WO2018068919A1 - Procédé de détection d'objets dans une région environnementale d'un véhicule automobile en tenant compte de données de capteur dans la plage de longueurs d'ondes infrarouges, appareil de détection d'objet, système d'aide à la conduite ainsi que véhicule automobile - Google Patents
Procédé de détection d'objets dans une région environnementale d'un véhicule automobile en tenant compte de données de capteur dans la plage de longueurs d'ondes infrarouges, appareil de détection d'objet, système d'aide à la conduite ainsi que véhicule automobile Download PDFInfo
- Publication number
- WO2018068919A1 WO2018068919A1 PCT/EP2017/069928 EP2017069928W WO2018068919A1 WO 2018068919 A1 WO2018068919 A1 WO 2018068919A1 EP 2017069928 W EP2017069928 W EP 2017069928W WO 2018068919 A1 WO2018068919 A1 WO 2018068919A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- sensor data
- motor vehicle
- wavelength range
- basis
- classified
- Prior art date
Links
Classifications
-
- 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
-
- 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
-
- 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
Definitions
- the present invention relates to a method for detecting objects in an environmental region of a motor vehicle, wherein by means of a detecting device of the motor vehicle first sensor data describing the environmental region in the visible wavelength range, and second sensor data describing the environmental region in the infrared wavelength range, are provided and the object is detected on the basis of the first sensor data and/or the second sensor data.
- the present invention relates to an object detection apparatus.
- the present invention relates to a driver assistance system.
- the present invention relates to a motor vehicle.
- the interest is presently directed to the detection of objects in an environmental region of a motor vehicle.
- objects in the environment are detected with sensors, which are arranged on the motor vehicle.
- the information about the objects can then be used by the driver assistance systems of the motor vehicle.
- a warning to the driver of the motor vehicle can be outputted if a collision with an object or obstacle is imminent.
- the objects can for example be detected by the aid of cameras that provide images of the environmental region. Using appropriate object detection methods, the objects can then be detected in the images.
- the objects are classified or assigned to a group.
- the objects can be detected as pedestrians, vehicles or road markings. Further, it may be provided that the objects are classified as static or dynamic objects.
- WO 2015/161208 A1 discloses a vehicle with a vision system comprising a stereo camera for light in the visible wavelength range.
- the vision system comprises an infrared camera.
- the vision system comprises a camera for the near infrared wavelength range.
- first sensor data describing in particular the environmental region in the visible wavelength range, and second sensor data describing in particular the environmental region in the infrared wavelength range are provided.
- the object is preferably detected on the basis of the first sensor data and/or the second sensor data.
- the object is preferably classified as a static object or as a dynamic object on the basis of the first sensor data and/or the second sensor data and if the object is classified as a static or as a dynamic object, it is preferably verified on the basis of the second sensor data, whether the object is a plant.
- a method according to the invention serves for detection in an environmental region of a motor vehicle.
- first sensor data describing the environmental region in the visible wavelength range
- second sensor data describing the environmental region in the infrared wavelength range
- the object is detected on the basis of the first sensor data and/or the second sensor data.
- the object is detected on the basis of the first sensor data and/or the second sensor data.
- the object is preferably classified as a static object or as a dynamic object on the basis of the first sensor data and/or the second sensor data. If the object is classified as a static or as a dynamic object, it is verified on the basis of the second sensor data, whether the object is a plant.
- the first sensor data are provided with the detection device.
- the first sensor data describe the environmental region in the visible wavelength range.
- the first sensor data describes in particular light in the visible wavelength range, which is reflected from the object and/or emitted by the object.
- the second sensor data are provided by means of the detection device, which describe the surrounding region and particularly the object in the infrared wavelength range.
- the second sensor data describe in particular the radiation in the infrared wavelength range, which is reflected by the object and/or is emitted from it.
- the object On the basis of the first sensor data and/or the second sensor data, the object can then be detected.
- a computing device can be used by means of which the first sensor data and/or the second sensor data can be evaluated. If the first sensor data and the second sensor data are provided each as an image, an appropriate object recognition algorithm can for example be carried out using the computing device to recognize the object. In this case, for example, methods of segmentation and/or classification can be used to detect the object in the images. It can also be provided that the detected object is compared with known forms. In particular, it is provided that the object is classified based on the first sensor data and/or the second sensor data.
- the classification it can in particular be differentiated between a static or non-moving object on the one hand and a dynamic or a moving object on the other hand.
