Side mirror camera monitoring

Abstract

A vehicle includes a rear-view camera system comprising a camera mounted on one side of the vehicle, the camera defining a rearward field of view. A screen is visible from a driver's position within the vehicle and is configured to display an image input captured by the camera. At least a portion of the vehicle lies within the rearward field of view. The portion of the vehicle within the rearward field of view includes at least one distinguishable vehicle feature that is fixed to the vehicle with respect to the rear-view camera.A controller comprises a memory and a processor, the memory storing instructions configured to cause the controller to operate at least one quality assurance subprocess in real time, and configured to cause the controller to notify the driver in response to at least one quality control metric determination.

Description

introduction

[0001] The present disclosure relates to vehicles and in particular to rear-facing side mirror camera systems.

[0002] Vehicles are equipped with one or more rear-facing side mirrors. These mirrors provide a view to the rear on the side of the vehicle where they are mounted, without requiring the driver to take their head completely off the road. This rearward visibility facilitates the safe execution of numerous driving operations and maneuvers, such as changing lanes, overtaking, turning, etc.

[0003] In some cases, vehicles can replace one or more of the conventional mirrors with a rear-facing camera and a corresponding display to provide a better view than can be achieved using a mirror. For example, the camera can be positioned to reveal aspects obscured by a mirror or to provide a wider field of view than is possible with a mirror. Alternatively, similar camera placements can be used in addition to the mirrors, with the resulting images complementing the views provided directly by the mirrors.

[0004] Typically, the screen displaying the rear view is not physically fixed to the camera's location in the same way as a mirror. As a result, the rear view screen can be positioned within the vehicle away from the camera, allowing the driver to see the rear view without distracting their attention from the road ahead.

[0005] It is desirable to provide a side-mounted rear-view camera that produces an accurate real-time image of the view to the rear, thus enabling the camera system to replace the functions of a rear-view mirror.

[0006] US 2021 / 0031705 A1 discloses a device and a method for monitoring a vehicle camera system and a camera monitoring system, as well as a method for fault detection for such camera monitoring systems. DE 1020107016 A1 discloses a method for verifying a recording area of ​​an image acquisition unit or a readout area of ​​an image sensor of the image acquisition unit, for a vehicle. Summary

[0007] A vehicle includes a rear-view camera system comprising a camera mounted on one side of the vehicle, the camera defining a rearward field of view. A screen is visible from a driver's position in the vehicle and configured to display an image feed captured by the camera. At least a portion of the vehicle lies within the rearward field of view. This portion of the vehicle within the rearward field of view includes at least one distinguishable vehicle feature that is fixed to the vehicle with respect to the rear-view camera.A controller comprises a memory and a processor, wherein the memory stores instructions configured to cause the controller to operate at least one quality assurance subprocess in real time, and configured to cause the controller to notify the driver in response to at least one quality control metric determination. The at least one quality assurance subprocess comprises a frozen-frame monitoring subprocess, wherein the frozen-frame monitoring subprocess is configured to determine semantic similarity between a first frame and a subsequent frame using a frozen-frame subprocess, the subsequent frame following the first frame with a delay of a plurality of intervening frames.

[0008] In addition to one or more of the features described herein, the frozen image subprocess includes providing the first image as input to a first neural network and providing the second input to a second neural network, as well as comparing an output of the first neural network with an output of the second neural network.

[0009] In addition to one or more of the features described herein, the first neural network and the second neural network are identically trained neural networks with exactly the same parameters and the same weights for these parameters.

[0010] In addition to one or more of the features described herein, the output of the first neural network is a first vector and the output of the second neural network is a subsequent vector.

[0011] In addition to one or more of the features described herein, the at least one quality assurance subprocess includes a latency monitoring subprocess configured to detect latency of an image feed provided by the camera by comparing an expected number of images in the image feed during a predetermined time window with an actual number of images received from the camera at the controller during the predetermined time window.

[0012] In addition to one or more of the features described herein, the latency monitoring subprocess is configured to cause the controller to notify the driver in response to the latency exceeding an acceptable latency threshold.

[0013] In addition to one or more of the features described herein, the at least one quality assurance subprocess includes a subprocess for monitoring a camera position and orientation, configured to detect any deviation of the actual position and orientation of the camera from an expected position and orientation of the camera.

