EP3391282A1 - Procédé d'identification d'un objet dans une zone environnante d'un véhicule automobile, système d'assistance à la conduite ainsi que véhicule automobile - Google Patents

Procédé d'identification d'un objet dans une zone environnante d'un véhicule automobile, système d'assistance à la conduite ainsi que véhicule automobile

Info

Publication number
EP3391282A1
EP3391282A1 EP16816262.6A EP16816262A EP3391282A1 EP 3391282 A1 EP3391282 A1 EP 3391282A1 EP 16816262 A EP16816262 A EP 16816262A EP 3391282 A1 EP3391282 A1 EP 3391282A1
Authority
EP
European Patent Office
Prior art keywords
pedestrian
contrast pattern
motor vehicle
detected
identified
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Withdrawn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
EP16816262.6A
Other languages
German (de)
English (en)
Inventor
Markus Heimberger
Perikles RAMMOS
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.)
Valeo Schalter und Sensoren GmbH
Original Assignee
Valeo Schalter und Sensoren GmbH
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 Valeo Schalter und Sensoren GmbH filed Critical Valeo Schalter und Sensoren GmbH
Publication of EP3391282A1 publication Critical patent/EP3391282A1/fr
Withdrawn legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/20—Movements or behaviour, e.g. gesture recognition
    • G06V40/23—Recognition of whole body movements, e.g. for sport training
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion
    • G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/103—Static body considered as a whole, e.g. static pedestrian or occupant recognition
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60T—VEHICLE BRAKE CONTROL SYSTEMS OR PARTS THEREOF; BRAKE CONTROL SYSTEMS OR PARTS THEREOF, IN GENERAL; ARRANGEMENT OF BRAKING ELEMENTS ON VEHICLES IN GENERAL; PORTABLE DEVICES FOR PREVENTING UNWANTED MOVEMENT OF VEHICLES; VEHICLE MODIFICATIONS TO FACILITATE COOLING OF BRAKES
    • B60T7/00—Brake-action initiating means
    • B60T7/12—Brake-action initiating means for automatic initiation; for initiation not subject to will of driver or passenger
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30196—Human being; Person
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30248—Vehicle exterior or interior
    • G06T2207/30252—Vehicle exterior; Vicinity of vehicle

