Method for reducing a glare effect

Abstract

The invention relates to a method for reducing a glare effect on a driver of a vehicle, said glare effect coming from light sources located outside the vehicle, wherein a glare effect on the driver is determined based on a pupil diameter determined for the driver's eye. According to the invention, the glare effect is reduced by triggering the pupillary light reflex of the driver's eye using light emitted by an activated lighting unit in the vehicle.

Description

[0001] Mercedes-Benz Group AG

[0002] Methods for reducing glare

[0003] The invention relates to a method for reducing the glare effect of a driver of a vehicle based on light sources located outside the vehicle, wherein the glare effect of the driver is determined on the basis of a determined pupil diameter of each eye of the driver.

[0004] German patent application DE 102013 010454 B4 discloses a method for operating a driver assistance device for a motor vehicle and a driver assistance device itself. The method comprises the following steps: detecting the eyelid position and pupil size and their respective changes as glare-relevant eye parameters of a driver's eye, evaluating the glare-relevant eye parameter, deciding whether the driver is experiencing glare, and if so, preparing the vehicle for braking.

[0005] Furthermore, EP 2 583 847 A1 describes a device and a method for reducing glare from external light sources when driving a vehicle. The method comprises: darkening an area of ​​a dimmable panel positioned between the driver's eyes and the light source; capturing an image of the driver; and determining, based on the image, whether the darkened panel casts a shadow on the driver's eyes.

[0006] From JP 2018-43 724 A, a luminance control system is known that is capable of displaying a difficult-to-recognize image. From JP 2007-1417 A, a visual driving assistance device is known which includes a light source for controlling the pupil diameter, emitting light with a predetermined light intensity for the driver's eyes.

[0007] The invention is based on the objective of providing a method for reducing the glare effect experienced by a driver of a vehicle, which is based on light sources located outside the vehicle.

[0008] The problem is solved according to the invention by a method which has the features specified in claim 1.

[0009] Advantageous embodiments of the invention are the subject of the dependent claims.

[0010] A method for reducing the glare effect of a driver of a vehicle based on light sources located outside the vehicle, wherein the glare effect of the driver is determined on the basis of a determined pupil diameter of each eye of the driver, provides according to the invention that the glare effect is reduced by a pupillary light reflex of each eye of the driver initiated on the basis of emitted light from an activated lighting unit in the vehicle.

[0011] By applying this method, it is possible to reduce the glare experienced by a vehicle driver, particularly in typical glare situations during vehicle operation, such as when exiting a tunnel. This optimizes driving comfort for the driver. The method can also proactively counteract subsequent glare for the occupants when exiting a tunnel. Furthermore, the method can be used, for example, to counteract the glare from an oncoming vehicle with its high beams activated.

[0012] No additional components, such as glasses, are required for the application, i.e., the execution, of the procedure, nor is any integration of components into the vehicle necessary. Furthermore, it is not necessary for the driver to perform any action to trigger and / or execute the procedure. In one embodiment, each of the driver's eyes is scanned at regular intervals, and a corneal reflection image is extracted from the captured image data. Specifically, this corneal reflection image is extracted to detect glare, particularly future glare, experienced by the driver.

[0013] In one approach, a histogram is created based on the corneal reflection image. This histogram determines whether a bimodal distribution is present, which is used to detect glare. It is assumed that glare leads to a bimodal distribution with two distinct local maxima, particularly with respect to frequency and pixel intensity. Pixel intensity refers specifically to the brightness values ​​of pixels in the corneal reflection image, resulting from contrasts and shading. Alternatively or additionally, a gradient is determined based on the corneal reflection image. This gradient spreads uniformly across the image, and the presence of glare is detected based on this gradient. Glare is thus identified.

[0014] Another design involves recording the position and diameter of the occupant's pupil at regular intervals. Based on this recorded pupil position and diameter, it can be determined which part of the eye needs to be illuminated to generate maximum light intensity on the retina.

[0015] In one implementation, the position and diameter of the pupil are determined at regular intervals using a corneal reflection image and / or an eye-tracking algorithm. Once the position and diameter of the eye are known, at least one segment of the lighting unit is activated based on image data captured by a driver observation camera relative to a vehicle coordinate system and the lighting unit. This segment illuminates the pupil of the eye according to its detected position, in order to generate the pupillary light reflex and reduce a predicted glare effect, for example, when exiting a tunnel.

