• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.12
• /
• pp.1919-1925
• /
• 2022

In this paper, we propose an image-to-image translation approach based on the conditional generative adversarial network for semantic segmentation. Semantic segmentation is the task of clustering parts of an image together which belong to the same object class. Unlike the traditional pixel-wise classification approach, the proposed method parses an input RGB image to its corresponding semantic segmentation mask using a pixel regression approach. The proposed method is based on the Pix2Pix image synthesis method. We employ residual connections-based convolutional neural network architectures for both the generator and discriminator architectures, as the residual connections speed up the training process and generate more accurate results. The proposed method has been trained and tested on the NYU-depthV2 dataset and could achieve a good mIOU value (49.5%). We also compare the proposed approach to the current methods in semantic segmentation showing that the proposed method outperforms most of those methods.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.12
• /
• pp.1588-1594
• /
• 2020

In this paper, we discuss a data balancing method for learning a neural network that generates digital holograms using a deep neural network (DNN). Deep neural networks are based on deep learning (DL) technology and use a generative adversarial network (GAN) series. The fringe pattern, which is the basic unit of a hologram to be created through a deep neural network, has very different data types depending on the hologram plane and the position of the object. However, because the criteria for classifying the data are not clear, an imbalance in the training data may occur. The imbalance of learning data acts as a factor of instability in learning. Therefore, it presents a method for classifying and balancing data for which the classification criteria are not clear. And it shows that learning is stabilized through this.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.22 no.5
• /
• pp.515-528
• /
• 2020

Underground concrete structures are usually designed to be used for decades, but in recent years, many of them are nearing their original life expectancy. As a result, it is necessary to promptly inspect and repair the structure, since it can cause lost of fundamental functions and bring unexpected problems. Therefore, personnel-based inspections and repairs have been underway for maintenance of underground structures, but nowadays, objective inspection technologies have been actively developed through the fusion of deep learning and image process. In particular, various researches have been conducted on developing a concrete crack detection algorithm based on supervised learning. Most of these studies requires a large amount of image data, especially, label images. In order to secure those images, it takes a lot of time and labor in reality. To resolve this problem, we introduce a method to increase the accuracy of crack area detection, improved by 0.25% on average by applying adversarial learning in this paper. The adversarial learning consists of a segmentation neural network and a discriminator neural network, and it is an algorithm that improves recognition performance by generating a virtual label image in a competitive structure. In this study, an efficient deep neural network learning method was proposed using this method, and it is expected to be used for accurate crack detection in the future.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.6
• /
• pp.622-628
• /
• 2020

This paper proposes a music generation technique that automatically generates trot music using a Generative Adversarial Network (GAN) model composed of a Recurrent Neural Network (RNN). The proposed method uses a method of creating a chord as a skeleton of the music, creating a melody and bass in stages based on the chord progression made, and attaching it to the corresponding chord to complete the structured piece. Also, a new chorus chord progression is created from the verse chord progression by applying the characteristics of a trot song that repeats the structure divided into an individual section, such as intro, verse, and chorus. And it extends the length of the created trot. The quality of the generated music was specified using subjective evaluation and objective evaluation methods. It was confirmed that the generated music has similar characteristics to the existing trot.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.420-422
• /
• 2022

인공지능 기술이 문서 분류, 얼굴 인식, 자율 주행 등 실생활 전반에 걸쳐 다양한 분야에 적용됨에 따라, 인공지능 모델에 대한 취약점을 미리 파악하고 대비하는 기술의 중요성이 높아지고 있다. 이미지 영역에서는 입력 데이터에 작은 섭동을 추가해 신경망을 속이는 방법인 적대적 공격 연구가 활발하게 이루어졌지만, 텍스트 영역에서는 텍스트 데이터의 이산적인 특징으로 인해 연구에 어려움이 존재한다. 본 논문은 텍스트 분야 인공지능 기술에 대한 적대적 공격 기법을 분석하고 연구의 필요성을 살펴보고자 한다.

• The Journal of the Acoustical Society of Korea
• /
• v.40 no.2
• /
• pp.161-168
• /
• 2021

In this paper, a Silent Speech Interface (SSI) technology was proposed in which Doppler frequency shifts of the reflected signal were used to synthesize the speech signals when 40kHz ultrasonic signal was incident to speaker's mouth region. In SSI, the mapping rules from the features derived from non-speech signals to those from audible speech signals was constructed, the speech signals are synthesized from non-speech signals using the constructed mapping rules. The mapping rules were built by minimizing the overall errors between the estimated and true speech parameters in the conventional SSI methods. In the present study, the mapping rules were constructed so that the distribution of the estimated parameters is similar to that of the true parameters by using Generative Adversarial Networks (GAN). The experimental result using 60 Korean words showed that, both objectively and subjectively, the performance of the proposed method was superior to that of the conventional neural networks-based methods.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.185-185
• /
• 2021