Definitions

  • the invention relates to a method for identifying an object in a
  • the invention also relates to a driver assistance system and a motor vehicle.
  • Driver assistance system may be, for example, an electronic brake assistant with a pedestrian recognition.
  • the brake assist is in particular designed to lower a speed of the motor vehicle when a distance of a detected pedestrian falls below a predetermined value. For example, a collision with the pedestrian can be prevented or at least one
  • a method for object recognition for example pedestrian detection (PD), is known, for example, from DE 10 2013 1 12 163 A1.
  • a method according to the invention serves to identify an object in a surrounding area of a motor vehicle as a pedestrian.
  • the surrounding area of the Motor vehicle captured in pictures.
  • a shadow projected from the object on a floor in the surrounding area and a contrast pattern formed on the floor and dependent on the shadow are recognized, and the object is compared with a predetermined pedestrian by comparing the formed contrast pattern
  • the method thus serves to classify a detected object.
  • the environmental region of the motor vehicle is captured in images by means of at least one camera.
  • four cameras are provided, wherein a camera may be arranged on a front region of the motor vehicle for detecting the surrounding area in front of the motor vehicle, a camera on a rear region of the motor vehicle for detecting the
  • Ambient area behind the motor vehicle can be arranged and in each case a camera can be arranged in each case a side mirror of the motor vehicle for detecting a driver side and a passenger side surrounding area of the motor vehicle.
  • the surrounding area of the motor vehicle can be monitored in particular completely and completely.
  • the object in the at least one image is not identified on the basis of the detected object itself, but on the basis of the contrast pattern on the ground, which depends on the shadow which the object throws on the ground.
  • the floor may be, for example, a roadway of the motor vehicle or a walkway for the pedestrian.
  • the contrast pattern is generated by the fact that the floor is partially darkened and partially illuminated by the shadow.
  • the contrast pattern comprises, in particular, the areas on the floor which are darkened by the shadow and the areas which are not darkened or illuminated and which adjoin the darkened areas.
  • Vehicle-side image processing device recognized and compared with the predetermined contrast pattern, which includes characteristic for a pedestrian light-shadow transitions.
  • the predetermined contrast pattern thus corresponds to a contrast pattern which would be generated by means of a shadow of a pedestrian on the ground. If the detected contrast pattern coincides with the characterizing contrast pattern or if the detected contrast pattern at most deviates a predetermined tolerance value from the characterizing contrast pattern, the detected object is identified as a pedestrian.
  • the invention is based on the finding that the light-dark transitions or light-dark progressions formed by the contrast pattern can be recognized quickly and easily. Thus, the object can be quickly and easily identified as a pedestrian.
  • the object is identified as a pedestrian if, in the contrast pattern generated by the shadow of the object, two substantially parallel or juxtaposed dark stripes and an intervening, lighter stripe are recognized as the predetermined contrast pattern.
  • the two parallel, dark stripes are in particular the shadows of legs of the pedestrian, through which the floor is partially darkened.
  • the intermediate light stripe is formed, for example, by the fact that light or sunlight shines through between the legs of the pedestrian. Through this sunlight shining between the legs sunlight is illuminated. From this embodiment, there is the advantage that only the legs of the pedestrian or only the shadow of the legs of the pedestrian must be detected to identify the object as a pedestrian. Thus, for example, at least one of the two cameras attached to the side mirrors of the motor vehicle can be used to detect the object.
  • the object is identified as a pedestrian if it is detected that an orientation of at least one of the strips deviates at most by a predetermined angle from a detection axis of the camera.
  • Detection axis of the camera or a camera axis corresponds to a main axis of a detection range of the camera. In the case of a camera in the front region of the motor vehicle or in the rear region of the motor vehicle, this runs
  • Detection axis in particular along a vehicle longitudinal axis of the motor vehicle.
  • an angle between the at least one strip and the camera axis, in particular a projection of the camera axis on the ground, can be detected. This angle can be with the
  • the contrast pattern can thus be reliably identified as the contrast pattern generated by a pedestrian and, for example, not be confused with a contrast pattern of a road marking. Since the angle between the camera axis and the at least one strip changes, in particular when the pedestrian is moving, for example crossing the road, or when the motor vehicle is moving, it can additionally be determined from the detected angle and a detected speed of the motor vehicle whether the pedestrian moves or stands.
  • a respective length and / or a respective width of at least one of the strips is detected and the object is identified as a pedestrian if the respective length and / or the respective width is at most one
  • the predetermined value deviates from a predetermined length and / or width for the characterizing contrast pattern.
  • the width and / or length which is dependent on a dimensioning of the legs, plausible, usual for a person or characteristic values can be specified.
  • the detected width and / or detected length can thus be plausible whether the strip is the shadow of a leg.
  • the contrast pattern of the pedestrian can be represented by a contrast pattern of a lane marking, which is shown for example as a white line on the lane and which likewise has a contrast pattern in the form of a dark light. Dark transition generated, be differentiated.