[0016] In one embodiment, the lighting unit emits light with a predetermined wavelength, particularly between 490 nm and 540 nm, and a predetermined pulse duration in the direction of the determined pupil position to achieve the maximum possible reduction in pupil diameter. The pupillary light reflex causes the pupil to constrict, thus reducing the amount of light received by the retina. This reflex is dependent on the wavelength and pulse duration. The method ensures that the vehicle's light-emitting unit generates a short pulse relative to the pupil position to constrict the pupil and thus largely prevent momentary retinal oversaturation in glare situations, such as when exiting a tunnel. The eye is preconditioned by this scenario-based method.In particular, a wavelength between 490 nm and 540 nm has proven suitable for achieving the highest reduction in pupil diameter. Therefore, a wavelength of 500 nm is set for those segments of the illumination unit that illuminate the pupil, with the light being emitted with a pulse duration of, for example, 5 s.

[0017] In a further development of the method, light with the specified wavelength and pulse duration is emitted as long as glare is present, based on a bimodal distribution determined from the corneal reflection image. This means that the pupil is illuminated as long as glare is present for the driver's eye, thereby increasing driver safety and, in particular, vehicle safety.

[0018] In one embodiment, depending on the determined pupil diameter, the light intensity is increased and / or the pulse duration is increased and / or the segment width of the illumination unit through which the light is emitted is enlarged in order to generate the pupillary light reflex and thus reduce glare.

[0019] In another possible embodiment of the method, the light is emitted by means of at least one segment of ambient lighting, which serves as the vehicle's lighting unit, to initiate the pupillary light reflex. This means that only the segment of the vehicle's ambient lighting that is positioned relative to the driver's pupil is activated in such a way as to illuminate the pupil and thus generate the pupillary light reflex. Exemplary embodiments of the invention are explained in more detail below with reference to a drawing.

[0020] This shows:

[0021] Fig. 1 schematically shows a process for reducing the glare experienced by a driver of a vehicle based on light sources located outside the vehicle.

[0022] The single figure shows the sequence of a procedure for reducing the glare effect on the driver of a vehicle based on light sources located outside the vehicle.

[0023] It is generally known that vehicle assistance systems support a driver in a variety of driving tasks. For example, adaptive cruise control assists with longitudinal control of the vehicle, an adaptive lane change assist provides lateral support, and a surround-view system can assist with driving, particularly maneuvering at low speeds and during parking. In addition, there are manual driving tasks that can be comparatively demanding for a driver. Examples of such tasks include driving under strong glare, for instance at night when an oncoming vehicle has its high beams activated, and / or exiting tunnels in daylight.Exiting a tunnel, in particular, can lead to a safety-critical driving situation. When a vehicle emerges from a relatively dark tunnel, the driver's eyes require an adaptation period to adjust to the brightness outside. This adaptation period can impair the driver's vision due to glare. This is caused by the detection of an increase in brightness on the retina and a regulatory mechanism that reduces the pupil diameter, i.e., closes the iris.

[0024] The method described below reduces glare by pre-conditioning the eye, particularly the pupil, thereby actively counteracting subsequent glare, for example, when a vehicle exits a tunnel. The method involves detecting glare and pre-adapting the pupil using an activated lighting unit, specifically the ambient lighting of a vehicle.

[0025] In a first process step S1, a driver observation camera of a vehicle is converted into a vehicle coordinate system in order to assign a segment of the ambient lighting to a driver's pupil illumination in a sixth process step S6. Thus, in the first process step S1, an extrinsic calibration of the driver observation camera with respect to the vehicle coordinate system takes place.

[0026] In a second process step S2, image data relating to the eye, particularly the pupil, is acquired using the driver observation camera. In a third process step S3, a corneal reflection image of the driver's eye is extracted from the acquired image data. Specifically, the position and diameter of the pupil are determined from the acquired image data. The position and diameter of the pupil can be determined using eye-tracking algorithms within the acquired image data from the driver observation camera. Furthermore, in the third process step S3, the corneal reflection image is post-processed, for example, to correct distortion in image data acquired with a fisheye camera.

[0027] Based on the corneal reflection image, a histogram, specifically an image histogram, is created (i.e., calculated) in a fourth process step (S4). Furthermore, the image histogram is used to determine whether a bimodal distribution is present, as it is assumed that glare leads to a biomodality.

[0028] If a bimodal distribution is detected in the fourth process step S4, process steps S2 to S4 are repeated, and the intensity of the two maxima forming the bimodal distribution is compared in the image histogram. If a pixel intensity increases in the image histogram, it is inferred that future glare for the driver will occur. Specifically, process steps S2 to S4 are repeated if the image histogram and the bimodal distribution indicate that no glare is present. In particular, in the fourth process step S4, driver glare is determined based on the image histogram, whereby the distance of the vehicle, and thus the driver, to a potential glare event decreases, causing the maxima of the bimodal distribution to change accordingly.For example, no glare is detected at a relatively large distance, whereas the glare effect increases with increasing reduction of the distance, as can be seen from the maxima of the bimodal distribution.