기후변화로 인한 돌발 강우 등 이상 기후 현상이 증가함에 따라 정확한 강우예측의 중요성은 더 증가하는 추세이다. 전통적인 강우예측의 경우 기상수치모델 또는 외삽법을 이용한 레이더 기반 강우예측 기법을 이용하며, 최근 머신러닝 기술의 발달에 따라 이를 활용한 레이더 자료기반 강우예측기법이 개발되고 있다. 기존 머신러닝을 이용한 강우예측 모델의 경우 주로 시계열 이미지 예측에 적합한 2차원 순환 신경망 기반 기법(Convolutional Long Short-Term Memory, ConvLSTM) 또는 합성곱 신경망 기반 기법(Convolutional Neural Network(CNN) Encoder-Decoder) 등을 이용한다. 본 연구에서는 생성적 적대 신경망 기반 기법(Generative Adversarial Network, GAN)을 이용해 미래 강우예측을 수행하도록 하였다. GAN 방법론은 이미지를 생성하는 생성자와 이를 실제 이미지와 구분하는 구별자가 경쟁하며 학습되어 현재 이미지 생성 분야에서 높은 성능을 보여주고 있다. 본 연구에서 개발한 GAN 기반 모델은 기상청에서 제공된 2016년~2019년까지의 레이더 이미지 자료를 이용하여 초단기, 단기 강우예측을 수행하도록 학습시키고, 2020년 레이더 이미지 자료를 이용해 단기강우예측을 모의하였다. 또한, 기존 머신러닝 기법을 기반으로 한 모델들의 강우예측결과와 GAN 기반 모델의 강우예측결과를 비교분석한 결과, 본 연구를 통해 개발한 강우예측모델이 단기강우예측에 뛰어난 성능을 보이는 것을 확인할 수 있었다.

• Electronics and Telecommunications Trends
• /
• v.35 no.4
• /
• pp.91-102
• /
• 2020

Recently, generative adversarial networks (GANs) is a field of research that has rapidly emerged wherein many studies conducted shows overwhelming results. Initially, this was at the level of imitating the training dataset. However, the GAN is currently useful in many fields, such as transformation of data categories, restoration of erased parts of images, copying facial expressions of humans, and creation of artworks depicting a dead painter's style. Although many outstanding research achievements have been attracting attention recently, GANs have encountered many challenges. First, they require a large memory facility for research. Second, there are still technical limitations in processing high-resolution images over 4K. Third, many GAN learning methods have a problem of instability in the training stage. However, recent research results show images that are difficult to distinguish whether they are real or fake, even with the naked eye, and the resolution of 4K and above is being developed. With the increase in image quality and resolution, many applications in the field of design and image and video editing are now available, including those that draw a photorealistic image as a simple sketch or easily modify unnecessary parts of an image or a video. In this paper, we discuss how GANs started, including the base architecture and latest technologies of GANs used in high-resolution, high-quality image creation, image and video editing, style translation, content transfer, and technology.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.01a
• /
• pp.17-18
• /
• 2023

비지도형 도메인 적응(UDA)에 대한 최근 연구는 도메인 적응에 대한 설명 및 전이 가능한 특징을 풀어 내기 위해 적대적 학습에 의존한다. 그러나 기존 방법에는 대상 도메인의 클래스 인식(class-aware) 정보를 고려하지 않고는 잠재 공간의 구별 가능성을 완전히 보장할 수 없다는 것과 소스 및 대상 도메인의 샘플만으로는 잠재 공간에서 도메인 불변(domain- invariant) 특성을 추출하기에 부족하다는 두 가지 문제가 있다고 알려져 있다. 본 논문에서는 기존 알려진 UDA의 도메인 적응시 발생되는 문제를 해결하기 위해 Adversarial Discriminative Domain Adaptation(ADDA)에서 mixup을 활용해 신경망의 로버스트네스를 향상시키는 것을 확인하였다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.6
• /
• pp.700-705
• /
• 2020

Generative adversarial networks are methods of generating images by opposing two neural networks. When generating the image, randomly generated noise is rearranged to generate the image. The image generated by this method is not generated well depending on the noise, and it is difficult to generate a proper image when the number of pixels of the image is small In addition, the speed and size of data accumulation in data classification increases, and there are many difficulties in labeling them. In this paper, to solve this problem, we propose a technique to generate noise based on random noise using real data. Since the proposed system generates an image based on the existing image, it is confirmed that it is possible to generate a more natural image, and if it is used for learning, it shows a higher hit rate than the existing method using the hostile neural network respectively.