- dynamic objects those objects are to be understood, which move relative to the motor vehicle.
- the dynamic objects can move in particular on a road or a floor.
- static objects those objects are to be understood, which do not move relative to the motor vehicle.
- the static objects can be located in particular on the road or near the road.
- the object is classified as a static object or as a dynamic object, it is verified on the basis of the second sensor data, whether the object is a plant.
- This is based on the finding that plants can move as a result of environmental influences. For example, the plants and in particular their leaves can move in wind or precipitation. This can lead to the situation that these plants or parts thereof are classified as a moving object.
- textures in static vegetation can mimic the patterns of objects or pedestrians used in classification methodologies. Thus, these textures of the static plants can be classified as static objects. If the object is classified as a static or as a moving object, the second sensor data describing the object can be used.
- the infrared radiation, and particularly the near infrared radiation which is reflected and/or emitted from the object is examined.
- plants typically reflect or scatter radiation in the near infrared wavelength region during photosynthesis.
- the plants reflect or scatter the radiation in the near infrared wavelength range in order to prevent overheating and damage to the cells.
- the second sensor data it is possible to distinguish the objects classified as static or as moving objects from plants.
- the false positive rate can be significantly reduced.
- the object can be classified as a static pedestrian. As already mentioned, this may be due to the texture of the plants.
- the verification based on the second sensor data it can be determined in a reliable manner, whether it is a pedestrian or a plant, for example a tree, a bush or the like. This information can then be used for example by a driver assistance system of the motor vehicle to avoid a collision between the motor vehicle and the pedestrian.
- the object that is classified as static or as dynamic is detected as a plant on the basis of the radiation emitted from the object in the near infrared wavelength range, in particular in a wavelength range of between 0.7 ⁇ and 1 .1 ⁇ .
- the knowledge is taken into account that plants especially during
- photosynthesis significantly reflect or scatter radiation in the near infrared region to prevent overheating and thus damage to the cells. If it is now recognized that the object emits radiation in this wavelength range, it can be assumed with a high probability that it is a plant.
- the first sensor data can be used to assess the object that is classified as a static object or as a dynamic object closer. For example, it can be verified, which colour the object has. If the object has a green colour, it can be assumed with a high probability that it is a plant.
- the colour of the object or a part of the object can be compared with predetermined colours, which describe different plants. This takes into account that leaves of trees and shrubs can have different colours. Thus, the false positive rate can be further reduced.
- the object that is classified as static or as dynamic is detected as a plant based on a radiation emitted from the object in a wavelength range of between 0.4 ⁇ and 0.7 ⁇ .
- a radiation emitted from the object in a wavelength range of between 0.4 ⁇ and 0.7 ⁇ .
- the chlorophyll in the plants and in particular in the leaves of plants reflects or scatters light in the visible wavelength range of between 0.4 ⁇ and 0.7 ⁇ .
- a normalized differenced vegetation index of the as object that is classified as dynamic is determined on the basis of the first sensor data and the second sensor data and it is verified whether the object is a plant on the basis of the normalized differenced vegetation index.
- the normalized differenced vegetation index is usually calculated on the basis of satellite data. This NDVI can now be determined based on the first sensor data and the second sensor data. In particular, the NDVI can be determined from reflection values in the near infrared range, which are included in the second sensor data, and reflection values in the red visible range, which are included in the first sensor data. Based on the NDVI plants in the environmental region can be determined in a simple and reliable manner.
- an image is determined on the basis of the first sensor data and the second sensor data and it is verified whether the object is a plant on the basis of the image.
- a composite image that contains information in the visible wavelength region and the infrared wavelength range can be provided based on the first sensor data and the second sensor data.
- this image can describe the NDVI.
- a so-called NRG image (NRG - Near- infrared/Red/Green) is provided, describing the environmental region in the near infrared wavelength range, in the red wavelength range and the green wavelength range.
- NRG image NRG - Near- infrared/Red/Green
- the proportion of plants or vegetation can be determined in the environmental region.
- the object detection apparatus is adapted for performing a method according to the invention and
- the computing device is in particular connected to the detecting device for data transmission and can receive the first sensor data and the second sensor data from the detection device.
- the computing device may also identify and classify the object in the surrounding area based on the first sensor data and/or the second sensor data. Further, the computing device can verify based on the second sensor data, if the object that is classified as static or as dynamic is a plant.
- the detection device comprises a camera for providing the first sensor data and a sensor for providing the second sensor data.
