• 대한임베디드공학회논문지
• /
• 제18권3호
• /
• pp.89-100
• /
• 2023

Recently, on-device artificial intelligence (AI) solutions using mobile devices and embedded edge devices have emerged in various fields, such as computer vision, to address network traffic burdens, low-energy operations, and security problems. Although vision transformer deep learning models have outperformed conventional convolutional neural network (CNN) models in computer vision, they require more computations and parameters than CNN models. Thus, they are not directly applicable to embedded edge devices with limited hardware resources. Many researchers have proposed various model compression methods or lightweight architectures for vision transformers; however, there are only a few studies evaluating the effects of model compression techniques of vision transformers on performance. Regarding this problem, this paper presents a performance evaluation of vision transformers on embedded platforms. We investigated the behaviors of three vision transformers: DeiT, LeViT, and MobileViT. Each model performance was evaluated by accuracy and inference time on edge devices using the ImageNet dataset. We assessed the effects of the quantization method applied to the models on latency enhancement and accuracy degradation by profiling the proportion of response time occupied by major operations. In addition, we evaluated the performance of each model on GPU and EdgeTPU-based edge devices. In our experimental results, LeViT showed the best performance in CPU-based edge devices, and DeiT-small showed the highest performance improvement in GPU-based edge devices. In addition, only MobileViT models showed performance improvement on EdgeTPU. Summarizing the analysis results through profiling, the degree of performance improvement of each vision transformer model was highly dependent on the proportion of parts that could be optimized in the target edge device. In summary, to apply vision transformers to on-device AI solutions, either proper operation composition and optimizations specific to target edge devices must be considered.

• International Journal of Advanced Culture Technology
• /
• 제10권4호
• /
• pp.569-575
• /
• 2022

This is a research thesis on the development of a monitoring and diagnosis device that prevents the risk of an accident through monitoring and diagnosis of a modular Main Circuit Breaker (MCB) using Vacuum Interrupter (VI) for Korean high-speed rolling stock. In this paper, a comprehensive MCB monitoring and diagnosis was performed by converging vacuum level diagnosis of interrupter, operating coil monitoring of MCB and environmental temperature/humidity monitoring of modular box. In addition, to develop an algorithm that is expected to have a similar data processing before the actual field test of the MCB monitoring and diagnosis device in 2023, the cluster analysis and factor analysis were performed using the WEKA data mining technique on the big data of Korean railroad transformer, which was previously researched by Tae Hee Evolution with KORAIL.

• 한국항해항만학회지
• /
• 제47권2호
• /
• pp.57-65
• /
• 2023

In 2017, the International Maritime Organization (IMO) adopted MSC.428 (98), which recommends establishing a cyber-risk management system in Ship Safety Management Systems (SMSs) from January 2021. The 27th International Association of Marine Aids to Navigation and Lighthouse Authorities (IALA) also discussed prioritizing cyber-security (cyber-risk management) in developing systems to support Maritime Autonomous Surface Ship (MASS) operations (IALA guideline on developments in maritime autonomous surface ships). In response to these international discussions, Korea initiated the Korea Autonomous Surface Ship technology development project (KASS project) in 2020. Korea has been carrying out detailed tasks for cybersecurity technology development since 2021. This paper outlines the basic concept of ship network security equipment for supporting MASS ship operation in detailed task of cybersecurity technology development and defines ship network security equipment interface for MASS ship applications.

• 한국정보전자통신기술학회논문지
• /
• 제13권6호
• /
• pp.470-476
• /
• 2020

본 논문에서는 인공지능 블록을 구동할 수 있도록 Edge Device와 서버를 분리하는 영상 딥러닝용 모듈 교체형 인공지능 서버의 설계와 데이터 송수신 방법을 제시한다. 영상 딥러닝용 모듈 교체형 인공지능 서버를 통해 소셜 로봇과 로봇의 플랫폼이 구동될 Edge Device 간의 종속성을 줄여 구동 안정성을 향상할 수 있다. 사용자가 소셜 로봇과의 상호작용을 위해서 인공지능 서버에 기능을 요청하면 모듈화된 기능들을 이용해 결과만을 반환받을 수 있다. 인공지능 서버에서 모듈화되어있는 기능들은 서버 관리자에 의해 모듈별로 유지 보수 및 변경이 쉽게 가능하다. 기존 서버 시스템과 비교했을 때 모듈 교체형 인공지능 서버는 수행되는 프로그램의 규모 차이와 서버 유지 보수 면에서 더 효율적인 성능을 낸다. 이를 통해 사람-로봇 간의 상호작용이 가능한 로봇 시나리오에 더 다양한 영상 딥러닝을 포함 시킬 수 있으며, 로봇 플랫폼 외에 영상 딥러닝을 위한 인공지능 서버에 적용할 때 더 효율적인 성능을 낼 수 있다.

