• 전자통신동향분석
• /
• 제33권5호
• /
• pp.121-134
• /
• 2018

The Von Neumann based architecture of the modern computer has dominated the computing industry for the past 50 years, sparking the digital revolution and propelling us into today's information age. Recent research focus and market trends have shown significant effort toward the advancement and application of artificial intelligence technologies. Although artificial intelligence has been studied for decades since the Turing machine was first introduced, the field has recently emerged into the spotlight thanks to remarkable milestones such as AlexNet-CNN and Alpha-Go, whose neural-network based deep learning methods have achieved a ground-breaking performance superior to existing recognition, classification, and decision algorithms. Unprecedented results in a wide variety of applications (drones, autonomous driving, robots, stock markets, computer vision, voice, and so on) have signaled the beginning of a golden age for artificial intelligence after 40 years of relative dormancy. Algorithmic research continues to progress at a breath-taking pace as evidenced by the rate of new neural networks being announced. However, traditional Von Neumann based architectures have proven to be inadequate in terms of computation power, and inherently inefficient in their processing of vastly parallel computations, which is a characteristic of deep neural networks. Consequently, global conglomerates such as Intel, Huawei, and Google, as well as large domestic corporations and fabless companies are developing dedicated semiconductor chips customized for artificial intelligence computations. The AI Processor Research Laboratory at ETRI is focusing on the research and development of super low-power AI processor chips. In this article, we present the current trends in computation platform, parallel processing, AI processor, and super-threaded AI processor research being conducted at ETRI.

• 한국전기전자재료학회논문지
• /
• 제35권3호
• /
• pp.215-222
• /
• 2022

Artificial neuromorphic devices are considered the key component in realizing energy-efficient and brain-inspired computing systems. For the artificial neuromorphic devices, various material candidates and device architectures have been reported, including two-dimensional materials, metal-oxide semiconductors, organic semiconductors, and halide perovskite materials. In addition to conventional electrical neuromorphic devices, optoelectronic neuromorphic devices, which operate under a light stimulus, have received significant interest due to their potential advantages such as low power consumption, parallel processing, and high bandwidth. This article reviews the recent progress in optoelectronic neuromorphic devices using various active materials such as two-dimensional materials, metal-oxide semiconductors, organic semiconductors, and halide perovskites

• 한국정보처리학회논문지
• /
• 제7권12호
• /
• pp.3910-3922
• /
• 2000

현재의 전력시스템은 컴퓨터와 통신기술의 발전으로 분산시스템인 EMS(Energy Menagement System)/SCADA(Super-vision Control and Data Acquisition) 형태로 운영되어 전력의 생산, 전송 그리고 분배가 효과적으로 이루어지고 있다. 그러나 각 시스템은 언어, 운영체제 그리고, 통신프로토콜이 서로 다른 제품으로 구성되어 시스템간에 데이터를 교환하는데 많은 어려움이 따르고 있다. 이러한 문제를 해결하기 위해서는 미국 전력연구소는 전력제어센터간의 통신을 담당하는 ICCP(Inter-Control Center Protocol)라는 새로운 형태의 통신규약을 발표하였다. ICCP는 자동화용 표준 통신규약인 MMS(Manufacturing Message Specification)를 응용계층의 하부 규약으로 지정함으로써 서로 다른 기종의 제어센터간의 원활한 통신을 지원한다. 본 논문은 ICCP의 특징과 MMS와 ICCP가 어떻게 상호 연관되는가를 밝힌다. ICCP에서 이용하는 MMS 라이브러리(library)의 86개의 서비스 중 일부 서비스를 구현한 후 이것을 이용하여 TCP/IP 환경 하에서 ICCP의 기본이면서 핵심적인 기능을 구현한다. 그 다음, ICCP 프로토콜을 이용하여 EMS간의 통신을 모델링하고, 전력센터간의 실제 데이터 교환을 윈도우 환경 하에서 구현하여 ICCP 프로토콜의 동작과 기능을 보여준다.

• ETRI Journal
• /
• 제42권1호
• /
• pp.54-66
• /
• 2020

Nowadays, the broad availability of cameras and embedded systems makes the application of computer vision very promising as a supporting technology for intelligent transportation systems, particularly in the field of vehicle tracking. Although there are several existing trackers, the limitation of using low-cost cameras, besides the relatively low processing power in embedded systems, makes most of these trackers useless. For the tracker to work under those conditions, the video frame rate must be reduced to decrease the burden on computation. However, doing this will make the vehicle seem to move faster on the observer's side. This phenomenon is called the fast motion challenge. This paper proposes a tracker called dynamic swarm particle (DSP), which solves the challenge. The term particle refers to the particle filter, while the term swarm refers to particle swarm optimization (PSO). The fundamental concept of our method is to exploit the continuity of vehicle dynamic motions by creating dynamic models based on PSO. Based on the experiments, DSP achieves a precision of 0.896 and success rate of 0.755. These results are better than those obtained by several other benchmark trackers.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권11호
• /
• pp.4065-4083
• /
• 2021

