DE102009054231A1 - Head-up-display for information indication in motor vehicle, comprises indicating unit, which has display unit, in which lighting points are produced - Google Patents

Abstract

The head-up-display comprises an indicating unit, which has a display unit, in which lighting points are produced, and a deflection unit, which is arranged between the display unit and a windshield in the path of rays of the head-up-display. An imaging unit is provided for producing an information indication on a detail screen for an eye of the passenger.

Description

The invention relates to a head-up display for displaying information in a motor vehicle.

Modern motor vehicles are increasingly being equipped with so-called head-up displays, which allow optical information to be reflected in the windshield of the respective motor vehicle. As a rule, the driver "sees" this information as hovering outside the motor vehicle or just ahead of the engine hood, which eliminates the accommodation that arises, for example, from changing viewing of display units in the vehicle interior and the roadway.

From the document DE 103 02 387 A1 a display device for a motor vehicle is known with means for changing the parallax, which arises when viewing the display by the distance of the eyes of the observer from each other. Also, a so-called contact-analog information display, so for example, the harmonious from the perspective of the driver augmenting superimposition of a real image of the vehicle environment with additional information shown by the head-up display is in the DE 103 02 387 A1 treated.

Also, electronic methods for eliminating image distortion in a conventional head-up display are known in the art, such as from the documents DE 100 46 019 A1 or WO 2007/000178 A1 ,

It is an object of the invention to provide an improved head-up display.

This object is achieved by a head-up display according to claim 1. Advantageous embodiments and modifications of the invention will become apparent from the dependent claims.

For a clear description of the invention, reference is made to a technical teaching relating to a novel head-up display for stereoscopic information display in a motor vehicle. The invention complements particularly advantageous with such a stereo head-up display. However, not all technical features of such a stereo head-up display are required to implement the invention. In particular, the invention is also applicable to a mono-head-up display. For a better understanding of the invention is still closer to the novel stereo head-up display (and various embodiments and developments of the same), to the invention then in particular with reference to the similarities and differences to the technical teaching, which this novel stereo Head-up display underlying, describe.

The technical teaching on which the novel stereo head-up display is based is referred to below as reference theory. In order to prevent misunderstandings, it should be mentioned at this point that to the knowledge of the applicant this reference teaching does not belong to the state of the art. Rather, the reference theory is pursued in a parallel patent application by the applicant.

Inset REFERENCE:

The reference teaching relates to a head-up display for stereoscopic information display in a motor vehicle, comprising the following components: display means with a display unit on which luminous points can be generated, the light rays are visible by reflection on the windshield of the motor vehicle for an occupant of the motor vehicle; Deflection means, which are arranged in the beam path of the head-up display between the display unit and the windshield and deflect light beams of different luminous dots in different emission directions; Image generating means for generating two different frames of a stereoscopic information display intended for the two eyes of the occupant; Pixel allocation means for assigning pixels of the individual images to luminous points of the display unit. In this case, the deflection means and the display unit (in particular a display surface of this display unit) are arranged and arranged relative to one another and the pixels of the individual images can be assigned to the illuminated points of the display unit by the pixel allocation means such that the generation of the luminance points for the left eye of the occupant the pixels of the individual image intended for the left eye are visible and the pixels of the individual image intended for the right eye are visible to the right eye of the occupant.

The pixels of the individual images can thereby, but not necessarily, be arranged in a regular grid. Decisive for the success of the reference theory is that the desired individual images are decomposed into individual pixels, which can each be generated by a luminous dot of the display unit. In that regard, the pixel allocation means may also comprise image decomposition means by which a decomposition into such pixels is vornehmbar. Alternatively, the individual images can first (apart from the pixel allocation means) be decomposed by independent image decomposition means into suitable pixels, which are then assigned by the pixel allocation means to the luminous points of the display unit. The pixel allocation means and the Of course, image decomposition means can be implemented by one and the same arithmetic unit.

• • dass ortsabhängige Unterschiede des Strahlengangs ausgleichbar sind, indem die Zuordnung durch die Bildpunktzuordnungsmittel in Abhängigkeit von einer Beschaffenheit, insbesondere der Wölbung, reflektierender Flächen des Head-up-Display, insbesondere der Windschutzscheibe, vornehmbar ist; und/oder
• • dass die Bildpunktzuordnungsmittel eine Zuordnungsvorschrift für die Zuordnung von Bildpunkten der Einzelbilder zu Leuchtpunkten der Anzeigeeinheit anwenden, welche in einem Kalibrationsvorgang ermittelt worden ist, der die Ansteuerung ausgewählter Leuchtpunkte der Anzeigeeinheit und die zeitsynchronisierte kamerabasierte Erfassung der jeweils resultierenden in der Windschutzscheibe reflektierten Abbildung umfasst; und/oder
• • dass die Ablenkmittel als Matrix optischer Körper ausgebildet sind; und/oder
• • dass die Anzeigeeinheit als TFT-Display oder LCD-Display oder Plasma-Display und/oder als Durchlicht-Display ausgebildet sind; und/oder
• • dass die Zuordnung durch die Bildpunktzuordnungsmittel auf veränderlichen, insbesondere in Abhängigkeit von einer Geometrie der Windschutzscheibe und/oder von der Verbauposition der Teile des Head-Up-Displays veränderlichen, Zuordnungsparametern basiert; und/oder
• • dass die Zuordnung durch die Bildpunktzuordnungsmittel in Abhängigkeit von der Kopfposition und/oder Augenposition des Insassen variierbar ist; und/oder
• • dass durch die Bilderzeugungsmittel für einen ersten Insassen des Kraftfahrzeugs, insbesondere den Fahrer, und einen zweiten Insassen des Kraftfahrzeugs, insbesondere den Beifahrer, zumindest jeweils ein Einzelbild erzeugbar ist, und dass die Ablenkmittel und die Anzeigeeinheit derart beschaffen und zueinander angeordnet sind und die Bildpunkte der Einzelbilder den Leuchtpunkten der Anzeigeeinheit derart durch die Bildpunktzuordnungsmittel zuordbar sind, dass für den ersten Insassen die Bildpunkte des für den ersten Insassen bestimmten Einzelbildes sichtbar sind und dass für den zweiten Insassen die Bildpunkte des für den zweiten Insassen bestimmten Einzelbildes sichtbar sind; und/oder
• • dass zur Vermeidung von Doppelbildern die Einzelbilder erzeugbar sind, indem auf Basis eines als Einzelbild darzustellenden Nutzbildes zumindest ein zu erwartendes Störbild bestimmt wird, indem dem darzustellenden Nutzbild ein im Wesentlichen helligkeitskonstantes Ausleuchtungsniveau hinzugefügt wird und indem das zumindest eine Störbild von dem Nutzbild bzw. von dem hinzugefügten Ausleuchtungsniveau bzw. von der Summe aus Nutzbild und hinzugefügtem Ausleuchtungsniveau im Wesentlichen abgezogen wird; und/oder
• • dass durch die Anzeigemittel und/oder durch zusätzliche Kompensationsmittel Kompensationslichtstrahlen erzeugbar sind, die zumindest teilweise gegenphasig kohärent zu den für die Darstellung eines Nutzbilds erzeugten Lichtstrahlen der Leuchtpunkte der Anzeigeeinheit sind, dass auf Basis des Nutzbilds zumindest ein zu erwartendes Störbild bestimmbar ist und dass das Störbild zumindest teilweise durch die Kompensationslichtstrahlen kompensierbar ist.

