• Journal of Multimedia Information System
• /
• 제8권4호
• /
• pp.203-210
• /
• 2021

Deep leaning convolutional neural networks (CNN) have successfully been applied to image super-resolution (SR). Despite their great performances, SR techniques tend to focus on a certain upscale factor when training a particular model. Algorithms for single model multi-scale networks can easily be constructed if images are upscaled prior to input, but sub-pixel convolution upsampling works differently for each scale factor. Recent SR methods employ multi-scale and multi-path learning as a solution. However, this causes unshared parameters and unbalanced parameter distribution across various scale factors. We present a multi-scale single-path upsample module as a solution by exploiting the advantages of sub-pixel convolution and interpolation algorithms. The proposed model employs sub-pixel convolution for the highest scale factor among the learning upscale factors, and then utilize 1-dimension interpolation, compressing the learned features on the channel axis to match the desired output image size. Experiments are performed for the single-path upsample module, and compared to the multi-path upsample module. Based on the experimental results, the proposed algorithm reduces the upsample module's parameters by 24% and presents slightly to better performance compared to the previous algorithm.

• International journal of advanced smart convergence
• /
• 제6권2호
• /
• pp.24-33
• /
• 2017

Since 'teaching coding' or 'programming' classes for computational thinking (CT) education in K-12 are renowned around the world, a pre-service teachers' readiness for integrating CT into their teaching subjects is important due to the fact that CT is considered to be another 'R' from algoRitm for 21st century literacy, in addition to the traditional 3R (Reading, Writhing, and Arithmetic) [2] and CT roles to other disciplines. With this rationale, we designed a guided instruction based CT course for pre-service teachers. We show the effectiveness of the program with respect to the teachers' attitude toward combining CT into their teaching subjects, and mindset changes of learning computing connected with the career development of the teacher themselves. The research focused on the instructional methodology of teaching programing for non-Computer Science Education (CSE) majors who are not familiar with computer science for alleviating the cognitive load of first exposure to programming course under the CT concepts.

• 한국디지털건축인테리어학회논문집
• /
• 제1권1호
• /
• pp.9-15
• /
• 2001

This research is to grasp the identify the curriculum development procedure for architectural computer graphics. In order to achieve the above purpose, this study was carried out job analysis of architectural graphic designer. Architectural Computer Graphics Curriculum is developed by tasks analysis. Each characteristics are as fellow: 1. job-related knowledges are based on architectural design parts, 2. skilled workers should be made on the base of job analysis and O.J.T(On the job training) 3. trained students are related many kinds of all industry parts of Architecture & interior Especially college students have needs for design part of architecture fit on study in these curriculum.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2017년도 춘계학술발표대회
• /
• pp.953-954
• /
• 2017

화재, 지진 등과 같은 대규모의 재난에 대비하는 것은 최근 매우 중요한 활동으로 인식되고 있다. 그러나 이러한 재난 대비 및 대처의 중요성 인식에도 불구하고, 많은 사람들이 실제 발생하는 재난 상황에서 적절한 대처를 하지 못하여 인명 또는 재산상의 피해가 발생하고 있다. 본 논문에서는 화재가 발생하는 경우의 신속한 대처를 지원하기 위한 가상현실 기반의 화재 대처 어플리케이션을 개발하고자 한다. 이는 화재 발생시, 소화전이나 소화기를 정확하고 올바르게 사용할 수 있도록 도와줄 뿐만 아니라, 대규모의 학생들에 대한 교육에도 매우 유용하게 활용될 수 있을 것이다.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2019년도 추계학술발표대회
• /
• pp.1026-1029
• /
• 2019

최근 제안되고 있는 Bert 등의 딥러닝 언어 모델 기반 pre-training 기법은 다양한 NLP 분야에서 활용되고 있다. 텍스트로 작성된 데이터 셋을 딥러닝 언어 모델이 학습하기 위해서는 토크나이징(tokenizing) 기술이 필요하다. 그러나 기존 토크나이징 방식은 한국어 및 한글이 가지는 고유한 특성(교착어적 특성과 모아쓰기 반영)을 반영하기 어렵다는 한계를 가지고 있다. 본 논문에서는 한국어와 한글이 가지는 고유한 특성을 고려하기 위하여 음소 단위의 임베딩 기법을 제안하며, 이를 기반으로 언어 모델을 설계 및 구현한다. 또한 음소 단위 임베딩 기반 한국어 모델이 실제 데이터 집합(구약성서)에서 나타나는 언어적 패턴을 학습할 수 있다는 것을 실험을 통하여 밝힌다.

