Methods for detecting objects in a means of transport

Abstract

The invention relates to a method (100) for detecting objects in a means of transport (200) comprising the steps of receiving (S1) a data set which includes an image of an interior (202) of the means of transport (200), determining (S2) a category of an object (300) based on the received data set, and depending on the category of the object (300) outputting (S3) a signal to adjust a property of a safety device (204) of the means of transport (200).

Description

[0001] The present invention relates to a method for detecting objects in a means of transport and to a means of transport.

[0002] Current technology offers a wide range of safety systems for passenger compartments in vehicles. Airbags, belt pretensioners, and other safety features are the preferred options. However, these airbags and belt pretensioners are not adapted to the vehicle's environment.

[0003] The aforementioned objects known in the state of the art do not satisfactorily exploit the technical possibilities of modern means of transport.

[0004] It is an object of the present invention to satisfy the need identified above.

[0005] To solve the aforementioned problem, a method for detecting objects in a means of transport is proposed. The means of transport can be, for example, a car, a van, a truck, an aircraft, and / or a watercraft. In certain scenarios outside of public road traffic, situations may arise in which occupants in the interior of the means of transport wear a helmet or similar device. For example, when driving on a racetrack, the driver may use a head restraint system in combination with a helmet. Common safety systems in means of transport, such as airbag systems, do not take the use of a helmet and / or a frontal head restraint device into account, even though these can have a strong influence on the movement behavior of a body in an emergency situation.Therefore, the use of a helmet or a head restraint system in the interior of the means of transport should advantageously be used by means of image-based recognition to adapt the safety systems or safety devices in the means of transport accordingly.

[0006] This is preferably achieved according to the invention in that the method for detecting objects on a means of transport comprises the steps: - Receiving a data set comprising an image of an interior of the means of transport, - Determine a category of an object based on the received data set, and depending on the category of the object: - Outputting a signal to adjust a property of a safety device of the means of transport.

[0007] In other words, for example, an image sensor or similar can be used to detect that an object such as a helmet or similar is being worn by a person in the interior of the means of transport. Based on the detection, the safety devices in the means of transport can be adapted to the corresponding object or to the helmet or similar. In particular, image-based recognition, preferably with machine learning, can be used to determine the category of the object. Adapting the property of the safety device of the means of transport can, for example, include adapting the airbag system to the movement behavior used by the driver based on the helmet. The interior can also mean the interior of a convertible, roadster, or similar.

[0008] The subclaims show preferred developments of the invention.

[0009] Preferably, the safety device of the means of transport is an airbag and / or a belt tensioner.

[0010] An advantage of this embodiment is that the safety devices, which are in direct contact with the driver and have a strong physical impact, can be adjusted to accurately adapt to the changed movement behavior due to a helmet or similar.

[0011] Further preferably, the adaptation of the property of the safety device comprises an adaptation of a number of pyro units of the airbag to be ignited and / or an adaptation of a triggering time of the airbag.

[0012] An advantage of this design is that the type of deployment of the safety device can be adapted to the helmet or similar, thus maximizing the cushioning effect of the safety device. For example, the number of airbag units to be deployed can be adjusted based on the helmet, as can the helmet's deployment time. Furthermore, the deployment time can be moved closer and / or further from the detection of the safety event, since the increased weight on the head caused by the helmet would cause the head to hit the airbag more quickly than without the helmet.

[0013] Preferably, the category of the object includes at least: - a helmet, and / or - head protection, and / or - a head restraint arrangement.

[0014] An advantage of this embodiment is that if the category of the object is restricted to one of these listed systems, the detection probability can be significantly improved, since these can be easily identified in the interior of the means of transport using common image recognition methods.

[0015] Preferably, determining the category of the object includes at least the following steps: - Determining a contour of the object using the data set and - Determine the category of the object based on the determined contour.

[0016] An advantage of this embodiment is that contour detection of the object allows for a simple determination of the type of object. For example, the contour of a helmet may differ significantly from the contour of a head restraint system, so the decision parameters for selecting the object category can be significantly simplified.

