Camera-based inspection assistance procedure

Abstract

Method for measuring wading depth in a vehicle (110) having at least one camera (116, 118, 120, 122) with a field of view (θ) covering at least part of a vehicle body (126) in a downward direction, comprising the steps of: calibrating a position of the at least one camera (116, 118, 120, 122) with respect to the body (126) of the vehicle (110) before using the vehicle (110); Learning a shape of the vehicle body (126) from the at least one camera (116, 118, 120, 122) in a downward direction; Receiving a camera image that covers the part of a vehicle body (126) from the at least one camera (116, 118, 120, 122) in a downward direction; Identifying a part of the vehicle body (126) from the at least one camera (116, 118, 120, 122) in a downward direction that is not covered with liquid (130);Comparing the part of the vehicle body (126) as seen by the at least one camera (116, 118, 120, 122) in a downward direction, which is not covered with liquid (130), with the shape of the vehicle body (126) as seen by the at least one camera (116, 118, 120, 122) in a downward direction; and determining a wading depth based on the comparison, characterized in that the step for learning a shape of the vehicle body (126) from the at least one camera (116, 118, 120, 122) in a downward direction comprises performing a self-learning process for learning the shape of the vehicle body (126) from the at least one camera (116, 118, 120, 122) in a downward direction.

Description

The present invention relates to a method for detecting weather depth in a vehicle having at least one camera.The present invention also relates to a driving assistance system for a vehicle, the driving assistance system being suitable for carrying out the above method.Furthermore, the present invention relates to a vehicle having the above-described driving assistance system.Existing solutions for water depth detection in a vehicle are based on ultrasonic sensors, which are typically mounted on side mirrors. The ultrasonic sensors are typically directed downwardly toward the ground to detect a distance from the ground to a water surface.FIG. 1 shows a conventional vehicle 10. The ultrasonic sensor 14 has a field of view α. As shown in FIG. 1, the vehicle is partially covered with water 16. Therefore, ultrasonic waves emitted from the ultrasonic sensor 14 are reflected by a surface 18 of the water 16. The depth of wake is determined based on reflections of ultrasonic pulses emitted from the ultrasonic sensor 14. The reflections are received by the ultrasonic sensor 14, and based on a travel time of the ultrasonic waves, a distance to the surface 18 of the water 16 is determined based on a known position of the ultrasonic sensor 14 on the vehicle 10. Additional contact sensors 20 for detecting the water 16 based on contact are provided on a lower side of the vehicle 10. The contact sensors 20 are used, for example, for validating the ultrasonic sensors 14.However, such ultrasonic sensors have a limited field of view for detecting the water level, i.e., the surface of the water. Thus, the ultrasonic sensors only perform a local measurement of the water level. Additionally, the ultrasonic sensors typically require a planar surface, which limits the usefulness of the ultrasonic sensors. Moreover, water is a liquid medium, so that its surface can move greatly in an unpredictable manner. Water is a dynamically changing medium because of its fluidic nature. This makes it difficult to determine the depth of wake using ultrasonic sensors.In this connection, a vehicle according to WO 2015 / 071 170 A1 has a system for assisting control of a vehicle by a driver when the vehicle is crossing a water station. The system includes a measurement device for determining a measured depth of water that is passing through the vehicle. The measuring device is positioned and arranged relative to the vehicle such that the measured depth indicates the water depth in a first measuring range relative to the actual vehicle. A processor is connected to the measurement device and is configured to calculate an estimated water depth as a function of the measured depth and as a function of the vehicle speed. Thus, a vehicle includes a system having a controller and at least one remote sensor including a first ultrasonic transducer sensor mounted on a left side mirror of the vehicle and a second ultrasonic transducer sensor mounted on a right side mirror of the vehicle such that the first and second ultrasonic transducer sensors are disposed on the vehicle. The first and second ultrasonic transducer sensors are configured to emit and receive a pulsed ultrasonic signal. The reception time of an ultrasonic echo signal indicates a distance detected between the ultrasonic transducer sensor and the surface level of a water spot in a measurement area adjacent to the vehicle.The publication US 2014 / 0 293 056 A1 describes a motor vehicle which comprises at least one camera with which an occupant of the vehicle can acquire an image of a part of an outer surface of the vehicle which is exposed to a liquid during the drying. The weather image is arranged so that the occupant can determine whether the liquid level is below a prescribed value.The present invention is based on the object of specifying a method for detecting the depth of wake in a vehicle, a driver assistance system for a vehicle and a vehicle having a driver assistance system of this type, which method solves at least some of the problems described above. In particular, it is an object of the present invention to specify a method for seat depth detection in a