스테레오 비전 기술을 이용한 도로 표지판의 3차원 추적

최창원(Chang-Won Choi),최성인(Sung-In Choi),박순용(Soon-Yong Park) 제어로봇시스템학회 2014 제어·로봇·시스템학회 논문지 Vol.20 No.12

Road signs provide important safety information about road and traffic conditions to drivers. Road signs include not only common traffic signs but also warning information regarding unexpected obstacles and road constructions. Therefore, accurate detection and identification of road signs is one of the most important research topics related to safe driving. In this paper, we propose a 3-D vision technique to automatically detect and track road signs in a video sequence which is acquired from a stereo vision camera mounted on a vehicle. First, color information is used to initially detect the sign candidates. Second, the SVM (Support Vector Machine) is employed to determine true signs from the candidates. Once a road sign is detected in a video frame, it is continuously tracked from the next frame until it is disappeared. The 2-D position of a detected sign in the next frame is predicted by the 3-D motion of the vehicle. Here, the 3-D vehicle motion is acquired by using the 3-D pose information of the detected sign. Finally, the predicted 2-D position is corrected by template-matching of the scaled template of the detected sign within a window area around the predicted position. Experimental results show that the proposed method can detect and track many types of road signs successfully. Tracking comparisons with two different methods are shown.

도로표지에 대한 LiDAR 검지영향요인 연구: 도로표지의 모양과 높이를 중심으로

김지윤,박범진 한국ITS학회 2022 한국ITS학회논문지 Vol.21 No.4

This study investigated the effect of the shape and height of road signs on detection performance when detecting road signs with LiDAR, which is recognized as an essential sensor for autonomous vehicles. For the study, four types of road signs with the same area and material and different shapes were produced, and a road driving test was performed by installing a 32Ch rotating LiDAR on the upper part of the vehicle. As a result of comparing the shape of the point cloud and the NPC according to the shape of the road sign, It is expected that a distance of less than 40m is required to recognize the overall shape of a road sign using 32Ch LiDAR, and shapes such as triangles and rectangles are more advantageous than squares in securing the maximum point cloud from a long distance. As a result of the study according to the height of the road sign, At short distances (within 20m), if the height of the sign is raised to more than 2m, it deviates from the vertical viewing angle of the LiDAR and cannot express the complete point cloud shape. However, it showed a negligible effect compared to the near-field height change. These research results are expected to be utilized in the development of road facilities dedicated to LiDAR for the commercialization of autonomous cooperative driving technology. 본 연구는 자율주행차량의 필수 센서로 인식되는 LiDAR로 도로표지를 검지할 시, 도로표지의 모양과 높이 등이 검지성능에 주는 영향에 대하여 알아보았다. 연구를 위해 면적과 재질은동일하고, 모양은 서로 다른 도로표지를 4종을 제작하였으며, 32Ch 회전형 LiDAR를 차량 상단부에 장착하여 도로주행실험을 수행하였다. 도로표지의 모양에 따른 점군데이터의 형상과NPC를 비교한 결과, 32ch LiDAR를 활용하여 도로표지의 전체 모양을 인식하려면 40m 이내의거리가 필요할 것으로 기대되며, 원거리에서 최대한 점군을 확보하는 데 있어서는 정사각형보다는 삼각형, 직사각형 등의 형상이 유리하였다. 도로표지의 높이에 따른 연구 결과, 근거리(20m이내)에서는 표지의 높이를 2m 이상으로 올리면 LiDAR의 수직시야각에서 이탈하여 완전한 점군 형상을 표현하지 못하게 되며, 차로변화로 센서와 표지 사이의 횡간격과 입사각이 커지게 되면 NPC가 소폭 감소하나 근거리 높이 변화에 비하면 미미한 영향을 보였다. 이러한연구결과는 자율협력주행기술 상용화를 위한 LiDAR 전용 도로시설물 개발에 활용될 수 있을것으로 기대된다.