- the camera may be a camera, which is typically used for object detection in the automotive field. With the camera, for example, images of the environmental region can be provided as the first sensor data. With the sensor then additionally the second sensor data in the infrared wavelength range, and in particular near-infrared wavelength region, can be provided.
- the sensor may be, for example, a corresponding infrared camera.
- the detection device can comprise a camera, which can provide both the first sensor data and the second sensor data.
- a camera can be used which does not have an infrared filter.
- this camera can provide, for example, images that provide in addition to the information in the visible wavelength range also information in the infrared wavelength range.
- a driver assistance system comprises an object detection apparatus according to the invention.
- objects and in particular pedestrians can be detected reliably.
- static objects such as walls, curbs, guardrails or the like can be detected with the object detection apparatus based on the first sensor data and/or the second sensor data.
- other motor vehicles, motorcycles or cyclists can be detected. This information can be used by the driver assistance system in order to assist the driver when driving the motor vehicle.
- a motor vehicle according to the invention includes a camera system according to the invention.
- the motor vehicle is in particular formed as a passenger car.
- Fig. 1 a motor vehicle according to an embodiment of the present invention, comprising a driver assistance system with an object detection apparatus;
- Fig. 2 an image which is provided by the object detection apparatus.
- Fig. 1 shows a motor vehicle 1 according to an embodiment of the present invention in a plan view.
- the motor vehicle 1 is configured as a passenger car.
- the motor vehicle 1 comprises a driver assistance system 2, which is used to assist a driver when driving the motor vehicle 1 .
- the driver assistance system 2 includes an object detection apparatus 3. With the aid of the object detection apparatus 3 objects 9 can be detected in an environmental region 8 of the motor vehicle 1 . In the present case, as objects 9 a pedestrian 10 and a plant 1 1 or a tree are located in the environmental region 8 of the vehicle 1 in the direction of travel in front of the motor vehicle 1 .
- the driver assistance system 2 can intervene in a steering system, a drive motor and/or a braking system of the motor vehicle 1 in order to prevent a collision between the motor vehicle 1 and the object 9.
- the object recognition device 3 includes a camera 5, by means of which the first sensor data can be provided.
- the first sensor data describe the environmental region 8 in the visible wavelength range.
- the first sensor data for example, image data or images can be provided by the camera 5.
- the object detection apparatus 3 comprises a sensor 6 by means of which the second sensor data can be provided.
- the second sensor data describe the environmental region 8 in the infrared wavelength range.
- the sensor 6 is embodied as an infrared camera.
- the object detection apparatus 3 includes a computing device 7.
- the computing device 7 is connected with the camera 5 and the sensor 6 for data transmission. By means of the calculating device 7, the first sensor data from the camera 5 and the second sensor data from the sensor 6 can be received.
- the objects 9 can be detected in the
- the first sensor data and the second sensor data can be provided, for example, as an image. It can also be provided that a composite image 12 is determined by means of the computing device 7 based on the first sensor data and the second sensor data.
- the objects 9 can be detected by means of a corresponding object detection algorithm. For example, a segmentation method can be used. By means of the computing device 7, the detected objects 9 are classified additionally. In particular, it is provided that the objects 9 are classified as static objects or as dynamic objects.
- an image 12 showing the environmental region 8 and the objects 9 is provided with the calculating device 7 based on the first sensor data and the second sensor data.
- Such an image 12 is exemplarily shown in Fig. 2.
- the image 12 shows the pedestrian 10 and a part of the plant 1 1 or the tree. It can be the case that both the pedestrian 10 and the plant 1 1 are classified as static or moving object, and in particular as a pedestrian. This may be due to the plant 1 1 and in particular the leaves of the plant 1 1 moving as a result of environmental conditions such as wind.
- the object 9 If the object 9 is classified as static or as dynamic, it can be verified on the basis of the second sensor data whether the object 9 reflects or scatters radiation in the near infrared wavelength region and in particular in a wavelength range between 0.7 ⁇ and 1 .1 ⁇ . This takes into account that plants 1 1 scatter radiation in this wavelength range in order to prevent overheating and thereby caused damage to the cells. Alternatively or additionally, it can be verified on the basis of the first sensor data, whether the object scatters light in a wavelength range of between 0.4 ⁇ and 0.7 ⁇ . This takes into account that the chlorophyll in the leaves of plants strongly absorbs light in this wavelength range.