vehicle, a driver assistance system for a vehicle and a vehicle having a driver assistance system of this type, which enable reliable seat depth detection in a simple manner.The object is achieved according to the invention by the features of the independent claims. Advantageous embodiments of the invention are specified in the dependent claims.In particular, according to the invention, a method for detecting weather depth in a vehicle is provided, which has at least one camera with a field of view covering at least part of a vehicle body of the at least one camera in a downward direction, comprising the steps of: calibrating a position of the at least one camera with respect to the body of the vehicle before use of the vehicle, learning a shape of the vehicle body of the at least one camera in a downward direction, receiving a camera image covering the part of the vehicle body of the at least one camera in a downward direction, identifying a part of the vehicle body of the at least one camera in a downward direction which is not covered with liquid, comparing the part of the vehicle body of the at least one camera in a downward direction which is not covered with liquid, with the shape of the vehicle body of the at least one camera in a downward direction, and detecting a wet depth based on the comparison.According to the invention, a driving assistance system for a vehicle is also specified, wherein the driving assistance system is suitable for carrying out the above method.Furthermore, according to the invention, a vehicle having the above driving assistance system is specified.The basic idea of the present invention is thus to use at least one camera, typically at least one optical camera, in order to carry out a wake-up state detection. The use of the at least one camera has the advantage that it has a field of view that is better than the field of view of a typical ultrasonic sensor. Due to the better field of view, a larger area of the surface may be monitored, so that the reliability of the wet condition detection may be increased. In addition, the wide field of view enables an analysis of the field of view already before reaching a weather range. Moreover, vehicles are frequently equipped with at least one camera nowadays. Therefore, using this camera, the method can be performed without additional camera hardware. Compared to ultrasonic sensors, cameras provide a huge amount of sensor data that allow detailed analysis, whereby reliable wet condition detection, in particular wet depth detection, can be implemented.Browsing may be a desired or accepted feature, for example in the case of off-road vehicles. However, it is also important for normal vehicles to monitor a wet depth in order to avoid damage caused by a liquid. The liquid is typically water or slurry having the property of a liquid.The depth of gravity refers to a height of the liquid in an area around the vehicle. For off-road vehicles, it is even possible to achieve wake depths of more than one meter, whereas in normal vehicles a wake depth of a few centimeters may already be dangerous due to possible damage to the vehicle, in particular to the engine, in particular due to water entering the cylinder. Based on an orientation of the vehicle, the depth of wake may be different, e.g., on the left and right sides of the vehicle or on the front and rear sides.The vehicle has at least one camera. Preferably, the vehicle comprises an all-round camera system having a plurality of cameras covering all directions around the vehicle. Therefore, the depth of wake around the vehicle can be determined.The at least one camera has a field of view covering a portion of the vehicle body from the at least one camera in a downward direction. Therefore, this part of the vehicle body can be used as a reference for detecting a liquid level around the vehicle. The at least one camera provides images covering the portion of the vehicle body from the at least one camera in a downward direction. The images may be provided at any suitable rate, e.g., depending on the vehicle speed.Calibrating a position of the at least one camera with respect to the vehicle body is required in order to generate an absolute reference for determining the depth of wake. Different cameras may have different references. However, in this case as well, the information can be combined based on the known reference positions of the camera. The calibration must be performed at least once before the vehicle is used. The calibration step S 100 may be repeated, e.g., to make an adjustment to a changed vehicle feature, for example, when the air pressure of wheels of the vehicle changes or when the air pressure changes according to the current driving conditions. In particular, off-road vehicles may require different air pressures when changing between road travel and off-road travel.According to the invention, the step of learning a shape of the vehicle body of the at least one camera in a downward direction comprises carrying out a self-learning process of the shape of the vehicle body of the at least one camera in a downward direction. Therefore, the method can be easily applied to various types of vehicles. Changes in the appearance of the vehicle can be easily considered and do not lead to false weather condition detection because the vehicle can adapt to such changes, for example, if the color of the vehicle is changed, dirt or water drops adhere to the vehicle body, labels are attached to the vehicle body, or the like. Since the vehicle body is a static object, it can be learned as background information during a simple training phase. The shape of the vehicle can