Headlight adjustment procedure

Abstract

Method for adjusting the headlights of a motor vehicle with a headlight designed to produce a vehicle-specific light distribution and a camera used to detect the light distribution of the headlight, characterized by that the camera detects the change in light distribution due to the control, especially the movement, of the headlight in order to detect the light-dark boundary of the light distribution, where, in order to detect the change in light distribution, the light distribution is captured as a binary image, where the change in light distribution is detected when a threshold is exceeded in order to filter out camera noise and lighting interference effects.

Description

[0001] The invention relates to a method for adjusting the headlights of a motor vehicle with a headlight designed to generate a vehicle-specific light distribution and a camera used to detect the light distribution of the headlight.

[0002] It is well known that headlights are used in motor vehicles to illuminate the road ahead. Headlights produce different lighting functions, such as high beam, low beam, and the like, which must comply with specific legal regulations depending on the traffic situation. However, over the course of vehicle use, and especially during operation, the headlight positions can deviate from their original settings. This can be caused, for example, by temperature-related volume changes and settling processes in the headlight materials. These deviations can ultimately lead to an undesirable shift in the light distribution and unacceptable glare. Therefore, after a certain period of time, the headlights must be readjusted.This can involve using a marked measuring wall that represents the desired target coordinates of a light-dark boundary in the light distribution, onto which the headlights can then be aligned. Alternatively, special camera-based alignment methods can be used, in which the camera measures the light intensities of the light distribution to determine the light-dark boundary. From this information, the deviation from the desired position of the light-dark boundary is calculated, and a corresponding correction value for the headlight position is then determined.

[0003] Conventional methods that determine the light-dark boundary based on image intensity are prone to errors when dealing with other light-dark transitions in the image. This can have various causes, such as poor lighting, poor quality of the measuring wall or road surface, camera resolution issues, etc. For example, road markings can be misinterpreted as light-dark boundaries. Consequently, headlight adjustment can only be performed inaccurately or not at all. A further disadvantage of these established methods is the high computational and time-consuming nature of determining the light-dark boundary, which makes real-time evaluation particularly difficult.

[0004] WO 2012 / 048 795 A1 discloses a method for adjusting and / or calibrating at least one headlight of a vehicle.

[0005] It is therefore the object of the present invention to overcome the disadvantages of conventional methods described above. In particular, it is the object of the present invention to provide a method for headlight adjustment that enables improved, more precise headlight adjustment.

[0006] The aforementioned problem is solved by a method with the features of independent claim 1. Advantageous embodiments of the invention are specified in the dependent claims.

[0007] The invention includes the technical teaching that the camera can detect the change in light distribution due to the control, in particular the movement, of the headlight in order to detect the light-dark boundary of the light distribution. The light-dark boundary of the light distribution can also be referred to as the headlight's light edge. The invention proposes controlling or moving the headlight in such a way as to generate a change in the light distribution. If it is a movable headlight, the headlight can be swiveled. If it is a static headlight, for example an LED headlight, individual LEDs or LED segments can be controlled, in particular switched on or off, or dimmed, depending on the size of the LED matrix. Thus, a first advantage of the invention is that the method according to the invention can be applied universally, i.e.,not only in movable swivel modules but also in rigid LED matrix beam systems.

[0008] The inventive concept lies in the fact that changes in light distribution can be detected more easily than with a single static image. Changes in light distribution can, for example, be captured as a difference image between two successive images. The invention is based on the idea that changes or moving objects between two images can be represented in a difference image. The difference image can thus provide information about the light-dark boundary of the light distribution. Advantageously, capturing a difference image between two images is easier and more precise than capturing the light-dark boundary of the light distribution using only one image. This follows from the fact that disturbances, such as...Ambient lighting, other light edges such as side stripes, the surface structure of the projection surface, and ambient brightness can be filtered out more easily than when measuring light intensity based solely on an image. This results in a further advantage of the invention, namely that the method according to the invention is less susceptible to interference compared to conventional methods for headlight alignment. It is advantageous that the measurement results are improved and more precise positioning of the headlight can be ensured.

