• 제어로봇시스템학회논문지
• /
• 제4권6호
• /
• pp.759-764
• /
• 1998

We propose a novel algorithm capable of recognizing the road lane by image processing. Considering the fact that the direction and location of road lane are maintained similarly in successive images we formulate a function to represent the property. However, as noises play the role of making a lot of similar patterns appear and disappear in the road image, keeping of robustness in the lane detection has been known a difficult work. To overcome this problem, we introduce the following three ideas: 1) design of a function based on an edge direction and magnitude, 2) construction of a recursive filter to estimate the function recursively for successive images, 3) principal axis-based line fitting. These concepts enhance the adaptability to cope with the random environment of traffic scene and eventually lead to the reliable detection of a road lane.

• Journal of KIEE
• /
• 제11권1호
• /
• pp.69-77
• /
• 2001

This paper describes an image processing algorithm capable of recognizing road lanes by using a CDF(cumulative distribution function). The CDF is designed for the model function of road lanes. Based on the assumptions that there are no abrupt changes in the direction and location of road lanes and that the intensity of lane boundaries differs from that of the background, we formulated the CDF, which accumulates the edge magnitude for edge directions. The CDF has distinctive peak points at the vicinity of lane directions due to the directional and the positional continuities of a lane. To obtain lane-related information a scatter diagram was constructed by collecting edge pixels, of which the direction corresponds to the peak point of the CDF, then the principal axis-based line fitting was performed for the scatter diagram. Noises can cause many similar features to appear and to disappear in an image. Therefore, to reduce the noise effect a recursive estimator of the CDF was introduced, and also to prevent false alarms or miss detection a scene understanding index (DUI) was formulated by the statistical parameters of the CDF. The proposed algorithm has been implemented in real time on video data obtained from a test vehicle driven on a typical highway.

• 대한의용생체공학회:의공학회지
• /
• 제29권2호
• /
• pp.122-131
• /
• 2008

For B-mode ultrasound images, we propose an image enhancement algorithm based on a multi-resolution approach, which consists of edge enhancing and noise reducing procedures. Edge enhancement processing is applied sequentially to coarse-to-fine resolution images obtained from wavelet-transformed data. In each resolution, the structural features of each pixel are examined through eigen analysis. Then, if a pixel belongs to an edge region, we perform two-step filtering: that is, directional smoothing is conducted along the tangential direction of the edge to improve continuity and directional sharpening is conducted along the normal direction to enhance the contrast. In addition, speckle noise is alleviated by proper attenuation of the wavelet coefficients of the homogeneous regions at each band. This region-based speckle-reduction scheme is differentiated from other methods that are based on the magnitude statistics of the wavelet coefficients. The proposed algorithm enhances edges regardless of changes in the resolution of an image, and the algorithm efficiently reduces speckle noise without affecting the sharpness of the edge. Hence, compared with existing algorithms, the proposed algorithm considerably improves the subjective image quality without providing any noticeable artifacts.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.1701-1704
• /
• 2005

We developed an machine vision method for navigation control of a traveling automatic guided vehicle(AGV) on desired trajectory with guided marks. The formulated EDF accumulates the edge magnitude for edge directions. The EDF has distinctive peak points at the vicinity of trajectory directions due to the directional and the positional continuities of desired trajectory. Examining the EDF by its shape parameters of the local maxima and symmetry axis results in identifying whether or not change in traveling direction of an AGV has occurred. Simulation results show that the presented method is useful for navigation control of AGV.

• 한국안광학회지
• /
• 제9권1호
• /
• pp.29-34
• /
• 2004

본 연구는 타각적 검사장비의 개발을 위한 연구로 PC를 기반으로 Foucault Test (Knife edge Test)를 이용하여 굴절상태의 측정을 목표로 하였다. Knife edge Test를 이용하여 입사광선이 망막에서 반사된 출사광선의 이미지를 CCD 카메라를 사용하여 촬영하고, 그 이미지를 분석하여 반사상의 지름을 측정하여 동공의 크기를 측정하였다. 또한 반사 이미지의 방향으로 굴절상태를 판정하고 반사 이미지의 높이를 측정하여 굴절이상도를 확인하였다. 구면수차는 반사상과 반대방향에 나타나는 부반사 이미지의 명도를 비교하여 측정하였고, 그 크기를 통해 수차값을 계산할 수 있었다.

