• 한국인공지능학회지
• /
• 제11권4호
• /
• pp.1-8
• /
• 2023

In this paper, we propose a black ice detection platform framework using Convolutional Neural Networks (CNNs). To overcome black ice problem, we introduce a real-time based early warning platform using CNN-based architecture, and furthermore, in order to enhance the accuracy of black ice detection, we apply a multi-scale dilation convolution feature fusion (MsDC-FF) technique. Then, we establish a specialized experimental platform by using a comprehensive dataset of thermal road black ice images for a training and evaluation purpose. Experimental results of a real-time black ice detection platform show the better performance of our proposed network model compared to conventional image segmentation models. Our proposed platform have achieved real-time segmentation of road black ice areas by deploying a road black ice area segmentation network on the edge device Jetson Nano devices. This approach in parallel using multi-scale dilated convolutions with different dilation rates had faster segmentation speeds due to its smaller model parameters. The proposed MsCD-FF Net(2) model had the fastest segmentation speed at 5.53 frame per second (FPS). Thereby encouraging safe driving for motorists and providing decision support for road surface management in the road traffic monitoring department.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2014년도 춘계학술대회 논문집
• /
• pp.588-595
• /
• 2014

In this paper, analysis of the vibration Characteristic for the Mine Detectable Test Platform is described. The test platform system is the multi-sensor mine detectable vehicle. This multi-sensor mine detectable unit is more efficient detection performance than other conventional methods. The test platform system has five subsystems, the UWB(Ultra Wide Band) sensor scanner, the MD(Metal Detector) sensor scanner, the neutron sensor scanner, and the detectable vehicle. We perform the vibration tests for the test platform and analyze the vibration characteristic, such as the max displacement, the max deformation and the max Von-Misses Stress.

• 한국소음진동공학회논문집
• /
• 제24권12호
• /
• pp.927-934
• /
• 2014

In this paper, analysis of the vibration characteristic for the mine detectable test platform is described. The test platform system is the multi-sensor mine detectable vehicle. This multi-sensor mine detectable unit is more efficient detection performance than other conventional methods. The test platform system has five subsystems, the UWB(ultra wide band) sensor scanner, the MD(metal detector) sensor scanner, the ND(neutron detector) sensor scanner, and the detectable vehicle. We perform the vibration tests for the test platform and analyze the vibration characteristic, such as the max displacement, the max deformation and the max Von-Misses stress.

• 디지털콘텐츠학회 논문지
• /
• 제9권2호
• /
• pp.225-233
• /
• 2008

정보통신 기술의 급격한 발달은 u-러닝, m-러닝, t-러닝과 같은 다양한 형태의 e-러닝을 유도 하였다. 각종 기술을 통하여 학습자는 고정된 학습 공간이 아니라 다양한 환경에서 연속성을 가지고 학습할 수 있게 되었다. 이러한 다중 환경에서의 학습을 위해서는 기본적으로 웹 콘텐츠에 접속하는 장치들의 성능과 상태를 나타내는 플랫폼 정보를 획득하고 처리할 수 있는 기능이 필수적이다. 본 논문에서는 다중 환경을 지원하는 학습 시스템의 필수요소인 플랫폼 분석기의 모델을 설계하고 구현하였다. 또한 DTV를 중심으로 PC, PDA 및 휴대폰을 이용하는 다중 환경 학습 프레임워크를 제안하고, 샘플 학습 시나리오와 콘텐츠를 통하여 다중 환경 학습의 가능성을 살펴보았다.

• 인터넷정보학회논문지
• /
• 제9권3호
• /
• pp.9-24
• /
• 2008

본 논문에서는 홈 환경에서 멀티 에이전트 기반의 헬스케어 상황정보 서비스를 제공하기 위한 소프트웨어 구조의 설계에 대해 기술한다. 본 플랫폼에서의 분산객체그룹 프레임워크(Distributed Object Group Framework, DOGF)는 수행객체들 및 헬스케어 지원 센서 또는 기기들의 논리적인 서비스별 그룹화를 지원하고, 멀티 에이전트 프레임워크인 JADE(Java Agent DEvelopment framework)는 사용자의 이동성과 이질적인 환경에서의 서비스를 지원한다. 플랫폼 상의 멀티 에이전트는 건강관리 및 유지와 관련된 여러 종류의 헬스케어 상황정보 서비스를 제공하기 위하여 각각의 환경에 대한 정보를 수집하는 에이전트와 모바일 기기의 특징에 따라 서로 다른 서비스를 하는 에이전트 그리고 이들을 관리하는 에이전트로 분류할 수 있다. 이러한 JADE의 에이전트와 분산객체그룹 프레임워크의 동작은 모바일 프락시에서 인터페이스를 하며 정보 교환을 제공하거나 이동 패턴을 지원한다. 본 논문에서는 이러한 동작을 통하여 분산객체그룹 프레임워크와 JADE가 헬스케어 상황정보 서비스를 제공하기 위해 멀티 에이전트에 기반을 둔 플랫폼과 서비스별 에이전트의 설계에 대해 기술하였고, 끝으로 헬스케어 상황 정보 서비스를 위한 물리적인 시스템 환경과 플랫폼 기반의 프로토타입을 보였다.