[0009] Since the resolution of the difference image is higher than that of a single image, the method according to the invention no longer relies on smooth, optimally positioned surfaces to detect the light-dark boundary, as is the case with conventional methods. The method is therefore independent of the nature or selection of the projection surface. Consequently, any wall or the road area in front of the vehicle can be used as the projection surface. Even objects with complex surfaces, such as the rear of a vehicle, can be used for evaluation. This achieves a further advantage of the invention: the method can be carried out by the driver himself or even automatically, completely independently of professional assistance, for example, in a workshop.

[0010] The method according to the invention is therefore not only robust but also fast, as it does not require significant computing power. The method can therefore also be carried out while the headlight is in operation. Advantageously, the implementation of the method requires no additional hardware and can be implemented in existing systems. An existing front camera can advantageously be used to capture the images. According to another advantage, the method according to the invention can be executed fully automatically, for example, periodically. Additionally or alternatively, the method according to the invention can be started selectively, for example, by manual activation by the user.

[0011] According to the invention, it can be provided that, to detect changes in the light distribution, the headlight can be vertically controlled, in particular moved, when the light distribution to be detected is a low beam distribution. Since the change in the light distribution only becomes apparent when the images being compared differ, the light distribution must be shifted in such a way that the light-dark boundary of the light distribution changes at all characteristic points. The low beam distribution comprises a horizontal area, an area inclined upwards, and a further higher area. All these areas can change when the headlight is controlled in the vertical direction, so that the change in the light distribution becomes visible at all characteristic points of the light-dark boundary.In the case of low beam distribution, it is therefore sufficient to detect the change in light distribution if the headlight is controlled or moved in such a way that the cut-off line of the light split is raised or lowered.

[0012] According to the invention, it can further be provided that, to detect changes in the light distribution of the headlight, it can be controlled vertically and horizontally, in particular moved, if the light distribution to be detected is a high beam distribution. In the case of a glare-free light distribution, the light-dark boundary comprises a horizontal and a vertical area. In order for the horizontal area to become visible as a change in the light distribution, the headlight must be controlled or moved in the vertical direction. Similarly, in order for the vertical area to become visible as a change in the light distribution, the headlight must be controlled or moved in the horizontal direction. Advantageously, a corresponding control or movement of the headlight can be implemented for each characteristic light distribution of the headlight in order to make the change in the light distribution visible in a difference image.

[0013] According to the invention, the camera can record and / or store successive image sequences to detect changes in light distribution. It is advantageous that the recorded images can be assigned to corresponding headlight positions and the image differences can be compared with existing patterns. For this purpose, suitable software for evaluating the images can advantageously be loaded into a central control unit, the lighting control system, or the camera. Existing control units can also be used to control the headlights according to the invention. Consequently, the method according to the invention can advantageously be implemented in existing systems and does not require any modification of the vehicle.

[0014] According to the invention, at least one of the following steps can be carried out to detect the change in the light distribution: a) Capturing an initial image at an initial point in time, b) Controlling or moving the headlight, c) Capturing a second image at a second time, d) Calculating a difference image between the first image and the second image.

