• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.39 no.6
• /
• pp.401-408
• /
• 2021

With the proposal of amendments to the Pedestrian Safety Act in 2021, when the amendment bill is passed in the near future, a general dimensional investigation of the sidewalks' physical condition, which is the basis of pedestrian safety, is expected to be legislated and made mandatory. Therefore, this study presented a affordable methodology for street environment survey using entry-level drones and examined the feasibility of conducting a complete survey of pedestrian paths by local governments nationwide. To this end, various street facilities in the experimental site were measured to compare and analyze the accuracy of the point cloud data. As a result of the analysis, it was found that the measurement error range satisfies the public surveying guidelines. If the methodology presented in this study is applied, it is expected that individual local governments will be able to make a significant contribution to monitoring the physical conditions of streets to improve the pedestrian environment in the near future.

• KIPS Transactions on Computer and Communication Systems
• /
• v.10 no.5
• /
• pp.137-144
• /
• 2021

The IETF has recently been standardizing the QUIC protocol for HTTP/3 services. It is noted that HTTP/3 uses QUIC as the underlying protocol, whereas HTTP/1.1 and HTTP/2 are based on TCP. Differently from TCP, the QUIC uses 0-RTT or 1-RTT transmissions to reduce the connection establishment delays of TCP and SCTP. Moreover, to solve the head-of-line blocking problem, QUIC uses the multi-streaming feature. In addition, QUIC provides various features, including the connection migration, and it is available at the Chrome browser. In this paper, we analyze the performance of QUIC for HTTP-based web and streaming services by comparing with the existing TCP and Streaming Control Transmission Protocol (SCTP) in the network environments with different link delays and packet error rates. From the experimental results, we can see that QUIC provides better throughputs than TCP and SCTP, and the gaps of performances get larger, as the link delays and packet error rates increase.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.1
• /
• pp.111-120
• /
• 2022

It is essential to make a purchase decision to make sure that the furniture matches well with other structures in the room. Moreover, in the Untact Marketing situation caused by the COVID-19 crisis, this is becoming an even more impact factor. Accordingly, methods of measuring length using AR(Augmented Reality) are emerging with the advent of AR open sources such as ARCore and ARKit for furniture arrangement interior simulation. Since this existing method using AR generates a Depth Map based on a flat camera image and it also involves complex three-dimensional calculations, limitations are revealed in work that requires the information of accurate room size using a smartphone. In this paper, we propose a method to accurately measure the size of a room using only the accelerometer and gyroscope sensors built in smartphones without using ARCore or ARKit. In addition, as an example of application using the presented technique, a method for applying a pre-designed room interior to each room is presented.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.32 no.5
• /
• pp.737-750
• /
• 2022

The demands of Unmanned Aerial Vehicles (UAVs) are growing very rapidly with the era of the 4th industrial revolution. As the technology of the UAV improved with the development of artificial intelligence and semiconductor technology, it began to be used in various civilian fields such as hobbies, bridge inspections, etc from being used for special purposes such as military use. MAVLink (Macro Air Vehicle Link), which started as an open source project, is the most widely used communication protocol between UAV and ground control station. However, MAVLink does not include any security features such as encryption/decryption mechanism, so it is vulnerable to various security threats. Therefore, in this study, the block cipher is implemented in UAV to ensure confidentiality, and the results of the encryption and decryption performance evaluation in the UAV according to various implementation methods are analyzed. In addition, we proposed the security requirements in accordance with Common Criteria, which is an international recognized ISO standard.

• KNOM Review
• /
• v.22 no.1
• /
• pp.11-19
• /
• 2019

As technology develops, the latest security threats on the network applied to users are increasing. By attacking industrial or corporate systems with malicious purposes, hackers cause many social problems such as confidential information leakage, cyber terrorism, infringement of information assets, and financial damage. Due to the complex and diversified threats, the current security personnel alone are not enough to detect and analyze all threats. In particular, the Supervisory Control And Data Acquisition (SCADA) used in industrial infrastructures that collect, analyze, and return static data 24 hours a day, 265 days a year, is very vulnerable to real-time security threats. This paper introduces security information and event management (SIEM), a powerful integrated security management system that can monitor the state of the system in real time and detect security threats. Next, we compare SIEM solutions from various companies with the open source SIEM (OSSIM) from AlienVault, which is distributed as an open source, and present cases using the OSSIM and how to utilize it.

• Smart Media Journal
• /
• v.11 no.11
• /
• pp.25-31
• /
• 2022

Deep learning-based image processing is essential for autonomous vehicles. To process road images in real-time in a System-on-Chip (SoC) environment, we need to execute deep learning models on a NPU (Neural Procesing Units) specialized for deep learning operations. In this study, we imported seven open-source image processing deep learning models, that were developed on GPU servers, to Texas Instrument Deep Learning (TIDL) NPU environment. We confirmed that the models imported in this study operate normally in the SoC virtual environment through performance evaluation and visualization. This paper introduces the problems that occurred during the migration process due to the limitations of NPU environment and how to solve them, and thereby, presents a reference case worth referring to for developers and researchers who want to port deep learning models to SoC environments.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.86-86
• /
• 2021

