• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.05a
• /
• pp.153-156
• /
• 2022

오늘날 이미지 및 오디오와 같은 디지털 미디어의 활용이 급격하게 증가함에 따라, 디지털 콘텐츠의 저작권을 보호하기 위한 워터마킹 기술의 중요성이 대두되고 있다. 최근 딥러닝 기반 이미지 워터마킹 기술에 대한 다양한 연구 결과가 발표되고 있는 반면, 딥러닝을 이용한 오디오 워터마킹에 관련된 연구는 미진한 것이 현실이다. 본 논문에서는 딥러닝을 기반으로 오디오 워터마킹 기술을 개발하기 위한 오토인코더 모델 및 생성적 적대 신경망 모델에 대해 제안한다.

• Journal of Internet Computing and Services
• /
• v.22 no.1
• /
• pp.23-30
• /
• 2021

This paper proposes a noise-tolerant image classification system using multiple autoencoders. The development of deep learning technology has dramatically improved the performance of image classifiers. However, if the images are contaminated by noise, the performance degrades rapidly. Noise added to the image is inevitably generated in the process of obtaining and transmitting the image. Therefore, in order to use the classifier in a real environment, we have to deal with the noise. On the other hand, the autoencoder is an artificial neural network model that is trained to have similar input and output values. If the input data is similar to the training data, the error between the input data and output data of the autoencoder will be small. However, if the input data is not similar to the training data, the error will be large. The proposed system uses the relationship between the input data and the output data of the autoencoder, and it has two phases to classify the images. In the first phase, the classes with the highest likelihood of classification are selected and subject to the procedure again in the second phase. For the performance analysis of the proposed system, classification accuracy was tested on a Gaussian noise-contaminated MNIST dataset. As a result of the experiment, it was confirmed that the proposed system in the noisy environment has higher accuracy than the CNN-based classification technique.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.11a
• /
• pp.643-646
• /
• 2021

경찰청에 따르면 도로교통법이 개정된 이후 3개월단 개인형 이동장치(PM)를 단속한 결과 무면허 운전이 3199건에 달하는 것으로 나타났다. 공유 킥보드 서비스의 경우 회원가입을 할 때 운전면허증 취득 여부를 확인하긴 하지만 서비스를 이용할 때는 별도의 확인 절차 없이 대여할 수 있기 때문에 운전면허증을 취득하지 않았어도 대여하는 경우가 발생한다. 본 논문에서는 공유 킥보드 서비스의 보안 취약점을 보완하기 위해 오토인코더와 변이형 오토인코더를 사용한 딥러닝 기반의 공유 킥보드 대리 대여 방지 시스템을 제안한다. 오토인코더는 지문 데이터로부터 특징만을 추출할 수 있어, 사용자의 지문 원본을 서버에게 노출시키지 않을 수 있다. 변이형 오토인코더는 생성형 모델로써, 사용자의 지문 데이터를 증폭 시켜 합성곱 신경망의 성능을 높이는데 도움을 준다. 이러한 오토인코더와 변이형 오토인코더의 특징을 이용해 사용자의 지문을 서버에 노출시키지 않으면서 적은 데이터로 신뢰성 높은 사용자 인증이 가능한 전동 킥보드 대여 시스템을 제안한다.

• Journal of Korea Society of Industrial Information Systems
• /
• v.24 no.1
• /
• pp.23-30
• /
• 2019

As deep learning with the network-based algorithms evolve, artificial intelligence is rapidly growing around the world. Among them, finance is expected to be the field where artificial intelligence is most used, and many studies have been done recently. The existing financial strategy using deep-run is vulnerable to volatility because it focuses on stock price forecasts for a single stock. Therefore, this study proposes to construct ETF products constructed through portfolio methods by calculating the stocks constituting funds by using deep learning. We analyze the performance of the proposed model in the KOSPI 100 index. Experimental results showed that the proposed model showed improved results in terms of returns or volatility.

• Journal of the Korea Computer Graphics Society
• /
• v.29 no.5
• /
• pp.11-20
• /
• 2023

In this paper, we propose a novel deep learning-based motion reconstruction approach that facilitates the generation of full-body motions, including finger motions, while also enabling the online adjustment of motion generation delays. The proposed method combines the Vive Tracker with a deep learning method to achieve more accurate motion reconstruction while effectively mitigating foot skating issues through the use of an Inverse Kinematics (IK) solver. The proposed method utilizes a trained AutoEncoder to reconstruct character body motions using tracker data in real-time while offering the flexibility to adjust motion generation delays as needed. To generate hand motions suitable for the reconstructed body motion, we employ a Fully Connected Network (FCN). By combining the reconstructed body motion from the AutoEncoder with the hand motions generated by the FCN, we can generate full-body motions of characters that include hand movements. In order to alleviate foot skating issues in motions generated by deep learning-based methods, we use an IK solver. By setting the trackers located near the character's feet as end-effectors for the IK solver, our method precisely controls and corrects the character's foot movements, thereby enhancing the overall accuracy of the generated motions. Through experiments, we validate the accuracy of motion generation in the proposed deep learning-based motion reconstruction scheme, as well as the ability to adjust latency based on user input. Additionally, we assess the correction performance by comparing motions with the IK solver applied to those without it, focusing particularly on how it addresses the foot skating issue in the generated full-body motions.

