EP2149107A1 - Procédé et dispositif pour reconnaître des panneaux de signalisation - Google Patents

Procédé et dispositif pour reconnaître des panneaux de signalisation

Info

Publication number
EP2149107A1
EP2149107A1 EP08758434A EP08758434A EP2149107A1 EP 2149107 A1 EP2149107 A1 EP 2149107A1 EP 08758434 A EP08758434 A EP 08758434A EP 08758434 A EP08758434 A EP 08758434A EP 2149107 A1 EP2149107 A1 EP 2149107A1
Authority
EP
European Patent Office
Prior art keywords
image
traffic sign
components
images
motor vehicle
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Withdrawn
Application number
EP08758434A
Other languages
German (de)
English (en)
Inventor
Lars-Peter Becker
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hella GmbH and Co KGaA
Original Assignee
Hella KGaA Huek and Co
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Hella KGaA Huek and Co filed Critical Hella KGaA Huek and Co
Publication of EP2149107A1 publication Critical patent/EP2149107A1/fr
Withdrawn legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T3/00—Geometric image transformations in the plane of the image
    • G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
    • G06T3/4053—Scaling of whole images or parts thereof, e.g. expanding or contracting based on super-resolution, i.e. the output image resolution being higher than the sensor resolution
    • G06T3/4076—Scaling of whole images or parts thereof, e.g. expanding or contracting based on super-resolution, i.e. the output image resolution being higher than the sensor resolution using the original low-resolution images to iteratively correct the high-resolution images
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
    • G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
    • G06V20/582—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads of traffic signs

