• Journal of the Semiconductor & Display Technology
• /
• v.17 no.4
• /
• pp.101-104
• /
• 2018

Augmented Reality (AR) is an emerging technology and the applications of technology are still not fully unveiled. This paper explores a new application of augmented reality for new direction in art exhibitions, which aims to bring interactive learning experience to life. The project takes printed images on book or exhibiting arts to the next level by applying AR technology to provide a unique fascinating experience to its readers on mobile devices. AR technology composing with animation brings new digital entertainment experience to the user of art exhibitions. The key feature of this paper uses the technology presents auxiliary information in the field of view of an object on art exhibitions automatically without human intervention.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.413-415
• /
• 2021

This paper proposes a nail detection algorithm using OpenPose and implements virtual nail art using it. Based on the key points detected by OpenPose, the finger area is detected using skin color characteristics for each finger. The nail region is detected from the edge image of the detected finger region. Then, a virtual nail art is implemented by synthesizing nail tips in the nail area. In a somewhat controlled shooting environment, simulation results show that the proposed algorithm detects nail areas well and implements virtual nail art well.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2013.05a
• /
• pp.179-180
• /
• 2013

This study aims to make continuedly the community culture and art development to promote the use of a cultural facility through the art program of an art museum. This increases the satisfaction of the citizens' life by it and excavates the faithful users' culture and art. It contributes the culture and art development of Pusan in the future. We analyzed the data of the social index related Pusan's culture and leisure to investigate the basic concept of art museum education based on previous research. We reviewed the degree of use of facility for culture and art based on it. We present a community culture and art development plan through connection program displayed art museum. We expect the basis to increase the community culture level in Pusan.

• Journal of the Korean Crystal Growth and Crystal Technology
• /
• v.17 no.1
• /
• pp.41-45
• /
• 2007

ART DECO is a suitable for the rational style and the variation in modeling techniques of artistic revival. At present, production using CAD & CAM by software have great potential adaptation in the industrial arts since it accomplishs the understanding of the plan and promote as realization of automation with rapid and accurate results. Until a recent date, geometrical, abstract mode was discovered through the advanced development, with the tendency of mechanical modification. This paper attempts to practice in accordance with liberal form's fascinating variety from above formative style the application of engineering science. As follows presenting the drawing and 3D MODELING/DATA accordingly it carries out the 3D shapes and techniques by CAD & CAM.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09b
• /
• pp.71-75
• /
• 2003

오류 역전파 학습 알고리즘을 이용하여 영상 인식에 적용 할 경우에는 은닉층의 노드 수를 경험적으로 설정하므로, 학습시간과 지역최소화 및 정체현상이 발생한다. 그리고 ARTI 알고리즘은 입력 패턴과 저장 패턴간의 측정 방법인 유사성 검증 방법과 경계 변수의 설정에 따라 인식률이 좌우된다. 경계 변수의 값이 크면 입력 패턴과 저장 패턴사이에 약간의 차이만 있어도 새로운 카테고리(Category)로 분류하고, 반대로 경계 변수의 값이 적으면 입력 패턴과 저장 패턴 사이에 많은 차이가 있더라도 유사성이 인정되어 입력 패턴들을 대략적으로 분류한다. 따라서 ART1 알고리즘을 영상 인식에 적용하기 위해서는 경계 변수를 경험적으로 설정하므로 인식률에 부정적인 영향을 갖는 문제점이 있다. 따라서 본 논문에서는 개선된 ART1 알고리즘과 지도 학습 방법을 결합하여 신경망의 은닉층 노드를 동적으로 변화시키는 자가 생성지도 학습 알고리즘을 제안한다. 제안된 신경망에서 입력층과 은닉층의 학습 구조에는 ART1 알고리즘을 개선하여 적용하고, 은닉층과 출력층의 학습 구조에는 은닉층에서 승자로 선택된 노드와 출력층 노드와 연결된 가중치만을 조정하고 Delta-Bar-Delta 알고리즘을 적용한다. 제안된 방법의 학습 성능을 분석하기 위하여 학생증 영상에서 추출한 학번 패턴 분류에 적용한 결과, 기존의 신경망 학습 알고리즘보다 학습 성능이 개선됨을 확인하였다.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.1287-1292
• /
• 2005

This paper presents an effective pattern classification model by designing an artificial neural network based pattern classifiers for face recognition. First, a RGB image inputted from a frame grabber is converted into a HSV image which is similar to the human beings' vision system. Then, the coarse facial region is extracted using the hue(H) and saturation(S) components except intensity(V) component which is sensitive to the environmental illumination. Next, the fine facial region extraction process is performed by matching with the edge and gray based templates. To make a light-invariant and qualified facial image, histogram equalization and intensity compensation processing using illumination plane are performed. The finally extracted and enhanced facial images are used for training the pattern classification models. The proposed H-ART2 model which has the hierarchical ART2 layers and F-LVQ model which is optimized by fuzzy membership make it possible to classify facial patterns by optimizing relations of clusters and searching clustered reference patterns effectively. Experimental results show that the proposed face recognition system is as good as the SVM model which is famous for face recognition field in recognition rate and even better in classification speed. Moreover high recognition rate could be acquired by combining the proposed neural classification models.

