• 대한임베디드공학회논문지
• /
• 제17권1호
• /
• pp.41-49
• /
• 2022

This paper proposes a novel lane detection algorithm based on inverse perspective transformation and machine learning in lightweight embedded system. The inverse perspective transformation method is presented for obtaining a bird's-eye view of the scene from a perspective image to remove perspective effects. This method requires only the internal and external parameters of the camera without a homography matrix with 8 degrees of freedom (DoF) that maps the points in one image to the corresponding points in the other image. To improve the accuracy and speed of lane detection in complex road environments, machine learning algorithm that has passed the first classifier is used. Before using machine learning, we apply a meaningful first classifier to the lane detection to improve the detection speed. The first classifier is applied in the bird's-eye view image to determine lane regions. A lane region passed the first classifier is detected more accurately through machine learning. The system has been tested through the driving video of the vehicle in embedded system. The experimental results show that the proposed method works well in various road environments and meet the real-time requirements. As a result, its lane detection speed is about 3.85 times faster than edge-based lane detection, and its detection accuracy is better than edge-based lane detection.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2000년도 추계학술대회논문집A
• /
• pp.596-601
• /
• 2000

In a robot/vision system, the vision sensor, typically a CCD array sensor, is mounted on the robot hand. The problem of determining the relationship between the camera frame and the robot hand frame is refered to as the hand-eye calibration. In the literature, various methods have been suggested to calibrate camera and for sensor registration. Recently, one-step approach which combines camera calibration and sensor registration is suggested by Horaud & Dornaika. In this approach, camera extrinsic parameters are not need to be determined at all configurations of robot. In this paper, by modifying the camera model and including the lens distortion effect in the perspective transformation matrix, a new one-step approach is proposed in the hand-eye calibration.

• 센서학회지
• /
• 제23권1호
• /
• pp.66-71
• /
• 2014

Vision is the most important sensing capability for both men and sensory smart machines, such as intelligent robots. Sensed real 3D world and its 2D camera image can be related mathematically by a process called camera calibration. In this paper, we present a novel linear solution of camera calibration. Unlike most existing linear calibration methods, the proposed technique of this paper can identify camera parameters explicitly. Through the step-by-step procedure of the proposed method, the real physical elements of the perspective projection transformation matrix between 3D points and the corresponding 2D image points can be identified. This explicit solution will be useful for many practical 3D sensing applications including robotics. We verified the proposed method by using various cameras of different conditions.

• 한국정밀공학회지
• /
• 제19권1호
• /
• pp.172-179
• /
• 2002

In a robot/vision system, the vision sensor, typically a CCD array sensor, is mounted on the robot hand. The problem of determining the relationship between the camera frame and the robot hand frame is refered to as the hand-eye calibration. In the literature, various methods have been suggested to calibrate camera and for sensor registration. Recently, one-step approach which combines camera calibration and sensor registration is suggested by Horaud & Dornaika. In this approach, camera extrinsic parameters are not need to be determined at all configurations of robot. In this paper, by modifying the camera model and including the lens distortion effect in the perspective transformation matrix, a new one-step approach is proposed in the hand-eye calibration.

• 제어로봇시스템학회논문지
• /
• 제15권11호
• /
• pp.1088-1095
• /
• 2009

In this paper, we describe the reorientation method of distorted road sign by using projection transformation for improving recognition rate of road sign. RSR (Road Sign Recognition) is one of the most important topics for implementing driver assistance in intelligent transportation systems using pattern recognition and vision technology. The RS (Road Sign) includes direction of road or place name, and intersection for obtaining the road information. We acquire input images from mounted camera on vehicle. However, the road signs are often appeared with rotation, skew, and distortion by perspective camera. In order to obtain the correct road sign overcoming these problems, projection transformation is used to transform from 4 points of image coordinate to 4 points of world coordinate. The 4 vertices points are obtained using the trajectory as the distance from the mass center to the boundary of the object. Then, the candidate areas of road sign are transformed from distorted image by using homography transformation matrix. Internal information of reoriented road signs is segmented with arrow and the corresponding indicated place name. Arrow area is the largest labeled one. Also, the number of group of place names equals to that of arrow heads. Characters of the road sign are segmented by using vertical and horizontal histograms, and each character is recognized by using SAD (Sum of Absolute Difference). From the experiments, the proposed method has shown the higher recognition results than the image without reorientation.

• 한국군사과학기술학회지
• /
• 제2권1호
• /
• pp.194-203
• /
• 1999

This paper presents a measuring technique for bursting height of proximity fuses. This technique uses camera calibration to obtain the perspective transformation matrix describing the projection of the world coordinates to image coordinates, and calculates the world coordinates of bursting points from their image coordinates. The surface approximation algorithm by polynomial functions are also implemented.