[0015] Changes in light distribution can be detected as a difference image. Advantageously, the steps can be performed either sequentially or at least partially, for example, steps a) and c) with step b). In the latter case, the camera can take the images one after the other, while the spotlight is controlled simultaneously. This allows for not only a flexible but also a fast process. Consequently, the process can be carried out during operation without causing significant disruption to the user and without long waiting times. During the movement of the spotlight or light source, image sequences are captured and stored using a camera. In subsequent preprocessing, the image sequences can be prepared in such a way that interference is largely reduced.Advantageously, a difference image is then calculated between the first and second images, in which changes or moving objects between the two images can be displayed. This effect is used to detect the headlight beam edge. A difference image is calculated by subtracting the values ​​of two consecutive individual images from each other. Thus, only the areas that have changed between these images are visible in the difference image. Image information that has not changed has the same value and is not displayed in the difference image. The advantage of the invention lies in the fact that the difference image can be displayed brighter and more clearly against the background than the representation of the headlight beam edge in a single image. This effect is used to more reliably detect the light-dark boundary and to more easily identify or filter out interfering factors.By combining or adding multiple difference images, motion detection can be further enhanced. Analyzing several difference images allows for a more reliable determination of the direction in which the headlight is being moved and thus the direction in which the light-dark boundary is shifting, enabling reliable verification of the headlight's light edge and its differentiation from other light edges or side stripes.

[0016] To detect changes in light distribution, the light distribution is captured as a binary image. Changes in the light distribution are detected when a threshold is exceeded, filtering out camera noise and lighting interference. Advantageously, irrelevant changes can be filtered out by the applicable threshold. The threshold can be advantageously adjusted according to ambient, weather, and / or day / night conditions to achieve better and more accurate results. Based on the adjustable threshold, it can be more reliably determined whether the change is relevant. If the difference between the images is greater than the threshold, a 1 is entered at position x in the difference image. The result is a binary difference image in which the changes in the image sequence become visible.This method of obtaining a difference image can be described as difference subtraction. Alternatively, it is also conceivable that the camera captures the images to be compared, especially the first and second images, as binary images. The gray levels in each image can then be entered as a 1 if a certain contrast value is exceeded. The change in contrast can then be recorded in the difference image.

[0017] According to the inventive method, at least one of the following steps can be performed to adjust the headlight position: e) Comparing the change in the light distribution, in particular the difference image, with the light-dark boundary of the given light distribution. f) Calculating the correction value for the current headlight position, g) Controlling, in particular moving, the headlight according to the correction value.

[0018] Advantageously, it is possible to detect whether the position of the cut-off line of the current light distribution deviates from the position of the predetermined cut-off line of the desired light distribution. The magnitude of the deviation, as well as the vehicle's position relative to the measuring wall or surface, can provide the required information about the correction value according to which the headlight can be controlled or moved to produce the desired, specified light distribution. In the case of a movable headlight, this can involve a horizontal and a vertical offset by which the headlight can be pivoted around the horizontal and vertical axes. In the case of an LED headlight, the correction value can, according to the invention, be based on rows and columns.In both cases, according to the invention, it may be sufficient to detect a deviation from the desired light distribution as described in step e) and to provide the user with corresponding information, for example in the form of a warning signal, that the headlight needs to be readjusted. Furthermore, steps f) and g) can be carried out by trained personnel in a workshop.

[0019] In a preferred embodiment of the invention, it may be provided that the method first calculates the current headlight position, whereby at least one of the following steps may be performed: h) Comparing the change in light distribution, in particular the difference image, with predefined light-dark boundary patterns of the light distribution as a function of the headlight position, i) Determining the headlight position of the pattern with the highest agreement with the change in light distribution, in particular with the difference image, as the current headlight position.

[0020] The vehicle can be positioned perpendicular to a projection surface. According to the invention, the headlight with a low beam distribution can be controlled or moved by a vertical angle setting, while the headlight with a high beam distribution requires both horizontal and vertical control or movement to make the light-dark boundary visible using the difference image. During movement, image sequences can then be recorded and stored by the camera. The motion detection approach used is advantageously based on continuous difference calculation with conversion to binary images via the threshold value. Using a downstream pattern recognition routine and the theoretically calculated target difference area, the current headlight position can be determined based on the difference images, according to the invention.An alternative approach involves comparing a predefined light-dark boundary of a low-beam distribution with the characteristic rise of the gusset against the difference image. The sample position with the highest agreement with the difference image ultimately determines the current headlight position.