• 제어로봇시스템학회논문지
• /
• 제11권3호
• /
• pp.214-221
• /
• 2005

For decades, airships have being developed in Europe (especially German) and America. Airships are planning to be used for advertisements and airliners as well. In Korea, KARI (Korea Aerospace Research Institute) is developing stratospheric communication airship and the similar research is carried out in Japan. Among them, Zeppelin of German has the cutting-edge airship technology with Zeppelin NT. In this paper, the flight performance and stability were evaluated by comparing mathematical theory and the real test. The stability was examined through dynamic modeling and assured by designing controllers at each flight mode. Elevator angle, rudder angle, magnitude of thrust and tilting angle of thrust vector were used as control inputs. Moreover, after measuring the airship velocity, flight direction, magnitude and direction of the wind, attitude angles and trajectories of the airship at each flight mode, the results were compared with the simulation. To get the reasonable data, low-pass filter and band-stop filter were designed to get rid of the sensor noise and engine vibration. The test was accomplished at cruise mode, turning mode, and deceleration. To conclude, with comparing the simulation data and flight test data, it could be known that the dynamic model used in this paper was reasonable.

• 대한전자공학회논문지SP
• /
• 제45권3호
• /
• pp.37-45
• /
• 2008

화질 평가는 원영상과 열화된 영상 간의 차이를 측정함으로써, 열화된 영상의 화질이 좋고 나쁨을 판단하는 것을 목표로 한다. 본 논문에서는 열화된 영상의 화질 평가를 위해, 원영상과 열화된 영상 전체를 비교하는 것 대신, 원영상과 열화된 영상, 각각의 특징으로 에지 투영을 이용하는 방법을 제안하였다. 여기서 에지 투영은 에지 맵에서 수직, 수평 방향으로 투영시킴으로써 얻을 수 있다. 에지 투영 시 수직, 수평 방향에 대한 그래디언트 크기를 고려함으로써, 보다 나은 화질 평가 방법을 제안하였다. 제안한 방법의 탁월함을 기존의 화질 평가 방법인 structural similarity(SSIM), edge peak signal-to-noise ratio(EPSNR), 그리고 edge histogram descriptor(EHD) 방법과 비교 실험을 통해 보였다.

• 응용통계연구
• /
• 제28권5호
• /
• pp.991-1001
• /
• 2015

영상은 렌즈를 통하여 형성된 이미지로 많은 응용 분야에서 사용된다. 디지털 기기로 획득한 디지털 영상은 수치화된 자료로 통계분석이 가능하며, 신속하고 효율적인 작업이 가능하게 한다. 영상처리 분야에서는 화질의 개선을 위해서 잡음을 제거하는 방법들이 연구되고 있다. 본 논문은 영상 잡음을 효과적으로 제거하는 방법으로 통계적 방법들을 사용하여, 에지의 방향과 크기를 적용한 새로운 잡음 제거 방법을 제안한다. 이 방법은 동일한 방향에 위치한 화소들에 대하여 로버스트 회귀모형을 적용하고 해당 화소의 밝기 값을 추정한다. 추정된 화소의 밝기 값은 에지의 크기가 가중값으로 사용되어 평균필터의 성능을 개선한다. 모의실험의 결과, 제안한 방법은 특징을 포함하는 화소를 잘 유지하며, 잡음 제거 성능도 기존의 방법보다 개선되는 것을 확인하였다.

• 한국농공학회지
• /
• 제35권4호
• /
• pp.76-85
• /
• 1993

The purpose of this study was to analyze the wind force distribution on the two single-span arched plastic house depending upon the house spacing and wind direction, which may provide the fundamental criteria for the structural design. In order to specify the wind force distribution, the variation of the wind force coefficients, the mean wind force coefficients and the drag force coefficients were estimated from the wind tunnel test data. The results obtained are as follows : 1. At the wind direction of 90$^{\circ}$, there was a typical span interval at which the maximum negative pressure was occured at the edge of the inside walls. 2. In the consideration of wind loads, the wind force coefficients estimated from independent single-span arched plastic house should not be directly applied to the structural design on the double houses separated. 3. The average maximum negative wind force on the inside walls was occured at the wind direction of 90$^{\circ}$, and the variations depending on the span intervals was not significant. 4. The average maximum drag force was occured at the wind direction of 300, and the magnitude of drag force was more significant at the first house. As the distance between two houses was increased, the drag force was slightly increased for every wind direction.

• 제어로봇시스템학회논문지
• /
• 제12권11호
• /
• pp.1088-1094
• /
• 2006

In this paper, an auto path generation and an active compliance grinding control using 3-axis farce sensor are presented. These control algorithms enable the grinding robot to follow unknown path of various workpiece shape pattern. The robot is able to go grinding along unknown paths by position controller managing tangential direction angle and cutting speed, with only information about the start position and the end position. Magnitude and direction of normal force are calculated using force data that go through low pass filter. Moreover, normal and tangential directions are separated for force control and velocity control, respectively.