• 대한임베디드공학회논문지
• /
• 제17권2호
• /
• pp.93-100
• /
• 2022

In this paper, we propose a machine learning-based method for supporting resource management of IoT software platforms in a multi-modal sensing scenario. We assume that an IoT device installed with a oneM2M-compatible software platform is connected with various sensors such as PIR, sound, dust, ambient light, ultrasonic, accelerometer, through different embedded system interfaces such as general purpose input output (GPIO), I2C, SPI, USB. Based on a collected dataset including CPU usage and user-defined priority, a machine learning model is trained to estimate the level of nice value required to adjust according to the resource usage patterns. The proposed method is validated by comparing with a rule-based control strategy, showing its practical capability in a multi-modal sensing scenario of IoT devices.

• Earthquakes and Structures
• /
• 제19권5호
• /
• pp.331-345
• /
• 2020

The multi-story timber structure with high platform base is one of the important architectural types in the traditional Chinese buildings. To study the dynamic characteristics and seismic responses on this kind of traditional structure, the 3-D finite element models of Xi'an drum tower which included the high platform base, upper timber structure and whole structure was established considering the structural form and material performance parameters of the structure in this study. By the modal analysis, the main frequencies and mode shapes of this kind of traditional building were obtained and investigated. The three kinds of earthquake excitations included El-Centro wave, Taft wave and Lanzhou wave were separately imposed on the upper timber structure model and the overall structure model, and the seismic responses on the tops of columns were analyzed. The results of time history analysis show that the seismic response of the upper timber structure is obviously amplified by high platform base. After considering the effect of high platform base, the mean value on the lateral displacement increments of the top column in the overall structure is more than 20.478% and the increase of dynamic coefficients was all above 0.818 under the above three different earthquake excitations. Obviously, it shows that the existence of high platform base has a negative influence on the seismic responses of upper timber structure. And the high platform base will directly affect the safety of the upper timber structure. Therefore, the influence of high platform base on the dynamic response of its upper timber structure cannot be neglected.

• 전기학회논문지P
• /
• 제65권4호
• /
• pp.326-334
• /
• 2016

In this paper for developing and implementing the realtime HD level video streaming transmitter which is operated on the multi-platform in all network and client environment compared to the exist video live streaming transmitter. We design the realtime HD level video streaming transmitter supporting the multi-platform using the TMS320DM386 video processor of T.I company and then porting the Linux kernel 2.6.29 and implementing the RTSP(Real Time Streaming Protocol)/RTP(Real Time Transport Protocol), HLS(Http Live Streaming), RTMP(Real Time Messaging Protocol) that can support the multi-platform of video stream protocol of the received equipments (smart phone, tablet PC, notebook etc.). For proving the performance of developed video streaming transmitter, we make the testing environment for testing the performance of streaming transmitter using the notebook, iPad, android Phone, and then analysis the received video in the client displayer. In this paper, we suggest the developed the Realtime HD(High Definition) level Video Streaming transmitter performance data values higher than the exist products.

• 전기전자학회논문지
• /
• 제21권3호
• /
• pp.185-194
• /
• 2017

본 논문에서는 다양한 분야와 환경에서 필요에 따라 기능을 쉽게 교환하고 조립이 가능한 모듈형 웨어러블 플랫폼을 제안한다. 제안된 플랫폼은 국내 CPU 코어 기반의 모듈형 플랫폼과 다양한 환경에 빠르게 대응하여 자유롭게 연결 가능한 plug & play 플랫폼으로 구성된다. 설계된 SoC는 32-bit RISC CPU, 32-bit symmetric multi-core processor, 그리고 16-bit DSP (CDSP)로 구성되고 여기에 필요에 따라 센서 모듈과 통신 모듈이 체인 형태로 연결된다. SoC 칩은 130nm 공정으로 개발되었고 온도와 습도 센서를 이용하여 제안된 모듈형 웨어러블 플랫폼의 기능의 동작을 검증하였다.

• Genomics & Informatics
• /
• 제5권2호
• /
• pp.83-87
• /
• 2007

The analysis of DNA microarray data is a rapidly evolving area of bioinformatics, and various types of microarray are emerging as some of the most exciting technologies for use in biological and clinical research. In recent years, microarray technology has been utilized in various applications such as the profiling of mRNAs, assessment of DNA copy number, genotyping, and detection of methylated sequences. However, the analysis of these heterogeneous microarray platform experiments does not need to be performed separately. Rather, these platforms can be co-analyzed in combination, for cross-validation. There are a number of separate laboratory information management systems (LIMS) that individually address some of the needs for each platform. However, to our knowledge there are no unified LIMS systems capable of organizing all of the information regarding multi-platform microarray experiments, while additionally integrating this information with tools to perform the analysis. In order to address these requirements, we developed a web-based LIMS system that provides an integrated framework for storing and analyzing microarray information generated by the various platforms. This system enables an easy integration of modules that transform, analyze and/or visualize multi-platform microarray data.