Advantageous embodiments and further developments of the reference teaching (which are pursued as dependent claims in the parallel patent application to the reference teaching) relate to the design of such a stereo head-up display in such a way that

• • that location-dependent differences of the beam path are compensated by the assignment by the Bildpunktzuordnungsmittel depending on a nature, in particular the curvature, reflective surfaces of the head-up display, in particular the windshield, vornehmbar; and or
• • that the pixel allocation means apply an assignment rule for the assignment of pixels of the individual images to display points that has been determined in a calibration process, which comprises the activation of selected display pixel and the time-synchronized camera-based detection of the respective resulting image reflected in the windscreen; and or
• • that the deflection means are formed as a matrix of optical bodies; and or
• • that the display unit as a TFT display or LCD display or plasma display and / or as a transmitted light display are formed; and or
• The assignment by the pixel allocation means is based on variable assignment parameters which vary, in particular, depending on a geometry of the windshield and / or on the installation position of the parts of the head-up display; and or
• The assignment by the pixel allocation means being variable in dependence on the head position and / or eye position of the occupant; and or
• • that by the image generating means for a first occupant of the motor vehicle, in particular the driver, and a second occupant of the motor vehicle, in particular the passenger, at least one individual image can be generated, and that the deflection means and the display unit arranged and arranged to each other and the pixels the individual images are assignable to the illuminated points of the display unit by the pixel allocation means such that the pixels of the individual image intended for the first occupant are visible to the first occupant and the pixels of the individual image intended for the second occupant are visible to the second occupant; and or
• • that to avoid double images, the individual images can be generated by at least one expected image is determined on the basis of an image to be displayed as a single image by the image to be displayed a substantially constant brightness level is added and by the at least one image from the Nutzbild or of is essentially subtracted from the added illumination level or from the sum of the utility image and the added illumination level; and or
• Compensation light beams can be generated by the display means and / or by additional compensating means which are at least partially coherent in phase opposition to the light beams of the display unit's luminous points generated for the representation of a useful image, that on the basis of the useful image at least one predicted glitch can be determined and that the Image is at least partially compensated by the compensation light beams.

The display unit can be designed substantially planar. In particular, it can be used as a substantially planar display, e.g. B. TFT, LCD or plasma display be designed. Alternatively, the display unit may be configured as a unit for outputting one or more deflectable light beams, in particular laser beams.

Due to the head-up display according to reference theory, a stereoscopic information display that is unadulterated by these conditions is made possible under the difficult physical conditions of the head-up display principle.

The display unit can in particular be designed as a self-illuminating display unit, for example as a TFT display or LCD display or plasma display. However, it can also be designed as a transmitted-light display (eg DLP module). The display unit may substantially correspond to that of a conventional head-up display.

The deflection means are preferably formed as a matrix or matrix-like arrangement of optical bodies. The optical bodies may in particular be small lenses and / or small plastic bodies, which in particular have a semicircular section. In particular, these can be configured as so-called cylindrical lenses. The matrix or matrix-like arrangement is preferably arranged essentially in one plane, hereinafter referred to as distribution plane. The optical bodies may be substantially similar and preferably identical with respect to Distribution level be arranged what z. B. is advantageous for a cost-effective production of the deflection. In particular, the optical bodies can be configured rotationally symmetrical with respect to a vertical of the distribution plane. Preferably, the columns of the matrix are arranged substantially equidistantly. Preferably, the rows of the matrix are arranged substantially equidistantly.

Alternatively, the optical bodies may be unevenly or regionally distributed unevenly. This may be particularly advantageous, since for different areas of the windshield on average a different levels of deflection of the beams and a different resolution is necessary. The optical bodies, which are mainly used for the generation of images in different directions, may also have different shapes.

Advantageously, the optical bodies of the deflection means may be connected to the display unit and / or to one another and / or to a common translucent bottom part. This can be z. B. can be realized by being cost-effectively pressed by means of a mold into one unit. Thus, the deflection can be made particularly inexpensive.

The distribution plane is preferably substantially parallel to a plane in which the substantially planar display unit extends. The matrix or matrix-like arrangement of optical bodies preferably lies at a short distance above a display surface composed of luminous points of the display unit.

In this case, the optical bodies are preferably dimensioned and arranged relative to the display unit such that the light beams of a plurality of luminous spots are deflected in different directions by a single optical body.

The image generating means generate - in the simplest case in a manner known per se - two different individual images of a stereoscopic information display intended for the two eyes of the occupant. However, the eyes of the inmate reach the frames only after a reflection on the windshield. The information can z. As symbols, numeric displays, objects and object constellations, which should be perceived by the occupant as being spatially.

Particular attention should be paid to the role of the pixel allocation means in the reference theory. They ensure the assignment of pixels of the individual images to luminous points of the display unit. This assignment can be understood as a kind of coding in which the physical conditions of the head-up display are taken into account and reversed in a certain sense. For example, a single image intended for the viewer's left eye is not emitted directly by contiguous or adjacent luminous dots of the display unit, but rather those luminous points of the display unit are driven whose emitted light is influenced, in particular, by various optical elements of the head-up display Distraction by the deflection means and reflection on several mirrors as well as the windshield - a coherent from the perspective of the left eye of the observer result image. The coding made by the pixel allocation means is effectively unbundled from the subsequent physical-optical mapping.

In this case, preferably, location-dependent differences of the beam path are compensated by the assignment by the pixel allocation means as a function of a nature (in particular of the local curvature) of reflective surfaces (in particular of the windshield or any further optical body, eg a crooked mirror) of the head-up Displays is made.

An assignment rule to be applied by the pixel allocation means can for example be calculated on the basis of geometric models and physical-optical mapping laws, if all geometrical conditions are known.

Alternatively, an assignment rule can be determined in a calibration process. Also a mixed form, d. H. An assignment rule that is based initially on geometric considerations and refined in a calibration process is conceivable.

In the simplest case, the following procedure can be used for calibration purposes: All the luminous dots of the display unit are lit individually (or selected locally representative luminous dots for a group of the luminous dots) are lit in a specific sequence. A camera, which is located in a typical position of an eye of the driver, is synchronized with the drive of the display unit and detects the resulting image reflected in the windshield. By evaluating the recorded images, it is then possible to easily determine which luminous points are visible in the recorded images and at which position, ie at which pixel, they appear in the respective recorded image. In addition, a measure of the focusing of the resulting image, for. B. based on the angle at which the light from each luminance points arrives at the camera, are measured. Thus, z. For example, it can be ensured that sufficient or insufficient focusing is present for a part of the image, in particular with respect to a specific observation position.

The inversion of the thus obtained association between luminous dots and pixels represents a simple assignment rule for the pixel allocation means. Such a calibration procedure can be carried out successively or (eg when using two cameras) simultaneously for the typical positions of both eyes of the driver.

By using prior knowledge of the geometrical-optical conditions of the overall arrangement, a process for obtaining an assignment rule can be defined which is simpler or less complicated than the calibration procedure described above. For example, the overall arrangement will generally be designed so that due to the geometric relationships of this overall arrangement for certain pixels in the individual image generated for a particular eye only the luminous points of a group of luminous points adjacent to the display unit or display surface "come into question". Of course, then only the luminous points of this group must be checked for their suitability for generating the corresponding pixel in the individual image generated for this eye. In addition, due to the spatial arrangement with respect to an overlying optical body, it can be ruled out for certain luminous points, for example, that its light appears in the image taken on behalf of a specific eye. Such luminous points then need not be checked for this eye.