Super-resolution can improve the clarity of low-resolution (LR) images, which can increase the accuracy of high-level compute vision tasks. Portable devices have low computing power and storage performance. Large-scale neural network super-resolution methods are not suitable for portable devices. In order to save the computational cost and the number of parameters, Lightweight image processing method can improve the processing speed of portable devices. Therefore, we propose the Enhanced Information Multiple Distillation Network (EIMDN) to adapt lower delay and cost. The EIMDN takes feedback mechanism as the framework and obtains low level features through high level features. Further, we replace the feature extraction convolution operation in Information Multiple Distillation Block (IMDB), with Ghost module, and propose the Enhanced Information Multiple Distillation Block (EIMDB) to reduce the amount of calculation and the number of parameters. Finally, coordinate attention (CA) is used at the end of IMDB and EIMDB to enhance the important information extraction from Spaces and channels. Experimental results show that our proposed can achieve convergence faster with fewer parameters and computation, compared with other lightweight super-resolution methods. Under the condition of higher peak signal-to-noise ratio (PSNR) and higher structural similarity (SSIM), the performance of network reconstruction image texture and target contour is significantly improved.

• International Journal of Fluid Machinery and Systems
• /
• 제4권1호
• /
• pp.179-190
• /
• 2011

The swirling flow developing in Francis turbine draft tube under part load operation leads to pressure fluctuations usually in the range of 0.2 to 0.4 times the runner rotational frequency resulting from the so-called vortex breakdown. For low cavitation number, the flow features a cavitation vortex rope animated with precession motion. Under given conditions, these pressure fluctuations may lead to undesirable pressure fluctuations in the entire hydraulic system and also produce active power oscillations. For the upper part load range, between 0.7 and 0.85 times the best efficiency discharge, pressure fluctuations may appear in a higher frequency range of 2 to 4 times the runner rotational speed and feature modulations with vortex rope precession. It has been pointed out that for this particular operating point, the vortex rope features elliptical cross section and is animated of a self-rotation. This paper presents an experimental investigation focusing on this peculiar phenomenon, defined as the upper part load vortex rope. The experimental investigation is carried out on a high specific speed Francis turbine scale model installed on a test rig of the EPFL Laboratory for Hydraulic Machines. The selected operating point corresponds to a discharge of 0.83 times the best efficiency discharge. Observations of the cavitation vortex carried out with high speed camera have been recorded and synchronized with pressure fluctuations measurements at the draft tube cone. First, the vortex rope self rotation frequency is evidenced and the related frequency is deduced. Then, the influence of the sigma cavitation number on vortex rope shape and pressure fluctuations is presented. The waterfall diagram of the pressure fluctuations evidences resonance effects with the hydraulic circuit. The influence of outlet bubble cavitation and air injection is also investigated for low cavitation number. The time evolution of the vortex rope volume is compared with pressure fluctuations time evolution using image processing. Finally, the influence of the Froude number on the vortex rope shape and the associated pressure fluctuations is analyzed by varying the rotational speed.

• 전자통신동향분석
• /
• 제33권5호
• /
• pp.103-110
• /
• 2018

Studies on artificial intelligence have been developed for the past couple of decades. After a few periods of prosperity and recession, a new machine learning method, so-called Deep Learning, has been introduced. This is the result of high-quality big- data, an increase in computing power, and the development of new algorithms. The main targets for deep learning are 1D audio and 2D images. The application domain is being extended from a discriminative model, such as classification/segmentation, to a generative model. Currently, deep learning is used for processing 3D data. However, unlike 2D, it is not easy to acquire 3D learning data. Although low-cost 3D data acquisition sensors have become more popular owing to advances in 3D vision technology, the generation/acquisition of 3D data remains a very difficult problem. Moreover, it is not easy to directly apply an existing network model, such as a convolution network, owing to the variety of 3D data representations. In this paper, we summarize the 3D deep learning technology that have started to be developed within the last 2 years.

• 지능정보연구
• /
• 제17권1호
• /
• pp.53-69
• /
• 2011

로봇 자체 또는 로봇에 탑재된 콘텐츠와의 상호작용을 위해 일반적으로 영상 또는 음성 인식 기술이 사용된다. 그러나 영상 음성인식 기술은 아직까지 기술 및 환경 측면에서 해결해야 할 어려움이 존재하며, 실적용을 위해서는 사용자의 협조가 필요한 경우가 많다. 이로 인해 로봇과의 상호작용은 터치스크린 인터페이스를 중심으로 개발되고 있다. 향후 로봇 서비스의 확대 및 다양화를 위해서는 이들 영상 음성 중심의 기존 기술 외에 상호보완적으로 활용이 가능한 인터페이스 기술의 개발이 필요하다. 본 논문에서는 로봇 인터페이스 활용을 위한 가속도 센서 기반의 제스처 인식 기술의 개발에 대해 소개한다. 본 논문에서는 비교적 어려운 문제인 26개의 영문 알파벳 인식을 기준으로 성능을 평가하고 개발된 기술이 로봇에 적용된 사례를 제시하였다. 향후 가속도 센서가 포함된 다양한 장치들이 개발되고 이들이 로봇의 인터페이스로 사용될 때 현재 터치스크린 중심으로 된 로봇의 인터페이스 및 콘텐츠가 다양한 형태로 확장이 가능할 것으로 기대한다.