• Nuclear Engineering and Technology
• /
• v.54 no.1
• /
• pp.310-317
• /
• 2022

The mass attenuation coefficient is the primary physical parameter to model narrow beam gamma-ray attenuation. A new machine learning based approach is proposed to model gamma-ray shielding behavior of composites alternative to theoretical calculations. Two fuzzy logic algorithms and a neural network algorithm were trained and tested with different mixture ratios of vanadium slag/epoxy resin/antimony in the 0.05 MeV-2 MeV energy range. Two of the algorithms showed excellent agreement with testing data after optimizing adjustable parameters, with root mean squared error (RMSE) values down to 0.0001. Those results are remarkable because mass attenuation coefficients are often presented with four significant figures. Different training data sizes were tried to determine the least number of data points required to train sufficient models. Data set size more than 1000 is seen to be required to model in above 0.05 MeV energy. Below this energy, more data points with finer energy resolution might be required. Neuro-fuzzy models were three times faster to train than neural network models, while neural network models depicted low RMSE. Fuzzy logic algorithms are overlooked in complex function approximation, yet grid partitioned fuzzy algorithms showed excellent calculation efficiency and good convergence in predicting mass attenuation coefficient.

• Journal of the Korea Convergence Society
• /
• v.13 no.4
• /
• pp.195-201
• /
• 2022

The core of the Fourth Industrial revolution, the topic of the 21st century, lies in the convergence of various technologies that are developing to the cutting edge. Therefore, convergence education is essential to solve problems from various perspectives. Above all, in order to solve the complex problem arising from the contact point between humans and technology in the future society, it was considered that insight through a humanities and artistic approach was necessary. Now, universities are trying to run a multidisciplinary comprehensive design to adapt to the new digital environment. In this study, an operation case of an engineering comprehensive design subject incorporating art operated by a researcher was described. The researcher presented an art convergence curriculum in a comprehensive design subject by applying 1. problem recognition and definition, 2. idea derivation and evaluation, 3. project development, 4. presentation, and demonstration process operation methods. Through this study, it is meaningful that it presented a method of operating art convergence in engineering comprehensive design subjects by allowing students to attempt an artistic aesthetic and emotional approach to project development.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2013.05a
• /
• pp.368-370
• /
• 2013

본 논문에서는 복부 초음파 영상에서 복부 근육을 추출하고 추출된 근육 영역에서 지방을 분석하는 방법을 제안한다. 복부 초음파 영상에서 밝은 명암도를 가지는 근막 영역과 어두운 명암도를 가지는 근육 영역의 명암 대비를 강조하기 위해서 앤드 인 탐색 스트레칭 방법과 Multiple 연산을 적용한다. 평균 명암도와 명암 대비가 강조된 복부 초음파 영상에서 수직 방향의 명암도가 200이상인 픽셀들은 퍼지 이진화 기법을 적용하여 이진화한다. 이진화된 영상에서 외복사근 상단선을 추출한 후, 퍼지 이진화 기법이 적용된 영상과 합성한다. 합성된 영상에서 최종 근막 영역을 추출한다. 추출된 각각의 복부 근육 영역에 ART2 알고리즘을 적용하여 복부 근육 영역을 양자화한다. 양자화된 복부 근육 내의 영역을 분석하여 최종 지방 영역을 추출한다. 제안된 복부 근육 추출 및 지방 분석 방법을 실제 복부 초음파 영상을 대상으로 실험한 결과, 추출된 복부 근육 영역에 ART2 알고리즘 기반 양자화 기법을 적용하여 지방을 추출하는 것이 복부비만을 분석하는데 도움이 되는 것을 영상 의학과 전문의를 통해 확인하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2003.05a
• /
• pp.603-606
• /
• 2003

본 논문에서는 ART(Adaptive Resonance Theory) 신경회로망을 이용하여 한글 모음을 인식하고, 그 유형을 분류하는 방법을 제안하였다. 기존의 연구들은 단순히 문자의 선분, 획 등의 정합만을 이용하여 한글의 자소 분류에 중점을 두었다. 그러나 인식 대상 운자의 특성이 각각 다르므로 효율적인 인식을 위해서는 먼저 포괄적인 특정적 유형 분류가 필요하다. 제안된 한글 유형 분류 시스템에서는 먼저 ART 신경회로망의 문제점인 증가분류 알고리즘의 단점을 최소화할 수 있도록 비교층에 최초 활성화패턴의 크기를 기억하는 메모리를 두고 각 층간 하향틀 변화를 경계인수 값을 "1" 이내로 제한하여 이미 입력된 패턴을 다시 입력할 때, 새로운 노드의 활성화를 방지하여 비교적 입력순서에 둔감한 분류가 가능하였다. 실험 결과 제안된 시스템에서는 한글의 6형식 중 1, 3, 4, 5형식 분류는 평균 97.3% 의 분류율을 보였으나, 나머지 2, 6형식 분류는 다소 떨어지는 평균 94.9% 분류율를 보였다.