by 5 times. We expect the research results will be an alternative to GPGPU solution for real-time applications, especially in data centers or server farms where energy consumption is a critical problem.
신종 감염병(COVID-19) 환자 간호의 행위 신념에 영향을 미치는 요인
박윤진(Yoonjin Park),이선라(Sun Ra Lee) 한국산학기술학회 2021 한국산학기술학회논문지 Vol.22 No.2
본 연구는 신종 감염병 확산과 관련한 간호사의 외상 후 스트레스와 불안 정도를 파악하고 간호의도에 영향을 미치는 간호사의 행위 신념과 통제 신념과의 관계를 분석하여 이들이 행위 신념에 미치는 영향을 파악하고자 한 서술적 조사연구이다. 본 연구의 대상자는 경기도에 소재한 의료기관에서 COVID-19환자를 직접 간호한 경험이 있는 간호사를 대상으로 하였다. 본 연구의 통계방법은 SPSS 22.0을 사용하여 SPSS/WIN 20.0 프로그램을 이용하여 기술통계, Independent t-test, one-way ANOVA, Pearson"s Correlation Coefficient, Hierarchical Multiple Regression 으로 분석할 예정이다. 자료는 SPSS/WIN 20.0 프로그램을 이용하여 분석하였다. 대상자의 특성과 외상 후 스트레스, 불안, 행위 신념, 통제 신념을 대상자의 특성에 따른 외상 후 스트레스, 불안, 행위 신념, 통제 신념의 차이는 t-test와 ANOVA로 파악하였으며, 제 변수 간의 상관관계는 Pearson’s correlation coefficients로 확인하였다. 행위신념에 영향을 미치는 요인은 다중회귀분석을 사용하였다. 본 연구에 참여한 대상자의 외상 후 스트레스는 평균 24.20±20.58, 불안은 48.31±6.61, 행위 신념은 평균 -1.00±17.12, 통제 신념은 3.41±11.66으로 조사되었다. 본 연구결과에 따르면 행위 신념과는 외상 후 스트레스와 음의 상관관계를 나타냈으며(r=-4.71, p<.001), 불안과도 유의미한 음의 상관관계를 나타냈다(r=-2.248, p<.05). 통제 신념과는 유의미한 상관관계가 나타나지 않았다(p>.05). 본 연구의 결과를 토대로 간호사의 행위 신념을 증진시키기 위한 적절한 심리적 중재 프로그램 개발을 제언한다. This is an investigative study to identify the degree of post-traumatic stress and anxiety of nurses in relation to the COVID-19 pandemic, and to analyze the relationship with the nurse"s belief in behavior and control that affects nursing intentions to determine their impact on the belief in behavior. The subjects of this study were nurses with experience of directly nursing COVID-19 patients at medical institutions located in Gyeonggi-do. The SPSS 22.0 version was used for the descriptive analysis, independent t-test, one-way ANOVA, Pearson"s correlation coefficients and multiple regression. Post-traumatic stress of the participants in this study was found to be 24.20±20.58, anxiety 48.31±6.61, behavior beliefs -1.00±17.12, and control beliefs 3.41±11.66. According to this study, the belief in conduct is negatively correlated with post-traumatic stress (r=-4.71, p<).001) and showed a significant negative correlation with anxiety (r=-2.248, p<.05). There was no significant correlation with control beliefs (p>.05). Based on the results of this study, it is proposed to develop an appropriate psychological arbitration program for mediating post-traumatic stress in order to promote the nurse"s behavior beliefs.
Development of a Botox-like recombinant neurotoxin miniature
Hye Rin KIM,Choongjin BAN,Joon-Bum PARK,Sora CHO,Yong Joon KIM,Jonghyeok SHIN,Myungseo PARK,Younghun JUNG,Seokoh MOON,Yunjeong PARK,Hooyeon KIM,Hyun Seok OH,Jinkyeong YANG,Jaehyeon HWANG,Yoonjin BAE,D 한국생물공학회 2019 한국생물공학회 학술대회 Vol.2019 No.4
Low Power Reconfiguration Technique for Coarse-Grained Reconfigurable Architecture
Yoonjin Kim,Mahapatra, R.N.,Ilhyun Park,Kiyoung Choi IEEE 2009 IEEE transactions on very large scale integration Vol.17 No.5
<P>Coarse-grained reconfigurable architectures (CGRAs) require many processing elements (PEs) and a configuration memory unit (configuration cache) for reconfiguration of its PE array. Although this structure is meant for high performance and flexibility, it consumes significant power. Specially, power consumption by configuration cache is explicit overhead compared to other types of intellectual property (IP) cores. Reducing power is very crucial for CGRA to be more competitive and reliable processing core in embedded systems. In this paper, we propose a reusable context pipelining (RCP) architecture to reduce power-overhead caused by reconfiguration. It shows that the power reduction can be achieved by using the characteristics of loop pipelining, which is a multiple instruction stream, multiple data stream (MIMD)-style execution model. RCP efficiently reduces power consumption in configuration cache without performance degradation. Experimental results show that the proposed approach saves much power even with reduced configuration cache size. Power reduction ratio in the configuration cache and the entire architecture are up to 86.33% and 37.19%, respectively, compared to the base architecture.</P>