By generating entire patterns and / or test images (instead of individual luminous luminous dots), the calibration process can be significantly accelerated. By using prior knowledge of the geometrical-optical conditions of the overall arrangement and / or the exact structure of the pattern / test image, an association with the lighting points causing the exposure is possible even with simultaneous or non-temporally resolved exposure of several pixels.

Advantageously, a plurality of points or groups of points can be provided with different temporal behavior, for. B. pulsed or encoded with a specific pulse pattern. Thus, the z. B. detected with a camera light from different pixels are distinguished from each other. Thus, the calibration process can be extremely accelerated because the measurement of the radiation angle for several or all pixels can be done virtually simultaneously.

A calibration method can also run in reverse, for example on the basis of a search step method: A camera records an image which is compared with a desired individual image. Depending on the result of the comparison, the assignment rule is changed step by step until there is a minimum of the deviation between the recorded image and the desired individual image.

Also by the use of interpolation methods, a process for obtaining an assignment rule can be defined which is simpler or less complicated than the calibration procedure described above. For example, a calibration process may only provide vertices for an assignment rule, while intermediate points are determined by spline interpolation or another interpolation method.

Preferably, the assignment by the pixel allocation means is based on variable, in particular depending on a geometry of the windshield and / or installation tolerances in the respective vehicle variable, assignment parameters. Thus, similar systems can be used in different vehicle models and / or despite manufacturing tolerances in different vehicles of the same vehicle model. The assignment parameters can then be determined as part of a model-specific or vehicle-specific calibration. If there is sufficient model knowledge of all geometric conditions, the assignment parameters can also be calculated on the basis of geometric models and physical-optical mapping laws.

The assignment by the pixel allocation means is preferably variable in dependence on the head position and / or eye position of the occupant. Thus, a tolerance of the display against head and eye movements of the viewer can be achieved.

The head position and / or the eye position of the occupant is preferably detected in such a manner in a conventional manner.

For a head position and / or eye position that has been changed compared to the normal position, one can determine complete complete assignment rules or a complete set of assignment parameters-in calibration procedures or model-based. The acquisition of assignment rules or assignment parameters can again be limited to selected interpolation points, while For intermediate positions, an interpolation method is used.

Instead of a separate complete assignment rule or a complete set of assignment parameters for compared to the normal position changed head positions and / or eye positions also (assignment) change rules can be defined, according to which an assignment rule in case of changes in the head position and / or eye position relative to the normal position is to be changed in order to achieve a tolerant to head and eye movements of the viewer representation.

Ideally, from the viewer's point of view, an unaffected representation of head and / or eye movements of the observer (as long as they remain within certain limits) is achieved.

According to a development of the reference theory, the suitability of the head-up display according to reference teaching to represent different images for different viewing angles can also be used to supply different occupants, in particular drivers and front-seat passengers, with different information displays.

The presentation to the second occupant, usually this will be the passenger, it can - as well as the representation for the first occupant - also done stereoscopically. A total of four individual images are then generated by the head-up display and displayed for the respectively suitable viewing angle: one image for each eye of the first occupant and one image for each eye of the second occupant.

However, the representation at the second occupant can also be monoscopic for the sake of simplicity. There are then generated by the head-up display a total of three frames and displayed for each suitable angle: one image for each eye of the first occupant and a third image for the second occupant.

If only a single image is generated for the passenger, it must be ensured that the deflection means used are designed to have such large "opening angles", especially in the lateral emission area, that light points whose light is deflected in the direction of the passenger for both eyes of the passenger are visible.

If two separate images are generated for the passenger, on the other hand, care must be taken to ensure that the deflection means used are designed to have such small "opening angles" in the lateral emission area that luminous points whose light is deflected in the direction of the passenger are only ever are visible to an eye of the passenger.

The head-up display according to reference teaching also allows adjustment of the subjectively perceived image distance in which the resulting image appears. In addition, in the case of two different occupants, this can be set separately for each of you.

According to two other - alternatively or additionally applicable - developments of the reference theory parasitic double images or multiple images are compensated in different ways. According to a first approach to compensate for such parasitic effects, a constant (ambient) level of illumination is added to a payload (actually displayed). This is to be understood as an image with a substantially constant brightness in the entire image area, in a sense a brightness offset. The illumination level is preferably chosen so that its brightness corresponds approximately to the maximum brightness of the parasitic effects to be compensated. If it is now possible to predict the parasitic effects on the basis of the useful image, the level of illumination or the sum of the useful image and the level of illumination can be reduced by precisely these parasitic effects. As a result, the useful image is superimposed with a kind of negative image of the parasitic double images or multiple images. Thus, although a (generally undesirable) (ambient) level of illumination is added to the payload. However, the parasitic double images or multiple images are compensated and thus invisible to the viewer.

In accordance with a second approach to compensate for such parasitic effects, a coherent light source generates anti-phase light which selectively eliminates the double images.

In the following, a preferred embodiment of the reference teaching will be described with reference to the accompanying drawings. This results in further details, preferred embodiments and refinements of the reference theory. In detail, show schematically

1 the principle of a deflection of light rays in different directions by means of a matrix of optical bodies (top: cross section, bottom: top view),

2 the principle of applying a deflection of light beams in different directions for a stereoscopic information display,

3 a first embodiment variant for deflecting the light beams of individual luminous dots of a display unit in different directions,

4 a second embodiment variant for deflecting the light beams of individual luminous dots of a display unit in different directions,

5 a schematic diagram for correcting the beam path of individual pixels in a stereoscopic display by means of a head-up display in a motor vehicle.

• – Anaglyphenverfahren: Die „Verteilung” unterschiedlicher Bilder (aus unterschiedlichen Perspektiven) auf die beiden Augen eines Betrachters basiert dabei auf einer Farbkodierung und nachfolgender Dekodierung mittels einer Rot-Grün-Brille.
• – Verfahren auf Basis einer Shutterbrille: Die „Verteilung” unterschiedlicher Bilder auf die beiden Augen eines Betrachters basiert dabei auf deren zeitlicher Abfolge im Bildstrom und dem mit dieser Abfolge synchronisierten Verschlussmechanismus der einzelnen Brillengläser:
• – Holografieverfahren: Die „Verteilung” unterschiedlicher Bilder auf die beiden Augen eines Betrachters basiert dabei auf Interferenz- und Defraktionseffekten, die mittels kohärenter Lichtquellen erzeugt werden.
• – Lentikularverfahren: Die „Verteilung” unterschiedlicher Bilder auf die beiden Augen eines Betrachters basiert dabei auf optischer bzw. räumlicher Kodierung der Information und ihrer Dekodierung mittels zylindrischer Linsen.

Methods for stereography or stereoscopic photography are well known in the art. A satisfactory implementation of such methods in the field of application of head-up displays in motor vehicles, however, is not known. Known methods for stereography or stereoscopic photography are, for example

• - Anaglyph method: The "distribution" of different images (from different perspectives) on the two eyes of a viewer based on a color coding and subsequent decoding using a red-green glasses.
• - Methods based on a shutter glasses: The "distribution" of different images on the two eyes of a viewer based on their temporal sequence in the image stream and synchronized with this sequence shutter mechanism of the individual lenses:
• - Holographic method: The "distribution" of different images on the two eyes of a viewer based on interference and Defraktionseffekten, which are generated by means of coherent light sources.
• Lenticular method: The "distribution" of different images onto the two eyes of a viewer is based on optical or spatial coding of the information and its decoding by means of cylindrical lenses.