[0017] Preferably, determining the category of the object includes at least the following steps: - Determining at least one part of a body, in particular a skeleton of the body, - Compare the determined part of the body with a body reference - Determine the category of the object by comparing the body to the body reference.

[0018] An advantage of this embodiment may be that if no body reference is detectable, such as the face, the type of object can be inferred, since a helmet may obscure the face. Further preferably, the dimensions of the body part can be detected in order to distinguish a helmet from other headgear, such as a sun hat.

[0019] Preferably, determining the category of the object comprises at least the steps: - Determining a position of the object in the interior of the means of transport, and - Determine the category of the object based on the determined position of the object.

[0020] An advantage of this embodiment is that the object's position can be used to determine its category. For example, a helmet can be positioned significantly closer to the headliner of the vehicle's interior than a head restraint system, which can be positioned significantly further away from the headliner of the vehicle's interior. Thus, the object's position can be used to determine its type.

[0021] Preferably, determining the category of the object further comprises the steps: - Determining a reference object in the interior of the vehicle using the data set, and - Determine the category of the object based on the determined position of the object and the determined reference object.

[0022] An advantage of this embodiment is that by identifying reference objects, such as a steering wheel, a B-pillar, or a seat, conclusions can be drawn about the type of object. For example, a steering wheel can be identified in an image of the interior of a vehicle, so the person is presumably the driver. In combination with the position of the object and the identified reference object, a decision can be made about the type of category of the object.

[0023] More preferably, the method comprises the step: - Assigning the object to a seating position in the interior and / or to the safety device of the means of transport based on the determined position and / or the determined reference object.

[0024] An advantage of this embodiment is that the object, such as a helmet, can be assigned to a seating position in the interior in order to adapt the safety devices that may interact with the corresponding seating position. For example, airbags and belt pretensioners can be adapted to the corresponding seating position based on the object's assignment to the seating position.

[0025] Preferably, the method comprises the step: - capturing an image with an image sensor to form the data set, wherein the captured image is a two-dimensional image and / or a three-dimensional image.

[0026] An advantage of this embodiment is that a variety of available cameras in the vehicle can be used to generate the data set. For example, the image sensor can be a two-dimensional RGB infrared image sensor or a 3D camera. Thus, the image of the interior can be captured, and in particular, both a two-dimensional image and a three-dimensional image can be captured.

[0027] Further preferably, the image sensor is arranged in a rearview mirror of the means of transport, part of a fatigue monitoring system and / or arranged in a portable user terminal.

[0028] An advantage of this embodiment is that, if the image sensor is located in the vehicle's rearview mirror, the entire interior of the vehicle can be monitored. Furthermore, the total number of required components can be reduced if image sensors from existing systems or user devices such as a smartphone can be used.

[0029] Preferably, the method comprises the step: - Triggering the capture of the image when a driving mode of the vehicle is changed.

[0030] An advantage of this embodiment is that when changing the driving mode, such as changing to a sport mode and / or racing mode, the image sensor captures an image of the interior in order to detect whether one of the persons in the interior of the means of transport is wearing an object, such as a helmet.

[0031] Preferably, determining the category of the object further comprises the steps: - Receiving a driver profile from a portable user device and / or selecting a driver profile by means of an on-board computer of the means of transport.

[0032] An advantage of this embodiment is that the category of the object can be specifically selected using user input. This can also reduce false positive detections. For example, the category of the object can be stored in a driver profile or selected from a multitude of existing objects using the vehicle's on-board computer.

[0033] Preferably, the method further comprises the steps: - comparing the determined category of the object with a variety of object references, and - Checking the determined category of the object based on the result of the comparison.

[0034] An advantage of this embodiment is that the determined category of the object can be compared against a plurality of object references, such as a database, in order to significantly reduce the number of false detections.

[0035] Preferably, the method comprises the steps: - Updating the plurality of object references and / or creating a new entry for the object reference based on the determined category of the object.

