• 비파괴검사학회지
• /
• 제23권1호
• /
• pp.21-29
• /
• 2003

실시간 로봇 위치 제어를 위해 비젼시스템을 사용할 때 이 모델에 포함된 매개변수를 추정하는데 있어 계산시간을 줄이는 것은 매우 중요하다. 불행히도 흔히 사용되고 있는 일괄 처리 기법은 반복적으로 계산이 수행되기 때문에 많은 계산 시간을 필요로 하여 실시간 로봇 위치 제어를 어렵게 한다. 반면에 본 연구에서 사용하고자 하는 화장 칼만 필터링은 사용하기 편리하고, 또한 순환적 방법으로 계산되기 때문에 비젼시스템의 매개변수를 계산하는데 있어 시간을 줄이는 커다란 장점을 가지고 있다. 따라서 본 연구에서는 실시간 로봇 위치 제어를 위해 사용하는 비젼 제어 기법에 확장 칼만 필터링을 적용되었다 여기서 사용된 비젼시스템 모델은 카메라 내부 매개변수(방향, 초점거리 등) 및 외부 매개변수(카메라와 로봇 사이의 상대적 위치)를 설명하기 위해 6개 매개변수를 포함하고 있다. 이러한 매개변수를 추정하기 위해 확장 칼만 필터링 기법이 적용되었다. 또한 이렇게 추정된 6개 매개변수를 사용하여 로봇을 구동시키기 위해 필요한 로봇 회전각 추정에도 화장 칼만 필터링 기법이 적용되었다. 최종적으로 확장 칼만 필터링을 사용하여 개발된 비젼 제어 기법의 타당성을 로봇 위치 제어 실험을 수행하여 확인하였다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 하계종합학술대회 논문집(4)
• /
• pp.206-209
• /
• 2000

Almost vision application systems use 2-D information by taking only one camera. Recently it arises to utilize 3-D information, which is distance from camera to object, because 2-D information is not sufficient. Therefore, we take stereo camera system. In motion detection algorithm using stereo vision, it operates like one camera system, which takes advantage of correlation, edge, and difference algorithm, when it detects any motion. At that time, to detect motion, it compares two images, which is from two cameras, to calculate disparity that contains distance information. By disparity, it can compute real distance and size of object information. We describe a motion detection algorithm which computes 3-D distance and object size in real time.

• Journal of Multimedia Information System
• /
• 제1권1호
• /
• pp.77-85
• /
• 2014

Usually, the skin pigmentation detection and diagnosis are made by clinicians. In this process it is subjective and non-quantitative. We develop an approach to detect and measure the different pigmentation lesions base on computer vision technology. In the paper we study several usually used skin-detecting color space like HSV, YCbCr and normalized RGB. We compare their performance with illumination influence for detecting the pigmentation lesions better. Base on a relatively stable color space, we propose an approach which is RGB channels vector difference characteristic for the detection. After the object region detection, we also use the difference to measure the difference between the lesion and the surrounding normal skin. From the experiment results, our approach can effectively detect the pigmentation lesion, and perform robustness with different illumination.

• International Journal of Control, Automation, and Systems
• /
• 제5권2호
• /
• pp.184-191
• /
• 2007

Autonomous navigation of mobile robot requires a ranging system for measurement of distance to environmental objects. It is obvious that the wider and the faster distance measurement gives a mobile robot more freedom in trajectory planning and control. The active omni-directional ranging system proposed in this paper is capable of obtaining the distance for all 3600 directions in real-time because of the omni-directional mirror and the structured light. Distance computation including the sensitivity analysis and the experiments on the omni-directional ranging are presented to verify the effectiveness of the proposed system.