Lenticular procedures are described in more detail below: The application of lenticular procedures for so-called "wobbly images" and / or stereography is known. Lenticular procedures are suitable for achieving a stereo effect without aids that "load" the viewer (such as special goggles).

The image coded in a specific manner is typically decoded in known lenticular processes by means of a series of semi-cylindrical lenses which are lined up in a very high density (40-90 pieces per inch). As a rule, these are produced together with the "underbody" of transparent plastic and processed into a flexible film / layer.

The principle on which lenticular stereography is based, namely the principle of deflecting light rays in different directions by means of a matrix of optical bodies, shows 1 , The upper part of 1 shows a cross section of the arrangement used, the lower part of a plan view. A stripe picture 1 is composed of a plurality of individual images (which can not be explained in more detail here, as is known per se), which show an object or a scene from different perspectives. The parts of the frames are in such a way in strips 4a . 4b . 4c . 4d decomposes the parts to be viewed from a given horizontal viewing angle to a specific band. The (at the upper part of 1 extending perpendicular to the plane) semi-cylindrical lenses 2 , direct the pictures of the individual strips 4a . 4b . 4c . 4d in the form of light rays 3a . 3b . 3c . 3d in different directions. Thus, the content of the image presented to a viewer depends on the direction in which the entity is viewed.

In stereoscopic lenticular imaging, the image information is "sorted" onto the strips in different parts of the lenticular image such that the viewer's right and left eyes are each presented with an image in the corresponding perspective on the scene or object. Thus, the viewer sees the scene represented spatially. There are representations of objects in different depths generated, d. H. objects that are virtually behind the surface of the stripe image and objects that virtually stand out from the surface of the stripe image, d. H. be seen in front of the picture plane. When changing the viewing angle, z. B. by head movement and / or eye movement, several pairs of images can be seen in succession.

2 illustrates the addressed "sorting" of the light beams for a stereoscopic information display. If the observer's head is in a position (not specifically shown) P1, then his left eye sees the image strips from which the light rays L1 emanate, his right eye sees the image strips from which the light rays R1 emanate. If the observer's head is in a position P2, his left eye sees the image strips from which the light rays L2 emanate, his right eye sees the image strips from which the light rays R2 emanate, etc.

In today's known head-up displays (HUD) in motor vehicles, the driver's impression that a virtual object represented by means of the HUD hovers in front of the driver at a distance of about 2 to 2.5 meters is generated by the fact that the image of a display with multiple mirrors "folded" (ie mirrored several times). Often, highly precise and costly curved mirrors are used. The functional principle of today's head-up displays is not suitable for displaying objects or images that are intended to cover or cover large parts of the windshield as far as the observer is concerned, since this would require an extremely complex and complex optical system. Currently, head-up displays in motor vehicles in particular allow the display of pictograms and symbols, which are perceived by the viewer as "suspended in the air" at a certain virtual distance. Head-up displays for displaying three-dimensional objects and / or for simultaneously displaying a plurality of planar objects in different depth levels are not available on the market. A disadvantage of some marketable head-up displays is a disturbing double image effect, caused by parasitic reflections. Such parasitic reflections arise in particular when using normal glass panes as a combiner of the head-up display. They are currently being eliminated, if at all, by a very elaborate and costly method, namely the provision of a wedge-shaped course of an intermediate layer incorporated into the windshield.

According to the embodiment of the reference teaching presented here, a head-up image is generated in a motor vehicle with a matrix of optical bodies. The optical bodies break the light generated by certain light-emitting elements or luminous dots in such directions of radiation that these rays after reflection on the windshield yield the views of the scene to be imaged from more than one perspective. In other words, the optical bodies represent deflection means which are arranged in the beam path of the head-up display between a display unit, which is essentially considered here as a matrix of luminous dots, and the windshield, and the optical bodies deflect light beams of different luminous spots in different directions from; The optical body and the display unit are arranged and arranged to one another and on the display unit such image information is generated that for the left eye of a vehicle occupant a particular image intended for the left eye is visible and that for the right eye of the vehicle occupant for the right Eye specific frame is visible.

To effect this, the scene to be imaged must be appropriately encoded into an image to be output from the display unit. This must be done in such a way that the distribution of the luminous intensity and color for each luminous element or each luminous point of the display unit is configured as a function of the radiation angle to be achieved.

Thus, in the case of the head-up display presented here by way of example, a projection of a map visible to at least one vehicle occupant takes place, taking into account a matrix of optical bodies which direct the beams from different luminous points into different emission directions assigned to them. In this case, the scene to be projected is converted into the image to be output by the display unit, so to speak encoded such that the resulting after the reflection on the windshield beam angle each point of the scene for the viewer in a - corresponding to the three-dimensionality of the scene - depth appear ,

In contrast to market-available 3D displays, the particular optical conditions of a head-up display, in particular the reflection on the windshield, are taken into account in the method explained here by way of example with reference to an example. A core idea of the reference theory is therefore not only to be seen in producing a stereoscopic image or video as such, but in creating a stereoscopic reflection on the windshield according to the real vehicle conditions - as well as the nature of the windshield. For this purpose, on the one hand, a set of emission angles is coded between the emission directions in such a way that the desired stereoscopic image results for the user after reflection on the windscreen. On the other hand, the special feature is taken into account that in a conventional 3D display, the beam angle difference for the right and left eye is always constant for a specific imaging depth, which is not the case with reflection on a curved or arched windshield.

The term coding of the image here stands for the determination of the luminous values of the luminous dots of the display unit for a resulting image to be displayed. This coding can be configured by an assignment of the picture elements to the corresponding required emission angles and determination of the luminous points of the display unit to be actuated so that the required beam angles result by means of the beam deflection in the deflection means. In the wording of the claims, the coding consists in the assignment of pixels of the (the individual eyes to be displayed in the result) individual images to luminous points of the display unit.

The depth at which each point of a scene is seen by the driver, for example, depends on the angle between the rays entering the right and left eyes of the driver (triangulation).

The presented method is suitable both as - in particular cost-effective and space saving variant of a conventional head-up display as well as representing multiple virtual objects located far from the driver. In this case, the number of different depth levels - in contrast to a full-fledged 3D display - may also be kept low in favor of a simple implementation.

Two preferred embodiments of the method according to reference theory are shown below.

design variant 1 ( 3 ) provides for the use of a (self-luminous) TFT display or LCD display or plasma display as a display unit, design variant 2 ( 4 ) the use of a transmitted-light display.

3 illustrates the embodiment variant 1 , The pixels serve as luminous points 11a . 11b . 11c . 11d a TFT, LCD or plasma display. The optical body 12 are here designed as hemispheres made of plastic, which are arranged in a regular grid. The optical bodies make sure that for each luminous point or each of the pixels 11a . 11b . 11c . 11d another emission direction of the light rays 13a . 13b . 13c . 13d results, which later impact on the (not specifically shown here) windshield.