[0021] Then, at least one of the following steps can be performed to adjust the headlights: j) Calculating the deviation between the current and the desired headlight position, k) Controlling, in particular moving, the headlight to compensate for the deviation from the desired headlight position.

[0022] Once the current headlight position has been determined, the deviation between the current and desired headlight positions can be calculated. This deviation can also be described as a correction value, which can comprise two components: one vertical and one horizontal. To produce the desired light distribution, the headlight must now be controlled, specifically moved, according to this correction value.

[0023] In summary, the invention provides a fast, robust method for adjusting the headlight, which has several further advantages, such as: • Determination of the horizontal and vertical position of the headlight, independent of the nature of the projection surface. Any wall or the road surface in front of the vehicle can be used as the projection surface. Objects with complex surfaces, such as the rear of a vehicle, can also be used for evaluation. • Insensitivity of the difference images to disturbances such as external lighting, other light edges like side stripes, surface structure and ambient brightness. • Universal application of the method in movable swivel modules and in rigid LED matrix beam systems. • Time and computational efficiency through the simple implementation when evaluating the differences between successive images without complex algorithms for determining the light-dark boundary as in conventional methods.

[0024] According to the invention, the features described in the invention's method can be combined in a wide variety of ways. Further measures improving the invention are described in more detail below, along with a description of preferred embodiments of the invention, with reference to the figures. The figures show: Fig. 1. Capturing a difference image of a low beam distribution, and Fig. 2. Capturing a difference image of a high beam distribution,

[0025] Fig. Figure 1 shows three images: Figure 1 (top left), depicting the light distribution at time t1; Figure 2 (bottom left), depicting the light distribution at time t2; and a difference image (d) on the right. Figures 1 and 2 show a typical profile of the light-dark boundary of a low-beam headlight distribution that a motor vehicle headlight can produce on a vertical measuring wall. A light-dark boundary forms between an illuminated and a dark area on the measuring wall. The profile of the light-dark boundary is formed by a horizontal line, a line rising to the right (in right-hand traffic) or to the left (in left-hand traffic), and another horizontal line that connects to each other. The profile of the light-dark boundary can advantageously be clearly shown in the difference image (d).According to the inventive method, the vehicle can first be positioned perpendicular to a projection surface in order to project images 1 and 2 onto the projection surface. To capture the low beam distribution, the headlight can be controlled vertically. During the movement, a camera can record and store images 1 and 2. To determine the change in the light distribution, a difference image d is calculated between image 2 and the previous image 1. Using a downstream pattern recognition routine, the headlight position can be determined according to the invention based on the binary difference image d. According to the inventive method, the difference image d can be compared with a series of predefined low beam patterns, each corresponding to a series of headlight positions.The headlight position of the pattern with the highest agreement with the difference image d can advantageously determine the current headlight position.

[0026] Fig. Figure 2 also shows three images: Figure 1 (top left), which depicts the light distribution at time t1; Figure 2 (bottom left), which depicts the light distribution at time t2; and a difference image (d) on the right. Figures 1 and 2 each show a typical progression of a light-dark boundary in a long-range distribution, formed by a horizontal and a vertical line intersecting at a point. In the case of the high-beam distribution, the headlight produces a so-called horizontal light-dark boundary, as shown on the right in Figure 2. Fig. Figure 2 shows the position of the headlight with a horizontal cut-off line. Therefore, determining the headlight's position with a horizontal cut-off line requires vertical and horizontal control of the headlight so that both sides are visible in the difference image d. To determine the change, the difference image d between the second image 2 and the previous image 1 is calculated. Using a subsequent pattern recognition routine, the headlight position can also be determined here based on the binary difference images d.

[0027] According to the two Fig. 1 and Fig. 2. The following steps are performed to record the change in light distribution: a) Capturing a first image 1 at a first time t1, b) Controlling or moving the headlight, c) Capturing a second image 2 at a second time t2, d) Calculating a difference image d between the first image 1 and the second image 2.