  • the contrast pattern can be assigned to a pedestrian in a particularly simple manner. The method is thus designed to be particularly reliable.
  • a current light incidence direction of sunlight is determined with respect to the motor vehicle and the characterizing contrast pattern is determined as a function of the determined light incidence direction.
  • the invention is based on the knowledge that the recognized contrast pattern is dependent on the light incidence direction of the sunlight, for example a current sun position of the sun.
  • an orientation of the strips of the contrast pattern is dependent on the Light direction.
  • a light incident direction-specific, a pedestrian-characterizing contrast pattern can be deposited for each direction of light incidence and, depending on the current light incidence direction, the corresponding deposited contrast pattern for comparison with the detected
  • the direction of light incidence may, for example, based on a basic brightness
  • an internet-based weather database can be called up by a vehicle-side control device of the driver assistance system, in which current
  • Weather data are stored. Based on the weather data stored in the weather database, a position of the sun at a current position of the motor vehicle at the current time and thus the current light incidence direction can be determined as a function of an orientation of the motor vehicle. The position of the
  • Motor vehicle and the orientation of the motor vehicle can be determined for example by means of a vehicle-side GPS device.
  • characterizing contrast pattern can be determined by means of which the object can be classified reliably.
  • Surrounding area performing map determined and tracking the pedestrian in at least one other image, the object based on the shadow of the object and the formed contrast pattern as a pedestrian and based on the at least one further image, a further position of the object in the
  • the environment map is in particular a virtual map, in which, for example, positions and geometric dimensions of detected objects in the surrounding area, in particular distances of the detected objects relative to the motor vehicle, are registered.
  • the driver assistance system of the motor vehicle assistance functions such as automatic emergency braking or autonomous parking operation, provide.
  • the positions of the pedestrian are tracked over time and entered into the environment map. With others In other words, this means that a course of movement of the pedestrian is determined.
  • the pedestrian is thus tracked or "tracked.”
  • the positions tracked over time can be used, for example, to determine whether the pedestrian is standing or moving, for example, if it is detected that the pedestrian is standing, then automatic emergency braking can be dispensed with but it was detected that the pedestrian is moving towards the motor vehicle and threatens a collision with the motor vehicle, the automatic emergency braking can be performed by the driver assistance system.
  • the object itself is detected in at least one of the images and the object is identified and confirmed as a pedestrian only if the object is identified on the basis of the object recognized in at least one of the images and in at least one of the images recognized contrast pattern is identified as a pedestrian.
  • this means that the identification of the object is verified on the basis of the recognized contrast pattern by additionally recognizing the object itself in at least one image and identifying it as a pedestrian.
  • the method of object identification thus comprises two sub-methods. In a first partial method, the object is identified on the basis of the contrast pattern and in a second
  • Sub-method based on the detected in the at least one image object itself is based on the detected in the at least one image object itself.
  • Procedures for pedestrian identification based on the detected object are
  • Such a classifier may be, for example, the so-called support vector machine (SVM).
  • SVM support vector machine Only if the object has been identified as a pedestrian by means of both partial procedures, the pedestrian identification is considered successful and the detected object is confirmed as a pedestrian.
  • the object was identified as a pedestrian only by the first partial procedure, but the object was identified by the second
  • the recognized contrast pattern was not the contrast pattern of a pedestrian, but the contrast pattern, which was generated for example by a road marking on the floor. If the object was identified as a pedestrian only on the basis of the second partial method, but the identification of the object was negated on the basis of the first partial method, then it is assumed that the detected object is another object, for example a tree.
  • the partial methods therefore provide a redundancy by means of which the object can be reliably identified.
  • the object itself and the contrast pattern are recognized, and the object is identified as a pedestrian on the basis of the same image. In other words, this means that for both sub-processes the same image acquired by the vehicle-side camera for
  • Identification of the object is used. Thus, based on a single recorded image reliably the object can be identified as a pedestrian.
  • an operation of identifying the object as a pedestrian on the basis of the recognized object and a process of identifying the object based on the detected contrast pattern are carried out at least simultaneously at the same time.
  • an object can be identified and confirmed very fast as a pedestrian.
  • a position of the detected object and a position of the contrast pattern in an environment map are determined and the detected object and the contrast pattern are entered into the area map. Based on the position of the detected object and the location of the object
  • a distance of the object relative to the motor vehicle can be determined. This distance can, for example, the as