4 illustrates the embodiment variant 2 , As luminous dots serve here the pixels or segments 21a . 21b . 21c . 21d a transmitted-light display. The segments can change the respectively transmitted proportion of light and / or its color pixel by pixel. The resulting transmitted light display can be configured as a DLP module, which allows the generation of high light levels in the head-up display, ie the use of the device even in the bright day for images with many color or gray levels (high dynamics). The optical body 22 (Substantially identical to the optical bodies of embodiment variant 1 ) are also configured here as hemispheres made of plastic, which are arranged in a regular grid. The optical bodies make sure that for each luminous point or each of the segments 21a . 21b . 21c . 21d another emission direction of the light rays 23a . 23b . 23c . 23d results, which later impact on the (not specifically shown here) windshield. The transmitted light display is backlit with uniform, preferably parallel, focused light 24 from a light source 25 , Optionally, a lens 26 or be preceded by a lens system and / or other optical elements.

The matrix of optical bodies 12 out 3 (as well as the matrix of optical bodies 22 out 4 ) is arranged in a plane in the present example. The optical body 12 . 22 are made similarly and are arranged identically with respect to this plane.

They are rotationally symmetric with respect to a vertical of this distribution plane. The columns of the matrix are arranged equidistant here, as are the rows of the matrix. The plane with the optical bodies is parallel to the plane of the display and the optical bodies are at a small distance above the pixels of the display. In this case, the light beams of several pixels are deflected in different directions by a single optical body.

The method or the device may be configured (in both embodiments) such that each optical body has a group of 4 (= 2 × 2) or 9 (= 3 × 3) or 16 (= 4 × 4) or 25 (= 5 × 5) or 36 (= 6 × 6) or 49 (= 7 × 7) or 64 (= 8 × 8) or 81 (= 9 × 9) or 100 (= 10 × 10) light pixels and / or transmitted light pixels Thus, for these pixels, the coding of the image to the radiation of different image information in different directions allowed.

Advantageously, due to the specific application situation in the HUD in the motor vehicle, in which the requirements for the resolution in the direction transverse to the direction of travel in terms of image coding are significantly higher than vertical - that of an optical body 12 . 22 Covered pixel groups be designed as rectangles with very different edge lengths. Ie. There are also asymmetric configurations, eg. Coverage of a group of 18 (= 6x3) pixels is possible and may be advantageous.

The optical body 12 . 22 Optionally, they can also be designed as cylindrical lenses and / or microlenses or microprisms etc. In any case, at least in one dimension, the translation of the position of a point should be guaranteed in a beam angle.

Most of today's marketable head-up displays compensate for distortion of the image due to a high-quality curved mirror, which is very costly. This is still practicable with relatively small imaging surfaces of the HUD, but for large imaging surfaces this approach appears impractical and unreasonably expensive. The correction of installation tolerances and focusing are achieved by very complex mechanical adjustment of optical elements.

To otherwise distort a distortion of the image presented to the viewer To avoid this, it is proposed not only to generate the stereo base for the image when coding the image, but also to electronically compensate the influence of the curvature of the windshield on the focusing. Thus, influences of differently built or differently curved windshields on the focusing of large-format content can be corrected much easier than z. B. by further optical body and their adjustment would be possible.

In particular, the location-dependent difference of the beam path due to the nature of reflective surfaces, in particular the windshield, at least partially offset by the coding of the image.

The method thus created or a device thus created makes it possible to represent each part of the image not only with a specific brightness and color, but also to "sort" several partial images according to radiation angles. Therefore, it is also possible to compensate for the location-dependent difference in the beam path on the windshield, by means of the consideration of a corresponding inverse function in the coding (image processing) of the scene to be displayed (see. 5 ). The inverse function or an assignment rule which takes into account this inverse function can be calculated on the basis of geometric models and physical-optical mapping laws, provided all relevant geometric conditions are known. Alternatively, it can be determined in a calibration process. It is also conceivable to use a hybrid form, ie an assignment rule which is initially based on geometrical considerations and is refined in a calibration process, in particular with the help of individual selected image regions.

5 shows an example of a correction of the beam path by incorporating corrective factors in the coding of the image. A (here highly schematic) projection element 1 projects rays of light onto the windshield 2 of the motor vehicle. In the uncorrected case, the radiation angles of the light rays would be both absolute and relative by the curvature of the windshield 2 falsified (see ray bundle 3 in 5 ) that the viewer would present a blurred and / or distorted image that does not appear in the desired position and / or orientation. The ray pairs 3L1 and 3L2 represent example rays intended for the left eye of the observer; 3R1 and 3R2 represent example rays intended for the right eye of the viewer. In the case corrected as proposed, on the other hand, the corrected beam path results in a displayed object or image being presented to the user both in a sharp and undistorted manner and in the predetermined position and orientation. It is both the representation of multiple objects in different depths as well as the representation of a "full-fledged" 3D image, which - in human perception - allows a nearly continuous transition between the depth effects of individual image areas possible. The ray pairs 4L1 and 4L2 represent example rays intended for the left eye of the observer; 4R1 and 4R2 represent example rays intended for the right eye of the viewer. The Rays 5L1 . 5L2 . 5R1 and 5R2 in 5 illustrate the virtual continuation of the light rays striking the eyes of the beholder. A displayed object or object part appears from the perspective of the viewer in a virtual position, for example, about 2.5 meters away from the eyes of the beholder.

The method presented here differs from known methods for electronic equalization, z. As a camera image, in particular, that the correction function is not only the position of a point on a display, but alternatively or additionally refers to the encoding of the emission angle, which is of course not the same.

The method presented here will in particular gain additional relevance if the expected functional growth in the field of head-up display displays will in future arouse the desire or demand for larger display areas of head-up displays, eg. B. for driver assistance systems with a full-surface and / or contact-analog information display on the entire windshield of a motor vehicle. Since it would be neither practicable nor affordable to provide a correspondingly curved mirror of the required size and quality, even with regard to local differences in curvature, the method described above is much simpler to implement for large areas. Suitable projection and / or deflection means can, for example, be "rolled out" along the dashboard or at least integrated with a low penetration depth into the dashboard.

In today known or marketable HUDs often a very complex calibration is performed, which is to certain optical parts or the actual HUD module, align such that manufacturing tolerances of the vehicle and the windshield are compensated and positioned as desired, sharp picture is reached for the viewer.

Preferably, in the proposed method by means of a suitable adaptation, in particular automatic determination, Software parameters of the image encoding carried out a calibration to the respective individual vehicle, by which manufacturing tolerances of the vehicle and the windscreen are compensated. Thus, identical or similar devices of the proposed type (ie, the same product) for different vehicle models with different windshield curvature apply, with only certain (possibly all) parameters of the angle encoding vehicle-individually adjusted by software. The calibration of the device in a vehicle preferably takes place by means of an automatic determination of the coding parameters. Such coding parameters may in particular be parameters for the distribution of the brightness and / or color on the pixels of the device which are required to make a scene to be displayed and / or an object appear after reflection on the windshield at the predetermined depth and sharpness of image ,

According to a preferred embodiment of the described method or the described device, the described type of compensation of optical properties of the windshield is carried out on the basis of a look-up table which contains the required change in the beam angle for certain reference points in the image.

These points can only be representative of a specific area of the image. For the determination of the angle change for points which are located between the interpolation points, an interpolation method, for example a spline interpolation, can then be used. Such a look-up table can be changed in the context of the above-described calibration procedure itself and / or the differences resulting from a fixed reference table can be stored separately. Such handling of the encoding values reduces the required computational power and memory size of a corresponding rendering system.

