• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.30 no.6
• /
• pp.1031-1041
• /
• 2020

Cloud computing technology plays an important role in the deep learning industry as deep learning services are deployed frequently on top of cloud infrastructures. In such cloud environment, virtualization technology provides logically independent and isolated computing space for each tenant. However, recent studies demonstrate that by leveraging vulnerabilities of virtualization techniques and shared processor architectures in the cloud system, various side-channels can be established between cloud tenants. In this paper, we propose a novel attack scenario that can steal internal information of deep learning models by exploiting the Meltdown vulnerability in a multi-tenant system environment. On the basis of our experiment, the proposed attack method could extract internal information of a TensorFlow deep-learning service with 92.875% accuracy and 1.325kB/s extraction speed.

• Journal of Broadcast Engineering
• /
• v.26 no.6
• /
• pp.748-757
• /
• 2021

In this paper, we propose a gaze estimation network in which eye landmark position detection and gaze direction vector estimation are integrated into one deep learning network. The proposed network uses the Stacked Hourglass Network as a backbone structure and is largely composed of three parts: a landmark detector, a feature map extractor, and a gaze direction estimator. The landmark detector estimates the coordinates of 50 eye landmarks, and the feature map extractor generates a feature map of the eye image for estimating the gaze direction. And the gaze direction estimator estimates the final gaze direction vector by combining each output result. The proposed network was trained using virtual synthetic eye images and landmark coordinate data generated through the UnityEyes dataset, and the MPIIGaze dataset consisting of real human eye images was used for performance evaluation. Through the experiment, the gaze estimation error showed a performance of 3.9, and the estimation speed of the network was 42 FPS (Frames per second).

• The Journal of Information Systems
• /
• v.31 no.4
• /
• pp.145-160
• /
• 2022

Purpose The purpose of this study is to develop a virtual try-on deep learning model that can efficiently learn front and back clothes images. It is expected that the application of virtual try-on clothing service in the fashion and textile industry field will be vitalization. Design/methodology/approach The data used in this study used 232,355 clothes and product images. The image data input to the model is divided into 5 categories: original clothing image and wearer image, clothing segmentation, wearer's body Densepose heatmap, wearer's clothing-agnosting. We advanced the HR-VITON model in the way of Mixed-Precison, Gradient Accumulation, and sharing model weights. Findings As a result of this study, we demonstrated that the weight-shared MP-GA HR-VITON model can efficiently learn front and back fashion images. As a result, this proposed model quantitatively improves the quality of the generated image compared to the existing technique, and natural fitting is possible in both front and back images. SSIM was 0.8385 and 0.9204 in CP-VTON and the proposed model, LPIPS 0.2133 and 0.0642, FID 74.5421 and 11.8463, and KID 0.064 and 0.006. Using the deep learning model of this study, it is possible to naturally fit one color clothes, but when there are complex pictures and logos as shown in <Figure 6>, an unnatural pattern occurred in the generated image. If it is advanced based on the transformer, this problem may also be improved.

• KNOM Review
• /
• v.24 no.1
• /
• pp.1-12
• /
• 2021

VM (Virtual Machine) live migration is a server virtualization technique for deploying a running VM to another server node while minimizing downtime of a service the VM provides. Currently, in cloud data centers, VM live migration is widely used to apply load balancing on CPU workload and network traffic, to reduce electricity consumption by consolidating active VMs into specific location groups of servers, and to provide uninterrupted service during the maintenance of hardware and software update on servers. It is critical to use VMlive migration as a prevention or mitigation measure for possible failure when its indications are detected or predicted. In this paper, we propose two VNF live migration methods; one for predictive load balancing and the other for a proactive measure in failure. Both need machine learning models that learn periodic monitoring data of resource usage and logs from servers and VMs/VNFs. We apply the second method to a vEPC (Virtual Evolved Pakcet Core) failure scenario to provide a detailed case study.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.521-521
• /
• 2023

지방상수도의 관 파손사고 감지 및 누수관리 방법에는 블록시스템 구축을 통한 소블록별 야간최소유량 감시방법이 가장 대표적이다. 야간최소유량은 새벽 2시와 4시 사이의 인구 활동 비율이 가장 낮은 새벽 시간대에 소블록에 공급된 유량을 의미하며, 대부분 유량 성분은 누수량일 것이라는 가정에서 출발한다. 그러나 아파트 중심의 주거 형태를 보이는 도심지의 경우, 새벽 시간대에도 다량의 물수요가 비정기적으로 발생하고 있어 관망의 이상 여부를 감시하기 위한 관리기준으로서 야간최소유량을 이용하기에는 높은 일간 변동성에 따른 한계가 있다고 할 수 있다. 즉, 야간최소유량은 관 파손사고 발생의 감시보다는 관로 연결 또는 급수전 분기 부위에서 발생하는 미량의 누수가 수개월에 걸쳐 누적되는 장기추세를 분석하여 누수탐사반의 투입 시점을 결정하기 위한 근거를 제시하기 위한 목적으로 사용되며, 아직까지 관 파손사고의 발생은 자체적인 감지보다는 민원에 의해 인지되는 경우가 많다. 최근, 스마트관망 구축사업(SWM) 등을 통해 관 파손 및 누수 감지를 위한 청음식 누수감지센서가 소블록 내 도입되고 있으나, 초기 시설투자에 큰 비용이 수반되며 주변 소음과 배터리 전원방식의 한계로 인하여 새벽 시간대에만 분석이 제한적으로 적용되는 경우가 많아 이 역시도 상시적인 관 파손사고의 감시기술이라 보기는 어렵다. 본 연구에서는 소블록 유입점에서의 유량·압력과 소블록 내에 설치된 대수용가 스마트미터, 그리고 사고감지를 위한 수압계 사이의 평상시 수리적 균형을 학습한 DNN(Deep Neural Network) 모델을 이용하여 관 파손사고를 실시간 감지하는 모델 개발연구를 수행하였다. 모델은 관 파손사고 감지를 위한 수압계의 최적 위치와 대수를 결정하기 위한 모듈과 관 파손사고 감지모듈로 구성되며, 1개 소블록 Test-Bed를 구축하여 모델을 생성하고 PDD 관망해석 모델을 통해 생성된 가상의 사고에 대한 감지 여부로서 개발 모델의 감지성능을 평가하였다.