  • Contrast pattern can be assigned to the object.
  • a projection of the detected object on the ground is determined in the surroundings map and the object identified as a pedestrian when the projection of the object and the contrast pattern of the object entered in the surroundings map overlap at least in a predetermined overlapping area.
  • the projection can be done taking into account the current light incidence direction at the time of the recorded Image to be determined.
  • the projection and the shadow of the object that is to say the contrast pattern dependent on the artificially created shadow and the contrast pattern dependent on the shadow of the object
  • overlap at least in the predetermined overlapping area
  • the contrast pattern on the ground was created and the detected contrast pattern is not a road marking, for example.
  • the pedestrians recognized from the first sub-method and the second sub-method can advantageously be associated to form an object, and thus the detected object can be reliably identified as a pedestrian.
  • the invention also relates to a driver assistance system for identifying an object in a surrounding area of a motor vehicle as a pedestrian.
  • the driver assistance system comprises at least one vehicle-side camera for detecting the surrounding area in images.
  • an image processing device of the driver assistance system is configured to detect in at least one of the captured images a shadow projected from the object onto a floor in the surrounding area and a contrast pattern dependent on the shadow on the floor and the object from a comparison of that from the shadow dependent
  • the driver assistance system can be designed, for example, as an automatic emergency brake assistant and, in the event of an imminent collision with an object identified as a pedestrian, automatically decelerate the motor vehicle.
  • a motor vehicle according to the invention comprises an inventive
  • the motor vehicle is designed in particular as a passenger car.
  • Embodiments and their advantages apply correspondingly for the driver assistance system according to the invention and for the motor vehicle according to the invention.
  • Fig. 1 is a schematic representation of an embodiment of a
  • FIG. 2 is a schematic representation of an image captured by a vehicle-mounted camera.
  • Fig. 1 shows a motor vehicle 1 with a driver assistance system 2.
  • Driver assistance system 2 is used to identify an object 3 in one
  • Driver assistance system 2 can be configured, for example, as an automatic emergency brake assist, which warns a driver of the motor vehicle of an imminent collision with the pedestrian 5 and / or automatically decelerates the motor vehicle 1 in the event of an impending collision with the pedestrian.
  • the driver assistance system 2 comprises at least one camera 6 on the vehicle.
  • the motor vehicle 1 comprises four cameras 6 on the vehicle, a camera 6 being arranged in a front region 7 of the motor vehicle 1.
  • Camera 6 the surrounding area 4 can be monitored in front of the motor vehicle 1.
  • a further camera 6 is arranged at a rear area 8 of the motor vehicle 1 in order to monitor the surrounding area 4 behind the motor vehicle 1.
  • Two further cameras 6 are integrated here in each case a side mirror 9 of the motor vehicle 1, so that the surrounding area 4 can be monitored laterally of the motor vehicle 1.
  • the cameras 6 are designed to detect the surrounding area 4 of the motor vehicle 1 in pictures B, one of the camera 6 in the front area 7 of the motor vehicle
  • I captured image B is exemplified in Fig. 2.
  • the at least one captured image B is from an on-vehicle image capture device 10 of
  • Driver assistance system 2 a shadow 1 1 detected, which projects the object 3 on a floor 12.
  • the floor 12 may, for example, a roadway for the
  • I I is on the floor 12 a dependent of the shadow 1 1 contrast pattern 13 generated, which is detected by the image processing device 10 of the driver assistance system 2 in the at least one image B.
  • This detected contrast pattern 13 is compared with a predetermined contrast pattern characterizing a pedestrian, and the detected object 3 is identified as a pedestrian 5 on the basis of this comparison.
  • the detected object 3 is identified on the basis of the comparison as a pedestrian 5, if the detected contrast pattern 13 at most by a predetermined value deviates from the characterizing contrast pattern, so if the detected
  • Contrast pattern 13 and the characterizing contrast pattern overlap or overlap at least in a predetermined overlap area.
  • the object 3 is identified in particular as a pedestrian 5 if the contrast pattern 13 has two substantially parallel or juxtaposed, dark stripes 14 or strip-shaped regions and a light strip 15 or strip-shaped region extending between the dark stripes 14.
  • the dark stripes 14 correspond to a darkened area on the floor 12, which is generated by the shadow of legs 16 of the pedestrian 5.
  • the light strip 15 extending between the dark stripes 14 corresponds to a non-darkened area on the floor 12.
  • This contrasting pattern 13 of alternating a dark, a light and a dark area on the floor 12 is defined as being characteristic of a pedestrian.
  • a width 17 of at least one of the strips 14, 15 and / or a length 18 of at least one of the strips 14, 15 is detected.
  • the object 3 is identified as pedestrian 5 only if the detected width 17 and / or the detected length 18 are at most a predetermined value of one for which characterizing
  • Contrast pattern of predetermined length and / or width deviates.
  • the width and the length of the strips 14, 15 are dependent on a width and length of the legs 16 of the pedestrian 5.
  • typical for a person values can be specified so that can be determined by the image processing device 10 Whether a detected width 17 and / or a detected length 18 in the contrast pattern 13 for a pedestrian is realistic or plausible.