be learned by itself, i.e., outside the factory. The step of learning a shape of the vehicle body must be robust to handle changes in illumination and the presence of reflections on the vehicle body. Typically, only a portion of the vehicle body will be visible to each camera and learned as a vehicle body. Preferably, an initial training step is performed prior to use of the vehicle.According to a modified embodiment of the invention, the step of identifying a part of the vehicle body of the at least one camera in a downward direction that is not covered with liquid comprises modelling the liquid beforehand as a dynamic texture with a color changing in space. Due to the nature of the liquid, e.g. water or even sludge, its surface may move unpredictablely. Liquids are dynamically changing substances because of their fluidic nature. For example, waves may form on a surface of the liquid. This generally makes it difficult to determine a correct depth of wake. However, if the liquid is properly modeled, a correct depth of wake can be determined despite movement of the liquid.According to a modified embodiment of the invention, the step of prior modelling the liquid as a dynamically space changing texture with a colour comprises modelling the liquid as a pixel by pixel time changing liquid using an auto regressive moving average process (auto regressive moving average process). The autoregressive moving average process is also referred to as an ARMA process. The ARMA process is used in statistical analysis of time sequences. The ARMA process provides an economical description of a (weak) stationary stochastic process with respect to two polynomials, one for autoregressive and the second for moving average.According to a modified embodiment of the invention, the method comprises a step for detecting surface water in a direction of travel and a step for automatically activating the water depth detection with detected surface water in the direction of travel. When the at least one camera is oriented in the direction of travel, the detection of surface water is activated. The surface water can generally be captured already in front of the vehicle depending on the type of camera used and / or an orientation of the camera. However, other vehicle environmental sensors may also be used to determine surface water in the direction of travel. Therefore, the water depth detection may be already started in advance, so that the water depth detection is active and runs when the vehicle enters the water. A manual interaction to start the water depth detection is not required.According to a modified embodiment of the invention, the step of detecting a wet depth based on the comparison includes performing subtraction of a part of the vehicle body from the at least one camera in a downward direction that is not covered with liquid from the shape of the vehicle body from the at least one camera in a downward direction. By carrying out the subtraction, it is possible to determine up to what height the water level in the region of the camera reaches. When the subtraction result is substantially zero, it is assumed that the current image represents the shape of the vehicle body as learned previously. Therefore, water is not present around the vehicle. The higher the subtraction result, the larger the difference between the current image and the shape of the vehicle body as learned previously. Based on the subtraction, non-body parts can be easily eliminated to determine the depth of wake. A learning phase must be robust to handle changes in illumination and the presence of reflections on the body part.According to a modified embodiment of the invention, the step of detecting a depth of wake based on the comparison further comprises a step of modelling pixels remaining after the subtraction by K Gaussian distributions having weights ωiand parameters μ (mean value) and σ (variance). There are many variants of the Gaussian mixture model (GMM) that are usable to model the pixels. Preferably, the step of modelling image points remaining after subtraction by K Gaussians comprises performing an adaptive mixture of Gaussians, the K Gaussians being sorted based on a ratio ωi / σ 2 whereby the smallest variance and thus more consistent Gaussians 'i' are chosen, the upper k Gaussians being selected from the sorted sequence. Furthermore, a Gaussian model with an adaptive mixture of Zivkovic Gaussian distributions (MOG) is preferably used. According to the Zikovic adaptive mixture of Gaussian distributions, the background subtraction is analyzed using a pixel-level approach. Recursive equations are used to continuously update the parameters and simultaneously also select an appropriate number of components for each pixel.According to a modified embodiment of the invention, the method comprises an additional step of tiling the image into different blocks, wherein the upper k Gaussian distributions are learned separately for each block. Therefore, for each block, a deviation from the learned model during the training time can be calculated. Training data can be artificially augmented with various noise effects such as reflection, illumination changes, etc. We use this method in particular because it includes a changing Gaussian function that can model a temporal change of the liquid surface.A probability that a pixel is modeled by these K Gaussian distributions may be calculated asParameter updates are performed as follows for the case that components match I:Parameter updates are performed as follows in the case that components do not match I:According to a modified embodiment of the invention, the method comprises an additional step of executing a vehicle-to-vehicle warning upon detection of a predefined depth of wake. Therefore, information on the depth of the seat can be supplied to vehicles that do not support the state of the seat. In order to execute the vehicle-to-vehicle warning, the vehicle has a communication device for transmitting the wet depth to other vehicles either directly or via a server which distributes the wet state detection information. The communication device may be configured to communicate according to any suitable mobile communication standard, such as Bluetooth, WiFi, GPRS, UMTS, LTE, 5G, or others, just to name a few. Vehicle-to-vehicle warning enables contactless wake condition detection or even wake depth detection for the warned vehicle.According to a modified embodiment of the invention, the method comprises an additional step of performing online dynamic background subtraction for other vehicles. The at least one camera is used to determine whether other vehicles are immersed. A depth of wake may be estimated based on a known or estimated size of the other vehicle, and background subtraction may be optionally performed to determine to what extent the vehicle body is covered by liquid.According to a modified embodiment of the invention, the method comprises an additional step for detecting the volume of the liquid displaced by the vehicle. Thereby, an increase in e.g. the water level based on the vehicle entering or moving in the liquid may be taken into account. For example, if a surface of the surface water is known, it can be easily determined based on the dimensions of the vehicle how much a liquid level will increase based on the presence of the vehicle and its inherent liquid displacement. For small wells and large vehicles, a volume of displaced liquid is substantially larger so that the immersion depth of the vehicle rapidly increases. Therefore, a predictive model can be added through the use of a particle filter, and the camera-based estimation provides a tracking and localization function due to a wide field of view (FoV).According to a modified embodiment of the invention, the vehicle has an ultrasonic distance sensor, and the method has an additional step of combining the detected wet depth based on the comparison and a wet depth detected by the ultrasonic distance sensor. The combined information about the wakefulness increases the reliability of wakefulness detection and in particular wakefulness detection. The ultrasonic distance sensor may be used to determine the depth of wet as is known in the art. The fusion may be performed by a heterogeneous Bayesian model, as data associated with the different sensors is very different and has different ranges.These and other aspects of the invention will be apparent and apparent from the embodiments described below. Individual features disclosed in the embodiments may form an aspect of the present invention alone or in combination. Features of the various embodiments may be transferred from one embodiment to another embodiment.The following are shown: FIG. 1 is a schematic illustration of a conventional vehicle having a driving assistance system for detecting a state of wake using an ultrasonic sensor in a side view; FIG. 2 is a schematic illustration of a vehicle having a driving assistance system for detecting a state of wake using a plurality of cameras according to a first preferred embodiment in a plan view with additional camera views of the plurality of cameras; FIG. 3 is a detailed schematic camera view of a left side mirror camera of FIG. 2, the camera view being shown once with and once without a wakefulness, according to the first embodiment; FIG. 4 is a schematic diagram of the vehicle having the driving assistance system according to the first embodiment for detecting a state of wake in a side view; FIG. 5 is a schematic diagram of the vehicle having the driving assistance system according to the first embodiment for detecting a state of wake in a side view; FIG. 6 is a schematic detailed camera view of a rear camera of FIG. 2, the view of the rear camera being illustrated with single blocks, according to the first embodiment; FIG. 7 is a camera perspective view of a front camera of FIG. 2, the front camera view showing a submergence vehicle, according to the first embodiment; and FIG. 8 is a flowchart illustrating a method of performing a seat depth detection with the vehicle and the driving assistance system according to the first embodiment.FIG. 2 shows a vehicle 110 having a driving assistance system 112 according to a first preferred embodiment of the present invention.The driving assistance system 112 has a processing unit 114 and an all-round vision camera system 116, 118, 120, 122. The around view camera system 116, 118, 120, 122 includes a front camera 116 covering a front direction of the vehicle 110, a rear camera 118 covering a rearward direction of the vehicle 110, a right mirror camera 120 covering a right direction of the vehicle 110, and a left mirror camera 122 covering a left direction of the vehicle 110.The cameras 116, 118, 120, 122 and the processing unit 114 are connected via a data bus connection 124.Each camera 116, 118, 120, 122 has a field of view β, e.g., as shown in FIG. 4 or 5, and covers a portion of a vehicle body 126 from the respective camera 116, 118, 120, 122 in a downward direction, as shown in FIG. 2 and FIG. 3.A method for wake depth detection by the vehicle 110 according to the first embodiment will be discussed below. The depth of gravity refers to the height of a liquid 130, typically water, around the vehicle 10. In particular, a wake condition refers to the presence of the liquid 130 around the vehicle, and the wake depth refers to a height of a surface 132 of the liquid 130. The method will be discussed with reference to Figure 8, which shows a flow chart of the method according to the invention. Obviously, some of the method steps may be performed in an order different from the order of the described embodiments.The method begins with step S 100, which relates to calibrating a position of the cameras 116, 118, 120, 122 relative to the body 126 of the vehicle 110. Calibrating a position of the cameras 116, 118, 120, 122 relative to the body 126 of the vehicle 110 involves generating an absolute reference for determining the depth of the seat. The step of calibrating a position of the cameras 116, 118, 120, 122 relative to the body 126 of the vehicle 110 is performed once prior to use of the vehicle 110. Later, step S 100 may be repeated, e.g., due to changing driving conditions.In step S 110, a shape of the vehicle body 126 is learned by the respective camera 116, 118, 120, 122 in a downward direction from the camera 116, 118, 120, 122. This includes self-learning the shape of the vehicle body 126 from the cameras 116, 118, 120, 122 in a downward direction. Since the vehicle body 126 is a static object, the shape is learned as background information during a training phase. Step S 110 may be performed at substantially any time. Step S 110 does not need to be performed continuously or each time the method is executed. The step of learning a shape of the vehicle body 126 from the respective cameras 116, 118, 120, 122 in a downward direction from the camera 116, 118, 120, 122 is performed once before using the vehicle 110 as initial training. Training may be continued later.In step S 120, surface water in a traveling direction is detected. If surface water is detected positively in the direction of travel, the further water depth detection is automatically started. The surface water is detected using the camera 116, 118, 120, 122 looking in the direction of travel. Most commonly, the front camera 116 is used for sensing surface water.In step S 130, the control unit 114 starts to receive camera images from the cameras 116, 118, 120, 122, each covering the corresponding part of the vehicle body 126 from the camera 116, 118, 120, 122 in a downward direction.In step S 140, a portion of the vehicle body 126 in a downward direction that is not covered with liquid 130 is identified by the camera 116, 118, 120, 122. To enable detection of the surface 132, the liquid 130 is modeled in advance as a dynamic texture changing in space. This involves modeling the liquid 130 as a pixelwise time-varying liquid by an autoregressive moving average process. The autoregressive moving average process is also referred to as an ARMA process. The ARMA process is used in statistical analysis of time sequences and provides an economical description of a (weak) stationary stochastic process with respect to two polynomials, one for the autoregressive and the second for the moving average.According to step S 150, the part of the vehicle body 126 from the camera 116, 118, 120, 122 in a downward direction that is not covered with liquid 130 is compared with the shape of the vehicle body 126 from the camera 116, 118, 120, 122 in a downward direction. Therefore, the vehicle body 126 is used as a reference for detecting a liquid level around the vehicle 110. The comparison is performed as a subtraction of the images so that the remaining image contents are related to a water depth. As can be seen in FIG. 3, an identical portion of the image provided by the camera 116, 118, 120, 122 is subtracted from the learned shape of the vehicle body 126, such that only different portions remain.Based on the comparison, the depth of the water is detected in step S 160. Therefore, pixels remaining after subtraction are modeled by K Gaussian distributions with the weights ωiand the parameters μ (mean value) and σ (variance). In particular, an adaptive mixing of Gaussian distributions is carried out, in which the K Gaussian distributions are sorted based on a ratio ωi / σ 2 whereby the smallest variance and thus a more consistent Gaussian distribution 'i' is selected. The upper k Gaussian distributions are selected from the sorted order. In addition, a Zivkovic Gaussian model with adaptive mixing of Gaussian distributions (MOG) is used. According to the adaptive mixture of Zivkovic Gaussian distributions, the background subtraction is analyzed according to a pixel-level approach. Recursive equations are used to continuously update the parameters and simultaneously also to select an appropriate number of components for each pixel.Thus, each image of each camera 116, 118, 120, 122 is tiled into different blocks 134, as shown in FIG. 6 by way of example, where the upper k Gaussian distributions are learned separately for each block 134. Therefore, for each block 134, a deviation from the learned model is calculated during the training phase. Training data is artificially supplemented with various noise effects such as reflection, illumination changes, etc.According to step S 170, a volume of the liquid 130 displaced by the vehicle 110 is detected. Therefore, an increase in liquid level is determined based on the vehicle 110 entering and moving within the liquid 130. Based on a surface 132 of the surface water and the dimensions of the vehicle 110, it is determined how much a wet depth increases based on the presence of the vehicle 110 and its displacement of the liquid 130.According to step S 180, online dynamic background subtraction is performed for other vehicles. In doing so, the cameras 