http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
중학교 우수 사이클 선수의 성별에 따른 신체구성, 최대근력 및 무산소성 파워의 비교분석
박지희 ( Park Jihee ),김영표 ( Kim Young-pyo ),서태범 ( Seo Tae-beom ) 제주대학교 체육진흥센터(구 제주대학교 체육과학연구소) 2019 체육과학연구 Vol.25 No.-
The purpose of this study is to investigate body composition, maximal muscle strength and anaerobic power in middle school elite cyclists. The experiment’s subjects were middle school male and female athletes (n=11) selected from 2016 korea cycling competition medal winner. The subjects were categorized into male (n=6) and female athlete (n=5) groups. The analysis factors were body composition (weight, FFM, %Fat, BMI), maximal muscle strength (squat, bench press, grip strength), and anaerobic power (peak power, mean power, fatigue index). In order to analyze collected data, independent t-test analysis was utilized. The results are as follows. First, in the body composition, FFM, %Fat, and BMI showed a significant differences between male and female groups (p<.005). Second, squat, bench press, and grip strength showed a significant differences between two groups (p<.001). Third, in the wingate test peak power, and mean power showed a significant difference between male and female groups (p<.001). Therefore, our findings suggest that male athletes in middle school may show higher pedaling power than female athletes, through increase of muscle mass and maximal muscle strength.
요약된 주소값 기반 메모리 분석 기법 및 이를 이용한 EVM 바이트코드 분석
박지희(Jihee Park),류석영(Sukyoung Ryu) 한국정보과학회 2020 정보과학회 컴퓨팅의 실제 논문지 Vol.26 No.4
암호화폐와 스마트 컨트랙트에 대한 관심이 커지면서 스마트 컨트랙트의 분석이 활발히 이루어지고 있다. 특히 이더리움을 기반으로 한 이더리움 가상머신(EVM) 위에서 실행되는 프로그램의 안전성을 증명하고자 하는 연구들이 진행되고 있는데, EVM 바이트코드로 컴파일되는 고급 언어들의 독특한 메모리 할당 구조는 포인터 분석을 어렵게 만드는 요인 중 하나이다. 이 연구에서는 고급 언어들의 메모리 모델이 어떻게 EVM 바이트코드로 컴파일되는지에 대해서 알아보고, 문법적으로 메모리 할당을 지원하지 않아 선형 메모리 모델에 포인터로 접근하는 프로그램을 분석하기 위해 요약된 주소값을 기반으로 하는 분석을 정의하였다. 본 연구를 통해 EVM 바이트코드에서 연구되어온 분석보다 더 정밀한 메모리 분석을 수행할 수 있음을 보인다. As interest in cryptocurrency and smart contracts grow, many studies are actively analyzing smart contracts. Especially, there are many studies about validating the safety of programs based on the Ethereum virtual machine, is one of the famous cryptocurrency frameworks. However, it is challenging to analyze memory contents of the Ethereum bytecode because of the complex behavior of memory allocation. In this study, we provide a summary of how memory models in high-level programming languages are compiled into EVM bytecodes, and propose a memory abstraction technique that uses abstract values as indices of abstract memory. It enables analyses of linear memory models with pointers which is difficult to analyze because they do not support memory allocation primitive. Through this study, we show that we can perform a more precise memory analysis than the analysis studied in the EVM bytecode.
박지희(Jihee Park),심재창(Jaechang Shim) 한국정보기술학회 2021 한국정보기술학회논문지 Vol.19 No.5
Recently, wild animals such as wild boars have been causing severe crop damage not only in urban areas but also in rural areas. Therefore, farmers have installed wild animal repelling devices on farmland to drive out the animals with LED lights and sounds. Existing wild animal repelling devices mostly rely on PIR and Doppler sensors. So, even when wild animals do not appear, repelling action is performed. In this paper, we propose a method that utilizes deep learning-based intelligent wild boar repellent device to implement repelling operation after determining whether it is wild boar with trained YOLOv4 model via camera. Existing wild animal repellent devices perform repelling action to both human and wild boar when human and wild boar appear. Deep learning-based intelligent wild boar repellent device repels only wild boar when human and wild boar appear. By installing deep learning-based intelligent wild boar repellent device, we can prevent unnecessary repelling action and more effectively repel the wild animals.
박지희 ( Jihee Park ),우지민 ( Jimin Woo ),조민지 ( Minji Cho ),김명주 ( Myuhng-joo Kim ) 한국정보처리학회 2020 한국정보처리학회 학술대회논문집 Vol.27 No.1
정보화시대로 넘어오면서 코딩 역시 중요한 교육으로 자리 잡고 있다. 초등학교는 물론 중, 고등학교도 교육적인 측면에서 프로그래밍을 중요시하고 있으며 C언어나 Java는 더이상 개발자만을 위한 언어로 인지되지 않는다. 하지만 코딩의 결과만 의존할 경우 정보보호를 하는 시큐어 코딩의 요소를 중요하게 여기지 않을 수 있다. 그 결과 DB는 물론 서버 에러를 비롯해 개인정보는 많은 위험에 노출될 수 있다. 기본적인 프로그래밍을 숙지한 개발자를 위해 더욱 안전한 코딩을 교육하고자 시큐어 C 코딩 앱을 기획하게 되었고, 정보보호를 코딩의 중요한 요소로 인지하기를 기대한다.
사이클 경기(Track, MTB, BMX) 종목에 따른 등속성 근력과 근지구력 및 무산소성 파워 특성비교
박지희 ( Park Jihee ),김영표 ( Kim Young-pyo ),서태범 ( Seo Tae-beom ) 제주대학교 체육진흥센터(구 제주대학교 체육과학연구소) 2020 체육과학연구 Vol.26 No.-
The purpose of this study is to investigate the characteristics of physical fitness among track, BMX and MTB athletes to improve their performance. The subjects were categorized into three groups; National track cyclist (n=3), National BMX athletes (n=3) and National MTB athletes (n=3) groups. The analysis factors were body composition (weight, FFM, %Fat, BMI), isokinetic knee muscle strength (absolute and relative strength values), and anaerobic power (peak power, mean power, fatigue index). In order to analyze collected data, statistical analysis was performed using one-way repeated ANOVA followed Tukey post hoc test. As results of this study, FFM and %Fat is not significant differences, but BMI showed a significant differences among National track, BMX and MTB groups. Isokinetic knee extension and flexion strength was higher in Track group compared to other groups as well as isokinetic knee muscle endurance were the highest in Track group. In the wingate test, peak power (PP) showed a significant difference among three groups. Therefore, our findings suggest that isokinetic knee strength and anaerobic pedaling power might be important factors to improve the performance of BMX and MTB athletes.