  • a road marking 19 on the ground which also produces a contrast pattern of alternating a dark, a light and a dark area on the floor 12, reliably distinguished from the shadow of the shadow 1 1 of the pedestrian 5 contrast pattern 13, as the Width of the dark stripes in the contrast pattern generated by the road marking 19 is significantly greater than the width of the dark stripes 14 in the contrast pattern 13 of a pedestrian. 5
  • an angle 22 between the at least one strip 14, 15 and a detection axis 23 or camera axis of the camera 6 is detected.
  • the object 3 is then identified as the pedestrian 5 if the orientation of the at least one strip 14 differs at most by a predetermined angle from the detection axis 23, that is, if the detected angle 22 does not exceed the predetermined angle.
  • the detection axis 23 corresponds to a main axis of a detection region 24 of the front-side camera 6, which runs along a vehicle longitudinal direction of the motor vehicle 1 here. If the pedestrian 5 in front of the front camera 6 crosses the lane or the pedestrian 5 is in front of the motor vehicle 1 and thus in front of the front camera 6, the strips 14, 15 of the contrast pattern 13 extend in particular along the detection axis 23.
  • the object 3 is thus identified as pedestrian 5, in particular, when at least one of the strips 14, 15 extends along the camera axis 23, when the angle 22 is thus approximately 0 ° than the predetermined angle.
  • the angle 22 between the at least one strip 14, 15 and the detection axis 23 may change over time, especially when the
  • Pedestrian 5 or the motor vehicle 1 move. Based on the angle change and Based on a detected speed of the motor vehicle 1 can thus be detected whether the pedestrian 5 moves or stands.
  • a light incident direction 20 of the light of the sun S can be determined. Based on the light incident direction 20, the characterizing
  • Contrast patterns are determined.
  • the invention is based on the knowledge that the projected on the floor 12 shadow 1 1 of the object 3 of the
  • Light incident direction 20 is dependent.
  • the orientation of the strips 14, 15 is dependent on the direction of light incidence 20.
  • the direction of light incidence 20 depends, in particular, on a position of the sun S.
  • the driver assistance system 2 may, for example, comprise a rain-light tunnel sensor 25 and / or have a sun position sensor, which can detect the sun's position of the sun S and thus the light incident direction 20.
  • the direction of light incidence 20 can also be determined on the basis of the image B captured by the camera 6, in that the camera 6 itself and / or the image processing device 10 detects a basic brightness or brightness of the image B. From the detected brightness, the light incident direction 20 can be determined.
  • the driver assistance system 2 is designed to be a
  • Internet-based weather database 21 call to receive, for example, current weather data at a position of the motor vehicle 1 at the current time or time of day.
  • the light incident direction 20 can be reliably determined.
  • the position and the orientation of the motor vehicle 1 can be determined for example by means of a GPS device of the motor vehicle 1, not shown here.
  • a position P of the pedestrian 5 can be detected and registered in an environment map, which represents the surrounding area 4.
  • the surrounding area 4 can be imaged for the driver assistance system 2; in particular, distances from detected objects 3 to the motor vehicle 1 can be stored.
  • the position P of the pedestrian 5 can be determined again and also entered in the map.
  • the position P of the pedestrian 5 over time, so a movement history of the pedestrian 5, are detected.
  • it can be determined whether the pedestrian 5 is moving, for example crossing the roadway of the motor vehicle 1, or whether the pedestrian 5 is standing.
  • it can be detected at which position P the pedestrian 5 stands and thus the distance of the pedestrian 5 relative to the motor vehicle 1 is determined.
  • an emergency braking of the motor vehicle 1 is for example necessary when the pedestrian 5 is on the roadway of the motor vehicle 1 or moves on the road and thus a collision with the on the
  • Road locomotive motor vehicle 1 threatens.
  • the emergency braking is not necessary if the pedestrian 5 stands on a sidewalk next to the roadway of the motor vehicle 1 and the motor vehicle 1 traveling on the roadway passes by the pedestrian 5.
  • the object 3 itself can be detected in the at least one image B.
  • the detected object 3 can be identified as a pedestrian 5, for example by means of a classifier.
  • the detected object 3 can for example be identified as the pedestrian 5 on the basis of a comparison with a stored template.
  • the object 3 itself can be identified as a pedestrian 5. Only if the object 3 based on the contrast pattern 13 and based on the object 3 itself as
  • the pedestrian identification is considered successful and the detected object 3 confirmed as a pedestrian 5. If the detected object 3 has been identified as a pedestrian 5 only on the basis of the contrast pattern 13, but not on the basis of the object 3 itself, it may be considered likely that the detected contrast pattern 13 was generated by the road marking 19. If the detected object 3 has been identified as a pedestrian 5 only on the basis of the object 3 itself, but not on the basis of the contrast pattern 13, then it may be considered probable that the detected object 3 is another object, for example a tree.
  • Image processing device 10 is provided for identifying the object 3.
  • the operations can be carried out at least temporarily simultaneously.
  • the image processing device 10 can thus perform the process of identification based on the object 3 itself and the process of identification using the contrast pattern 13 in parallel by means of the image B.
  • the process is designed especially fast.
  • a projection of the object 3 are determined on the ground 12.
  • Image processing device 10 a shadow of the object 3 on the floor 12 generated or determined. If the projection of the object 3 and the contrast pattern 13, ie the virtually generated shadows and the detected shadows 1 1 of the object 3, overlap at least in regions, the object 3 is identified as a pedestrian 5.

