뉴로모픽 구조 기반 FPGA 임베디드 보드에서 이미지 분류 성능 향상을 위한 특징 표현 방법 연구

정재혁,정진만,윤영선 한국소프트웨어감정평가학회 2021 한국소프트웨어감정평가학회 논문지 Vol.17 No.2

Neuromorphic architecture is drawing attention as a next-generation computing that supports artificial intelligence technology with low energy. However, FPGA embedded boards based on Neuromorphic architecturehave limited resources due to size and power. In this paper, we compared and evaluated the image reduction method using the interpolation method that rescales the size without considering the feature points and the DCT (Discrete Cosine Transform) method that preserves the feature points as much as possible based on energy. The scaled images were compared and analyzed for accuracy through CNN (Convolutional Neural Networks) in a PC environment and in the Nengo framework of an FPGA embedded board.. As a result of the experiment, DCT based classification showed about 1.9% higher performance than that of interpolation representation in both CNN and FPGA nengo environments. Based on the experimental results, when the DCT method is used in a limited resource environment such as an embedded board, a lot of resources are allocated to the expression of neurons used for classification, and the recognition rate is expected to increase.

IoT 응용을 지원하기 위한 뉴로모픽 하드웨어 기반 지능형 컴포넌트 설계

정재혁,김서연,정진만,윤영선 한국정보과학회 2021 한국정보과학회 학술발표논문집 Vol.2021 No.6

경계조건 분석을 통한 LCCBS 연계방안

정재혁,김태희 한국건축시공학회 2013 한국건축시공학회지 Vol.13 No.4

Costs and expenses are intertwined and incurred throughout an entire construction project, even from the pre-construction phase, and each phase has a different impact on the life cycle cost (LCC). However, the cost breakdown structure (CBS) is different in each phase of a building construction project, which makes it hard to reasonably calculate construction cost. For this reason, the boundary conditions were analyzed in this study based on the life cycle cost break structure (LCCBS). In addition, breakdown factors were analyzed based on the boundary conditions to derive a linkage method. The validity of the linkage method was verified through application to actual construction projects. Through the analysis, it was found that the problem of items being left out was reduced by more than 97.2 percent, and the work was done an average of 6 hours faster compared to the conventional method. It is expected that by applying the new LCC system, LCC will be both reduced and calculated in a more efficient manner 건축물의 생애주기비용은 시행과정 전 단계에서 상호 연관되어 비용이 발생하며 각 단계마다 서로 간에 영향력이 존재한다. 하지만 건축 프로젝트의 각 단계별 CBS가 서로 상이하여 합리적인 공사비를 산정하는데 문제점이 발생한다. 따라서 본 연구에서는 기존에 연구되었던 LCCBS를바탕으로 건축 프로젝트 각 단계 간에 경계조건을 분석하였다. 또한 분석된 경계조건을 바탕으로 분절요인을 분석하고 이에 따른 연계방안을 모색하였다. 연계방안의 유효성 검증을 위해 실제 건축프로젝트를 바탕으로 사례적용을 실시하였다. 분석결과 누락되는 항목이 97.2%이상을개선하였고 평균 6시간 빠르게 업무를 수행할 수 있음을확인할 수 있었다. 향후, LCC 시스템에 적용함으로써LCC의 효율적인 산정 및 LCC 절감을 계획할 수 있을 것으로 사료다.

Real-time PCR을 이용한 원유시료 유래 황색포도상구균의 신속 검출

정재혁,정순영,이상진,최성숙 한국식품위생안전성학회 2008 한국식품위생안전성학회지 Vol.23 No.2

The aim of this study was to develop a LightCycler-based real time PCR (LC-PCR) assay and toevaluate its diagnostic use for the detection of Staphylococcus aureus in raw milk samples. Folowing amplificationof 113 bp of coa gene encoding an coagulase precursor specific for Staphylococcus aureus, melting curve and DNAsequencing analysis was performed to verify the specificity of the PCR products. Amplification of 209 bp gene encod-mecA), melting curve analysis and DNA sequencing analysis wasperformed to verify methicillin resistance Staphyloccus aureus (MRSA). According to this study, 6 of 647 raw milksamples showed S. aureus positive and 2 of them showed a mecA positive and the detection limit was 10 fg of DNA.And we also isolated Staphylococcus chromogenes a causative agent of exudative epidermitis in pigs and cattle from3 samples.