[0029] Glare detection based on the image histogram is determined primarily by fulfilling a number of criteria. One criterion is a bimodal distribution, such that the image histogram contains two maxima. Another criterion is that the maxima lie within a comparatively high pixel intensity range, specifically exceeding a predefined pixel value. Furthermore, a further criterion is that the frequency of a maximum increases with each determination of the bimodal distribution.

[0030] In a fifth step (S5), the pupil is extracted from the corneal reflection image, and its position and diameter are determined. Alternatively, the pupil's position and diameter can be determined using at least one eye-tracking algorithm. The pupil's position and diameter are necessary to determine which part of the eye needs to be illuminated to generate maximum intensity on the retina.

[0031] In a sixth process step S6, the segment of the vehicle's ambient lighting is then determined which, based on its position relative to the determined position of the pupil, can illuminate it maximally in order to generate the pupillary light reflex. If this geometry is unknown, the segment for maximum illumination can alternatively be determined adaptively by comparing segment activation and detection relative to the corneal reflection image.

[0032] A seventh process step, S7, involves activating the ambient lighting segment to initiate the pupillary light reflex. Light with a wavelength of, for example, 500 nm (green light) is activated for a pulse duration of, for example, 5 seconds. The ambient lighting thus emits a brief light pulse in the direction of the determined pupil position to constrict the pupil and prevent temporary oversaturation of the retina, such as when exiting a tunnel in daylight. The eye is therefore pre-conditioned.

[0033] In an eighth process step S8, the pupil diameter is determined and monitored again, whereby if it is determined that the pupil diameter does not decrease due to the light pulse towards the pupil, it is specified to increase a light intensity and / or a pulse duration and / or a segment width of the lighting unit, in particular the ambient lighting, through which the light is emitted, in order to generate the pupil light reflex, thereby preconditioning the driver's eye.

[0034] The process then jumps back to the seventh process step S7 to control the ambient lighting segment accordingly, so that the pupillary light reflex is generated.

[0035] In a ninth process step S9, an adaptive repetition of a chain of effects takes place based on the fourth process step S4 and the eighth process step S8.

[0036] The process steps S1 to S9 are repeated cyclically, thus actively counteracting subsequent glare for a vehicle driver, for example, when exiting a tunnel in daylight. The process can also be used, for example, when exiting a darkened garage, entering an urban area with correspondingly intense lighting, and when a vehicle ahead with adaptive brake lights brakes relatively hard.

Claims

Mercedes-Benz Group AG Patent claims 1. A method for reducing the glare experienced by a driver of a vehicle based on light sources located outside the vehicle, wherein the glare effect of the driver is determined based on a measured pupil diameter of each driver's eye, characterized in that the glare effect is reduced by a pupillary light reflex of each driver's eye initiated based on light emitted from an activated lighting unit in the vehicle.

2. Method according to claim 1, characterized in that each eye of the driver of the vehicle is captured at regular intervals and a corneal reflection image is extracted from the captured image data.

3. Method according to claim 2, characterized in that a histogram is created based on the corneal reflection image, by which it is determined whether a bimodal distribution is present, by means of which glare is detected and / or a gradient is determined based on the corneal reflection image, which spreads uniformly over the corneal reflection image, wherein the gradient is used to detect whether glare is present.

4. Method according to one of the preceding claims, characterized in that the position of a pupil and the pupil diameter of the occupant's eye are recorded at regular intervals.

5. Method according to claim 4, characterized in that the position of the pupil and the pupil diameter of the eye are determined at regular time intervals using a corneal reflection image and / or using an eye-tracking algorithm.

6. Method according to claim 4 or 5, characterized in that light with a predetermined wavelength and a predetermined pulse duration is emitted by means of the illumination unit in the direction of the determined position of the pupil in order to achieve a maximum possible reduction of the pupil diameter.

7. Method according to claim 6, characterized in that light with a predetermined wavelength of 490 nm to 540 nm and a predetermined pulse duration is emitted by means of the illumination unit in the direction of the determined position of the pupil in order to achieve a maximum possible reduction of the pupil diameter.

8. Method according to claim 6 or 7, characterized in that the light with the predetermined wavelength and the predetermined pulse duration is emitted as long as glare is detected based on a detected bimodal distribution and / or based on a determined gradient.

9. Method according to one of the specified claims, characterized in that, depending on the determined pupil diameter, a light intensity is increased and / or a pulse duration is increased and / or a segment width of the illumination unit through which the light is emitted is increased.

10. Method according to one of the preceding claims, characterized in that the light is emitted by means of at least one segment of ambient lighting as a lighting unit of the vehicle to initiate the pupillary light reflex.

Images