- the image 12 a composite image is provided that shows both components in the visible wavelength range and in the near infrared wavelength range.
- the image 12 can show the normalized differenced vegetation index (NDVI).
- NDVI normalized differenced vegetation index
- the pedestrian 10 and the plant 1 1 have different values of the NDVI.
- the plant can have an NDVI of 1 . This is schematically illustrated by the hatching 13.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Traffic Control Systems (AREA)
Abstract
L'invention concerne un procédé de détection d'objets (9) dans une zone environnementale (8) d'un véhicule automobile. Au moyen d'un dispositif de détection (4) du véhicule automobile (1), des premières données de capteur décrivant la région environnementale (8) dans la plage de longueurs d'onde visibles, et des secondes données de capteur décrivant la région environnementale (8) dans la plage de longueurs d'onde infrarouge, sont fournies et l'objet (9) est détecté sur la base des premières données de capteur et/ou des deuxièmes données de capteur, l'objet (9) étant classé comme étant un objet statique ou un objet dynamique sur la base des premières données de capteur et/ou des deuxièmes données de capteur et si l'objet (9) est classé comme un objet statique ou dynamique, il est vérifié sur la base des secondes données de capteur, si l'objet est une plante (11).
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP17757697.2A EP3526725A1 (fr) | 2016-10-14 | 2017-08-07 | Procédé de détection d'objets dans une région environnementale d'un véhicule automobile en tenant compte de données de capteur dans la plage de longueurs d'ondes infrarouges, appareil de détection d'objet, système d'aide à la conduite ainsi que véhicule automobile |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
DE102016119592.8 | 2016-10-14 | ||
DE102016119592.8A DE102016119592A1 (de) | 2016-10-14 | 2016-10-14 | Verfahren zum Erkennen von Objekten in einem Umgebungsbereich eines Kraftfahrzeugs unter Berücksichtigung von Sensordaten im infraroten Wellenlängenbereich, Objekterkennungsvorrichtung, Fahrerassistenzsystem sowie Kraftfahrzeug |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2018068919A1 true WO2018068919A1 (fr) | 2018-04-19 |
Family
ID=59702673
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/EP2017/069928 WO2018068919A1 (fr) | 2016-10-14 | 2017-08-07 | Procédé de détection d'objets dans une région environnementale d'un véhicule automobile en tenant compte de données de capteur dans la plage de longueurs d'ondes infrarouges, appareil de détection d'objet, système d'aide à la conduite ainsi que véhicule automobile |
Country Status (3)
Country | Link |
---|---|
EP (1) | EP3526725A1 (fr) |
DE (1) | DE102016119592A1 (fr) |
WO (1) | WO2018068919A1 (fr) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE102020105821A1 (de) | 2020-03-04 | 2021-09-09 | Connaught Electronics Ltd. | Verfahren und System zum Führen eines Fahrzeugs |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE102022127833A1 (de) | 2022-10-21 | 2024-05-02 | Bayerische Motoren Werke Aktiengesellschaft | Fahrassistenzsystem und Fahrassistenzverfahren für ein Fahrzeug |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100253543A1 (en) * | 2009-04-02 | 2010-10-07 | Gm Global Technology Operations, Inc. | Rear parking assist on full rear-window head-up display |
WO2010144877A1 (fr) * | 2009-06-11 | 2010-12-16 | Petroalgae, Llc | Indices de végétation pour mesurer la densité et la croissance de microcultures multicouches |
WO2015161208A1 (fr) | 2014-04-18 | 2015-10-22 | Cnh Industrial America Llc | Vision stéréoscopique pour détecter un environnement d'exploitation de véhicules |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CA2454891C (fr) * | 2001-07-24 | 2009-07-21 | The Board Of Regents For Oklahoma State University | Procede d'application d'un element nutritif en haute-saison fonde sur le potentiel de rendement prevu |