온천천은 부산광역시 금정구, 동래구, 연제구를 흐르는 도심 하천으로 부산 시민들의 도심 속 산책길, 자전거 길 등으로 활용되는 도시하천이다. 그러나 온천천 양안의 동래 곡저 평야가 시가지화 되고 온천천 발원지인 금정산 주변에서 무허가 상수도를 사용하고 각종 쓰레기와 하수의 유입으로 인해 하천 전체가 하수관으로 변해왔다. 이에 따라 부산광역시는 온천천 정비 계획을 시행하여 하천 정비와 함께 자동측정망을 설치하여 하천의 DO (dissolved oxygen), 탁도, TDS농도 등 자료를 수집하고 있다. 그러나 자동측정망으로 쌓여가는 데이터를 활용하여 DO농도 예측은 거의 이뤄지지 않고 있다. DO는 하천의 수질 오염 정도를 판단하는 수질인자로 역사적으로 하천 연구의 주요 연구 대상이 되어 왔다. 본 연구에서는 일 자료 뿐만 아니라 시 자료를 기반으로 RNN-LSTM 알고리즘을 활용한 DO예측을 시도하였다. RNN-LSTM은 시계열 학습에 뛰어난 알고리즘으로 인공신경망의 발전된 형태인 순환신경망이다. 연구에 앞서 부산광역시 보건환경정보 공개시스템으로부터 받은 자료 중에서 교정, 보수 중, 비사용, 장비전원단절 등으로 인해 누락데이터를 2014년 1월 1일부터 2018년 12월 31일의 데이터 전수조사 후 이상데이터를 확인하여 선형 보간하여 데이터를 사용하였다. 연구에서는 Google에서 개발한 딥러닝 오픈소스 라이브러리인 텐서플로우를 활용하여 부산광역시 금정구 부곡동에 위치한 부곡교 관측소의 DO농도를 시간 또는 일 예측을 하였다. 일 예측 학습에는 2014년~ 2018년의 기상자료(기온, 상대습도, 풍속, 강수량), DO농도 자료를 사용하였고, 시 예측 학습에는 연속된 자료가 가장 많은 2015년 3월 ~ 12월까지의 데이터를 활용하여 연구를 진행하였다. 모형의 검증을 위해 결정계수(R square)를 이용하여 통계분석을 실시하였다.

• Journal of Practical Engineering Education
• /
• v.15 no.1
• /
• pp.31-38
• /
• 2023

In this paper, a basic machine learning model learning and utilization education curriculum for non-majors is proposed, and an education method using Orange machine learning model learning and analysis tools is proposed. Orange is an open-source machine learning and data visualization tool that can create machine learning models by learning data using visual widgets without complex programming. Orange is a platform that is widely used by non-major undergraduates to expert groups. In this paper, a basic machine learning model learning and utilization education curriculum and weekly practice contents for one semester are proposed. In addition, in order to demonstrate the reality of practice contents for machine learning model learning and utilization, we used the Orange tool to learn machine learning models from categorical data samples and numerical data samples, and utilized the models. Thus, use cases for predicting the outcome of the population were proposed. Finally, the educational satisfaction of this curriculum is surveyed and analyzed for non-majors.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.97-99
• /
• 2023

소프트웨어를 개발하거나 실행하는 환경은 매우 다양하다. 최근에 혁신을 이끌고 있는 인공지능 모델은 오픈소스 프로젝트룰 통해 공개되는 코드나 라이브러리를 활용하여 구현하는 경우가 많다. 하지만 실행을 위한 환경 설치 과정이 쉽지 않고, 데이터 혹은 기학습된 모델 사이즈가 대용량일 경우에는 로컬 컴퓨터에서 실행하는 것이 불가능한 경우도 발생하고, 동료와 작업을 공유하거나 수동 배포의 어려움 등 다양한 문제에 직면한다. 이러한 문제를 해결하기 위하여, 소프트웨어가 유연하게 동작할 수 있도록 효율적인 리소스를 관리할 수 있는 컨테이너 기술을 많이 활용한다. 이 기술을 활용하는 이유는 AI 모델이 시스템에 관계없이 정확히 동일하게 재현될 수 있도록 하기 위함이다. 본 연구에서는 인공지능 모델 개발과 관련하여 코드가 실행되는 환경을 편리하게 관리하기 위하여 소프트웨어를 컨테이너화하여 배포할 수 있는 기능을 제공하는 연구소프트웨어 개발 통합 프레임워크를 제안한다.

• Convergence Security Journal
• /
• v.23 no.5
• /
• pp.81-89
• /
• 2023

Kubernetes is a representative open-source software for container orchestration, playing a crucial role in monitoring and managing resources allocated to containers. As container environments become prevalent, security threats targeting containers continue to rise, with resource exhaustion attacks being a prominent example. These attacks involve distributing malicious crypto-mining software in containerized form to hijack computing resources, thereby affecting the operation of the host and other containers that share resources. Previous research has focused on detecting resource depletion attacks, so technology to respond when attacks occur is lacking. This paper proposes a reinforcement learning-based dynamic resource management framework for detecting and responding to resource exhaustion attacks and malicious containers running in Kubernetes environments. To achieve this, we define the environment's state, actions, and rewards from the perspective of responding to resource exhaustion attacks using reinforcement learning. It is expected that the proposed methodology will contribute to establishing a robust defense against resource exhaustion attacks in container environments