• 한국공작기계학회:학술대회논문집
• /
• 한국공작기계학회 1999년도 추계학술대회 논문집 - 한국공작기계학회
• /
• pp.119-124
• /
• 1999

This paper presents new approach to technology pattern recognition of camera back cover and test of soldering. In real-time implementing of pattern recognition camera back cover and test of soldering, the MVB-03 vision board has been used. Image can be captured from standard CCD monochrome camera in resolutions up to 640$\times$480 pixels. Various options re available for color cameras, a synchronous camera reset, and linescan cameras. Image processing os performed using Texas Instruments TMS320C31 digital signal processors. Image display is via a standard composite video monitor and supports non-destructive color overlay. System processing is possible using c30 machine code. Application software can be written in Borland C++ or Visual C++

• 한국공작기계학회:학술대회논문집
• /
• 한국공작기계학회 1996년도 춘계학술대회 논문집
• /
• pp.200-204
• /
• 1996

It is a great number of problem for the man to perform maintenance and repairing work owing to radioactive effusion for a nuclear fuel and the pollution of an related equipment in nuclear power plants. Therefore, the vision processing system presented in this research requires to maintain the good performance under the radioactive circumstances and to safety the real time processing system presented in this research requires to maintain the good performance under the radioactive circumstances and to safety the real time processing. The proposed vision scheme adapts the gradient and Laplacian operator to perform the high speed processing in an edge detection and the centroid formula at each direction to obtain the center position of a holes using DSPs

• 한국컴퓨터정보학회논문지
• /
• 제24권6호
• /
• pp.67-72
• /
• 2019

This paper proposes a computer vision-based banknote recognition system intended to assist the blind. This system is robust and fast in recognizing banknotes on videos recorded with a smartphone on real-life scenarios. To reduce the computation time and enable a robust recognition in cluttered environments, this study segments the banknote candidate area from the background utilizing a technique called Pixel-Based Adaptive Segmenter (PBAS). The Speeded-Up Robust Features (SURF) interest point detector is used, and SURF feature vectors are computed only when sufficient interest points are found. The proposed algorithm achieves a recognition accuracy of 98%, a 100% true recognition rate and a 0% false recognition rate. Although Korean banknotes are used as a working example, the proposed system can be applied to recognize other countries' banknotes.

• International Journal of Korean Welding Society
• /
• 제3권2호
• /
• pp.53-59
• /
• 2003

Visual inspection using laser vision sensor was proposed for fast and economic inspection and was verified experimentally. Welding is one of the most important manufacturing processes for automotive and electronics industries as well as heavy industries. The weld zone influences the reliability of the products. There are two kinds of weld inspection tests, destructive and non­destructive test. Even though the destructive test is much more reliable, the product should be destroyed, and hence the non­destructive test such as ultrasonic or X­ray test was used to overcome this problem. However, these tests are not used for real time inspection.

• 국제학술발표논문집
• /
• The 7th International Conference on Construction Engineering and Project Management Summit Forum on Sustainable Construction and Management
• /
• pp.252-259
• /
• 2017

Nowadays, the existing manual method for recording actual progress of the construction site has some drawbacks, such as great reliance on the experience of professional engineers, work-intensive, time consuming and error prone. A method integrating computer vision and BIM(Building Information Modeling) is presented for indoor automatic progress monitoring. The developed method can accurately calculate the engineering quantity of target component in the time-lapse images. Firstly, sample images of on-site target are collected for training the classifier. After the construction images are identified by edge detection and classifier, a voting algorithm based on mathematical geometry and vector operation will divide the target contour. Then, according to the camera calibration principle, the image pixel coordinates are conversed into the real world Coordinate and the real coordinates would be corrected with the help of the geometric information in BIM model. Finally, the actual engineering quantity is calculated.

• 한국항행학회논문지
• /
• 제27권5호
• /
• pp.685-689
• /
• 2023

The proposed system in the study aims to detect forest fires in real-time stream data received from the drone-camera. Recently, the number of wildfires has been increasing, and also the large scaled wildfires are frequent more and more. In order to prevent forest fire damage, many experiments using the drone camera and vision analysis are actively conducted, however there were many challenges, such as network speed, pre-processing, and model performance, to detect forest fires from real-time streaming data of the flying drone. Therefore, this study applied image data processing works to capture five good image frames for vision analysis from whole streaming data and then developed the object detection model based on YOLO_v2. As the result, the classification model performance of forest fire images reached upto 93% of accuracy, and the field test for the model verification detected the forest fire with about 70% accuracy.