Definitions

  • the invention relates to a method and a device for traffic sign recognition, in which at least two images, each with an image of a region of the surroundings of a motor vehicle, are detected as an image sequence by an image acquisition system of a motor vehicle.
  • the image acquisition system generates corresponding image data for each captured image of the image that is processed and evaluated for traffic sign recognition.
  • the images provided by an image acquisition system for traffic sign recognition have insufficient quality, particularly in unfavorable environmental conditions, unfavorable exposure angles, distant traffic signs and / or high vehicle speed, to perform traffic sign recognition with high recognition accuracy.
  • Various systems for traffic sign recognition are known, for example, from the documents EP 1 145 186 B1, EP 1 383 098 B1 and EP 1 131 803 B1. These systems analyze the captured images and perform traffic sign recognition for the traffic signs contained in the images.
  • the object of the invention is to provide the traffic signal recognition images provided by an image acquisition system of the traffic signs. improve their quality and to achieve a high recognition quality in traffic sign recognition.
  • a classification of the object by means of a known classification method is carried out for traffic sign recognition.
  • the classification method for classifying the object as a traffic sign it is first determined whether the object belongs to the group of relevant traffic signs and, if so, to which specific traffic sign it is.
  • the object detected in each of the images is a traffic sign or could be a traffic sign.
  • the determination of the image components of the object in the individual images and the superimposition of the image components is preferably carried out only when it has been determined that the object is or could be a traffic sign.
  • the pictorial components particular super-resolution procedure.
  • the object is assumed to be stationary, since the relevant objects are fixed traffic signs. This makes it possible to determine the position and size change of the object between the first and the second image due to the known proper motion of the motor vehicle between two images.
  • the position change in the image of an object can be determined by comparing specific features, such as shape, color and / or texture properties, of the object in the first image with candidates of objects in the second image, wherein the object in the second image is selected has the most similar properties and thus has the greatest agreement with the object found in the first image.
  • the image components assigned to the object in the first image and the image components assigned to the object in at least one second image are adjusted by scaling and translation such that differences in size and / or position of the image components are compensated. Furthermore, when superimposing the image constituents, the local image resolution or image quality of the image constituents of the object, preferably by means of a super-resolution method, can be increased. Through the over- storing the image components of the object, in particular, the image sharpness of the image of the object is increased in the image generated by the superimposition of the image components of the object.
  • the device with the features of claim 9 can be further developed in the same manner as specified for the method according to claim 1.
  • the device with the features of claim 9 can be further developed with the features specified for further development of the method according to claim 1 in the dependent claims or corresponding device features.
  • Fig. 1 is a block diagram of a system for image recording
  • FIGS. 2a to 2e show a plurality of successively recorded images with the image of the same traffic sign
  • Figures 2a to 2f generated illustration of the traffic sign; F ig. 3 a to 3 e successively taken pictures of a traffic sign; and
  • Fig. 3f is a composite of image components of the images 3a to 3e illustration of the traffic sign.
  • FIG. 1 shows a block diagram of an image acquisition and evaluation system 10 which is integrated in a motor vehicle, in particular in a car or a truck.
  • the image acquisition and evaluation system 10 has a camera system 12 which captures an image sequence with images of at least one area in front of the vehicle and respectively generates corresponding image data.
  • This image data is transmitted to an image processing unit 14 of the system 10 and analyzed and processed by means of control and processing modules 16a to 16d of the image processing unit 14.
  • the image processing unit 14 is connected via a vehicle bus 18 of the vehicle with further control and processing units 20 to 26, such as the navigation system, the speed detection and / or control system of the vehicle, the image processing unit 14 via the vehicle bus 18 data with these other control units 20 to 26 can exchange.
  • the image processing unit 14 is further connected via a suitable data line with a display and warning unit 28 and with a speed control and / or limiting controller 30.
  • the image processing unit 14 is connected to a non-volatile memory 32 via a data line.
  • a non-volatile memory 32 may be, for example, a hard disk memory, a flash memory or another control unit.
  • the permanent memory 32 can also be replaced by a database system of the motor vehicle be provided.
  • the camera system 12 can comprise a monocamera, a plurality of monocameras, a stereo camera system and / or a plurality of stereo camera systems, wherein the individual cameras generate grayscale images or color images.
  • a vehicle bus 18 conventional known bus systems, such as a LIN bus, a CAN bus and / or a Flexray bus can be used.
  • the optics of the camera or of the cameras of the camera system 12 can in particular have different fixed focal lengths or a focal length which is adjustable in particular via a revolving objective system or a variable objective system.
  • the image processing unit 14 also communicates with other sensors and receiving units, which can in particular record an ad hoc communication with other vehicles and / or with traffic facilities, such as traffic lights, gantries and / or individual traffic signs. Furthermore, information about signal states of individual actuators and sensors, in particular about the activation of the turn signals of the vehicle, is transmitted via the vehicle bus 18.
  • the traffic signs in the detection area of the camera system 12 are determined and classified by an analysis of the image data with the aid of the image processing unit 14.
  • the imaged traffic sign is identified as a concrete traffic sign.
  • the display and warning unit 28 displays the image of the traffic sign and / or a symbol corresponding to this traffic sign. Additionally or alternatively, text information and / or acoustic information can also be output.
  • traffic information is transmitted to the speed control and / or limiting control 30, in particular via a speed limit or a minimum speed, wherein the speed control and / or limiting control 30 can output a warning signal when the permissible maximum speed is exceeded and / or the minimum speed is not reached.
  • the speed limits may be used as the maximum limit values of a speed control and / or limit possibly with an offset.
  • a further processing and control module 16a to 16d of the image processing unit 14 determines the current lane of the vehicle based on the images of the image sequence.
  • a further processing module 16a to 16d allocates the traffic signs which can be assigned to one of the determined lanes only to this traffic lane, and these lane traffic signals assigned to the traffic lane are stored in the permanent memory 32. Further, in the non-volatile memory 32, comparison data for a pattern comparison for recognizing road signs, particularly country-specific patterns of traffic signs, symbols that can be output instead of the images of the recognized traffic signs on the display unit 28, as well as texts and / or acoustic information to be output are stored.
  • the traffic sign is detected as an object in a plurality of images taken one after the other as an image sequence. This is done in particular by means of a known tracking method. In each of these images or at least in a part of these images, image elements of the traffic sign are determined, which are then superimposed so that an improved overall image of the traffic sign is generated. Image data of the improved overall image is then further processed, the traffic sign being classified based on the image data of the improved overall image. By classifying a recognition of the traffic sign takes place. The superposition of the image components is carried out in particular by means of a super-resolution method.
  • a size and / or positional adjustment takes place in order to compensate for the size and positional differences resulting from the vehicle movement during and between the image recordings.
  • the determined object can be preclassified, the further processing for traffic sign recognition or for image improvement being carried out with the help of the superposition of the image components only if the object is or could be a traffic sign.
  • FIGS. 2 a to 2 e show successively recorded images with the aid of the camera system 12 of the vehicle, which have a relatively large image noise as a consequence of the relatively small image of the traffic sign in the respective overall image recorded by the camera system 12.
  • the distance between the traffic sign and the vehicle or the camera system 12 is continuously reduced between the images 2a to 2e recorded at the same time intervals, preferably with a constant vehicle speed between the recording times of the images 2a to 2e.
  • FIGS. 3 a to 3 e show further images of other images taken by the camera system 12 under different recording conditions. 2a to 2e shown, wherein the images shown in Figures 3a to 3e have a relatively large motion blur, especially as a result of relatively large exposure times and the proper movement of the camera system 12 during exposure.
  • the image components of the images associated with the traffic sign of the images shown in FIGS. 3a to 3e are superimposed to the image shown in FIG. 3f. Both the image shown in FIG. 2f and in FIG.
  • the objects to be classified located in the driving path of the vehicle move due to the relative movement of the vehicle between the recording times of the images with respect to the image plane of the camera system 12. This movement is visible through the different positions of the same object in several sequentially recorded images.
  • the position and / or size of the image constituents is therefore preferably adjusted such that an overlaying of an improved representation of the traffic sign is achieved.
  • high-quality images of a traffic signal arranged in the driving lane can be generated for traffic sign recognition.
  • the objects in the individual images are often displayed only blurred as a result of the described relative movement between the vehicle and the object.
  • image signal-specific information is utilized in the corresponding local image sections, the object being in particular the traffic sign to be classified and thus causing a qualitative improvement of the image of the traffic sign.
  • an advantageous utilization of the local image area e.g with respect to a pixel or a limited plurality of pixels
  • the image signal always being derived from the image signal same location of an object is emitted.
  • the image data of several independent images of the image sequence are used to determine the image signal and the variation.
  • the variation of the image signal superimposed at least parts of an undisturbed image signal or an undisturbed image.
  • the determination of all undistorted local image signals as a whole results in a qualitatively improved image component.
  • image signal-specific information which is contained in the individual images of the image sequence, when localizing the image components of the object, in particular a local image resolution can be increased and the motion blur of the image of the object can be reduced (eliminated).
  • the image enhancement can in particular be carried out with the aid of a suitable program by a data processing unit, in particular a signal processor or another control unit, and is fundamentally applicable to any camera system 12 or any other image acquisition system.
  • a data processing unit in particular a signal processor or another control unit
  • the image enhancement can be done both for traffic signs and for additional traffic signs.
  • a robust classification with a higher performance and better probability of detection and recognition can be achieved.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Traffic Control Systems (AREA)
  • Image Processing (AREA)