[0036] An advantage of this embodiment is that updating the plurality of object references allows a database or the like to be updated, so that, in particular, the means of transport or the safety device can be adapted to new helmets or head restraint systems. Furthermore, an entry can also be added to the plurality of object references if the determined category of the object is not yet included in the plurality of object references.

[0037] A further aspect of the invention relates to a means of locomotion which has a control device which is designed to carry out the method as described above and below.

[0038] The means of transport can implement all features of the described method as well as all combinations of features and advantages.

[0039] Further details, features, and advantages of the invention will become apparent from the following description and the figures. They show: Fig. 1 a flowchart illustrating the steps of an embodiment of a method according to the invention for detecting objects in a means of transport, Fig. 2 a flowchart illustrating the steps of an embodiment of a method according to the invention for detecting objects in a means of transport, Fig. 3 a schematic representation of an embodiment of a means of transport according to the invention, Fig. 4 a schematic representation of an embodiment of a means of transport according to the invention.

[0040] Preferably, all identical components, units and / or elements in all figures are provided with the same reference numerals.

[0041] Fig. 1 shows a method 100 for detecting objects in a means of transport 200, which comprises the steps: - Receiving S1 a data set comprising an image of an interior 202 of the means of transport 200, - Determining S2 a category of an object 300 based on the received data set, and depending on the category of the object 300: - Outputting S3 of a signal for adapting a property of a safety device 204 of the means of transport 200.

[0042] Fig. 2 shows a further embodiment of the method 100. The method 100 comprises the same steps S1 to S3 as already described with regard to the Fig. 1. Furthermore, the method 100 comprises the steps of determining a contour S4 of the object 300 using the data set and determining the category S5 of the object 300 based on the determined contour. Furthermore, the method 100 may comprise the steps of determining a position S6 of the object 300 in the interior 202 of the means of transport 200 and determining the category S7 of the object 300 based on the determined position of the object 300.

[0043] In addition, the method 100 can comprise the steps of determining S8 a reference object 212 in the interior 202 of the means of transport 200 using the data set and determining the category S9 of the object 300 based on the determined position of the object 300 based on the determined reference object 212. Furthermore, the method 100 can comprise the step of assigning S10 the object 300 to a seating position in the interior 202 and / or to the safety device 204 of the means of transport 200 based on the determined position and / or the determined reference object 202.

[0044] Preferably, the method 100 comprises the step of capturing S11 an image with an image sensor 214 to form the data set, wherein the captured image is a two-dimensional image and / or a three-dimensional image. Further preferably, the method 100 comprises the step of triggering S12 the capture of the image when a driving mode of the means of transport 200 is changed. Furthermore, the method 100 may comprise the step of receiving a driver profile S13 from a portable user device 500 and / or selecting a driver profile using an on-board computer 220 of the means of transport 200.

[0045] Further preferably, the method 100 comprises the steps of comparing S14 the determined category of the object 300 with a plurality of object references and checking S15 the determined category of the object 300 based on a result of the comparison. Further preferably, the method 100 comprises the step of updating S16 the plurality of object references and / or creating a new entry for the object reference based on the determined category of the object 300.

[0046] Fig. 3 shows an embodiment of the means of transport 200. The means of transport 200 has a control device 210, which can execute the method 100 as described above and below. The control device 210 can receive S1 a data set comprising an image of an interior 202 of the means of transport 200, determine S2 a category of an object 300 based on the received data set, and, depending on the category of the object 300, output S3 a signal for adapting a property of a safety device 204 of the means of transport 200.