머신러닝 기법을 활용한 고혈압 환자의 건강 관련 삶의 질 요인 예측

정재혁,조성현 대한통합의학회 2024 대한통합의학회지 Vol.12 No.3

Purpose : This study aims to identify the factors influencing health-related quality of life through machine learning of the general characteristics of patients with hypertension and to provide a basis for related research on patients, such as intervention strategies and management guidelines in the field of physical therapy for health promotion. Methods : Annual data from the second Korean Health Panel (Version 2.0) from 2019 to 2020, conducted jointly by the Korea Health and Social Research Institute and the National Health Insurance Service, were analyzed (Korea Health Panel, 2024). The data used in this study was collected from January to July 2020, and the data was collected using computer-assisted face-to-face interviews. Of the 13,530 household members surveyed, 1,368 were selected as the final study participants after removing missing values from 3,448 individuals diagnosed with hypertension by a doctor. Results : The results showed that walking (P2) was the most significant factor affecting health-related quality of life in random forest, followed by perceived stress (HS1), body mass index (BMIc), total household income (TOTc), subjective health status (SRHc), marital status (Marr), and education level (Edu). Conclusion :To prevent and manage chronic diseases such as hypertension, as well as to provide customized interventions for patients in advanced stages of the disease, research should be conducted in the field of physical therapy to identify influencing factors using machine learning. Based on the findings of this study, we believe that there is a need for additional content that can be utilized in the field of physical therapy to improve the health-related quality of life of patients with hypertension, such as diagnostic assessment and intervention management guidelines for hypertension, and education on perceived stress and subjective health status.

RGB 비디오 데이터를 이용한 Slowfast 모델 기반 이상 행동 인식 최적화

정재혁,김민석 한국멀티미디어학회 2022 멀티미디어학회논문지 Vol.25 No.8

HAR(Human Action Recognition) such as anomaly and object detection has become a trend in research field(s) that focus on utilizing Artificial Intelligence (AI) methods to analyze patterns of human action in crime-ridden area(s), media services, and industrial facilities. Especially, in real-time system(s) using video streaming data, HAR has become a more important AI-based research field in application development and many different research fields using HAR have currently been developed and improved. In this paper, we propose and analyze a deep-learning-based HAR that provides more efficient scheme(s) using an intelligent AI models, such system can be applied to media services using RGB video streaming data usage without feature extraction pre-processing. For the method, we adopt Slowfast based on the Deep Neural Network(DNN) model under an open dataset(HMDB-51 or UCF- 101) for improvement in prediction accuracy.

Expulsion of Fibroids to the Endometrial Cavity after Magnetic Resonance Imaging-guided High Intensity Focused Ultrasound Surgery (MRgFUS) Treatment of Intramural Uterine Fibroids

정재혁,홍길표,김유리,홍다교,하재은,염정인,김은정,김형일,이규섭 대한폐경학회 2016 대한폐경학회지 Vol.22 No.3

Objectives: This report seeks to introduce some cases of the patients who received magnetic resonance imaging (MRI)-guided high intensity focused ultrasound (HIFU) surgery (MRgFUS)-based intramural uterine fibroids treatment where the post-MRgFUS intramural uterine fibroids decreased in its volume and protruded towards the endometrial cavity to be expelled by hysteroscopy. Methods: Of the 157 patients who had received MRgFUS treatment in the Obstetrics and Gynecology of the Hospital from March, 2015 to February, 2016; this study examined 6 of the cases where, after high intensity focused ultrasound treatment, intramural uterine fibroids protruded towards the endometrial cavity to be removed by hysteroscopic myomectomy. The high intensity focused ultrasound utilized in the cases were Philips Achieva 1.5 Tesla MR (Philips Healthcare, Best, The Netherlands) and Sonalleve HIFU system.Results: The volume of fibroids ranged from 26.0 cm3 to 199.5 cm3, averaging 95.6 cm3. The major axis length ranged from 4.0 cm to 8.2 cm, averaging 6.3 cm. Fibroid location in all of the patients was in intramural uterine before treatment but after the high intensity focused ultrasound treatment, the fibroids were observed to protrude towards the endometrial cavity in at least Day 5 or up to Day 73 to allow hysteroscopic myomectomy.Conclusions: In some cases, after an intramural uterine fibroid is treated with MRgFUS,, fibroid volume is decreased and the fibroid protrudes towards the endometrial cavity. In this case, hysteroscopic myomectomy can be a useful solution. (J Menopausal Med 2016;22:139-145)

United Arab Emirates 사막환경에서 벼 재배를 위한 재배기간, 유전자원 및 수량 특성 연구

정재혁,황운하,이현석,양서영,최명구,김준환,김재현,정강호,이수환,오양열,이광승,서정필,정기열,이재수,최인찬,유승화,최순군,이슬비,이은진,이충근,이충근 한국농림기상학회 2022 한국농림기상학회지 Vol.24 No.3