Retrieval of Contextual Information on Korean Road Sign

Jong-Eun Ha,Jin-Bum Shim,Keun-Ho Choi 제어로봇시스템학회 2008 제어로봇시스템학회 국제학술대회 논문집 Vol.2008 No.10

Automatic extraction of contextual information on road sign can be used in various applications such as scene interpretation and autonomous navigation. This paper proposes a method for the automatic detection of contextual information on Korean road sign. It usually consists of various information such as directional symbol, the name of place according to direction, and corresponding route number. All this information is automatically detected using image processing. First, target area containing a road sign on image is found using learning based object detection algorithm. . Next, grouping into Korean and English characters, direction symbol, and road’ digit is done. Finally, contextual information is obtained by analyzing each group using color and geometric features.

DRIFT COMPENSATION OF MONO-VISUAL ODOMETRY AND VEHICLE LOCALIZATION USING PUBLIC ROAD SIGN DATABASE

Chanhee Jang,Young-Keun Kim 한국자동차공학회 2019 International journal of automotive technology Vol.20 No.6

This paper proposes a novel localization method based on a camera that can estimate the absolute position of a vehicle using a public online database of road signs. The estimated absolute position near a road sign is used to compensate the drift error of visual odometry (VO). In the first phase, the relative position between a road sign and a vehicle is estimated by matching a detected road sign image with the reference image from a public online database. Subsequently, the absolute position of the vehicle is calculated using the data from the database. Once the absolute position of the vehicle is estimated near a road sign, the current position of VO is updated to compensate the accumulated error. From a 24-km driving road test, it is validated that the proposed algorithm can estimate the absolute position of a vehicle within an error of 1.5 m. Moreover, a test of trajectory 3 km showed that it can maintain the drift error of VO within tens of meters. Our method is easy to be deployed, has low computation cost, and is accessible to a wide range of driving environments such as highways.

인공지능 : 날씨,조명 판단 및 적응적 색상모델을 이용한 도로주행 영상에서의 이정표 검출

김태형 ( Tae Hung Kim ),임광용 ( Kwang Yong Lim ),변혜란 ( Hye Ran Byun ),최영우 ( Yeong Woo Choi ) 한국정보처리학회 2015 정보처리학회논문지. 소프트웨어 및 데이터 공학 Vol.4 No.11

도로주행 영상에서의 객체 검출에 관한 기존의 연구들은 날씨 및 조명 상태에 따른 객체 검출의 어려움 때문에 대부분 맑은 날씨의 영상을 대상으로 연구가 진행되었다. 본 논문에서는 도로주행 영상의 다양한 날씨 및 조명 상태를 먼저 판단하고, 이를 기반으로 도로 이정표에 대한 색상모델을 설정하여 이정표 객체를 찾는 방법을 제안한다. 제안한 방법은 5종류의 도로 이미지 특징을 이용하여 맑음, 흐림, 비, 야간, 역광으로 날씨 및 조명 상태를 먼저 분류하고, 각각의 상태에서 대상 이정표 색상의 픽셀값의 범위를 추출하여 GMM(Gaussian Mixture Model)을 생성하고 이를 객체 추출에 사용한다. 날씨 및 조명이 다양하게 변하는 도로주행 영상에 제안한 방법을 적용하여 이정표 영역이 안정적으로 찾아지는 것을 확인할 수 있었다. Road-view object classification methods are mostly influenced by weather and illumination conditions, thus the most of the research activities are based on dataset in clean weathers. In this paper, we present a road-view object classification method based on color segmentation that works for all kinds of weathers. The proposed method first classifies the weather and illumination conditions and then applies the weather-specified color models to find the road traffic signs. Using 5 different features of the road-view images, we classify the weather and light conditions as sunny, cloudy, rainy, night, and backlight. Based on the classified weather and illuminations, our model selects the weather-specific color ranges to generate Gaussian Mixture Model for each colors, Green, Yellow, and Blue. The proposed method successfully detects the traffic signs regardless of the weather and illumination conditions.

Broken Detection of the Traffic Sign by using the Location Histogram Matching

Yang, Liu,Lee, Suk-Hwan,Kwon, Seong-Geun,Moon, Kwang-Seok,Kwon, Ki-Ryong Korea Multimedia Society 2012 멀티미디어학회논문지 Vol.15 No.3

The paper presents an approach for recognizing the broken area of the traffic signs. The method is based on the Recognition System for Traffic Signs (RSTS). This paper describes an approach to using the location histogram matching for the broken traffic signs recognition, after the general process of the image detection and image categorization. The recognition proceeds by using the SIFT matching to adjust the acquired image to a standard position, then the histogram bin will be compared preprocessed image with reference image, and finally output the location and percents value of the broken area. And between the processing, some preprocessing like the blurring is added in the paper to improve the performance. And after the reorganization, the program can operate with the GPS for traffic signs maintenance. Experimental results verified that our scheme have a relatively high recognition rate and a good performance in general situation.