Abstract

L'invention concerne un procédé et un dispositif permettant la reconnaissance de panneaux de signalisation. Au moins deux images comprenant chacune un cliché d'une partie de l'environnement d'un véhicule sont déterminées en tant que suite d'images par un système de collecte d'images (12) du véhicule. Ledit système de collecte d'images (12) produit, pour chaque image collectée du cliché, des données d'images correspondantes. Les données d'images sont ensuite traitées, les composantes d'images associés à cet objet dans chaque image étant déterminées pour au moins un objet. Les composantes d'images déterminées des images sont superposées, un cliché amélioré de l'objet étant produit à l'aide des composantes superposées. Une reconnaissance des panneaux de signalisation est ainsi obtenue à l'aide du cliché amélioré de l'objet.
EP08758434A 2007-05-08 2008-05-08 Procédé et dispositif pour reconnaître des panneaux de signalisation Withdrawn EP2149107A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
DE102007021577A DE102007021577A1 (de) 2007-05-08 2007-05-08 Verfahren und Vorrichtung zur Verkehrszeichenerkennung
PCT/EP2008/003745 WO2008135295A1 (fr) 2007-05-08 2008-05-08 Procédé et dispositif pour reconnaître des panneaux de signalisation

Publications (1)

Publication Number Publication Date
EP2149107A1 true EP2149107A1 (fr) 2010-02-03

Family

ID=39687083

Family Applications (1)

Application Number Title Priority Date Filing Date
EP08758434A Withdrawn EP2149107A1 (fr) 2007-05-08 2008-05-08 Procédé et dispositif pour reconnaître des panneaux de signalisation

Country Status (3)

Country Link
EP (1) EP2149107A1 (fr)
DE (1) DE102007021577A1 (fr)
WO (1) WO2008135295A1 (fr)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102009030594A1 (de) 2009-06-26 2010-12-30 Volkswagen Ag Kraftfahrzeug mit Verkehrszeichenerkennung
DE102015210117A1 (de) 2015-06-02 2016-12-08 Conti Temic Microelectronic Gmbh Fahrerassistenzvorrichtung zur Verkehrszeichenerkennung für ein Kraftfahrzeug sowie ein Verfahren zur Überprüfung einer etwaigen Fehlerkennung eines mit der Fahrerassistenzvorrichtung erkannten Verkehrszeichens