• 대한전자공학회논문지SP
• /
• 제39권4호
• /
• pp.429-441
• /
• 2002

본 논문은 실시간으로 3차원 공간상에서의 움직임 정보를 추출할 수 있는 입력 장치를 제안한다. 제안하는 3차원 입력 장치는 스테레오 카메라의 기하학적 구조와 색상, 움직임, 형태상의 특성을 이용하여 복잡한 환경에서 사전 카메라 캘리브레이션 없이 3차원 움직임 정보를 추출할 수 있다. 움직임 추출을 위해서 perspepctive projection 행렬과 perspective distortion 행렬을 이용한 스테레오 카메라의 기하학적 특성을 이용하며, 효과적인 좌우 영상의 특징점 추적 및 추출을 위해 색상 변환(Color transform)과 UPC(Unmatched Pixel Count) 및 이산 칼만 필터(Discrete Kalman Filter)의 효과적인 결합으로 이루어진 MAWUPC(Motion Adaptive Weighted Pixel Count)과 PCA(Principal Component Analysis)로 구성된 알고리즘을 제안한다. 추출된 3차원 공간상에서의 움직임은 가상환경에서의 가상 물체를 제어하거나 사용자 시점의 이동을 나타내는 인터페이스로 사용한다. 스테레오 비전을 이용한 입력 장치는 선으로 연결되지 않기 때문에 사용자가 가상환경에서 작업하기가 편리하며 몰입감을 높일 수 있는 등 보다 효율적인 상호작용을 가능하게 해준다.

• Journal of the Korean Data and Information Science Society
• /
• 제26권2호
• /
• pp.399-408
• /
• 2015

본 논문은 대기오염 자료를 여러 가지 방법의 데이터 시각화를 통해 나타내었고, 데이터 시각화 방법별로 통계적인 방법을 활용한 분석과 연계하여 어떤 특징을 알아 볼 수 있는지를 나타냈다. 데이터 시각화 도구로는 통계 패키지인 R을 사용하였다. 분석에 사용된 데이터는 뉴욕시에서 1973년 5월부터 9월까지 공기의 질을 측정한 자료이다. 먼저 단변량 분석과 단순회귀분석을 실시하여 데이터 시각화를 통해 자료의 기본적인 특성을 파악하고 시각화 방법으로 산점도행렬 등을 통해 특성을 한눈에 볼 수 있게 나타내었다. 다중 회귀 분석을 실시하여 로그변환 등을 이용하여 최적의 모형을 찾고 설명변수들을 범주화하여 상자그림이나 3차원 투시도, 3차원 산점도 등 여러 데이터 시각화 방법을 이용해 대기오염 데이터의 전체적인 특성들을 알아보았다.

• 한국멀티미디어학회논문지
• /
• 제17권7호
• /
• pp.818-828
• /
• 2014

After emerging of Microsoft Kinect, the interest in three-dimensional (3D) depth image was significantly increased. Depth image data of an object can be converted to 3D coordinates by simple arithmetic calculation and then can be reconstructed as a 3D model on computer. However, because the surface coordinates can be acquired only from the front area facing Kinect, total solid which has a closed surface cannot be reconstructed. In this paper, 3D registration method for multiple Kinects was suggested, in which surface information from each Kinect was simultaneously collected and registered in real time to build 3D total solid. To unify relative coordinate system used by each Kinect, 3D perspective transform was adopted. Also, to detect control points which are necessary to generate transformation matrix, 3D randomized Hough transform was used. Once transform matrices were generated, real time 3D reconstruction of various objects was possible. To verify the usefulness of suggested method, human arms were 3D reconstructed and the volumes of them were measured by using four Kinects. This volume measuring system was developed to monitor the level of lymphedema of patients after cancer treatment and the measurement difference with medical CT was lower than 5%, expected CT reconstruction error.

• 한국수학교육학회지시리즈E:수학교육논문집
• /
• 제37권1호
• /
• pp.65-84
• /
• 2023

라그랑주 승수법(Method of Lagrange Multipliers)은 등식 제약조건하에서 미분가능한 함수의 최대, 최소를 구하는 대표적인 방법이다. 선형대수학, 최적화 이론, 제어 이론을 포함하여 최근에는 인공지능 기초수학에서도 널리 활용되고 있다. 특히 라그랑주 승수법은 미분적분학과 선형대수학을 연결하는 중요한 도구이며, 주성분 분석(Principal Component Analysis, PCA)을 포함한 인공지능 알고리즘에 많이 활용되고 있다. 따라서 교수자는 대학 미분적분학에서 처음 라그랑주 승수법을 접하는 학생들에게 구체적인 학습 동기를 제공할 필요가 생겼다. 이에 본 논문에서는 교수자가 학생들에게 라그랑주 승수법을 효과적으로 교육하는데 필요한 통합적인 시야를 제공한다. 먼저 다양한 전공의 학생들이 계산에 대한 부담을 덜고 원리를 쉽게 이해할 수 있도록 개발한 시각화 자료 및 파이썬(Python) 기반의 SageMath 코드를 제공한다. 또한 라그랑주 승수법으로 행렬의 고윳값과 고유벡터를 유도하는 과정을 상세히 소개한다. 그리고 라그랑주 승수법을 간단한 경우에 대한 증명에서 시작하여 일반화된 최적화 문제로 확장하고, 수업에서 학생들이 라그랑주 승수와 PCA를 활용하여 실제 데이터를 분석한 결과를 추가하였다. 본 연구는 대학수학을 지도하는 다양한 전공의 교수자들에게 도움이 될 기초자료가 될 것이다.