As already discussed, it is advantageous for certain driver assistance systems, in particular those based on the display of contact-analogue information, if the HUD used has the largest possible effective display area. In such driver assistance systems, in the sense of an augmentation of the real image of the vehicle environment, certain symbols can be superimposed into the HUD, which correspond closely with the real world, for example, in the HUD. B. navigation arrows of a navigation system can then be displayed such that they match the real lanes of the road and / or obstacles in the real world can be "marked". As already discussed, however, a mere areal enlargement of a HUD according to the current state of the art is difficult to realize and affordable. Moreover, the mere areal magnification of a prior art HUD would have the disadvantage that the position of a display generated by the HUD in relation to the real world would depend on the driver's eye position, i. H. an indication generated independently of the observer's head / eye position would migrate from the point of view of the real world and / or the correction of mirror distortion on the windshield would only work satisfactorily for certain head / eye positions.

In order to prevent this, it is known to permanently adapt the image represented by the HUD to the position of the eyes or the head of the observer.

As a development of the method according to the reference theory, it is proposed to adapt the beam path of the HUD permanently to the position of the user's eyes accordingly. In other words, d. H. in the claim text, the assignment is varied by the pixel allocation means in dependence on the head position and / or eye position of the occupant. By means of the head-up display according to reference theory, this can be done in such a way that the angular relationship between the beams remains constant or changes according to predetermined criteria. The position of the eyes of the observer, in particular of the driver, can be detected by means of an interior camera in an eye-tracking method or it can be adapted in the simplified case, the beam path depending on the seating position and / or personal settings of the user. This can be carried out particularly advantageous and simple in the method according to reference theory because the relevant parameters can be adjusted very quickly by software. In particular, the image coding can be adapted to the current head position of the observer by - if such a table is used - the values of the above look-up table or another interpolation table are adapted accordingly.

Another advantage of such a refinement of the reference theory relates to the case of the presentation of information to a plurality of observers, in particular driver and front passenger. The images directed towards the driver and front passenger can thus be corrected separately from one another, ie separately for each observer and depending on their respective current situation and sitting position. It should be noted that a mere shift of the image content in the representation of a 3D scene is not sufficient for this purpose; required is alternatively or additionally rather an angle change. This is based on the above-described adaptation of the coding simply feasible by performing the adaptation separately for each viewer.

In the case of information presentation to a plurality of viewers, in particular driver and front passenger, the same view can be presented to these multiple viewers in the simplest case. However, a head-up display of the type presented also makes it possible to represent different image contents for different radiation directions. Thus, different image content can also be presented to different users, i. H. from different observation positions in the motor vehicle, different image contents can be seen.

Thus, for example, a display form could be achieved in which the - predominantly the driving task dedicates - driver relevant for this driving task navigational arrows on the windscreen, while the passenger the displays of an onboard computer, a travel guide or the graphics output of a video game - also format filling - be displayed on the windshield. The image content for different viewers can therefore be completely different; Of course, they can also be of similar type, but with viewer-specific additions / options.

A designed according to the proposed method imaging device can be made very flat, see. 1 , The adaptation of the device a specific vehicle model and its calibration, for example, to compensate for manufacturing and installation tolerances, can be done by software. Therefore, the use of the method is also and especially in retrofitting a vehicle with a HUD of the proposed kind of advantage. Original equipment or retrofitting can be carried out particularly easily if the vehicle to be upgraded or upgraded has a windshield which prevents the formation of parasitic double or multiple images in a manner known per se by means of a specially shaped PVB interlayer (butyrol-polyvinyl).

If parasitic double or multiple images can not or only insufficiently be avoided by the structure of the windshield, according to a preferred development of the reference theory, mirror effects on the windscreen are suppressed by projecting the projection surface within the useful radiation angle, in other words the angular range in which for illustration purposes is illuminated with an offset light intensity. In places where - without further measures - a double image would arise, this can be deducted from the image to be projected mathematically, in particular in an image preprocessing, d. H. be subtracted. A double image on the windshield is thus at least partially compensated according to this preferred development of the reference theory by adding a largely constant illumination level to a displayed payload image and subtracting the precalculated double image by calculation from the result. Analogously, it is possible to proceed with other interference images, in particular multiple images and / or multiply reflected parasitic images.

Such a refinement of the reference theory can also be used together with the above-mentioned - based on the profile of the windshield - known optical methods for avoiding parasitic double or multiple images.

The electronic compensation must then "only" refer to remaining parasitic double or multiple images remaining after optical compensation (and still relevant to human perception).

Since the double image is slightly differently curved compared to the payload, the compensating subtraction of the pre-calculated double image is advantageously carried out depending on the emission angle.

When carrying out the reference teaching according to the above development, the overall image produced would also "outside" the Nutzbildes easily - with the brightness of the constant illumination levels - "shine", which can be particularly disturbing at night.

To avoid this, you can proceed as follows. The luminous dots are designed by means of one or more coherent light sources. Thus, the value of the illumination at a certain point in principle - as is known, for example, from holography - be neutralized with antiphase coherent light.

Using this principle, in the case of using coherent light sources, such as laser diodes, clutter images can be canceled out by means of a phase shifted image, i. H. be selectively subtracted. Offset lighting is not required.

A practical implementation is already technically and economically feasible using a single laser light source, for example according to the arrangement 2 , The required phase shift can then be generated by passing a portion of the radiation of the light source along a longer beam path or through a medium with a changed optical density is directed (which leads to a reduction of the speed of light within the medium).

End of the module REFERENCE.

The invention follows on from the above reference teaching, or in many (but not necessarily all) details is consistent with this reference teaching. A combination of the invention with the reference teaching or with the individual embodiments and / or further developments of the reference teaching described above appears to be particularly advantageous.

The invention provides a head-up display for displaying information in a motor vehicle, which has display means with a display unit on which luminous points can be generated whose light beams are visible by reflection on the windshield of the motor vehicle for an occupant of the motor vehicle, which has deflection means, which are arranged in the beam path of the head-up display between the display unit and the windshield and deflect light beams of different luminous dots in different radiation directions, which image generating means for generating at least one specific to an eye of the occupant frame of an information display and which Bildpunktzuordnungsmittel comprises for the assignment of pixels of the single image to luminous points of the display unit. According to the invention, the deflection means and the display unit are arranged and arranged relative to one another and the pixels of the individual images can be assigned to the luminous points of the display unit by the pixel association means such that the light beams of a plurality of different luminous points impinging on the retina of the occupant's eye are substantially superimposed at a point of impact.

In particular, the head-up display (or its components and their relative arrangement) can be configured such that light beams of different luminous spots, which are mirrored at different points of the windshield, substantially coincide with their impact on the retina of the occupant's eye superimpose a point of impact.

If the plurality of different luminous points, whose light beams are superimposed, are driven in a substantially similar manner, then an information display which can be intensified by the occupant can be achieved. If the different luminous dots are driven in different ways, in particular in different colors, an improved color representation can alternatively or additionally be achieved. In both said cases, the invention is able to increase the display quality of the information display by the head-up display.

The statements made above for the reference teaching on the preferred design and arrangement of the deflection means apply equally to the invention. In particular, the deflection means may be formed as a matrix of optical bodies and / or arranged in a plane. The other remarks made above regarding the reference teaching on the preferred representational design of the head-up display apply equally to the invention.