• Journal of Platform Technology
• /
• v.11 no.2
• /
• pp.15-25
• /
• 2023

This study analyzes the limitations of the insider threat datasets used for insider threat detection research and compares and analyzes the solution-based insider threat data with public insider threat data using a security solution to overcome this. Through this, we design a data format suitable for insider threat detection and implement a system that can safely share insider threat information between different institutions and companies using blockchain technology. Currently, there is no dataset collected based on actual events in the insider threat dataset that is revealed to researchers. Public datasets are virtual synthetic data randomly created for research, and when used as a learning model, there are many limitations in the real environment. In this study, to improve these limitations, a private blockchain was designed to secure information sharing between institutions of different affiliations, and a method was derived to increase reliability and maintain information integrity and consistency through agreement and verification among participants. The proposed method is expected to collect data through an outflow threat collector and collect quality data sets that posed a threat, not synthetic data, through a blockchain-based sharing system, to solve the current outflow threat dataset problem and contribute to the insider threat detection model in the future.

• Journal of Broadcast Engineering
• /
• v.27 no.1
• /
• pp.20-30
• /
• 2022

Recently, as VR (Virtual Reality) technology has been in the spotlight, 360° panoramic images that can view lively VR contents are attracting a lot of attention. Image stitching technology is a major technology for producing 360° panorama images, and many studies are being actively conducted. Typical stitching algorithms are based on feature point-based image stitching. However, conventional feature point-based image stitching methods have a problem that stitching results are intensely affected by feature points. To solve this problem, deep learning-based image stitching technologies have recently been studied, but there are still many problems when there are few overlapping areas between images or large parallax. In addition, there is a limit to complete supervised learning because labeled ground-truth panorama images cannot be obtained in a real environment. Therefore, we produced three fisheye images with different camera centers and corresponding ground truth image through carla simulator that is widely used in the autonomous driving field. We propose image stitching model that creates a 360° panorama image with the produced fisheye image. The final experimental results are virtual datasets configured similar to the actual environment, verifying stitching results that are strong against various environments and large parallax.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.8 no.9
• /
• pp.647-658
• /
• 2018

Modern society is complex and rapidly changing, and knowledge sharing is needed to acquire and create knowledge. Knowledge sharing is the act of providing information knowledge and know-how of their own in order to cooperate with or help their colleagues. This study presents a research model using social capital theory to explain the mobile knowledge sharing behavior of virtual community members. Based on previous studies, social capital theory is divided into structural, relational, and cognitive aspects. It was composed of social interaction ties as a measure of structural aspect, trust as a measure of cognitive aspect, shared language, shared vision and relational aspect. After collecting survey data, factor analysis and regression analysis were performed using SPSS 22. In this way, we examined how the detailed factors of social capital affect information sharing behavior and how the level of knowledge sharing affects community promotion. The results showed that social interaction ties, shared language, shared vision, and trust affect knowledge sharing. Knowledge sharing has had a positive impact on community promotion.

• Journal of the Korea Computer Graphics Society
• /
• v.29 no.3
• /
• pp.79-84
• /
• 2023

Image generation models have been applied in various fields to overcome data sparsity, time and cost issues. However, it has limitations in generating images from regular pattern images and detecting defects in such data. In this paper, we verified the feasibility of the image generation model to generate pattern images and applied it to data augmentation for defect detection of OLED panels. The data required to train an OLED defect detection model is difficult to obtain due to the high cost of OLED panels. Therefore, even if the data set is obtained, it is necessary to define and classify various defect types. This paper introduces an OLED panel defect data acquisition system that acquires a hypothetical data set and augments the data with an image generation model. In addition, the difficulty of generating pattern images in the diffusion model is identified and a possibility is proposed, and the limitations of data augmentation and defect detection data augmentation using the image generation model are improved.

• Convergence Security Journal
• /
• v.22 no.5
• /
• pp.107-113
• /
• 2022

The AR/VR-based education and training system is expected to contribute greatly to accident prevention and budget reduction as well as practical training effects similar to the battlefield environment. Research to use AR/VR for learning is ongoing, and technology can be improved without experiencing failures that can occur in the real world. Major advanced countries in defense recognized the advantages of AR/VR technology early on, and developed and utilized systems using them in various fields, from mastery of individual weapon system operation to comprehensive combat training systems, war history education, and post-traumatic stress treatment. Therefore, the purpose of this study is to examine the cases of AR/VR application education and training in advanced defense countries and to draw implications for the South Korean military.