• Journal of Information Processing Systems
• /
• 제12권2호
• /
• pp.183-195
• /
• 2016

This paper presents a complete method for vehicle detection and tracking in a fixed setting based on computer vision. Vehicle detection is performed based on Scale Invariant Feature Transform (SIFT) feature matching. With SIFT feature detection and matching, the geometrical relations between the two images is estimated. Then, the previous image is aligned with the current image so that moving vehicles can be detected by analyzing the difference image of the two aligned images. Vehicle tracking is also performed based on SIFT feature matching. For the decreasing of time consumption and maintaining higher tracking accuracy, the detected candidate vehicle in the current image is matched with the vehicle sample in the tracking sample set, which contains all of the detected vehicles in previous images. Most remarkably, the management of vehicle entries and exits is realized based on SIFT feature matching with an efficient update mechanism of the tracking sample set. This entire method is proposed for highway traffic environment where there are no non-automotive vehicles or pedestrians, as these would interfere with the results.

• International Journal of Automotive Technology
• /
• 제5권2호
• /
• pp.135-144
• /
• 2004

In this paper, the model-matching control in the longitudinal autonomous driving system is investigated by vehicle dynamics simulation, which contains nonlinear subcomponents and simplified subcomponents. The design of the robust model-matching controller is performed by the characteristics of the 2 degrees of freedom controller, which is composed of the feedforward compensator and the feedback compensator. It makes the characteristics of tractive and brake force to be equivalent to the specific transfer function, which is suggested as the reference model. Mathematical models of vehicle dynamic analysis including the model-matching control are constructed for computer simulation. Then, simple examples on open-loop simulation without any controller and closed loop simulation with the model-matching controller are applied to check the validity of the robust controller. As the practical example, the autonomous driving system in the longitudinal direction is adopted. It is proved that the model-matching control is effective and adequate to the disturbances and the perturbations, which are shown in the responses of the change of a vehicle mass and a road gradient.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제54권2호
• /
• pp.97-103
• /
• 2005

Most car navigation systems(CNS) estimate the vehicle's location using global positioning system(GPS) or dead reckoning(DR) system. However, the estimated location has undesirable errors because of various noise sources such as unpredictable GPS noises. As a result, the measured position is not lying on the road, although the vehicle is known to be restricted on the road network. The purpose of map matching is to locate the vehicle's position on the road network where the vehicle is most likely to be positioned. In this paper, we analyze some general map matching algorithms first. Then, we propose a map matching method using compensation vectors to improve the performance of map matching. The proposed method calculates a compensation vector from the discrepancy between a measured position and an estimated position. The compensation vector and a newly measured position are to be used to determine the next estimation. To show the performance improvement of the map matching using compensation vectors, the real time map matching experiments are performed. The real road experiments demonstrate the effectiveness and applicability of the proposed map matching.

• ETRI Journal
• /
• 제42권3호
• /
• pp.411-419
• /
• 2020

In real-world intelligent transportation systems, accuracy in vehicle license plate detection and recognition is considered quite critical. Many algorithms have been proposed for still images, but their accuracy on actual videos is not satisfactory. This stems from several problematic conditions in videos, such as vehicle motion blur, variety in viewpoints, outliers, and the lack of publicly available video datasets. In this study, we focus on these challenges and propose a license plate detection and recognition scheme for videos based on a temporal matching prior network. Specifically, to improve the robustness of detection and recognition accuracy in the presence of motion blur and outliers, forward and bidirectional matching priors between consecutive frames are properly combined with layer structures specifically designed for plate detection. We also built our own video dataset for the deep training of the proposed network. During network training, we perform data augmentation based on image rotation to increase robustness regarding the various viewpoints in videos.

• 반도체디스플레이기술학회지
• /
• 제13권2호
• /
• pp.7-12
• /
• 2014

A vehicle license plate includes the most important information for recognition and classification of the vehicle. In this paper, we propose a vehicle license plate recognition system using image binarization and template matching. In the proposed system, an image of the vehicle license plate is converted into a gray scale image and the gray image undergoes the binarization process. Finally, the numbers on the plate are extracted from the binary image using the template matching algorithm.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2010년도 춘계학술대회
• /
• pp.193-196
• /
• 2010

본 논문에서는 Chamfer Matching을 이용하여 차량 영상의 형태를 대형, 중형, 소형으로 분류하는 방법을 제안하며, 다음과 같이 두 단계로 구성된다. 첫 번째, 템플릿 후보 영상에서 차량 윤곽선을 추출한 후, 윤곽선으로부터의 거리변환을 통해 에지 거리 템플릿을 생성한다. 두 번째, 입력 영상과 템플릿 간의 거리 값 차이가 최소인 템플릿의 형태로 차량형태를 분류한다. 거리 값이란 입력 영상의 차량 윤곽선의 한 픽셀이 템플릿의 경계와 얼마나 떨어져 있는가를 나타낸 것이다. 실험 결과 제안된 방법은 실험 영상에 대해 80%의 만족할만한 성능을 나타내었다.