[0014] In addition to one or more of the features described herein, the subprocess for monitoring a camera position and orientation is configured to detect a deviation of an actual camera position and orientation from an expected camera position and orientation by comparing an expected position of the at least one distinguishable vehicle feature within an image generated by the rear-view camera with an actual position of the at least one distinguishable vehicle feature within the image.

[0015] In addition to one or more of the features described herein, at least one distinguishable vehicle feature includes a rear light.

[0016] In addition to one or more of the features described herein, the subprocess for monitoring a camera position and orientation is further configured to crop the image before comparing the expected position of the at least one distinguishable vehicle feature within the image generated by the rear-view camera with the actual position of the at least one distinguishable vehicle feature within the image.

[0017] In another exemplary embodiment, a method for monitoring an image produced by a rearward-facing, side-mounted camera comprises operating a plurality of quality assurance subprocesses in real time and is configured to cause the controller to notify the driver in response to at least one quality control metric determination. The plurality of quality assurance subprocesses includes a subprocess for monitoring a frozen image, a subprocess for monitoring camera position and orientation, and a latency monitoring subprocess.

[0018] In addition to one or more of the features described herein, the frozen image monitoring subprocess determines a semantic similarity between a first image and a subsequent image using a frozen image subprocess that includes providing the first image as input to a first neural network and providing the second input to a second neural network, as well as comparing an output of the first neural network with an output of the second neural network.

[0019] In addition to one or more of the features described herein, the first neural network and the second neural network are identically trained neural networks with exactly the same parameters and the same weights for these parameters.

[0020] In addition to one or more of the features described herein, the output of the first neural network is a first vector and the output of the second neural network is a subsequent vector.

[0021] In addition to one or more of the features described herein, the following image immediately follows the first image.

[0022] In addition to one or more of the features described herein, the subsequent image follows the first image with a delay of a large number of intervening images.

[0023] The above features and advantages, as well as other features and advantages of the disclosure, are readily apparent from the following detailed description when it is made in conjunction with the accompanying drawings. Brief description of the drawings

[0024] Further features, advantages and details appear only as examples in the following detailed description, which refers to the drawings in which: Fig. 1A is a schematic representation of a vehicle that has a side-mounted camera system for rearward viewing; Fig. 1B is a perspective view of the side-mounted camera system for the rear view from the driver's point of view; Fig. 2 a monitoring process to ensure that an image presented to the driver by means of the rear view camera system is not frozen; Fig. 3 an exemplary procedure for comparing subsequent images in the process of Fig. 2 using machine learning; Fig. 4. An exemplary process for monitoring the latency of an image presented by the rear-view camera system is; and Fig. 5 is an exemplary process for monitoring the position and orientation of a camera using an image presented by the camera for the view to the rear. Detailed description

[0025] The following description is merely exemplary and is not intended to limit the present disclosure, its application, or uses. It should be understood that in the drawings, corresponding reference numerals denote identical or corresponding parts and features.

[0026] As used herein, the term module refers to a processing circuit arrangement that may include an application-specific integrated circuit (ASIC), an electronic circuit, a processor (common, dedicated or group) and memory executing one or more software or firmware programs, a combinational logic circuit and / or other suitable components providing the described functionality.

[0027] As used herein, the terms controller and control system refer to dedicated individual controllers, general controllers containing control modules dedicated for specific purposes, software systems within general controllers, networks of controllers interconnected and configured to work together to control one or more systems, and / or any similar configuration of processors and memory systems capable of implementing a control process.

[0028] According to an exemplary embodiment, a vehicle includes one or more side-mounted rear-view camera systems. These rear-view camera systems can be provided in addition to or instead of rear-facing side mirrors and offer the driver a similar view via a corresponding screen mounted inside the vehicle. To ensure that the image presented to the driver can be reliably used instead of, or in addition to, a controller inside the vehicle monitors the image feed from the camera(s) by performing quality assurance processes on the image feed. This quality assurance process comprises several simultaneous subprocesses.In one example, the quality assurance subprocesses include a process to ensure that the image is not frozen, a process to ensure that the image latency is below an acceptable latency level, and a process to ensure that the field of view captured by the camera(s) has not shifted.

[0029] Fig. 1A is a schematic top view of a vehicle 10 comprising a body 12 and a passenger compartment 14. Fig. Figure 1B illustrates a perspective view 40 of a side-mounted rear-view camera system 42 from the driver's point of view 18. The vehicle body 12 includes rear lights 16 arranged at the rear of the vehicle 10. The driver 18 is seated in a driver's position in the passenger compartment 14. The rear-view camera system 42 is configured to provide the driver 18 with views that replicate those provided by conventional side-mounted rear-view mirrors.