• 한국정보과학회 언어공학연구회:학술대회논문집(한글 및 한국어 정보처리)
• /
• 한국정보과학회언어공학연구회 2019년도 제31회 한글 및 한국어 정보처리 학술대회
• /
• pp.460-464
• /
• 2019

번역 품질 예측은 기계번역 시스템이 생성한 번역문의 품질을 정답 번역문을 참고하지 않고 예측하는 과정으로, 번역문의 사후 교정을 위한 번역 오류 검출의 역할을 담당하는 중요한 연구이다. 본 논문은 문장 수준의 번역 품질 예측 문제를 HTER 구간의 분류 문제로 간주하여, 번역 품질 말뭉치의 HTER 분포 불균형으로 인한 성능 제약을 완화하기 위해 인조 사후 교정 말뭉치를 이용하는 방법을 제안하였다. 결과적으로 HTER 분포를 균등하게 조정한 학습 말뭉치가 그렇지 않은 쪽에 비해 번역 품질 예측에 더 효과적인 것을 보였다.

• 한국컴퓨터정보학회:학술대회논문집
• /
• 한국컴퓨터정보학회 2023년도 제68차 하계학술대회논문집 31권2호
• /
• pp.695-696
• /
• 2023

본 논문은 뇌-컴퓨터 인터페이스(BCI)를 활용한 게임 개발을 통해 집중력 부족 문제를 해결하기 위한 방안을 제시한다. BCI 기술은 사용자의 뇌파 신호를 분석하여 게임에 적용할 수 있으며, 그에 따라 뇌파 신호를 활용한 집중력 향상을 도모해 볼 수 있는 게임을 설계하였다. Unity 게임 개발 환경과 Emotiv Insight 장비를 사용하여 게임을 구현하였으며, 사용자는 뇌파 신호를 통해 플레이어를 제어하여 게임을 즐길 수 있다. 연구 결과는 뇌파 기반 게임이 사용자의 집중력 향상에 도움을 줄 수 있는 잠재력을 보여준다.

• International Journal of Computer Science & Network Security
• /
• 제24권6호
• /
• pp.89-98
• /
• 2024

Due to the increase in population and consequently the increase in the production of plastic waste, recovery of this part of the waste is an undeniable necessity. On the other hand, the recycling of plastic waste, if it is placed in a systematic process and controlled, can be effective in creating jobs and maintaining environmental health. Waste collection in many large cities has become a major problem due to lack of proper planning with increasing waste from population accumulation and changing consumption patterns. Today, waste management is no longer limited to waste collection, but waste collection is one of the important areas of its management, i.e. training, segregation, collection, recycling and processing. In this study, a systematic method based on machine vision for sorting plastic bottles in different colors for recycling purposes will be proposed. In this method, image classification and segmentation techniques were presented to improve the performance of plastic bottle classification. Evaluation of the proposed method and comparison with previous works showed the proper performance of this method.

• Journal of Multimedia Information System
• /
• 제4권1호
• /
• pp.9-18
• /
• 2017

Ubiquitous computing is a novel paradigm that is rapidly gaining in the scenario of wireless communications and telecommunications for realizing smart world. As rapid development of sensor technology, smart sensor system becomes more popular in automobile or vehicle. In this study, a new vehicle detection mechanism in real-time for blind spot area is proposed based on imaging sensors. To determine the position of other vehicles on the road is important for operation of driver assistance systems (DASs) to increase driving safety. As the result, blind spot detection of vehicles is addressed using an automobile detection algorithm for blind spots. The proposed vehicle verification utilizes the height and angle of a rear-looking vehicle mounted camera. Candidate vehicle information is extracted using adaptive shadow detection based on brightness values of an image of a vehicle area. The vehicle is verified using a training set with Haar-like features of candidate vehicles. Using these processes, moving vehicles can be detected in blind spots. The detection ratio of true vehicles was 91.1% in blind spots based on various experimental results.

• Journal of Information Processing Systems
• /
• 제15권4호
• /
• pp.904-919
• /
• 2019

Many security systems rely solely on solutions based on Artificial Intelligence, which are weak in nature. These security solutions can be easily manipulated by malicious users who can gain unlawful access. Some security systems suggest using fingerprint-based solutions, but they can be easily deceived by copying fingerprints with clay. Image-based security is undoubtedly easy to manipulate, but it is also a solution that does not require any special training on the part of the user. In this paper, we propose a multi-factor security framework that operates in a three-step process to authenticate the user. The motivation of the research lies in utilizing commonly available and inexpensive devices such as onsite CCTV cameras and smartphone camera and providing fully secure user authentication. We have used technologies such as Argon2 for hashing image features and physically unclonable identification for secure device-server communication. We also discuss the methodological workflow of the proposed multi-factor authentication framework. In addition, we present the service scenario of the proposed model. Finally, we analyze qualitatively the proposed model and compare it with state-of-the-art methods to evaluate the usability of the model in real-world applications.