Abstract

L'invention concerne un procédé d'identification d'un objet (3) dans une zone environnante (4) d'un véhicule automobile (1) en tant que piéton (5), selon lequel la zone environnante (4) est capturée dans des images (B) au moyen d'au moins une caméra (6) à bord du véhicule. Dans au moins une des images capturées (B), une ombre (11) projetée par l'objet (3) sur un sol (12) dans la zone environnante (4) et un motif de contraste (13) formé sur le sol (12) et dépendant de l'ombre (11) sont reconnus, et l'objet (3) est identifié à l'aide d'une comparaison du motif de contraste (13) à un motif de contraste prédéfini caractérisant un piéton (5). L'invention concerne en outre un système d'aide à la conduite (2) ainsi qu'un véhicule automobile (1).
EP16816262.6A 2015-12-16 2016-12-16 Procédé d'identification d'un objet dans une zone environnante d'un véhicule automobile, système d'assistance à la conduite ainsi que véhicule automobile Withdrawn EP3391282A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
DE102015121952.2A DE102015121952A1 (de) 2015-12-16 2015-12-16 Verfahren zum Identifizieren eines Objektes in einem Umgebungsbereich eines Kraftfahrzeugs, Fahrerassistenzsystem sowie Kraftfahrzeug
PCT/EP2016/081409 WO2017103096A1 (fr) 2015-12-16 2016-12-16 Procédé d'identification d'un objet dans une zone environnante d'un véhicule automobile, système d'assistance à la conduite ainsi que véhicule automobile

Publications (1)

Publication Number Publication Date
EP3391282A1 true EP3391282A1 (fr) 2018-10-24

Family

ID=57590515

Family Applications (1)

Application Number Title Priority Date Filing Date
EP16816262.6A Withdrawn EP3391282A1 (fr) 2015-12-16 2016-12-16 Procédé d'identification d'un objet dans une zone environnante d'un véhicule automobile, système d'assistance à la conduite ainsi que véhicule automobile

Country Status (6)

Country Link
US (1) US10635896B2 (fr)
EP (1) EP3391282A1 (fr)
JP (1) JP6698163B2 (fr)
CN (1) CN108496176B (fr)
DE (1) DE102015121952A1 (fr)
WO (1) WO2017103096A1 (fr)

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102017200574B4 (de) * 2017-01-16 2019-12-12 Robert Bosch Gmbh Verfahren zum Überwachen eines Umfelds eines Fahrzeugs
DE102017206974A1 (de) * 2017-04-26 2018-10-31 Conti Temic Microelectronic Gmbh Verfahren zur indirekten Detektion eines verdeckten Verkehrsteilnehmers
DE102017010731A1 (de) 2017-11-20 2018-05-30 Daimler Ag Verfahren zur Erfassung eines Objektes
US20190164309A1 (en) * 2017-11-29 2019-05-30 Electronics And Telecommunications Research Institute Method of detecting shooting direction and apparatuses performing the same
DE102018203242A1 (de) 2018-03-05 2019-09-05 Robert Bosch Gmbh Messung von Fahrzeugabmessungen aus einem Schattenwurf
KR20210136567A (ko) * 2020-05-08 2021-11-17 현대자동차주식회사 차량의 커뮤니케이션 조명 시스템
DE102020206829A1 (de) 2020-06-02 2021-12-02 Robert Bosch Gesellschaft mit beschränkter Haftung Verfahren und Vorrichtung zur Bestimmung der Höhe von Pflanzen und Computerprogrammprodukt
US11900695B2 (en) 2021-03-04 2024-02-13 Nec Corporation Of America Marking and detecting road marks
US11881033B2 (en) * 2021-03-04 2024-01-23 Nec Corporation Of America Reliable visual markers based on multispectral characteristics
DE102021213881A1 (de) 2021-12-07 2023-06-07 Robert Bosch Gesellschaft mit beschränkter Haftung Verfahren und Vorrichtung zum Betreiben eines Fahrzeugs
GB2621822A (en) * 2022-08-17 2024-02-28 Skystrm Ltd Monitoring system