[0028] According to one embodiment of the invention, steps a) and c) are performed simultaneously with step b). The camera takes images sequentially while the headlight is controlled at the same time. This enables not only a rapid execution of the inventive method but also a quick correction of an incorrect headlight position. The difference image d on the right in the Fig. 1 and Fig. Image 2 is the subtraction of image 2 from image 1, in which only the areas that have changed between images 1 and 2 are visible. Advantageously, static image information is not displayed in the difference image d, so that the difference image d can be displayed more sharply than a single image. According to the invention, the difference image d yields the light-dark boundary, which can be displayed more clearly and is easier to recognize than in conventional methods.

[0029] According to a preferred embodiment of the method, the current headlight position is first calculated, whereby the following steps are performed: h) Comparing the difference image d with predefined patterns of the light-dark boundary of the light distribution as a function of the headlight position, i) Determining the headlight position of the pattern with the highest agreement with the difference image d as the current headlight position.

[0030] The following steps are then carried out according to the headlight adjustment procedure: j) Calculating the deviation between the current and the desired headlight position, k) Controlling, in particular moving, the headlight to compensate for the deviation from the desired headlight position.

[0031] It should be obvious to a person skilled in the art that the invention is not limited to the preferred embodiments described above. Rather, many different light distributions can be analyzed using the method according to the invention. Furthermore, it can be provided that several difference images are analyzed to further improve the detection of the light-dark boundary. It is also conceivable that various other methods of difference image generation and correction value calculation can be applied. To implement the method according to the invention and to control the headlight, suitable software can be uploaded directly to the camera, the existing light control system, or the central control unit. Alternatively or additionally, a separate control unit can be provided for executing the method and / or controlling the headlight.Furthermore, the inventive method may be not limited to the headlight systems of a motor vehicle but may also be applied in other different lighting systems. The camera may also be located outside the vehicle – in particular, data exchange via a mobile camera may also take place. Reference symbol list 1 first image 2 second picture d difference image

Claims

Method for adjusting the headlights of a motor vehicle with a headlight designed to generate a vehicle-specific light distribution and a camera used to detect the light distribution of the headlight, characterized in that the camera detects the change in the light distribution due to the control, in particular the movement, of the headlight in order to detect the light-dark boundary of the light distribution, wherein the light distribution is captured as a binary image to detect the change in the light distribution, wherein the change in the light distribution is detected when a threshold value is exceeded in order to filter out camera noise and lighting interference effects. Method according to claim 1, characterized in that, in order to detect the change in the light distribution, the headlight is vertically controlled, in particular moved, when the light distribution to be detected is a low beam distribution. Method according to one of the preceding claims, characterized in that, in order to detect the change in the light distribution, the headlight is controlled vertically and horizontally, in particular moved, if the light distribution to be detected is a glare-free light distribution. Method according to one of the aforementioned claims, characterized in that, in order to detect the change in the light distribution, the camera records and / or stores temporally successive image sequences. Method according to one of the preceding claims, characterized in that at least one of the following steps is carried out to detect the change in the light distribution: a) capturing a first image (1) at a first time (t1), b) controlling or moving the headlight, c) capturing a second image (2) at a second time (t2), d) calculating a difference image (d) between the first image (1) and the second image (2). Method according to one of the preceding claims, characterized in that at least one of the following steps is carried out for headlight adjustment: e) Comparing the change in the light distribution, in particular the difference image (d), with the light-dark boundary of the specified light distribution, f) Calculating the correction value for the current headlight position, g) Controlling, in particular moving, the headlight according to the correction value. Method according to one of the preceding claims, characterized in that at least one of the following steps is carried out to determine the current headlight position: h) Comparing the change in the light distribution, in particular the difference image (d), with predetermined patterns of the light-dark boundary of the light distribution as a function of the headlight position, i) Determining the headlight position of the pattern with the highest agreement with the change in the light distribution, in particular with the difference image (d), as the current headlight position.

Images