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE19852631C2 (de) * 1998-11-14 2001-09-06 Daimler Chrysler Ag Vorrichtung und Verfahren zur Verkehrszeichenerkennung
EP1383098B1 (fr) 2002-07-09 2006-05-17 Accenture Global Services GmbH Dispositif de détection automatique de panneaux de signalisation routière
KR100504594B1 (ko) * 2003-06-27 2005-08-30 주식회사 성진씨앤씨 데이터 압축 처리된 저해상도 영상으로부터 초해상도 영상복원 및 재구성 방법
US7466841B2 (en) * 2004-08-16 2008-12-16 Siemens Corporate Research, Inc. Method for traffic sign detection
US7492962B2 (en) * 2005-08-25 2009-02-17 Delphi Technologies, Inc. System or method for enhancing an image

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See references of WO2008135295A1 *

Also Published As

Publication number Publication date
DE102007021577A1 (de) 2008-11-13
WO2008135295A1 (fr) 2008-11-13

Similar Documents

Publication Publication Date Title
EP2150939B1 (fr) Procédé et dispositif pour déterminer la position d'un panneau de signalisation
EP2179381B1 (fr) Procédé et dispositif servant à la reconnaissance de panneaux de signalisation routière
EP3053097A1 (fr) Procédé et dispositif de reconnaissance de panneaux de signalisation
EP2573709A2 (fr) Procédé et dispositif destinés à identifier des véhicules automobiles pour la surveillance du trafic
DE102018212655A1 (de) Erkennung der Bewegungsabsicht eines Fußgängers aus Kamerabildern
EP3642758B1 (fr) Procédé d'évaluation d'un aspect visuel dans un environnement du véhicule et véhicule
DE102014106506A1 (de) Verfahren zum Durchführen einer Diagnose eines Kamerasystems eines Kraftfahrzeugs, Kamerasystem und Kraftfahrzeug
EP3488607A1 (fr) Dispositif de prise de vues servant à capter une zone environnante d'un véhicule équipé dudit dispositif et procédé permettant d'obtenir une fonction d'aide à la conduite
DE102011082477A1 (de) Verfahren und System zur Erstellung einer digitalen Abbildung eines Fahrzeugumfeldes
DE102018004782A1 (de) Verfahren und Vorrichtung zum Erfassen einer Umgebung, sowie Fahrzeug mit einer solchen Vorrichtung
DE102006037600B4 (de) Verfahren zur auflösungsabhängigen Darstellung der Umgebung eines Kraftfahrzeugs
WO2014146864A1 (fr) Procédé et dispositif d'obtention d'une vue d'un objet s'allumant de manière pulsée
DE102017208994A1 (de) Verfahren zur Ermittlung von Ergebnisbilddaten
EP2149107A1 (fr) Procédé et dispositif pour reconnaître des panneaux de signalisation
DE102007021579A1 (de) Verfahren und Vorrichtung zur Klassifikation mindestens eines Verkehrszeichens
DE102016103037A1 (de) Verfahren zum Überprüfen einer Funktionstüchtigkeit eines Kamera-Monitor-Systems, Kamera-Monitor-System sowie Kraftfahrzeug
DE102019003061B4 (de) Vorrichtung zum Prüfen eines optischen Erfassungssystems für ein Kraftfahrzeug, sowie ein Verfahren hierzu
DE102012020778A1 (de) Verfahren zum Labeln einer Sequenz von in zeitlicher Abfolge aufgenommenen Bildern mit integrierter Qualitätsprüfung
DE102009047437A1 (de) Verfahren und Vorrichtung zur Anpassung von Bildinformationen eines optischen Systems
DE102007008543A1 (de) Verfahren und Vorrichtung zum Ermitteln des Bewegungszustands von Objekten
DE102023104419A1 (de) Verfahren und vorrichtung zur korrektur eines bewegtbilds einer umgebung eines kraftfahrzeugs
DE102008014630A1 (de) Verfahren und Vorrichtung zum Verarbeiten von Bilddaten
DE102014218627A1 (de) Verfahren und Steuergerät zum Betreiben eines Bildsensors
DE102020213267A1 (de) Helligkeits-Umwandlung von Stereobildern
DE102015005363A1 (de) Verfahren und Vorrichtung zur augmentierten Darstellung sowie Fahrzeug

Legal Events

Date Code Title Description
PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

17P Request for examination filed

Effective date: 20091124

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MT NL NO PL PT RO SE SI SK TR

AX Request for extension of the european patent

Extension state: AL BA MK RS

DAX Request for extension of the european patent (deleted)
17Q First examination report despatched

Effective date: 20110302

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE APPLICATION IS DEEMED TO BE WITHDRAWN

18D Application deemed to be withdrawn

Effective date: 20131203