• Journal of Digital Convergence
• /
• v.19 no.6
• /
• pp.251-258
• /
• 2021

In this paper, we proposed an encoder-decoder model utilizing residual learning to improve the accuracy of the U-Net-based semantic segmentation method. U-Net is a deep learning-based semantic segmentation method and is mainly used in applications such as autonomous vehicles and medical image analysis. The conventional U-Net occurs loss in feature compression process due to the shallow structure of the encoder. The loss of features causes a lack of context information necessary for classifying objects and has a problem of reducing segmentation accuracy. To improve this, The proposed method efficiently extracted context information through an encoder using residual learning, which is effective in preventing feature loss and gradient vanishing problems in the conventional U-Net. Furthermore, we reduced down-sampling operations in the encoder to reduce the loss of spatial information included in the feature maps. The proposed method showed an improved segmentation result of about 12% compared to the conventional U-Net in the Cityscapes dataset experiment.

• Annual Conference on Human and Language Technology
• /
• 2019.10a
• /
• pp.333-338
• /
• 2019

인코더-디코더(Encoder-decoder)는 현대 기계 번역(Machine translation)의 가장 기본이 되는 모델이다. 인코딩은 마치 인간의 뇌가 출발어(Source language) 문장을 읽고 이해를 하는 과정과 유사하고, 디코딩은 뇌가 이해한 의미를 상응하는 도착어(Target language) 문장으로 재구성하는 행위와 비슷하다. 그렇다면 벡터로 된 인코더 표현은 문장을 읽고 이해함으로써 변화된 뇌의 상태에 해당한다고 볼 수 있다. 사람이 어떤 문장을 잘 번역하기 위해서는 그 문장에 대한 이해가 뒷받침되어야 하는 것처럼, 기계 역시 원 문장이 가진 의미를 제대로 인코딩해야 향상된 성능의 번역이 가능할 것이다. 본 논문에서는 뇌과학에서 뇌 동기화(Brain-to-brain coupling)라 일컫는 현상을 모방해, 출발어와 도착어의 공통된 의미를 인코딩하여 기계 번역 성능 향상에 도움을 줄 수 있는 이중 번역 기법을 소개한다.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.27 no.1
• /
• pp.115-127
• /
• 2024

This research aimed to construct models with various structures based on the Transformer module and to perform land cover classification, thereby examining the applicability of the Transformer module. For the classification of land cover, the Unet model, which has a CNN structure, was selected as the base model, and a total of four deep learning models were constructed by combining both the encoder and decoder parts with the Transformer module. During the training process of the deep learning models, the training was repeated 10 times under the same conditions to evaluate the generalization performance. The evaluation of the classification accuracy of the deep learning models showed that the Model D, which utilized the Transformer module in both the encoder and decoder structures, achieved the highest overall accuracy with an average of approximately 89.4% and a Kappa coefficient average of about 73.2%. In terms of training time, models based on CNN were the most efficient. however, the use of Transformer-based models resulted in an average improvement of 0.5% in classification accuracy based on the Kappa coefficient. It is considered necessary to refine the model by considering various variables such as adjusting hyperparameters and image patch sizes during the integration process with CNN models. A common issue identified in all models during the land cover classification process was the difficulty in detecting small-scale objects. To improve this misclassification phenomenon, it is deemed necessary to explore the use of high-resolution input data and integrate multidimensional data that includes terrain and texture information.

• Geophysics and Geophysical Exploration
• /
• v.25 no.4
• /
• pp.227-241
• /
• 2022

Full waveform inversion (FWI) in the field of seismic data processing is an inversion technique that is used to estimate the velocity model of the subsurface for oil and gas exploration. Recently, deep learning (DL) technology has been increasingly used for seismic data processing, and its combination with FWI has attracted remarkable research efforts. For example, DL-based data processing techniques have been utilized for preprocessing input data for FWI, enabling the direct implementation of FWI through DL technology. DL-based FWI can be divided into the following methods: pure data-based, physics-based neural network, encoder-decoder, reparameterized FWI, and physics-informed neural network. In this review, we describe the theory and characteristics of the methods by systematizing them in the order of advancements. In the early days of DL-based FWI, the DL model predicted the velocity model by preparing a large training data set to adopt faithfully the basic principles of data science and apply a pure data-based prediction model. The current research trend is to supplement the shortcomings of the pure data-based approach using the loss function consisting of seismic data or physical information from the wave equation itself in deep neural networks. Based on these developments, DL-based FWI has evolved to not require a large amount of learning data, alleviating the cycle-skipping problem, which is an intrinsic limitation of FWI, and reducing computation times dramatically. The value of DL-based FWI is expected to increase continually in the processing of seismic data.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.05a
• /
• pp.441-442
• /
• 2021

본 논문에서는 고해상도의 시계열 위성영상을 딥러닝 알고리즘으로 학습하여 도시 변화탐지를 수행한다. 고해상도 위성영상을 활용한 서비스는 4 차 산업혁명 융합 신사업 중 하나인 스마트시티에 적용하여 도시 노후화, 교통 혼잡, 범죄 등 다양한 도시 문제 해결 및 효율적인 도시를 구축하는데 활용이 가능하다. 이에 본 연구에서는 도시 변화탐지를 위한 딥러닝 알고리즘으로 DeepLabV3+를 사용한다. 이는 인코더-디코더 구조로, 공간 정보를 점진적으로 회복함으로써 더욱 정확한 물체의 경계면을 찾을 수 있다. 제안하는 방법은 DeepLabV3+의 레이어와 loss function 을 수정하여 기존보다 좋은 결과를 얻었다. 객관적인 성능평가를 위해, 공개된 데이터셋 LEVIR-CD 으로 학습한 결과로 평균 IoU 는 0.87, 평균 Dice 는 0.93 을 얻었다.