Family Cites Families (21)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2901112B2 (ja) * 1991-09-19 1999-06-07 矢崎総業株式会社 車両周辺監視装置
JP3574780B2 (ja) 2000-08-29 2004-10-06 株式会社日立製作所 車両の安全運転支援システム
DE10160719B4 (de) * 2001-12-11 2011-06-16 Deutsches Zentrum für Luft- und Raumfahrt e.V. Verfahren und Vorrichtung zur Erkennung und Wiedererkennung von sich bewegenden Objekten
DE10324895A1 (de) * 2003-05-30 2004-12-16 Robert Bosch Gmbh Verfahren und Vorrichtung zur Objektortung für Kraftfahrzeuge
DE102006003538B3 (de) * 2006-01-24 2007-07-19 Daimlerchrysler Ag Verfahren zum Zusammenfügen mehrerer Bildaufnahmen zu einem Gesamtbild in der Vogelperspektive
CN101149794A (zh) * 2007-10-16 2008-03-26 王海燕 基于视频的车辆检测跟踪及系统
DE102008062915B4 (de) * 2008-12-23 2019-10-24 Continental Safety Engineering International Gmbh Erkennung von Personen, insbesondere von Fußgängern
JP2013532858A (ja) * 2010-07-17 2013-08-19 ヴァレオ・シャルター・ウント・ゼンゾーレン・ゲーエムベーハー 自動車の側面に隣接する側面域に存在する障害物について自動車の運転者に警告する方法、および運転者支援システムを備える自動車
DE102010034139A1 (de) * 2010-08-12 2012-02-16 Valeo Schalter Und Sensoren Gmbh Verfahren zur Unterstützung eines Parkvorgangs eines Kraftfahrzeugs, Fahrerassistenzsystem und Kraftfahrzeug
US10081308B2 (en) * 2011-07-08 2018-09-25 Bendix Commercial Vehicle Systems Llc Image-based vehicle detection and distance measuring method and apparatus
JP5906696B2 (ja) 2011-11-30 2016-04-20 アイシン精機株式会社 車両周辺撮影装置および車両周辺画像の処理方法
WO2013169601A2 (fr) * 2012-05-07 2013-11-14 Honda Motor Co., Ltd. Procédé pour générer des surfaces d'affichage virtuelles à partir d'une imagerie vidéo d'un paysage routier
US9460354B2 (en) 2012-11-09 2016-10-04 Analog Devices Global Object detection
EP3007149B1 (fr) * 2013-05-31 2021-01-20 Toyota Jidosha Kabushiki Kaisha Dispositif d'assistance à la conduite pour véhicules et ordinateur de bord
CN103325255B (zh) * 2013-06-29 2016-01-20 佘若凡 基于摄影测量技术进行区域交通状况检测的方法
JP6155963B2 (ja) * 2013-08-21 2017-07-05 株式会社デンソー 衝突緩和装置
JP6187817B2 (ja) 2013-10-09 2017-08-30 アイシン精機株式会社 顔検出装置、方法およびプログラム
US9158985B2 (en) * 2014-03-03 2015-10-13 Xerox Corporation Method and apparatus for processing image of scene of interest
JP6407626B2 (ja) * 2014-08-26 2018-10-17 日立オートモティブシステムズ株式会社 物体認識装置及び車両制御システム
US10055643B2 (en) * 2014-09-19 2018-08-21 Bendix Commercial Vehicle Systems Llc Advanced blending of stitched images for 3D object reproduction
US9990550B2 (en) * 2014-09-19 2018-06-05 Bendix Commercial Vehicle Systems Llc Wide baseline object detection stereo system

Also Published As

Publication number Publication date
JP6698163B2 (ja) 2020-05-27
CN108496176A (zh) 2018-09-04
US10635896B2 (en) 2020-04-28
CN108496176B (zh) 2021-12-31
US20190012537A1 (en) 2019-01-10
JP2019500695A (ja) 2019-01-10
DE102015121952A1 (de) 2017-06-22
WO2017103096A1 (fr) 2017-06-22