[0030] The rear-view camera system 42 comprises cameras 30 mounted on the vehicle body 12 via wings 32. The cameras 30 generate corresponding rearward-facing fields of view 31. Due to the positioning of the cameras 30, the rearward-facing fields of view 31 encompass a corresponding part 33 of the vehicle body 12. In the example of Fig. 1A, the corresponding part 33 includes the taillights 16. In alternative examples where the taillights 16 are not within the fields of view 31, the vehicle body 12 may have markings or other features on the corresponding part 33, wherein the markings or other features have a fixed position in relation to the body 12, so that the markings or other features do not move in relation to the camera and always appear in a fixed position in the generated image.

[0031] A controller 20 comprises a memory 22 and a processor 24 and is connected to each camera 30. Although in the example of Fig. As illustrated by 1A as a dedicated controller, it is understood that the controller 20 may include additional features and elements for the additional control of one or more systems within the vehicle 10. For example, the controller 20 may include motor controls for steering the wings 32 and / or actuators within the wings 32, thus enabling the controller 20 to adjust the position and angle of the cameras 30 relative to the vehicle body 12.

[0032] Furthermore, the controller 20 is configured to output processed images generated by the cameras 30 to a screen 34, allowing the driver 18 to see the generated images in real time.

[0033] Memory 24 contains one or more modules that include subprocesses for monitoring the images generated by the cameras 30 and presented to the driver 18 on the screen 34. This monitoring continuously tests the functionality of the cameras 30 and detects any problems that could impair the functionality of the respective camera 30 in real time by running several simultaneous subprocesses on the controller 20. Detectable problems can include a camera freezing, unacceptably high latency, and / or a camera misalignment. When such a problem is detected, the controller 20 issues a corresponding warning or notification to the driver 18 via one or more onboard systems (e.g., the screen 34), informing the driver 18 of the loss or limitation of functionality.In some cases, the controller 20 also includes one or more modules configured to adjust or set the positioning and / or angle and orientation of the cameras 30 in order to correct or mitigate a limitation of functionality after the limitation has been detected.

[0034] Controller 20 is configured to provide the images generated by cameras 30 to additional vehicle systems, such as driver assistance systems, object recognition systems, pedestrian detection systems, and the like. Controller 20 is further configured to inform these additional systems of any loss or impairment of functionality when detected. Following notification, the additional systems can take appropriate corrective action, as warranted by the respective additional system.

[0035] During normal operation, the camera 30 captures an image of the surroundings within its field of view 31. The captured image is corrected by the controller 20 to account for any distortions due to the shape, angle, positioning, etc., of a camera lens and prepared for display on the screen 24 using image processing systems. One or more machine learning algorithms are used to compare the dynamic components in two consecutive (or sequential) images to detect camera freeze. Additionally, the taillight 16 (or another feature of the vehicle body 12) is always visible within the image and should be in a fixed position within the image.This knowledge is used to detect misalignment or displacement of the image by determining how far the actual position of the rear light 16 is from the expected position of the rear light 16. In another monitoring subprocess, an expected frequency of images is compared with an actual received frequency to monitor for latency of the video feed provided by the camera 30.

[0036] In some cases, the above process can be affected by flickering of the rear light 16, which, although not visible to the human eye, is discernible at the recording speeds of the cameras 30. To prevent the rear light flickering from affecting the monitoring subprocesses that compare the images, the signal processing performed on the image includes overlaying a solid block of color over the rear light 16. This prevents rear light flicker from unintentionally producing inaccurate results. After the monitoring subprocesses and before the images are displayed on the screen 34, the overlaid solid block of color is removed from the image.

[0037] With continued reference to Fig. 1A and Fig. 1B illustrates Fig. Process 200, one of the concurrent subprocesses used to monitor the image input from camera 30, determines whether the image is frozen. First, camera 30 generates an initial image (image A) and provides it to controller 20 in step 210, "Receive image A". Image A is temporarily stored in memory 24 in step 220, "Buffer image A". Then, camera 30 provides the controller with the next image (image B) in step 230, "Receive image B".

[0038] After receiving image B, image A and image B are preprocessed in step 240, "Process Images," using image processing techniques stored in memory 24. The image processing is performed according to established image processing techniques and prepares the images for display to the driver on screen 34. Among other processing steps, during step 240, "Process Images," images A and B are corrected for distortion using a calibration device.