저가형 센서융합기반 정밀측위를 위한 코너검출 기반 도로표지판 검출

이성주(Sung Joo Lee),서재규(Jae Kyu Suhr),정호기(Ho Gi Jung) 한국자동차공학회 2018 한국 자동차공학회논문집 Vol.26 No.1

To overcome the limitations of satellite navigation and inertial navigation, low-cost, precise localization systems that utilize road facilities as landmarks are being developed. Road signs are regarded as important landmarks because they are installed on all roads, and they can be seen even during congestion, unlike road surface markings. In this paper, we have proposed a method to detect road signs with various aspect ratios by identifying corners and combining such elements. By verifying each step to maximize detection performance and by tracking the bottom corners, the corners at close range can be used. The proposed method is applied to the images acquired on the highway, and high detection performance and real-time operation can be confirmed.

도로표지의 효율적인 데이터베이스 구축방안

김의명(Kim, Eui Myoung),조두영(Cho, Du Young),정규수(Chong, Kyu Soo),김성훈(Kim, Seong Hoon) 대한공간정보학회 2011 대한공간정보학회지 Vol.19 No.3

도로표지는 운전자에게 안전하고 편안하게 목적지까지 안내를 목적으로 하는 교통시설이다. 도로표지는 신규노선이나 노선의 변경 그리고 도로표지의 노후화 등에 의해 지속적으로 현지조사와 이를 데이터베이스화하는 노력이 필요하다. 본 연구에서는 이러한 도로표지의 현지조사와 데이터베이스 구축을 효율적으로 수행할 수 있는 방안을 제시하는 것을 목적으로 하였다. 이를 위해 현지조사를 위한 모바일 매핑시스템을 설계하였다. 설계된 모바일매핑 시스템은 도로표지 영상정보를 획득할 수 있는 3대의 카메라, 차량의 위치와 자세를 알 수 있는 GPS/IMU/DMI, 그리고 도로표지 지점위치와 노선정보를 획득할 수 있는 레이저스캐너로 구성하였다. 또한 도로표지 영상에서 자동으로 도로표지 영역을 검출하고 이로부터 문자인식을 수행하는 절차를 제시하였다. Road signs are part of the traffic facilities intended to guide drivers to their destinations in a safe and comfortable manner. Due to the creation of new routes, changes to the old routes, and the deterioration of road signs, road signs do require efforts to do ongoing field investigations and put the results in a database. The purpose of this study was to propose methodologies to do field investigations and build a database for road signs efficiently. For that purpose, a mobile mapping system was designed for field investigations. The designed mobile mapping system was comprised of three cameras to produce image information about road signs, GPS/IMU/DMI to obtain information about the position and attitude of a vehicle, and a laser scanner to generate information about the locations of road signs and routes. Also proposed in the study was a procedure to automatically detect the areas of road signs in the road signs images and recognize their characters.

영상처리 기법을 이용한 차선 및 원형 도로 표지판 검출 알고리즘

유진영, 김주현, 최경주 충북대학교 컴퓨터 정보통신 연구소 2017 컴퓨터정보통신연구 Vol.25 No.1

본문참고

자동차 외부 및 내부 대상 검출 모델

원웅재(Woong-Jae Won),손준우(JoonWoo Son),정우영(Wooyoung Jung) 한국자동차공학회 2009 한국자동차공학회 부문종합 학술대회 Vol.2009 No.4

In this paper we proposed a simple vehicle in-out object detection model to implement vision-based interactive intelligent vehicle. In order to simply localize vehicle in-out target object, we consider target object feature maximizing method as reflecting target color feature characteristic. Moreover, we adopt gaussian pyramid image based center surround and normalization algorithms to not only reinforce target object area, but also inhibit background noise influence. We also limit two target objects which are road traffic sign for vehicle out object and driver's face for vehicle for describing proposed model and making experiment. In experiment result, the proposed model can successfully localize task specific in-out target object areas.