• 방송공학회논문지
• /
• 제23권4호
• /
• pp.526-535
• /
• 2018

많은 스테가노그래피 알고리즘들이 연구되어왔고 스테가노그래피의 연구로 인해서 스테가노그래피 알고리즘이 적용된 스테고 영상을 검출하기 위한 스테그아날리시스 또한 연구되어왔다. 특히 영상 스테그아날리시스의 경우에 ALE, SPAM, SRMQ와 같은 통계적인 특성에 기반한 수제 특징이 영상으로부터 추출되고 이 특징을 기계학습 알고리즘을 사용하여 스테고 영상의 분류에 사용하였다. 하지만 이러한 연구들은 단지 단일 영상의 크기, 비밀 정보 삽입 비율에 대해서만 고려하였으며 다양한 영상의 크기, 가로세로 비, 비밀 정보의 비율들에 대해서는 고려하지 않았다. 결과적으로 SPAM 특징이 다양한 조건하에서도 범용성있게 사용가능한지에 대한 검증은 이루어지지 않았다. 본 논문에서는 SPAM 기반의 영상 스테그아날리시스를 사용하여 영상의 크기, 가로 세로 비 비밀 정보 삽입 비율의 다양한 조건에 대한 분류율을 분석하여 SPAM 특징이 일관적인 분류 결과를 보여줄 수 있는지 확인해본다.

• 인터넷정보학회논문지
• /
• 제21권3호
• /
• pp.113-121
• /
• 2020

Human Pose Estimation (HPE) which localizes the human body joints becomes a high potential for high-level applications in the field of computer vision. The main challenges of HPE in real-time are occlusion, illumination change and diversity of pose appearance. The single RGB image is fed into HPE framework in order to reduce the computation cost by using depth-independent device such as a common camera, webcam, or phone cam. However, HPE based on the single RGB is not able to solve the above challenges due to inherent characteristics of color or texture. On the other hand, depth information which is fed into HPE framework and detects the human body parts in 3D coordinates can be usefully used to solve the above challenges. However, the depth information-based HPE requires the depth-dependent device which has space constraint and is cost consuming. Especially, the result of depth information-based HPE is less reliable due to the requirement of pose initialization and less stabilization of frame tracking. Therefore, this paper proposes a new method of HPE which is robust in estimating self-occlusion. There are many human parts which can be occluded by other body parts. However, this paper focuses only on head self-occlusion. The new method is a combination of the RGB image-based HPE framework and the depth information-based HPE framework. We evaluated the performance of the proposed method by COCO Object Keypoint Similarity library. By taking an advantage of RGB image-based HPE method and depth information-based HPE method, our HPE method based on RGB-D achieved the mAP of 0.903 and mAR of 0.938. It proved that our method outperforms the RGB-based HPE and the depth-based HPE.

• 방송공학회논문지
• /
• 제18권3호
• /
• pp.401-408
• /
• 2013

본 논문에서는 영상 내의 문맥 특징(context feature)과 외형 특징(appearance feature)을 함께 학습함으로써 의료영상 내의 비슷한 외형 특징을 가지는 장기들을 자동으로 검출하는 기법을 제안한다. 기존 검출 기법들은 외형 특징 정보만을 학습하여 분류기(classifier)를 생성하였기 때문에 의료영상 내에 외형이 비슷한 장기들이 다수 포함되어 있는 경우 검출 오류가 발생하였다. 제안하는 기법은 외형 특징을 이용하여 학습된 분류기를 통해 얻은 확률 값들을 바탕으로 관심 복셀(voxel) 주변의 확률 분포 특징을 반복적으로 학습함으로써 문맥 정보를 포함하는 분류기를 생성한다. 또한, 실험 단계(test stage)에서 '지역 기반 투표 방식'(region based voting scheme)을 도입함으로써 효율성과 정확성을 향상시킨다. 제안하는 기법의 성능 평가를 위해 SKI10 무릎 관절 데이터 셋 내에서 외형 특징이 비슷한 대퇴골(femur)과 경골(tibia)을 검출하는 실험을 진행하였다. 실험 결과를 통해 제안하는 기법이 외형 특징만을 이용했던 검출 기법에 비해 개선된 검출 성능을 보이고 있음을 확인할 수 있었다.

