http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
추천 다양화 방법을 적용한 콜드 아이템 추천 정확도 향상
한정규,천세진,Han, Jungkyu,Chun, Sejin 한국멀티미디어학회 2022 멀티미디어학회논문지 Vol.25 No.8
When recommending cold items that do not have user-item interactions to users, even we adopt state-of-the-arts algorithms, the predicted information of cold items tends to have lower accuracy compared to warm items which have enough user-item interactions. The lack of information makes for recommender systems to recommend monotonic items which have a few top popular contents matched to user preferences. As a result, under-diversified items have a negative impact on not only recommendation diversity but also on recommendation accuracy when recommending cold items. To address the problem, we adopt a diversification algorithm which tries to make distributions of accumulated contents embedding of the two items groups, recommended items and the items in the target user's already interacted items, similar. Evaluation on a real world data set CiteULike shows that the proposed method improves not only the diversity but also the accuracy of cold item recommendation.
객체인식 알고리즘 분석을 통한 YOLOv4를 이용한 웹캠 사물인식
한정규(Jung-Gyu Han),한정규(Jung-Gyu Han) 대한전기학회 2022 대한전기학회 학술대회 논문집 Vol.2022 No.11
With the recent development of Internet technology, as COVID-19 is prevalent, more and more people are facing each other online rather than directly talking to each other at work or meetings. In addition, sensor-based object recognition technologies such as front and rear object recognition used in automobile research used in daily life are being introduced in autonomous vehicles, and this shows a huge object recognition rate. However, it is very expensive to introduce sensor-based object recognition technology that shows excellent object recognition rate in everyday life, and mass production is also difficult, so it is very burdensome for students who explore or study basic object recognition. This paper analyzes various object recognition algorithms that can recognize objects through webcams mounted on desktops rather than expensive sensor-based object recognition technology, and implements a program that recognizes basic objects and objects based on YOLOv4-based object recognition algorithm.
걷기운동이 여성노인의 신체조성, 혈압 및 대뇌피질활성화에 미치는 영향
한정규(Joung Kyue Han),이현우(Hyun Woo Lee) 한국사회체육학회 2012 한국사회체육학회지 Vol.0 No.48
A total of 14 healthy elderly females participants with the age between 59~63 were enrolled in this study, who did not show any medical complication or take muscle and skeletal disease medicine and underwent walking exercise for 12 weeks (60~65%HRmax). Also the results of this study were as follow. Thus, the importance of the exercise related to aging is one of the most important health management aspects for elderly female patients. Hence, the purpose of this study is to analyze waking exercise for the changes of body composition, blood pressure and the activation of cerebral cortex in elderly female patients and search for the health management promotion and the method for elderly people. According to the analysis results of the change differences between pre-study and post-study, there showed no change in the pre-study, while there showed the changing aspects of cerebral cortex activation in Rt. Frontal lobe cortex, Lt. Frontal lobe cortex, Rt. Parietal lobe subcortical white matter and corpus callosum. The body weight was decreased from 57.92±2.92kg(pre-study), middle 55.42±2.71kg to 54.64±2.83kg(post-study), Body fat 29.72±2.48%(pre-study), middle 28.51± 3.45% to 26.38±4.10%(post-study), SBP was decreased from 119.71±5.82mmHg(pre-study), middle 116.85 ±4.28mmHg to 116.00±6.08mmHg(post-study)and which were statistically significant(p<.05). According to the analysis results of the change differences between cerebral cortex activation and anatomical position, there showed the changing aspects of brain activation regarding motor area and perception area after the walking exercise directly and indirectly-Rt. Frontal lobe cortex, Lt. Frontal lobe cortex, Rt. Parietal lobe subcortical white matter and corpus callosum.
모바일 멀티미디어 데이타를 위한, 의사난수생성기와 순열 기법을 결합한 효율적인 암호화 기법
한정규(JungKyu Han),조유근(YooKun Cho) 한국정보과학회 2007 정보과학회논문지 : 시스템 및 이론 Vol.34 No.11·12
In Digital Right Management, symmetric cipher is used for content encryption to reduce encryption cost, AES, advanced encryption standard is usually used to multimedia encryption under desktop environment because of its reasonable security level and computation cost. But mobile handheld device often uses slow speed processor and operates under battery-powered environment. Therefore it requires low computation cost and low energy consumption. This paper proposes new stream cipher scheme which combines pseudo random number generator(PRNG) and dynamically generated permutations. Proposed scheme activates PRNG and generates original key streams. Then it generates extended key streams by applying permutation to original sequence. These extended key streams are XORed with plaintext and generate ciphertext. Proposed scheme reduces the usage of PRNG. Therefore this scheme is fast and consumes less energy in comparison with normal stream cipher. Especially, this scheme shows great speed up (almost 2 times) than normal stream cipher scheme in random access. 디지털 저작권 관리 기법은 계산비용의 경감을 목적으로 컨텐츠 암호화를 위해 대칭 키 암호화 기법을 채택하였으며 데스크탑 환경에서는 강한 보안성과 적절한 암호화 속도를 가지는 AES를 주로 사용하고 있다. 그러나 낮은 성능의 프로세서와 제한된 전력환경에서 동작하는 모바일 기기에서는 더욱 낮은 계산 비용 실현과 에너지 소모 경감을 요구한다. 이에 본 논문에서는 모바일 기기에서 사용 가능한 효율적인 스트림 암호화 기법을 제안한다. 제안 기법은 의사 난수 생성기를 사용하여 원본 키 스트림을 생성한 다음 이에 동적 생성한 순열을 적용하여 확장 키 스트림을 생성한다. 확장 키 스트림을 평문과 논리합하여 암호문을 생성한다. 순열을 이용하여 키 스트림 생성에 사용되는 의사 난수 생성기의 사용 횟수를 줄였기 때문에 일반 스트림 암호화 기법에 비해 멀티미디어 데이타의 암/복호화 속도가 빠르며 에너지 소모를 줄였다. 특히 제안 기법은 멀티미디어 파일의 임의 접근 시 일반 스트림 암호화 기법에 비해 약 2배의 속도향상을 보인다.
볼링 동호인의 운동상해 실태와 예방 및 처치 유형 분석
한정규(Han, Joung-Kyue),김미정(Kim, Mi-Jeong) 한국체육과학회 2021 한국체육과학회지 Vol.30 No.2
The purpose of this study is to find out the types of injuries, parts, timing, and first aid measures according to gender of bowling enthusiasts. As a result, there were elbow, wrist, finger, and shoulder injuries in each body part injuries, and in the type of injury, skin injuries were abrasions, fissures, and blisters. In muscle injuries, myositis, muscle bruises, and muscle stiffness were found a lot, and in joint injuries, sprains and ligament kidneys were found a lot. The season of heavy injury is winter, and the time of injury is frequent at night and in the afternoon. The cause of the injury was excessive movement, excessive practice, and lack of warm-up exercise, and psychological factors were caused by excessive desire to win and excessive tension. The treatment places after injury were orthopedic clinics, oriental medicine clinics, and home treatments, and the treatments that showed the effect were physical therapy, acupuncture and moxibustion. Most of them maintained the same state as before after injury, and as a method to prevent injury, warm-up exercise, cleanup exercise, adequate rest, and expert guidance appeared.