본 연구는 UAE 사막환경에서 벼 재배를 위해 재배기간, 적정 벼 유전자원, 생육발달 양상, 물 사용량 등을 조사하고자 수행하였다. UAE 사막환경에서 벼의 재배기간은 UAE 겨울의 저온을 벼의 영양생장기간 동안에 포함하는 11월 하순부터 이듬해 4월 하순까지이었다. UAE의 기온과 일장에 적응할 수 있는 유전자원을 국내의 인공기상 시설과 간척지 모래토양에서 사전 시험한 결과, 아세미와 FL478이 선정되었다. 사막 환경에서 아세미는 생육초기의 황화현상을 극복하여 수확까지 하였으나, FL478은 생육 초기에 황화현상과 생육이 불량하여 고사하였다. UAE 사막환경에서 아세미는 12월 초순부터 이듬해 3월 초순까지 영양생장기, 3월 초순부터 3월 하순까지 생식생장기, 3월 하순부터 4월 하순까지 등숙기의 분포를 보였다. 아세미의 백미 수량은 763kg/10a이었고, 한국과 비교하여 약 41.8% 증가한 수량으로 생식생장기와 등숙기의 풍부한 일사량의 영향으로 추정된다. 물 사용량은 UAE 재배기간 동안 2,619 ton/10a 수준으로 한국보다 약 3배 많아 물 절약 기술이 필요한 상황이다. UAE에서 벼 재배는 경제성과 관련이 매우 높기 때문에, 물 절약을 위해서 관개기술, 재배방법 개발 등이 보완되어야 한다. 또한 UAE 사막환경에서 안정적인 벼 재배를 위해서 다양한 적응 유전자원 선정, 생육초기의 황화현상 최소화 방법 등의 추가 연구가 필요할 것으로 판단된다.

염농도에 따른 갯질경(Limonium tetragonum) 종자의 발아와 식물체의 생장 및 항산화 활성

정재혁,황운하,안승현,정한용,이현석,백정선,최경진,이건휘,나지은,정남진,이승재,윤성중 한국자원식물학회 2017 한국자원식물학회지 Vol.30 No.4

Limonium tetragonum is a halophyte grown naturally in the coastal region in South Korea. This study was conducted to investigate the effects of salt concentrations on seed germination, seedling growth, and antioxidant capacity of L. tetragonum. Seeds were collected from naturally grown plants of L. tetragonum and those at full maturity were used in this experiment. All experiments were performed at 0%, 0.5%, 1.0%, or 2.0% of salt concentrations. Seed germination rate was highest as 86% at 20℃ and followed as higher in order of 25℃, 30℃ and 15℃. The germination rate was about 80% at 0% or 0.5% of salt concentration, but it was very low at the salt concentrations higher than 1%. Growth of L. tetragonum seedlings showed no difference in Hoagland solution containing NaCl in the range of 0% to 1.0% and seedlings survived at 2.0% of NaCl concentration. As the salt concentration increased, the content of Na+ in the shoot increased, but that of K+, Ca++, or Mg++ decreased. The antioxidant activity and the content of total polyphenol and total flavonoid in the shoot were similar at 0% and 0.5% of NaCl and were highest at 2.0% of NaCl concentration. In conclusion, performance of seed germination and plant growth of L. tetragonum was highest at 0% and 0.5% of NaCl concentration, and showed no difference in antioxidant activity, total polyphenol contents, and total flavonoid contents at the same salt concentrations. 본 연구는 염생식물인 갯질경의 염농도에 따른 종자 발아, 식물체의 생장 및 식물체 내 항산화 활성, 총 폴리페놀 함량, 총플라보노이드 함량을 조사하고자 수행하였다. 갯질경 종자는20℃에서 발아율이 86%로 가장 높았고, 이어서 25℃, 30℃, 1 5℃ 순서로 발아율이 높았다. 20℃에서 갯질경 종자의 발아율은 염농도 0%와 0.5%에서 80% 이상으로 높았지만, 염농도1.0% 이상에서는 발아율이 매우 낮았다. 갯질경 식물체의 생장은 염농도 0%에서 1.0%까지는 차이가 없었고, 염농도 2%에서는 염농도 0%와 비교하여 60% 수준의 건물중량을 보였다. 식물체 내 무기성분 함량은 염농도가 높을수록 Na+ 함량은 증가하였고, K+, Ca++, Mg++ 함량은 감소하였다. 항산화 활성 및 총 폴리페놀과 총 플라보노이드 함량은 염농도 0%와 0.5%에서 비슷하였고, 염농도 2%에서 가장 높은 경향을 보였다. 따라서 염농도0.5% 이하의 토양에서 갯질경을 재배하면 안정적인 발아와 생장이 가능하고, 항산화 활성 및 총 폴리페놀과 총 플라보노이드함량의 변화가 적기 때문에 균일한 품질의 항산화 관련 재료를얻는데 유용할 것으로 판단된다.

腦卒中患者의 血壓

정재혁,김유국,이판규,민병일 최신의학사 1983 最新醫學 Vol.26 No.3