• 방송공학회논문지
• /
• 제25권2호
• /
• pp.242-251
• /
• 2020

최근 단일 영상 초해상도에 깊은 합성 곱 신경망을 적용한 알고리듬이 많이 연구되었다. 현존하는 딥러닝 기반 초해상도 기법들은 네트워크의 후반부에 해상도를 업샘플링 하는 구조를 가진다. 이러한 구조는 저해상도에서 고해상도로 한 번에 매핑을 하기에 많은 정보를 예측하는 높은 확대율에서 비효율적인 구조를 가진다. 본 논문에서는 반복적인 업-다운 샘플링 구조를 기반으로 하여 채널 집중 잔여 밀집 블록을 이용한 단일 영상 초해상도 기법을 제안한다. 제안한 알고리듬은 저해상도와 고해상도의 매핑 관계를 효율적으로 예측하여 높은 확대율에서 기존의 알고리듬에 비해 최대 0.14dB 성능 향상과 개선된 주관적 화질을 보여준다.

• 한국산업정보학회:학술대회논문집
• /
• 한국산업정보학회 1999년도 추계공동학술대회 논문집:21세기지식경영과 정보기술
• /
• pp.301-313
• /
• 1999

Recently, the rabid development of information technology has brought enormous changes in education. Consolidation of communications and multimedia technologies are enabling the new educational paradigm such as distance learning and virtual education. Furthermore, many studies in the education engineering field report that teaching using multimedia technologies more enhances students' performance than the traditional instructor-teaching method. However, little research regarding the education using multimedia has been done in the MIS filed. None of multimedia-related studies could be found in the top-ranked MIS journals published in Korea for the last five years, and only a few studies were found even worldwide. In this regard, the purpose of this study is to investigate which features of multimedia software are most important to enhance the teaching results of students. From the previous research, we found out the specific features of the educational multimedia software which are considered to affect the students'performance, and defined the research variables related to those educational software features. And, based on the constructivism and motivation theory of the education engineering field a theoretical research model and research hypotheses were developed. Perceived usefulness of the class and a student's perceived interests in the class were used as surrogate variables to measure teaching performance. Total 277 students participated separately in one of the two multimedia classes which have continued for three weeks. One was C programming language class and the other was multimedia CD-title development class. Each student listened for the multimedia session of the class using multimedia software and, at the end of the multimedia session, answered the survey questionnaire. The results of the study show that motivation to the class and the contents of education were statistically significant to the students'performance in the class. That implies, not only in the traditional instructor-teaching method but also in the multimedia class, that the contents of education itself and student's motivation to the class are most important to raise instructional results.

• 대한방사선기술학회지:방사선기술과학
• /
• 제41권5호
• /
• pp.487-491
• /
• 2018

The purpose of this study was to analyze questionnaires of 161 college students attending radiology departments in order to investigate the actual condition of internet use of radiology students. As a result, 95% of college students using the Internet showed 5.8% of general knowledge, 56.9% of radiation major, and 45.8% of general education. In the field of Internet use, basic medicine was 71.2%, anatomy 59.5% and physiology 51.6%. Radiation theory was 39.9% in radiation physics, 31.4% in radiation biology, and 18.3% in radiation management. The radiological applications were followed by radiography and radiography in order of 31.4% and 20.3%, respectively. The radiological imaging was 45.8%, MRI was 37.9%, CT was 37.3%, ultrasound was 24.2%, And radiation nuclear medicine 25.5%. The results of the descriptive statistics of the satisfaction of the contents using the Internet media showed that the overall satisfaction was below 2.5 Based on the results of this study, it is necessary to develop a program with high accessibility to provide various opportunities for internet-based opportunities to increase the academic achievement value of major subjects through the internet and to solve the difficulties in the major subject.

• 방송공학회논문지
• /
• 제23권5호
• /
• pp.598-605
• /
• 2018

행동 인식은 데이터를 통해 인간의 행동을 인식하는 기술로서 비디오 감시 시스템을 통한 위험 행동과 같은 어플리케이션에 활용되어 질 수 있다. 기존의 행동 인식 알고리즘은 2차원 카메라를 통한 영상이나 다중모드 센서, 멀티 뷰와 같은 장비를 이용한 방법을 사용하거나 3D 장비를 이용하여 이루어져 왔다. 2차원 데이터를 사용한 경우 3차원 공간의 행위 인식에서는 가려짐과 같은 현상으로 낮은 인식율을 보였고 다른 방법은 복잡한 장비의 구성이나 고가의 추가적인 장비로 인한 어려움이 많았다. 본 논문은 RGB와 Depth 정보만을 이용하여 추가적인 장비 없이 CCTV 영상만으로 인간의 행동을 인식하는 방법을 제안한다. 먼저 RGB 영상에서 스켈레톤 추출 알고리즘을 적용하여 관절과 신체부위의 포인트를 추출한다. 이를 식을 적용하여 변위 벡터와 관계 벡터를 포함한 벡터로 변형한 후 RNN 모델을 통하여 연속된 벡터 데이터를 학습한다. 학습된 모델을 다양한 데이터 세트에 적용하여 행동 인식 정확도를 확인한 결과 2차원 정보만으로 3차원 정보를 이용한 기존의 알고리즘과 유사한 성능을 입증할 수 있었다.

