• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.3
• /
• pp.109-114
• /
• 2021

GAN(Generative Adversarial Network) is an image generation model, which is composed of a generator network and a discriminator network, and generates an image similar to a real image. Since the image generated by the GAN should be similar to the actual image, a loss function is used to minimize the loss error of the generated image. However, there is a problem that the loss function of GAN degrades the quality of the image by making the learning to generate the image unstable. To solve this problem, this paper analyzes GAN-related studies and proposes an edge GAN(eGAN) using edge detection. As a result of the experiment, the eGAN model has improved performance over the existing GAN model.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.7
• /
• pp.27-33
• /
• 2022

In this paper, we propose a novel generator architecture for multi-domain style transfer method not an image to image translation, as a method of generating a styled image by transfering a style to the content image. A latent vector and Gaussian noises are added to the generator of GAN so that a high quality image is generated while considering the characteristics of various data distributions for each domain and preserving the features of the content data. With the generator architecture of the proposed GAN, networks are configured and presented so that the content image can learn the styles for each domain well, and it is applied to the domain composed of images of the four seasons to show the high resolution style transfer results.

• Journal of the Korea Convergence Society
• /
• v.11 no.8
• /
• pp.99-106
• /
• 2020

The purpose of this study is to find out whether image similarity between celebrities and brands on social network service be a determinant to predict advertising effectiveness. To this end, an advertising effect prediction model for celebrity endorsed advertising was created and its validity was verified through a machine learning method which is a big data analysis technique. Firstly, the celebrity-brand image similarity, which was used as an independent variable, was quantified by the association network theory with social big data, and secondly a multiple regression model which used data representing advertising effects as a dependent variable was repeatedly conducted to generate an advertising effect prediction model. The accuracy of the prediction model was decided by comparing the prediction results with the survey outcomes. As for a result, it was proved that the validity of the predictive modeling of advertising effects was secured since the classification accuracy of 75%, which is a criterion for judging validity, was shown. This study suggested a new methodological alternative and direction for big data-based modeling research through celebrity-brand image similarity structure based on social network theory, and effect prediction modeling by machine learning.

• Smart Media Journal
• /
• v.12 no.1
• /
• pp.9-16
• /
• 2023

Recently, a convolutional neural network (CNN) based system is being developed to overcome the limitations of human resources in the apple quality classification of farmhouse. However, since convolutional neural networks receive only images of the same size, preprocessing such as sampling may be required, and in the case of oversampling, information loss of the original image such as image quality degradation and blurring occurs. In this paper, in order to minimize the above problem, to generate a image patch based graph of an original image and propose a random walk-based positional encoding method to apply the graph transformer model. The above method continuously learns the position embedding information of patches which don't have a positional information based on the random walk algorithm, and finds the optimal graph structure by aggregating useful node information through the self-attention technique of graph transformer model. Therefore, it is robust and shows good performance even in a new graph structure of random node order and an arbitrary graph structure according to the location of an object in an image. As a result, when experimented with 5 apple quality datasets, the learning accuracy was higher than other GNN models by a minimum of 1.3% to a maximum of 4.7%, and the number of parameters was 3.59M, which was about 15% less than the 23.52M of the ResNet18 model. Therefore, it shows fast reasoning speed according to the reduction of the amount of computation and proves the effect.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.36 no.5
• /
• pp.323-330
• /
• 2023

The development of an analysis model that reflects the microstructure characteristics of polyvinyl alcohol (PVA) fiber-reinforced cementitious composites, which have a highly complex microstructure, enables synergy between efficient material design and real experiments. PVA fiber orientations are an important factor that influences the mechanical behavior of PVA fiber-reinforced cementitious composites. Owing to the difficulty in distinguishing the gray level value obtained from micro-CT images of PVA fibers from adjacent phases, fiber segmentation is time-consuming work. In this study, a micro-CT test with a voxel size of 0.65 ㎛³ was performed to investigate the three-dimensional distribution of fibers. To segment the fibers and generate training data, histogram, morphology, and gradient-based phase-segmentation methods were used. A U-net model was proposed to segment fibers from micro-CT images of PVA fiber-reinforced cementitious composites. Data augmentation was applied to increase the accuracy of the training, using a total of 1024 images as training data. The performance of the model was evaluated using accuracy, precision, recall, and F1 score. The trained model achieved a high fiber segmentation performance and efficiency, and the approach can be applied to other specimens as well.