Particular attention may be paid in the case of the invention to the image decomposing means - possibly increased in relation to the reference theory. The image decomposition means can be understood as standalone or as part of the image allocation means (as in the invention). The image decomposition into pixels in the invention necessarily takes place differently than is customary in the prior art, namely superimposed (that is to say at the same position in the individual image) pixels.

In the simplest case, the pixels into which the at least one single image is decomposed are distributed to all raster points of a regular two-dimensional raster. In once again simplest case of this simplest case meet the same number of pixels on each of the halftone dots, z. B. three pixels. For example, in such an instance, the frame may be decomposed into a red image, a green image, and a blue image, all having the same spatial pixel count and pixel distribution. Thus, at each grid point, a red pixel, a green pixel and a blue pixel are superimposed. However, in more complex cases, the pixels into which the frame is decomposed may be distributed irregularly and / or varying numbers of pixels may overlap. The individual image can also be decomposed in such a way that only a part of the pixels overlaps with other pixels, while the other part of the pixels each have a specific position within the individual image as the only pixel. Thus, for example, the display quality of certain selected areas within the frame can be improved by superposition of pixels.

The superimposing of the light rays of a plurality of luminous dots in a point of impact of the retina may initially appear to be unrealisable (in particular with the aid of the common simplified optical models of the human eye, eg a perforated camera model), can be achieved with a correspondingly exact modeling of the human eye or of the human visual system, however, explain and implement. First and foremost, it can be explained by the fact that the pupil and the The lens of the human eye is modeled differently than often simplistic - not just punctiform (for example, the human eye lens is about 10-14 mm in size).

In order to obtain the facts necessary for the implementation of the invention, a modeling of the human eye can be carried out using known physiological and anatomical facts and / or existing models can be used. However, model knowledge can also be obtained empirically in test series or in a calibration method, for example using an artificial objective eye model (eg with glass lens and partially transparent plastic retina). This will be discussed in more detail later.

In view of the concatenation of difficult to mathematically modelable track parts (a) head-up display and (b) human eye, it appears to be particularly advantageous to determine in test series on which points of impact of the retina the light beams of the individual luminous dots of the display unit are imaged (according to their Deflection by the deflection means and after their processing by several optical means, eg mirrors, in the head-up display and after their entry into the eye). From the reversal of this relationship it can be deduced which luminous points of the display unit-with components of the head-up display otherwise considered to be given and unchangeable-have to be controlled in which way to produce a specific image on the retina.

The invention is based on the premise that the number and density or spatial resolution of the luminous dots of the display unit is so high (or the spatial resolution of human visual perception is so low) that (when using deflecting means that are not too strong in their entirety) in each case light beams of a plurality of illuminated points of the display unit are deflected onto a point of incidence (which can be resolved in human visual perception). The boundary conditions imposed by physiological and possibly anatomical prerequisites of the human visual system with respect to, for example, the required accuracy with regard to angle of entry and offset can be maintained by a device according to the invention based on components which are currently available on the market, in particular display components. The pixel sizes of common displays today are about 0.25 mm (even lower for special displays). The producible angular resolution is thus sufficient for implementing the invention.

In view of the above-described physical feasibility of superimposing the light beams in each case a plurality of luminous points can be in the perception of the occupant create a single image by composition of pixels, each of which are arranged at the same position within the perceived single image, but which are generated by different luminous points.

As already explained above, in the invention image decomposition means decomposes a single image to be displayed into pixels, each of which is arranged at the same position within the individual image. The invention proves to be particularly advantageous when a skilled image decomposition of a single image to be displayed to the occupant is made.

If the decomposition takes place, for example, in such a way that substantially uniform, in particular the same color, pixels result as a result of the decomposition (at one and the same position within the individual image), this (compared to a single image composed of non-overlapping or non-overlapping pixels ) a brightness gain of the frame can be achieved.

If the decomposition of colored individual images takes place in such a way that different monochrome pixels result as a result of the decomposition (at one and the same position within the individual image), then a reduction in expenditure with regard to the generation of color images can be achieved. In this case, monochrome luminous dots (eg of an LED display) would suffice for generating the pixels of colored individual images.

If the decomposition of colored individual images with high color depth such that result as the result of the decomposition (at one and the same position within the frame) different pixels with low color depth, this can also be achieved a reduction in the cost of generating color images.

Since the success or the feasibility of the invention is primarily ensured by the relationship between the emission position on the display unit and the emission angle assigned by the pixel allocation means, movements of the eye do not preclude the feasibility of the invention. The said relationship can also be observed during movements (at least with slight movements) of the eye. For larger displacements of the eye or head or observation position, which can be detected metrologically in a conventional manner, an adjustment of the pixel-luminous point assignment can be made depending on this eye or head or observation position.

The exact point of impact at which a light beam incident on the human eye strikes the retina depends inter alia also on the current focusing state of the eye, in particular on the muscularly variable curvature of the eye lens. Preferably, the pixel assignment by the pixel allocation means is executable such that for different focusing states of the human eye, a sharp image on the retina is achieved, ie that ensures for different focusing states that the image-like positionally identical pixels are assigned to luminous points whose light rays are incident upon essentially overlaying the retina at a point of impact. In this case, the image decomposition is preferably independent of the focusing state, while the pixel assignment is adapted to the variable optical properties (changed lens curvature) of the human eye. The variable optical properties are preferably taken into account in an optionally used eye model or in a possibly used combined optical model for head-up display and eye.

The invention makes it possible to focus the individual images perceivable by the occupant without optical focusing (which is generally to be carried out according to the prior art and usually extremely complex, especially in the case of large-area, eg contact-analogue head-up displays). Preferably, in the case of an image display both to a driver and to a front passenger, different focusses are set for the driver and for the front passenger and / or their different seat or head positions.

In a preferred refinement of the method according to the invention, the assignment of pixels of the individual image of the display unit made by the image association means is carried out as a function of the detected visual course of the occupant. This can in particular be done in such a way that the current focus state is assumed as a function of the distance of the object or of the objects to which the view of the occupant is directed or z. B. in the last seconds, was directed.

The focusing or the pixel assignment contributing to the focusing can also be location-dependent, d. H. be performed separately for different areas of the defined field of view. Ie. It is preferably for different parts of the image frame of the image, an adaptation of the assignment to different focusing states of the eye vornehmbar. In particular, two or more different parts of the image are preferably simultaneously displayable in the head-up display, which focus or focussed on different distances when the occupant's eye is focused or accomodated. Thus, for example, two different image parts, in particular displays of driver assistance systems with different meanings, eg. B. in relation to the near field or far field, are displayed in the head-up display, of which a first image part in the perception of the occupant in accommodation to the near area is sharply displayed and a second image part in the perception of the occupant in accommodation in the long-range is shown sharply. This results in great advantages, in particular in augmented reality applications, in which different objects in the vehicle environment, which are at different distances from the vehicle, are to be marked by insertions into the head-up display. Therefore, the distance of each object to be marked from the vehicle is preferably determined metrologically by the vehicle for the focused insertion of such markings. In other words, the head-up display preferably has the ability to at least two objects in the vehicle environment, the distance between the object and a reference point on the motor vehicle can be detected metrologically, to augment at the same time depending on the detected distance focusing.

According to a preferred embodiment of the invention, a visual impairment (in particular short-sightedness with a known diopter power) of the occupant is taken into account in the pixel assignment or in an underlying eye model, and the pixel assignment is made as a function of the visual impairment of the occupant. However, this aspect of the invention can contribute nothing or no solution to improve the view of a visually impaired in the traffic.