[0039] Once processed, images A and B are compared in step 250 "Compare image A with image B" to determine their similarity (illustrated in more detail in Fig. 3) For the purposes of process 200, images A and B are considered similar if they differ only in contrast, brightness, and rotation. This degree of similarity can alternatively be referred to as semantically identical. Semantically identical images depict the same objects in the same positions.

[0040] To determine whether the images represent a static image (e.g., a frozen image output by camera 30) and not merely similar scenes (e.g., successive images of a long straight stretch of an empty road), the processed images are compared using a machine learning process such as a deep neural network (DNN) to determine semantic identity.

[0041] With continued reference to Fig. 2 illustrates Fig. 3 a process flow 300 of the comparison using machine learning from step 250. First, image A and image B (images 302) are provided to a twin network 310, which compares the images 302 in real time.

[0042] Twin Network 310 is a neural network comprising two identical subnetworks (DNNs, Twin Networks 310, 310'). Each of the identical subnetworks contains exactly the same parameters and the same weights for those parameters. Subnetworks 310, 310' can be any neural network configured for image analysis. In one example, DNN subnetworks 310, 310' are convolutional neural networks. In other examples, other types of neural networks can be used for a similar effect. Each subnetwork 310, 310' receives the corresponding image 302 as input and outputs a vector 312, 314. The vectors 312, 314 are then compared in a step 316, "Vector Comparison," to determine how similar the images 302 are.Semantically identical images have identical image vectors 312, 314, which allows the controller to easily distinguish between superficially similar scenes resulting from minimally changing backgrounds and semantically identical images resulting from a frozen camera image.

[0043] Back on Fig. 2 are referring if the compared vectors 312 and 314 are identical, the images 302 are semantically identical, and a frozen image is detected. This detection results in a notification of the driver 18 in step 260, "Notify Operator." In some examples, step 260, "Notify Operator," also includes notifying one or more additional vehicle systems, such as driver assistance systems, pedestrian detection systems, and the like, that the image is frozen. In such cases, the additional vehicle systems are configured with appropriate responses to a frozen image. The appropriate responses depend on the specific additional vehicle system and may include responses such as using a last known valid image, removing the camera 30 providing the frozen image from a set of image sensors, locking, or...This includes disabling the additional vehicle system and / or any similar reactions.

[0044] In the exemplary process 200, image A 302 and image B 304 follow each other immediately in the image feed from the corresponding camera 30. In alternative examples, a delay can be introduced between the compared images by separating them with a certain number of additional images. Such a delay may be desirable if only minimal deviations from image to image are expected, and a delay can improve the ability of the controller 20 to detect a frozen image.

[0045] With reference to Fig. 4. Simultaneously with process 200, which determines whether an image is frozen, controller 20 executes process 400 to determine the latency of an image stream provided by camera 30. Process 400 operates on the underlying assumption that camera 30, when operating at peak capacity, will provide the controller with a known number of images within a specified timeframe. For example, if camera 30 has a frame rate of 180 frames per second, then controller 20 should receive 180 frames within any given one-second interval. The latency is a measure of how much the actual number is reduced compared to the expected number.

[0046] When initiated by controller 20, process 400 begins in step 410, "Receive image," by receiving an initial image. In the first loop of process 400, controller 20 reacts to the reception of the first image by starting a timer. Furthermore, upon receiving the image in step 410, a counter is incremented in step 410, "Increment counter." After the counter is incremented, process 400 checks the elapsed time in step 430, "Check timer."

[0047] If the timer is below a threshold time, e.g., one second, process 400 returns to step 410 "Receive image" and waits for a new image. As soon as the new image is received, a subsequent loop of process 400 begins at step 410 "Receive image".

[0048] If the timer is equal to or exceeds the threshold time, process 400 transitions from step 430, "Check Timer," to step 440, "Compare Counter to Expected Count." In step 440, the controller identifies the number of frames received within the time threshold as the counter value and compares this value to a known expected number of frames. In the example where 180 frames per second is the expected rate, the expected count is 180 for a period of one second.

[0049] Based on this comparison, the image feed latency is determined, and a corresponding latency warning is issued to the operator. For example, the latency can be set to several acceptance levels with corresponding warnings. In this example, a latency of 90% or higher (meaning that 90% or more of the expected image count is received within the timeframe) is considered good, and no warning is issued. A latency between 70% and 90% is considered acceptable but slow, and the image feed can still be used, provided the operator is informed. In this case, the operator receives a warning indicating that the image feed is delayed or laggy. A latency below 70% is considered unacceptably slow, and the image feed cannot be used.In such a case, the driver receives a warning (18) indicating that the image input is unacceptably slow, and the screen (34) stops displaying the image. The latency figures and ranges described herein are exemplary. Practical implementations will use different figures and ranges depending on the specific characteristics of the vehicle (10).

[0050] The latency warning can be provided to the driver 18 directly on screen 34, which is used to display the relevant video feed, on one or more auxiliary screens within the vehicle, or a combination thereof. Furthermore, the latency warning can be a combination of audiovisual warnings, including text overlays on screen 34, symbolic warnings, or any other suitable human-machine interface.

[0051] In some implementations, multiple subsequent iterations of process 400 can be analyzed by controller 20 to provide further context and thus communicate latency warnings. For example, if subsequent iterations of process 400 indicate that the latency is increasing (meaning that fewer images are received in each subsequent iteration), a corresponding warning can be provided to driver 18 preventively before the latency reaches unacceptable levels, allowing driver 18 to take any necessary or appropriate corrective action before the image feed becomes unreliable.

[0052] Alternatively, if the latency improves, the controller 20 can apply a lower acceptance threshold, allowing an image feed with improving latency to continue to be used even if the image feed is currently below the acceptable level.

[0053] Simultaneously with sub-processes 200 and 400, the controller 20 ensures in sub-process 500, for monitoring a position, that the camera 30 is in a correct position and orientation. Fig. Figure 5 illustrates an exemplary subprocess 500 for monitoring the position and orientation of the camera 30 based on an expected positioning of one or more features within the received image.

[0054] Subprocess 500 can be performed on a single received image. However, in some examples, subprocess 500 can be repeated on subsequent images, combining the outputs (e.g., by averaging) to verify the accuracy of the results.

[0055] First, in step 502 “Receive image”, the controller 20 receives an image from the camera 30. The received image then undergoes an initial image processing sequence in step 504 “Process image”.

[0056] The fixed features are immovably or rigidly mounted on the vehicle 10 or parts thereof and can be detected using automated image analysis. Based on the expected position and orientation of the camera 30, the controller 20 can identify where the fixed features are expected to appear in the received image. In step 506, "Crop to Feature Coordinates," the controller crops the image to a smaller area that directly surrounds the expected position of one or more fixed features, such as the taillight 16. Cropping the received image reduces the size of the image used in subsequent steps, thereby reducing the computational load of subprocess 500 and increasing the speed at which subprocess 500 can be completed.

[0057] After cropping the image to the area where the fixed features are expected to appear, the cropped image is compared to a target image in step 508, "Compare to Target." The target image is a precise position within the received image where the fixed features are expected to appear based on the anticipated angle and position of camera 30. The comparison determines any deviation from the expected position in pixels, and this deviation is then compared to a target threshold and a maximum threshold.

[0058] If the variation is less than the target threshold (check 510), the controller 20 determines that the camera 30 is within an acceptable range of the desired position and orientation and that any variation is negligible. Such a variation could, for example, be a result of vehicle vibrations, minor alignment errors, etc. If check 510 is passed, subprocess 500 returns to step 502 "Receive image" and is ready for another iteration.

[0059] If the variation is greater than or equal to the target threshold but less than a maximum threshold (check 512), the controller 20 determines that the camera 30 is not in the correct position and attempts to correct the position in step 514, “Include Correction.” In some examples, the controller 20 incorporates the correction by activating control mechanisms such as actuators within the wing 32 and the camera 30. The control mechanisms adjust the position and angle of the camera 30. The magnitude and orientation of the adjustments can be determined by the controller 20 based on the variation according to known techniques.

[0060] Once set, subprocess 500 returns to step 502 “Receive image” and immediately determines again whether the setting has adequately corrected the positioning.

[0061] If step 508, "Compare to Target," determines that the variation is above the target threshold and above a maximum threshold (check 516), or if step 508, "Compare to Target," determines that an attempt by a previous iteration to make corrections did not correct the position and / or orientation of camera 30, subprocess 500, in step 518, "Warn Operator and Controller," warns the operator and any additional systems that rely on images from camera 30 that camera 30 is not in the correct position and should not be relied upon. After warning the operator and any additional systems, controller 20 terminates subprocess 500 until a new motor cycle begins.

[0062] By monitoring the image input using the quality assurance subprocesses, the controller 20 can ensure that accurate and reliable images are provided to the driver 18 and that the driver 18 is notified if the camera 30 cannot provide reliable images.

[0063] The terms "a / an / an" do not denote a quantity restriction, but rather indicate the presence of at least one of the elements being referred to. The term "or" means "and / or" unless the context clearly indicates otherwise. A reference in the entire description to "an aspect" means that a specific element (e.g., a feature, a structure, a step, or a property) described in connection with that aspect is contained in at least one aspect described herein and may or may not be present in other aspects.

[0064] Furthermore, it is understood that the described elements can be combined in any suitable way in the various aspects.

[0065] When it is stated that an element, such as a layer, film, area, or substrate, is located "on" another element, it can be located directly on top of the other element, or there can be intermediate elements. Conversely, when it is stated that an element is located "directly on" another element, there are no intermediate elements.

[0066] Unless otherwise stated herein, all test standards or norms are the latest applicable norm as of the filing date of this application or, if priority is claimed, as of the filing date of the earliest priority application in which the test standard appears.

[0067] Unless otherwise defined, the technical and scientific terms used herein have the same meanings as generally understood by a person skilled in the field to which this disclosure relates.

[0068] Although the above disclosure has been described with reference to exemplary embodiments, it is understood by the person skilled in the art that various modifications can be made and equivalent elements can be substituted without altering the scope of the disclosure. Furthermore, many modifications can be made to adapt a particular situation or material to the teachings of the disclosure without altering its essential scope. Therefore, the present disclosure is not intended to be limited to the specific embodiments disclosed, but rather to encompass all embodiments that fall within its scope.

Claims

[1] Vehicle (10), comprising: a rear-view camera system (42) comprising a camera (30) mounted on one side of the vehicle (10), wherein the camera (30) defines a rearward field of view (31), a screen (34) visible from a driver's position in the vehicle (10) and configured to display an image input captured by the camera (30), and wherein at least a portion of the vehicle (10) is located within the rearward field of view (31), the portion of the vehicle (10) within the rearward field of view (31) comprising at least one distinguishable vehicle feature fixed to the vehicle (10) with respect to the rear-view camera (30); and a controller (20) comprising a memory (22) and a processor (24), wherein the memory (22) stores instructions configured to cause the controller (20) to operate at least one quality assurance subprocess in real time, and configured to cause the controller (20) to notify the driver in response to at least one quality control metric determination; wherein at least one quality assurance subprocess includes a subprocess for monitoring a frozen image; wherein the subprocess for monitoring a frozen image is configured to determine a semantic similarity between a first image and a subsequent image using a frozen image subprocess; and where the subsequent image follows the first image with a delay of a large number of intervening images. [2] Vehicle (10) according to claim 1, wherein the frozen image subprocess comprises providing the first image as input to a first neural network and providing the second input to a second neural network, as well as comparing an output of the first neural network with an output of the second neural network, and wherein the first neural network and the second neural network are identically trained neural networks with exactly the same parameters and the same weights for these parameters. [3] Vehicle (10) according to claim 2, wherein the output of the first neural network is a first vector and the output of the second neural network is a subsequent vector. [4] Vehicle (10) according to claim 1, wherein the at least one quality assurance subprocess comprises a latency monitoring subprocess configured to detect a latency of an image feed provided by the camera (30) by comparing an expected number of images in the image feed during a predetermined time window with an actual number of images received by the camera (30) at the controller (20) during the predetermined time window, and causes the controller (20) to notify the driver in response to a latency exceeding an acceptable latency threshold. [5] Vehicle (10) according to claim 1, wherein the at least one quality assurance sub-process comprises a sub-process for monitoring a camera position and orientation, which is configured to detect a deviation of the actual position and orientation of the camera (30) from an expected position and orientation of the camera (30) by comparing an expected position of the at least one distinguishable vehicle feature within an image generated by the camera (30) for the view to the rear with an actual position of the at least one distinguishable vehicle feature within the image. [6] Vehicle (10) according to claim 5, wherein the at least one distinguishable vehicle feature comprises a rear light and wherein the subprocess for monitoring a camera position and orientation is further configured to crop the image before comparing the expected position of the at least one distinguishable vehicle feature within the image generated by the camera (30) for the view to the rear with the actual position of the at least one distinguishable vehicle feature within the image.

Images