• Annual Conference on Human and Language Technology
• /
• 2018.10a
• /
• pp.389-393
• /
• 2018

딥러닝의 발달로 기계번역, 대화 시스템 등의 자연언어처리 분야가 크게 발전하였다. 딥러닝 모델의 성능을 향상시키기 위해서는 많은 데이터가 필요하다. 그러나 많은 데이터를 수집하기 위해서는 많은 시간과 노력이 소요된다. 본 연구에서는 이미지 생성 모델로 좋은 성능을 보이고 있는 생성적 적대 네트워크(Generative adverasarial network)를 문장 생성에 적용해본다. 본 연구에서는 긍/부정 조건에 따른 문장을 자동 생성하기 위해 SeqGAN 모델을 수정하여 사용한다. 그리고 분류기를 포함한 SeqGAN이 긍/부정 감성인식 학습데이터를 자동 생성할 수 있는지 실험한다. 실험을 수행한 결과, 분류기를 포함한 SeqGAN 모델이 생성한 문장과 학습데이터를 혼용하여 학습할 경우 실제 학습데이터만 학습 시킨 경우보다 좋은 정확도를 보였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2009.04a
• /
• pp.1013-1016
• /
• 2009

본 논문에서는 교육용 가상환경에서 학습자들에게 좀 더 효과적인 학습 효과를 주기 위해 학습 중에 보고 촬영했던 사진들을 자동으로 정리해서 사용자가 교육용 가상환경에서 체험했던 학습내용을 사진을 통해서 복습할 수 있도록 해주는 알고리즘을 소개한다. 기존의 날짜, 장소, 키워드 등의 정보를 이용하여 사진을 정리하는 알고리즘과는 달리, 본 논문에서는 사용자가 학습을 하면서 기억해야 할 중요한 내용이나 사용자의 관심도에 의해 사진 정리를 함으로써 사용자의 학습 효과를 높이는 것을 목적으로 하는 사진 정리 알고리즘을 소개한다. 이에 따라 알고리즘에서 학습적으로 중요한 사진을 뽑는 기준과 사용자의 관심도, 인지율 계산에 대해 설명하고 이 알고리즘을 기반으로 구현한 시스템을 설명한다. 또한 사용자 실험 분석을 하고 향후 연구방향에 대해 논한다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.07a
• /
• pp.573-576
• /
• 2022

본 논문에서는 삼각형 구조로 구성된 3차원 메쉬(Mesh)에서 합성곱 신경망(Convolutional Neural Network, CNN)의 정확도를 개선시킬 수 있는 새로운 학습 표현 기법을 제시한다. 우리는 메쉬를 구성하고 있는 삼각형의 넓이와 그 로컬 특징을 기반으로 학습을 진행한다. 일반적으로 딥러닝은 인공신경망을 수많은 계층 형태로 연결한 기법을 말하며, 주요 처리 대상은 오디오 파일과 이미지이었다. 인공지능에 대한 연구가 지속되면서 3차원 딥러닝이 도입되었지만, 기존의 학습과는 달리 3차원 학습은 데이터의 확보가 쉽지 않다. 혼합현실과 메타버스 시장으로 인해 3차원 모델링 시장이 증가가 하면서 기술의 발전으로 데이터를 획득할 수 있는 방법이 생겼지만, 3차원 데이터를 직접적으로 학습 표현하는 방식으로 적용하는 것은 쉽지 않다. 그렇기 때문에 본 논문에서는 산업 현장에서 사용되는 데이터인 삼각형 메쉬 구조를 바탕으로 기존 방법보다 정확도가 높은 학습 기법을 제안한다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.01a
• /
• pp.369-372
• /
• 2023

본 논문에서는 삼각형 구조로 구성된 3차원 메쉬(Mesh)에서 합성곱 신경망(Convolution Neural Network, CNN)을 응용하여 정확도가 높은 새로운 학습 표현 기법을 제시한다. 우리는 메쉬를 구성하고 있는 폴리곤의 edge와 face의 로컬 특징을 기반으로 학습을 진행한다. 일반적으로 딥러닝은 인공신경망을 수많은 계층 형태로 연결한 기법을 말하며, 주요 처리 대상은 1, 2차원 데이터 형태인 오디오 파일과 이미지였다. 인공지능에 대한 연구가 지속되면서 3차원 딥러닝이 도입되었지만, 기존의 학습과는 달리 3차원 딥러닝은 데이터의 확보가 쉽지 않다. 혼합현실과 메타버스 시장의 확대로 인해 3차원 모델링 시장이 증가하고, 기술의 발전으로 데이터를 획득할 수 있는 방법이 생겼지만, 3차원 데이터를 직접적으로 학습에 이용하는 방식으로 적용하는 것은 쉽지 않다. 그렇게 때문에 본 논문에서는 산업 현장에서 이용되는 데이터인 메쉬 구조를 폴리곤의 최소 단위인 삼각형 형태로 구성하여 학습 데이터를 구성해 기존의 방법보다 정확도가 높은 학습 기법을 제안한다.

• Journal of the Korea Convergence Society
• /
• v.13 no.4
• /
• pp.39-44
• /
• 2022

Mathematics franchise education companies are developing various online learning systems to provide on-off integrated education to learners. Most online learning systems deliver one-way lecture content to learners and perform quantitative problem-solving learning for learning results. However, each learner has different academic achievement competencies, and it is impossible to determine exactly where the level of understanding fell when solving a math method. and based on this, establish an online learning system to discover the weak points of learners and propose an effective learner management method. Through the developed learning method and system, it is expected to cultivate balanced problem-solving ability for learners and provide differentiated brand image and counseling service to franchise companies.