[0047] Fig.4 shows a means of transportation 200 according to a further embodiment. The means of transportation 200 comprises an interior 202 and a safety device 204, which may be an airbag or a belt tensioner or the like. Furthermore, the means of transportation 200 may comprise a reference object 212, such as a steering wheel or a B-pillar. Furthermore, the vehicle 200 may comprise an image sensor 214, which is preferably configured to capture a two-dimensional or three-dimensional image of the interior 202. Furthermore, the means of transportation 200 may have a rearview mirror 216 in which the image sensor 214 is arranged. Furthermore, the means of transportation 200 may have a fatigue monitoring system 218 in which the image sensor 214 can be arranged. Furthermore, the image sensor 214 may also be part of a portable terminal 400. In addition, the means of transport 200 can have an on-board computer 220 with the aid of which a driver profile can be selected.In addition, the means of transport 200 can be in data communication with a user device 500 in order to transmit a driver profile to the means of transport 200. The user device 500 can be, for example, a smart key, a smartphone, or the like. List of reference symbols: 100 procedures 200 means of transport 202 Interior 204 Safety device 210 Control device 212 Reference object 214 image sensor 216 rearview mirrors 218 Fatigue Monitoring System 220 on-board computers 300 objects 400 user terminals 500 user devices S1-S16 process steps

Claims

[1] Method (100) for detecting objects in a means of transport (200) comprising the steps: - receiving (S1) a data set comprising an image of an interior (202) of the means of transport (200), - Determining (S2) a category of an object (300) based on the received data set, and depending on the category of the object (300): - Outputting (S3) a signal for adapting a property of a safety device (204) of the means of transport (200). [2] Method (100) according to claim 1, wherein the safety device (204) of the means of transport (200) is an airbag and / or a belt tensioner. [3] Method (100) according to claim 2, wherein the adaptation of the property of the safety device (204) comprises an adaptation of a number of pyro units of the airbag to be ignited and / or an adaptation of a triggering time of the airbag. [4] Method (100) according to one of the preceding claims, wherein the category of the object (300) comprises at least: - a helmet, and / or - head protection, and / or - a head restraint arrangement. [5] Method (100) according to one of the preceding claims, wherein determining the category of the object (300) comprises at least the steps: - Determining a contour (S4) of the object (300) using the data set and - Determining the category (S5) of the object (300) based on the determined contour. [6] Method (100) according to one of the preceding claims, wherein determining the category of the object (300) comprises at least the steps: - Determining a position (S6) of the object (300) in the interior (202) of the means of transport (200) and - Determining the category (S7) of the object (300) based on the determined position of the object (300). [7] The method (100) of claim 6, wherein determining the category of the object (300) further comprises the steps of: - Determining (S8) a reference object (212) in the interior (202) of the means of transport (200) by means of the data set and - Determining the category (S9) of the object (300) based on the determined position of the object (300) and on the basis of the determined reference object (212). [8] Method (100) according to one of claims 6-7 comprising the step: - Assigning (S10) the object (300) to a seating position in the interior (202) and / or to the safety device (204) of the means of transport (200) based on the determined position and / or the determined reference object (202). [9] Method (100) according to one of the preceding claims, further comprising the step: - capturing (S11) an image with an image sensor (214) to form the data set, wherein the captured image is a two-dimensional image and / or a three-dimensional image. [10] Method (100) according to claim 9, wherein the image sensor (214) - is arranged in a rearview mirror (216) of the means of transport (200) and / or - is part of a fatigue monitoring system (218) and / or - is arranged in a portable user terminal (400). [11] Method (100) according to one of claims 9 to 10 comprising the step: - triggering (S12) the capture of the image when a driving mode of the means of transport (200) is changed. [12] Method (100) according to one of the preceding claims, wherein determining the category of the object (300) further comprises the steps of: - Receiving a driver profile (S13) from a portable user device (500) and / or selecting a driver profile by means of an on-board computer (220) of the means of transport (200). [13] Method (100) according to one of the preceding claims, further comprising the steps: - comparing (S14) the determined category of the object (300) with a plurality of object references and - Checking (S15) the determined category of the object (300) based on a result of the comparison. [14] Method (100) according to claim 13 comprising the steps: - Updating (S16) the plurality of object references and / or forming a new entry for the object reference based on the determined category of the object (300). [15] Means of transport (200) comprising a control device (210) which is designed to carry out the method (100) according to one of the preceding claims.

Images