• 한국실내디자인학회논문집
• /
• 제22권3호
• /
• pp.195-204
• /
• 2013

Today, the children's museum evolved from place of exhibition for seeing and feeling, to that of exhibition for searching and touching, or so-called the hands-on exhibition. This will allow for grabbing the interest from children and provide an opportunity for learning by an actual experience at this intellectual and benign environment. But in this modern society that constantly undergoes evolution even as we speak, it is not an easy task to bring our children to museums, because they are vastly exposed to various media. Moreover, if the children who visited museums carelessly bypass the specific areas intended for exhibition, the educational purpose of "enhanced learning effect with hands-on experience" is easily underachieved. According to children's visual stimuli, their behavioral charactoristics are appeared to freewill curiosity but they show shorter elapsed time(impermanence), intensity(stubbornness), frequent occurrence(frequentness) and changeable (translatability). So, We need Improvement of visual image affordance through the measurement of stay time. Firstly, five factors are extracted by factor analysis on twenty questions based on visual image factors; Color accessability and Satistaction(factor1), brightness and color harmony(factor2), feeling on harmonization of color and stay time(factor3), simplex & complex of space(factor4), feeling on scale(factor5) Secondly, the following result are derived through a distribution chart on an exhibition room of K-museum. As shown above, this study based on various analyzed aspects proposes the directions of a color image plan to improve stay time in exhibition space of children museums with a hope to support educational goals of experience education-focused children museums.

• 대한예방한의학회지
• /
• 제19권2호
• /
• pp.159-167
• /
• 2015

Objective : Recently many kinds of food therapies have sprung up to prevent or manage disease and to promote health. Seeing Korean history, the Korean medical doctors have been applying food therapy based on the thought that the dietary sources were as important as the medicine. Therefore, in this study, we designed a mobile edutainment game prototype with the goal of providing education about healthy food knowledge for users who belongs to different constitutional types. Materials and Method : We adopted the Sasang Constitutional Medicine as the medical background knowledge for our edutainment game design. Based on the user study, we developed the process of edutainment system which is composed of 'My game', 'My constitution', and 'My food'. Among the whole process, we developed a prototype for the core module - the 'My game' part. This prototype used a jumping game for mobile devices that is composed of training, level 1 and level 2 stages. Results : From the target user evaluation, it was proved that 1) in terms of the learning effect of healthy diet, the edutainment game we developed has a significant advantage over the conventional learning media. 2) after playing the edutainment game, the good and bad food identification accuracy based on picture and text format were increased by 44% and 42% respectively, and 3) target users perceived enjoyment while using this prototype, as well as showed positive intention to use this game as edutainment tool in the future. Conclusion : We designed and developed mobile edutainment game prototype to educate healthy food knowledge based on Sasang Constitutional Medicine. Through user evaluation, we proved that our prototype enhanced healthy food knowledge and that user accepted the prototype as a beneficial edutainment tool.

• 한국콘텐츠학회논문지
• /
• 제18권9호
• /
• pp.55-68
• /
• 2018

본 연구는 세계시민으로서의 책임과 실천의식 함양을 효과적으로 가져오는 세계시민교육 프로그램 개발을 목적으로 하고, 글로벌 이슈에 대한 공감, 몰입, 성찰과 인식의 변화를 가져오는 방법으로서 스토리텔링을 활용한 세계시민교육 프로그램을 개발하였다. 프로그램은 총 8차시로 구성되었으며, 고등학교 2학년 학생 40명을 대상으로 실시되었다. 스토리텔링은 교사가 학생들에게 스토리를 들려주거나 학생들이 직접 만들어 나가는 방법으로 활용되었다. 프로그램 적용 결과 세계시민의식과 이에 대한 학습흥미의 긍정적 변화를 확인할 수 있었다. 본 연구를 통하여 스토리텔링이 추상적 학습내용에 대한 구체적이고 맥락적 이해를 돕고, 세계시민의식 학습에 흥미로움을 더해주며, 다양한 교수학습활동과 매체를 적용하는 것을 용이하게 하는 데 좋은 학습자료가 될 수 있음을 확인하였다.