• 디지털정책학회지
• /
• 제1권2호
• /
• pp.1-8
• /
• 2022

위험상황 발생 시 비상구 역할을 하는 옥상은 소방법상 화재를 대비하여 개방되어 있어야 한다. 그러나, 옥상 문 개방 시 불법적인 출입, 범죄, 투신 등 각종 사건, 사고 발생의 장소가 되고 있다. 그러다보니, 범죄 및 각종 사건, 사고의 방지를 위하여 시설관리 측면에서 옥상 출입문을 폐쇄하는 것이 현실이다. 이에, 정부에서는 옥상 출입문에 전자식 자동개폐장치 설치를 의무화하는 내용의 주택건설기준 등에 관한 규정 입법을 추진하고 있다. 따라서, 본 논문에서는 지능형 비상문 개폐장치 시스템을 제안하였다. 이를 위해, 유무선 출입 카운팅 및 AI 영상분석을 연동하여 지능형 비상문 개폐 시스템을 개발하였다. 최종적으로, 장치상태 실시간 모니터링 이벤트 알람 중앙 집적 방식의 원격제어 및 이력관리 기능이 제공되는 무선통신 기반 통합 관리 플랫폼 구축이 가능하다.

• 전자통신동향분석
• /
• 제35권2호
• /
• pp.1-16
• /
• 2020

"Artificial Intelligence+" is a key strategic direction that has garnered the attention of several global medical device manufacturers and internet companies. Large hospitals are actively involved in different types of medical AI research and cooperation projects. Medical AI is expected to create numerous opportunities and advancements in areas such as medical imaging, computer aided diagnostics and clinical decision support, new drug development, personal healthcare, pathology analysis, and genetic disease prediction. On the contrary, some studies on the limitations and problems in current conditions such as lack of clinical validation, difficulty in performance comparison, lack of interoperability, adversarial attacks, and computational manipulations are being published. Overall, the medical AI field is in a paradigm shift. Regarding international standardization, the work on the top 10 standardization issues is witnessing rapid progress and the competition for standard development has become fierce.

• 한국멀티미디어학회논문지
• /
• 제25권2호
• /
• pp.345-352
• /
• 2022

Conducting AI-based fusion business due to the increment of ICT fusion medical device has been expanded. In addition, AI-based medical devices help change existing medical system on treatment into the paradigm of customized treatment such as preliminary diagnosis and prevention. It will be generally promoted to the change of medical device industry. Although the current demand forecasting of medical biotechnology commercialization is based on the method of Delphi and AHP, there is a problem that it is difficult to have a generalization due to fluctuation results according to a pool of participants. Therefore, the purpose of the paper is to predict demand forecasting for identifying promising technology based on building up big data in medical biotechnology. The development method is to employ candidate technologies of keywords extracted from SCOPUS and to use word2vec for drawing analysis indicator, technological distance similarity, and recommended technological similarity of top-level items in order to achieve a reasonable result. In addition, the method builds up academic big data for 5 years (2016-2020) in order to commercialize technology excavation on demand perspective. Lastly, the paper employs global data studies in order to develop domestic and international demand for technology excavation in the medical biotechnology field.

• ETRI Journal
• /
• 제44권5호
• /
• pp.859-874
• /
• 2022

This paper presents a low-power and lightweight human body communication (HBC) receiver with an embedded dummy electrode for improved signal acquisition. The clock data recovery (CDR) circuit in the receiver operates with a low supply voltage and utilizes a clock phase inversion scheme. The receiver is equipped with a main electrode and dummy electrode that strengthen the capacitive-coupled signal at the receiver frontend. The receiver CDR circuit exploits a clock inversion scheme to allow 0.9-V operation while achieving a shorter lock time than at 3.3-V operation. In experiments, a receiver chip fabricated using 130-nm complementary metal-oxide-semiconductor technology was demonstrated to successfully receive the transmitted signal when the transmitter and receiver are placed separately on each hand of the user while consuming only 4.98 mW at a 0.9-V supply voltage.

• 4차산업연구
• /
• 제4권1호
• /
• pp.19-27
• /
• 2024

Purpose: This study is based on the results of the National Information Society Agency's the Report on the Digital Divide in 2022. This study sought to develop digital literacy education programs for senior citizens, a digitally disadvantaged group, and to utilize smart devices to enhance their digital capabilities. Research design, data and methodology: Based on Gagné's nine events of instruction, a total of 7-session educational programs using smart devices were developed, and teaching-learning goals were set at a level that older learners can realistically perform. In preparation for the era of digital transformation, AI utilization methods are introduced and utilized in some sessions of the educational program. Results: Among a total of 7 sessions of the educational program, 5 sessions using KakaoTalk and Naver App, and 2 sessions using other apps were developed. There are a total of three sessions using AI. Conclusions: This study presented a digital literacy education program that combined AI, addressing the insufficiency of AI-based education programs targeting senior citizens. It is expected that this educational program will be able to improve the digital literacy skills and provide a basis for fulfilling their responsibilities as digital citizens by suggesting a direction for AI utilization education for senior citizens.

• 디지털산업정보학회논문지
• /
• 제17권3호
• /
• pp.17-26
• /
• 2021

Internet of Things is an infrastructure of the interconnected devices. In Internet of Things environment, many smart devices are used in daily life. It requires a new retrieval method using multiple devices. We propose an information retrieval method using both device characteristics and device usage characteristics in multi-device environments. Firstly, information retrieval is performed using a general purpose device. And then, it is performed using dedicated devices. Our method uses both characteristics of the devices and usage characteristics of them. Moreover, it considers queries on the general purpose device. This paper proposes a new retrieval method and describes algorithms. Then, it presents smart home scenarios. Performance evaluation is performed using the scenarios. The evaluation results show higher precision and efficiency than previous researches. The proposed method gets information more accurately and quickly in IOT multiple device environments.