DE102006059033A1 (de) * | 2006-12-14 | 2008-06-19 | Volkswagen Ag | Verfahren und System zum Erkennen eines Verkehrsteilnehmers und zum Erzeugen einer Warnung |
DE102014223741A1 (de) * | 2014-11-20 | 2016-05-25 | Conti Temic Microelectronic Gmbh | Erfassen von Terahertz-Strahlung zum Unterstützen eines Fahrers eines Fahrzeugs |
DE102014224857A1 (de) * | 2014-12-04 | 2016-06-09 | Conti Temic Microelectronic Gmbh | Sensorsystem und Verfahren zur Klassifikation von Fahrbahnoberflächen |
-
2016
- 2016-10-14 DE DE102016119592.8A patent/DE102016119592A1/de active Pending
-
2017
- 2017-08-07 EP EP17757697.2A patent/EP3526725A1/fr not_active Withdrawn
- 2017-08-07 WO PCT/EP2017/069928 patent/WO2018068919A1/fr unknown
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100253543A1 (en) * | 2009-04-02 | 2010-10-07 | Gm Global Technology Operations, Inc. | Rear parking assist on full rear-window head-up display |
WO2010144877A1 (fr) * | 2009-06-11 | 2010-12-16 | Petroalgae, Llc | Indices de végétation pour mesurer la densité et la croissance de microcultures multicouches |
WO2015161208A1 (fr) | 2014-04-18 | 2015-10-22 | Cnh Industrial America Llc | Vision stéréoscopique pour détecter un environnement d'exploitation de véhicules |
Non-Patent Citations (1)
Title |
---|
ESS A ET AL: "Moving obstacle detection in highly dynamic scenes", 2009 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION : (ICRA) ; KOBE, JAPAN, 12 - 17 MAY 2009, IEEE, PISCATAWAY, NJ, USA, 12 May 2009 (2009-05-12), pages 56 - 63, XP031510118, ISBN: 978-1-4244-2788-8 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE102020105821A1 (de) | 2020-03-04 | 2021-09-09 | Connaught Electronics Ltd. | Verfahren und System zum Führen eines Fahrzeugs |
Also Published As
Publication number | Publication date |
---|---|
DE102016119592A1 (de) | 2018-05-03 |
EP3526725A1 (fr) | 2019-08-21 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US9499171B2 (en) | Driving support apparatus for vehicle | |
US8301344B2 (en) | Device for classifying at least one object in the surrounding field of a vehicle | |
CN106537180B (zh) | 用于用针对行人的主动制动的摄影机输入缓解雷达传感器限制的方法 | |
US20170297488A1 (en) | Surround view camera system for object detection and tracking | |
CN110065494B (zh) | 一种基于车轮检测的车辆防碰撞方法 | |
US9359009B2 (en) | Object detection during vehicle parking | |
CN108263279A (zh) | 基于传感器整合的行人检测和行人碰撞避免装置及方法 | |
CN107093180B (zh) | 使用轮胎向后飞溅的基于视觉的潮湿路面状况检测 | |
CN107273785B (zh) | 多尺度融合的道路表面状况检测 | |
CN110033621B (zh) | 一种危险车辆检测方法、装置及系统 | |
CN106845332B (zh) | 利用轮胎旁溅的基于视觉的潮湿路面状况检测 | |
US20190065878A1 (en) | Fusion of radar and vision sensor systems | |
KR20170127036A (ko) | 차도 위의 반사체를 인식하고 평가하기 위한 방법 및 장치 | |
US20060038885A1 (en) | Method for detecting the environment ahead of a road vehicle by means of an environment detection system | |
EP3282392A1 (fr) | Système et procédé de vision pour véhicule à moteur | |
CN109677402A (zh) | 自动驾驶工具的安全防护系统及方法 | |
JP4116643B2 (ja) | 車両周囲の少なくとも1つの物体を分類する装置 | |
CN111505617B (zh) | 车辆定位方法、装置、设备及存储介质 | |
KR102017958B1 (ko) | 철도차량용 증강현실 헤드업 디스플레이 시스템 | |
EP1854666B1 (fr) | Systeme destine a la detection d'objets situes dans une zone exterieure frontale d'un vehicule, applicable a des vehicules industriels | |
WO2018068919A1 (fr) | Procédé de détection d'objets dans une région environnementale d'un véhicule automobile en tenant compte de données de capteur dans la plage de longueurs d'ondes infrarouges, appareil de détection d'objet, système d'aide à la conduite ainsi que véhicule automobile | |
CN104228831A (zh) | 用于防止机动车的轮胎损伤的方法 | |
CN111066024B (zh) | 用于识别车道的方法和装置、驾驶员辅助系统以及车辆 | |
LU100761B1 (en) | Method for obstacle identification | |
WO2016079117A1 (fr) | Détection de gradient basee sur une image transformée en perspective |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 17757697 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
ENP | Entry into the national phase |
Ref document number: 2017757697 Country of ref document: EP Effective date: 20190514 |