• 제어로봇시스템학회논문지
• /
• 제17권5호
• /
• pp.451-459
• /
• 2011

Fundamentally, there are 5 systems are needed for autonomous navigation of unmanned ground vehicle: Localization, environment perception, path planning, motion planning and vehicle control. Path planning and motion planning are accomplished based on result of the environment perception process. Thus, high reliability of localization and the environment perception will be a criterion that makes a judgment overall autonomous navigation. In this paper, via map matching using vehicle dynamic model and LIDAR sensors, replace high price localization system to new one, and have researched an algorithm that lead to robust autonomous navigation. Finally, all results are verified via actual unmanned ground vehicle tests.

• 한국생산제조학회지
• /
• 제22권1호
• /
• pp.150-155
• /
• 2013

This work present a new method of sealing inspection system for vehicle in which foam rubber materials are used for sealing the vehicle parts. This system is composed from a devices comprising non-contact and real-time scanning on visual inspection in machine parts. We have been investigated qualitative factors that influenced on sealing system of vehicle structure which flexibly attenuated vibration and plenty of foam rubber materials having elastic property. However, there are different factors which still depended on outdated technique (personnel subjective judgment) in the performance inspection of rubber parts, specially for cross section inspection. Through a newly developed inspection system which recently applied for the production line, we successfully achieved more effective results of matching rate by about 80 % in the sealing performance inspection with 0.7% to 1.4% in the repeated errors. These are resulted from non-contacted response by CCD camera and vision program using geometry matching method. We expect that this system may be widely applied in the strict inspection parts of more diverse cross-section in future.

• 한국컴퓨터정보학회논문지
• /
• 제14권12호
• /
• pp.63-73
• /
• 2009

본 논문에서는 단순화한 색상 정보에 기반한 컨벌류션 정합(Convolution Matching)을 이용하여 차량을 검출하는 기법을 제안한다. 입력 영상을 화소 색상 벡터의 방향을 고려해 8방향 색상(Red, Green, Blue Cyan, Yellow, Magenta, White, Black)으로 표현한다. 8 방향 색상의 표현은 조명이나 환경 변화에 강인한 영상을 제공한다. 본 논문의 차량 검출 단계는 크게 후보 영역 검출 단계와 차량 검증 단계로 구성된다. 후보 영역 검출 단계에서는 수직 에지와 그림자 등을 고려하여 차량의 후보 영역을 결정한다. 차량 검증 단계에서는 차량을 판별하기 위해 컨벌류션 정합과 후보 영역내의 에지 복잡도를 사용한다. 제안하는 차량 검출 알고리즘은 조명이나 환경이 변화하는 다양한 실험들에서 빠르고 높은 검출률을 보였다.

• 한국ITS학회 논문지
• /
• 제21권2호
• /
• pp.132-151
• /
• 2022

차량 측위 기술은 차량의 정확한 위치 정보를 제공한다는 점에서 자율주행을 위한 핵심 기술 중 하나로 평가되고 있다. 이미지 기반의 측위 기술은 위치 정보를 효율적으로 제공할 수 있을 것으로 판단되어 다양한 관련 연구가 진행되고 있다. 하지만, 기존 특징점 또는 차선 정보를 이용한 이미지 기반 측위 방법론은 도로 및 운행 환경에 측위 정확도가 큰 영향을 받을 수 있다는 한계가 있다. 선분 매칭은 특징점에 비하여 텍스쳐 반복에 강건하고 주변 환경 전체에서 추출된 선분을 활용하기 때문에 기존 방법론의 단점을 해결할 수 있다. 하지만, 차량 운행환경을 대상으로 한 선분 매칭 방법론을 다루는 연구는 거의 진행된 바 없다. 따라서 본 연구에서는 정확한 차량 측위 지원을 위한 선분 매칭 프레임워크를 제안한다. 또한 도로 주행 환경에서의 알고리즘 성능 비교 분석을 통하여 최적 선분 매칭 알고리즘을 결정하였다. 최종적으로 제안된 프레임워크는 선분 추출, 병합, 중첩 영역 탐지 및 MSLD 기반 선분 매칭의 4단계로 구성되었다. 제안된 프레임워크는 차량의 속도, 운행 방식, 주변 환경에 상관없이 차량 측위에 충분한 수준의 선분 매칭을 안정적으로 수행하였다.