As can already be seen from the reference theory, a single image can be generated for the two eyes of the occupant. In particular, different individual images can be generated for the two eyes of the occupant in order to create a stereoscopic impression. Accordingly, a focusing, and in particular a visual impairment, can be performed separately for each eye.

It can also, as also already performed in the reference teaching, individual images are presented to several vehicle occupants. Here, too, a separate focusing and, if appropriate, visual impairment can be carried out in each case.

The focusing or the pixel assignment by the pixel allocation means contributing to the focusing may also depend on be adapted to a metrologically detected line of sight of the occupant. Depending on whether the occupant directs the view of the near field or lets it wander into the distance, the focus can be adjusted accordingly.

According to one embodiment of the invention, a targeted steering of the attention of the occupant can be made by intentionally or intentionally changed the assignment made by the pixel mapping means of pixels to the luminous points of the display unit, that at least temporarily impaired focus in the visual perception of the Inmates (in fact a deteriorated focus of the retinal image) results. Preferably, such an attention steering is made when in the vehicle environment, an object (or other matter) is detected, to which the attention of the occupant should be directed and which is at such a distance from the motor vehicle that for his sharp, d , H. Focused, visual perception by the occupant a change in the current focus state of the occupant's eye appears necessary. To detect such a need to change the focus state of the occupant's eye, appropriate detection means are preferably provided. In other words, the head-up display preferably comprises on-demand detection means and / or is connected to such demand-recognition means by which a need to change the focus state of the occupant's eye is recognizable, and the assignment of image-frame pixels to luminance points by the pixel-allocation means the display unit is variable in response to the detection of such need to effect steering of the attention of the occupant. For example, the demand detection means may consist essentially in an image processing environment analysis device for the motor vehicle and signal a demand when a danger object is detected at a distance from the vehicle that is in the focus state of the eye that is in the current mapping of pixels to light spots is expected to be out of focus.

If, for example, information is first displayed in the head-up display in such a way that it is clearly perceptible when the occupant's eye is focused on the near area, and is the current focus state of the eye tuned to this information representation (of which - insofar as the information representation of the occupant even takes into account In view of the "autofocus" characteristics of the human eye, it is generally assumed that a transition to a representation perceptible when the occupant's eye is focused on the far field will cause the occupant's eyes to open to accomodate the far range. In the present example, therefore, the focusing of the eyes of the occupant and thus the attention of the occupant are deliberately directed into the distance. Such attention control can be done by sudden or gradual "de-focusing". Preferably, the association is changed in one or more steps toward an association that correlates with a focus state of the eye that is used to sharply perceive the object or other subject matter to which the attention of the occupant is to be directed. However, otherwise by a temporary or permanent defocusing the "auto-focus" function of the human eye can be stimulated and thus the likelihood and / or the speed of focusing on a relevant in the current situation distance can be increased.

Although already mentioned above, at this point it will once again be explained in more detail about a possible method for obtaining an assignment rule from pixels to luminous points:
As already mentioned, model knowledge can be obtained in particular empirically in test series or in a calibration method, for example using an artificial objective eye model (eg with glass lens and partially transparent plastic retina).

A simple calibration method for a head-up display according to the invention can proceed quite similarly to the calibration method described in the reference teaching. In the simplest case, the following procedure can be used for calibration purposes: All the luminous dots of the display unit are lit individually (or selected locally representative luminous dots for a group of the luminous dots) are lit in a specific sequence. An artificial objective eye model is placed in a typical position of the driver's eye. The eye model includes a glass lens and a semi-transparent plastic retina so that the light rays striking the artificial retina can be picked up by a camera located behind this retina. Alternatively, a camera chip can be arranged directly on the retina of the artificial eye. The camera is synchronized with the control of the display unit and detects the resulting on the retina from the activation of the luminous or the points. By evaluating the images taken by the camera, it can then be easily determined which luminous points are visible in the recorded images and at which position, ie on which Pixel, they appear in the respective captured image. The inversion of the thus obtained association between luminous dots and pixels represents a simple assignment rule for the pixel allocation means. Such a calibration procedure can be carried out successively or (eg when using two cameras) simultaneously for the typical positions of both eyes of the driver.

Such a calibration process can also be carried out successively for different focusing states of the eye. In this case, for example, different objective eye models can be used, which each embody a focusing state. If only an objective eye model is used, then this should embody a "medium" or "average" focus state.

Possible simplifications of the calibration method can be deduced analogously from the remarks on the reference theory.

The assignment rule obtained in the calibration process is preferably stored in a look-up table and used by the pixel allocation means.

The idea of the invention includes a vehicle with the head-up display according to the invention, a method for displaying information on which the head-up display according to the invention is based and the described method for calibrating the head-up display.

The following is based on the attached drawing 6 a preferred embodiment of the invention described. This results in further details, preferred embodiments and further developments of the invention. It shows

6 the beam path in a head-up display according to the invention.

The beam path in a head-up display according to the invention is in 6 highly schematized and exemplified.

The head-up display off 6 is used for information display in a motor vehicle and has a display unit and deflection means, each in the training already described on. The display unit and the deflection means arranged directly in the display unit in the present example are shown in FIG 6 for the sake of simplicity, as a single block 62 shown. The light beams emitted by the display unit and deflected by the deflection means - for the sake of clarity, shows 6 only four exemplarily selected light rays 65a . 65b . 66a . 66b - be on a (deliberately exaggeratedly curved) windshield 61 of the motor vehicle and thus reach the eye of a vehicle occupant.

The latter eye is in 6 very simplified represented by a (deliberately exaggerated) front part 63 especially the eye lens 63a includes, and a behind arranged (also deliberately exaggerated large) retina 64 ,

6 illustrates the fact that light rays 65a and 66a , which come from very different, here even relatively far away from each other luminescent points of the display unit, after their reflection at (different places) of the windshield 61 and their entry into the eye 63 . 64 at the same point Pa of the retina 64 can hit. The beam path of other light rays 65b and 66b , which also come from very different, relatively far apart luminous points of the display unit, ends after their reflection at (different locations) of the windshield 61 and their entry into the eye 63 . 64 likewise at the same point of impact of the retina, here the point Pb.

There are in 6 Thus, the deflection means and the display unit arranged and arranged to each other, that superimpose the incident on the retina of the eye of the occupant light rays of several different luminous points substantially in a point of impact.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• DE 10302387 A1 [0003, 0003]
• DE 10046019 A1 [0004]
• WO 2007/000178 A1 [0004]

Claims (10)

Head-up display for displaying information in a motor vehicle, comprising - Display means with a display unit on which luminous points can be generated, whose light rays are visible by reflection on the windshield of the motor vehicle for an occupant of the motor vehicle, Deflection means, which are arranged in the beam path of the head-up display between the display unit and the windshield and deflect light beams of different luminous dots in different emission directions, - Image generating means for generating at least one intended for an eye of the occupant frame of an information display and Pixel assignment means for assigning pixels of the single image to luminous dots of the display unit, wherein the deflection means and the display unit are arranged and arranged relative to one another and the pixels of the individual images are assignable to the luminous spots of the display unit by the pixel association means such that the light rays of a plurality of different luminous points incident on the retina of the occupant's eye are substantially superimposed at a point of impact. