• 한국컴퓨터정보학회논문지
• /
• 제29권8호
• /
• pp.157-164
• /
• 2024

본 논문에서는 제한된 자원을 가진 보드에서 딥러닝 기반 검출기 구축에 대한 실현 가능성에 대해 논의한다. 많은 연구에서 고성능 GPU 환경에서 검출기를 평가하지만, 제한된 연산 자원을 가진 보드에서의 평가는 여전히 미비하다. 따라서 본 연구에서는 검출기를 파싱하고 최적화하는 것으로 보드에 딥러닝 기반 검출기를 구현하고 구축한다. 제한된 자원에서의 딥러닝 기반 검출기의 성능을 확인하기 위해, 여러 검출기를 다양한 하드웨어 자원에서 모니터링하고, COCO 검출 데이터 셋에서 On-Board에서의 검출 모델과 On-GPU의 검출 모델을 mAP, 전력 소모량, 실행 속도(FPS) 관점으로 비교 및 분석한다. 그리고 군사 분야에 검출기를 적용한 효과를 고려하기 위해 항공 전투 시나리오를 고려할 수 있는 열화상 이미지로 구성된 자체 데이터 셋에서 검출기를 평가한다. 결과적으로 우리는 본 연구를 통해 On-Board에서 모델을 실행하는 딥러닝 기반 검출기의 강점을 조사하고, 전장 상황에서 딥러닝 기반 검출기가 기여할 수 있음을 보인다.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 1996년도 춘계학술대회 논문집
• /
• pp.281-285
• /
• 1996

In this paper, I made visual inspection system using Vision Board and it is consist of an illuminator (a fluorescent lamp), image input device(CCD(Charge)Coupled Device) camera), image processing system(Vision Board(FARAMVB-02), image output device(videomonitor, printer), a measuring instrument(TELMN1000). Length measurement by visual inspection system is used 100mm gauge block instead of calculating distance between camera and object, it measured horizontal and vertical length factor from 400mm to 650mm by increasing 50mm. In this place, measured horizontal and vertical length factor made use of length measurement of a injection. A measuring instrument used to compare a measured length of a injection visual inspection system with it. In conclusion, length measurement of a injection compared a measuring instrument withvisual inspecion system using length factor of 100mm guage block. Maximum error of length compared two devices a measuring instrument with visual inspection system is 0.55mm. And operation program is made up Borland C++ 3.1. By changing, it is applied to various uses.

• 대한전자공학회논문지
• /
• 제27권6호
• /
• pp.944-953
• /
• 1990

This paper describes design and implementation of a real-time high-precision vision system for an automation of SMT(surface mounting technology ). Also, a part inspection algorithm which calculates the position and direction of SMD(surface mounted device) accurately and performs the ruling using those information are presented, and a parallel processing technique for implementing those algorithms is also described. For a real-time implementation of iage acquisition and processing, several hardware modules, namely, multi-functional A/D-D/A board, frame memory board are developed. Particularly, a PE (processing element) board which employs the DSP56001 DSP (digital signal processor) is developed for the purpose of concurrent processing of part inspection algorithms. A stand-alone vision system is built by integration of the developed hardware modules and related softwares.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1987년도 한국자동제어학술회의논문집; 한국과학기술대학, 충남; 16-17 Oct. 1987
• /
• pp.101-104
• /
• 1987

This paper shows the robot-vision system which consists of robot, vision system, single board computer and IBM-PC. IBM-PC based system has a great flexibility in expansion for a vision system interfacing. Easy human interfacing and great calculation ability are the benefits of this system. It was carried to interface between each component. The calibration between two coordinate systems is studied. The robot language for robot-vision system was written in "C" language. User also can write job program in "C" language in which the robot and vision related functions reside in the library.side in the library.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1996년도 한국자동제어학술회의논문집(국내학술편); 포항공과대학교, 포항; 24-26 Oct. 1996
• /
• pp.5-7
• /
• 1996

We propose a triangulation method based on stereo vision angles. We setup stereo vision systems which extract the depth information to a moving object by detecting a moving object using difference image method and obtaining the depth information by the triangulation method based on stereo vision angles. The feature point of a moving object is used the geometrical center of the moving object, and the proposed vision system has the accuracy of 0.2mm in the range of 400mm.

• 한국전자통신학회논문지
• /
• 제17권4호
• /
• pp.633-640
• /
• 2022

다양한 분야에 서비스 로봇이 적용됨에 따라 각 임무에 적합한 영상처리 알고리즘을 빠르고 정확하게 수행할 수 있는 영상처리 프로세서에 관한 관심이 높아지고 있다. 본 논문에서는 로봇에 적용 가능한 영상처리 프로세서 설계방법을 소개한다. 제안한 프로세서는 CPU, GPU, FPGA가 융합된 형태로 AGX 보드, FPGA 보드, LiDAR-Vision 보드, Backplane 보드로 구성된다. 제안한 방법은 시뮬레이션 실험을 통해 검증한다.

• Journal of Biosystems Engineering
• /
• 제32권1호
• /
• pp.44-49
• /
• 2007

A stereo vision system was developed for robot milking system (RMS) using two monochromatic cameras. An algorithm for inverse perspective transformation was developed for the 3-D information acquisition of all teats. To verify performance of the algorithm in the stereo vision system, indoor tests were carried out using a test-board and model teats. A real cow and a model cow were used to measure distance errors. The maximum distance errors of test-board, model teats and real teats were 0.5 mm, 4.9 mm and 6 mm, respectively. The average distance errors of model teats and real teats were 2.9 mm and 4.43 mm, respectively. Therefore, it was concluded that this algorithm was sufficient for the RMS to be applied.

• 전자공학회논문지 IE
• /
• 제47권3호
• /
• pp.1-6
• /
• 2010

본 논문에서는 문자나 그래픽 같은 매우 간단한 메시지만을 표출하고 있는 소형 LED 전광판에서도 동영상이 표출될 수 있도록 임베디드 시스템을 이용한 프로세서를 설계하였다. 구현 방법으로는 임베디드 시스템에서 출력되는 24Bit의 디지털 동영상을 표출할 수 있도록 하기위하여 비디오 프로세서와 LED Display Panel을 설계한 동영상 LED 전광판을 제작하였다. 감마 보정, 밝기, 색 대비조정, 스케줄 기능, 인터넷에 의한 표출영상 변환 및 저장장치를 내장하였으며, 그래픽, 동영상 등을 소형LED 전광판에서 표출할 수 있도록 Windows CE 기반의 응용 프로그램을 설계하였다.

• 제어로봇시스템학회논문지
• /
• 제13권9호
• /
• pp.896-902
• /
• 2007

This paper proposes a vision-based kinematic control method for mobile robots with camera-on-board. In the previous literature on the control of mobile robots using camera vision information, the forward velocity is set to be a constant, and only the rotational velocity of the robot is controlled. More efficient motion, however, is needed by controlling the forward velocity, depending on the position in the corridor. Thus, both forward and rotational velocities are controlled in the proposed method such that the mobile robots can move faster when the comer of the corridor is far away, and it slows down as it approaches the dead end of the corridor. In this way, the smooth turning motion along the corridor is possible. To this end, visual information using the camera is used to obtain the perspective lines and the distance from the current robot position to the dead end. Then, the vanishing point and the pseudo desired position are obtained, and the forward and rotational velocities are controlled by the LOS(Line Of Sight) guidance law. Both numerical and experimental results are included to demonstrate the validity of the proposed method.

• 한국정밀공학회지
• /
• 제14권11호
• /
• pp.126-134
• /
• 1997

In this study, We made visual inspection system using Vision Board. It is consist of an illuminator (a fluorescent lamp), image input device (CCD (Charge-Coupled Device) camera), image processing system(Vision Board(FARAMVB-02)), image output device (video monitor, printer), and a measuring instrument(TELMN1000). Length measurement by visual inspection system make use of 100mm guage block(instead of calculating distance between a camera and a object). It measured horizontal and vertical length factor from 400mm to 650mm by increasing 50mm. In this place, measured horizontal and vertical length factor made use of length measure- ment of a injection. A measuring instrument used to ompare a measured length of a injection visual inspection system with it. In conclusion, length measurement of a injection compared a measuring instrument with visual inspecion system using length factor of 100mm gauge block. We find that maximum error of length is 0.55mm when it compar with the measuring value of two devices(FARAMVB-02, TELMN1000). Program of visual inspection system is made up Borland C++3.1.