Similar Documents

Publication Publication Date Title
EP3391282A1 (fr) Procédé d'identification d'un objet dans une zone environnante d'un véhicule automobile, système d'assistance à la conduite ainsi que véhicule automobile
EP3329462B1 (fr) Procédé d'identification d'un objet dans une zone environnante d'un véhicule automobile, système d'aide à la conduite ainsi que véhicule automobile
DE102011088130B4 (de) Verfahren und Vorrichtung zur Erkennung einer Bremssituation
DE102018203807A1 (de) Verfahren und Vorrichtung zur Erkennung und Bewertung von Fahrbahnzuständen und witterungsbedingten Umwelteinflüssen
DE102014201159A1 (de) Verfahren und Vorrichtung zum Klassifizieren eines Verhaltens eines Fußgängers beim Überqueren einer Fahrbahn eines Fahrzeugs sowie Personenschutzsystem eines Fahrzeugs
WO2017008801A1 (fr) Détection de feu de freinage sur des véhicules circulant à l'avant pour adapter le déclenchement de mécanismes de sécurité actifs
EP1928687A1 (fr) Procede et systeme d'aide a la conduite pour la commande de demarrage d'un vehicule automobile basee sur un capteur
EP2710573B1 (fr) Procédé et dispositif permettant d'identifier un objet de collision éventuel
EP3046076B1 (fr) Procédé destiné à étalonner un dispositif de détection d'image
DE102017223160B4 (de) Verfahren zum Erkennen von zumindest einem an einem Kraftfahrzeug anliegenden Objekt sowie Steuervorrichtung und Kraftfahrzeug
DE102006060893A1 (de) Vorrichtung und Verfahren zum Bestimmen eines Freiraums vor einem Fahrzeug
DE102017118651A1 (de) Verfahren und System zur Kollisionsvermeidung eines Fahrzeugs
DE102015010746A1 (de) Verfahren zur Selbstlokalisation eines Fahrzeugs
EP2080684B1 (fr) Procédé et dispositif destinés à la détection d'un véhicule passant devant dans l'obscurité
DE102015223500B4 (de) Verfahren und Vorrichtung zur Prüfung der Funktionalität einer außenseitigen Lichteinrichtung eines Fahrzeugs
DE102006037600B4 (de) Verfahren zur auflösungsabhängigen Darstellung der Umgebung eines Kraftfahrzeugs
WO2015090691A1 (fr) Procédé de génération d'un modèle d'environnement d'un véhicule automobile, système d'assistance au conducteur et véhicule automobile
WO2009097883A1 (fr) Procédé pour faire fonctionner un dispositif d'enregistrement d'images, et dispositif d'enregistrement d'images
DE102013022050A1 (de) Verfahren zum Verfolgen eines Zielfahrzeugs, insbesondere eines Motorrads, mittels eines Kraftfahrzeugs, Kamerasystem und Kraftfahrzeug
EP3520020B1 (fr) Classification de signaux routiers dans une zone environnante d'un véhicule automobile
WO2013026599A1 (fr) Procédé et dispositif de détection d'objets perturbateurs dans l'environnement d'un véhicule
EP1962245B1 (fr) Procédé et dispositif de détermination de l'état de déplacement d'objets
DE102018202356A1 (de) Verfahren und Vorrichtung zum Prüfen einer Freigabe einer Anfahrt für ein Fahrzeug
DE102017216802A1 (de) Verfahren und vorrichtung zum erkennen von fahrspuren, fahrerassistenzsystem und fahrzeug
DE102020003924A1 (de) Verfahren zum kamerabasierten Erkennen wenigstens einer Fahrspur sowie ein das Verfahren nutzendes Fahrzeug

Legal Events

Date Code Title Description
STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: UNKNOWN

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

Free format text: STATUS: THE INTERNATIONAL PUBLICATION HAS BEEN MADE

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

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

Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE

17P Request for examination filed

Effective date: 20180611

AK Designated contracting states

Kind code of ref document: A1

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

AX Request for extension of the european patent

Extension state: BA ME

DAV Request for validation of the european patent (deleted)
DAX Request for extension of the european patent (deleted)
STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: EXAMINATION IS IN PROGRESS

17Q First examination report despatched

Effective date: 20210324

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

Free format text: STATUS: EXAMINATION IS IN PROGRESS

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: 20211005