116, 118, 120, 122 are used to determine whether other vehicles 138 are submerged, as can be seen with reference to FIG. 7. A depth of wake may be estimated and approximated based on a known or estimated size of the other vehicle 138 by performing a background subtraction to determine to what extent the vehicle body 126 is covered by the liquid 130.According to step S 190, a vehicle-vehicle warning is carried out when detecting the seat depth, i.e. the warning is carried out if the seat depth exceeds a predefined seat depth. The driving assistance system 12 has a communication device for transmitting the depth of wake to other vehicles either directly or via a server which distributes the wake-up detection information. The communication device is configured to communicate according to a suitable mobile communication standard, such as Bluetooth, WiFi, GPRS, UMTS, LTE, 5G, or others.List of Reference Numerals10 Vehicle (prior art) 12 side mirror (prior art) 14 ultrasonic sensor (prior art) 16 water (prior art) 18 surface (prior art) 20 contact sensor (prior art) 110 vehicle 112 driving assistance system 114 processing unit 116 front camera 118 rear camera 120 right mirror camera 122 left mirror camera 124 data bus connection 126 vehicle body 130 liquid 132 surface 134 block 136 identical part

Claims

A method for depth sensing in a vehicle (110) having at least one camera (116, 118, 120, 122) with a field of view (β) covering at least a portion of a vehicle body (126) of the at least one camera (116, 118, 120, 122) in a downward direction, comprising the steps of: calibrating a position of the at least one camera (116, 118, 120, 122) relative to the body (126) of the vehicle (110) prior to use of the vehicle (110); learning a shape of the vehicle body (126) from the at least one camera (116, 118, 120, 122) in a downward direction; receiving a camera image covering the portion of a vehicle body (126) from the at least one camera (116, 118, 120, 122) in a downward direction; identifying a portion of the vehicle body (126) of the at least one camera (116, 118, 120, 122) in a downward direction that is not covered with liquid (130); comparing the portion of the vehicle body (126) of the at least one camera (116, 118, 120, 122) in a downward direction that is not covered with liquid (130) with the shape of the vehicle body (126) of the at least one camera (116, 118, 120, 122) in a downward direction; and detecting a depth of wake based on the comparison, characterized in that the step of learning a shape of the vehicle body (126) from the at least one camera (116, 118, 120, 122) in a downward direction includes performing a self-learning operation of learning the shape of the vehicle body (126) from the at least one camera (116, 118, 120, 122) in a downward direction.The method of claim 1, characterized in that the step of identifying a portion of the vehicle body (126) of the at least one camera (116, 118, 120, 122) in a downward direction that is not covered with liquid (130) comprises modeling the liquid (130) in advance as a dynamically spatially varying texture having a color.The method of claim 2, characterized in that the step of modeling the liquid (130) as a room-dynamic changing texture with a color in advance comprises modeling the liquid (130) as a pixel-by-pixel time changing liquid using an autoregressive moving average process.Method according to one of Claims 1 to 3, characterized in that the method has a step for detecting surface water in a direction of travel and a step for automatically activating the water depth detection when surface water in the direction of travel is detected.The method of any preceding claim, characterized in that the step of detecting a wet depth based on the comparison comprises performing a subtraction of a portion of the vehicle body (126) of the at least one camera (116, 118, 120, 122) in a downward direction that is not covered with liquid (130) from the shape of the vehicle body (126) of the at least one camera (116, 118, 120, 122) in a downward direction.The method according to claim 5, characterized in that the step of detecting a depth of wake based on the comparison further comprises a step of modelling pixels remaining after the subtraction by K Gaussian distributions with the weights ωi and the parameters μ (mean value) and σ (variance).A method according to claim 6, characterized in that the method comprises the step of tiling the image into different blocks, wherein the upper k Gaussian distributions are learned separately for each block.Method according to one of the preceding claims, characterized in that the method has an additional step for carrying out a vehicle-vehicle warning when detecting a predefined depth of wake.A method according to any preceding claim, characterized in that the method comprises an additional step of performing online dynamic background subtraction for other vehicles.Method according to any of the preceding claims, characterized in that the method comprises an additional step of performing a detection of a volume of liquid (130) displaced by the vehicle (110).Method according to one of the preceding claims, characterized in that the vehicle has an ultrasonic distance sensor and the method has an additional step for combining the wake depth detected on the basis of the comparison and a wake depth detected by the ultrasonic distance sensor.Driving assistance system (12) for a vehicle (10), characterized in that the driving assistance system (12) is configured to carry out the method according to one of the preceding claims.Vehicle (10) having a driving assistance system (12) according to Claim 12.

Images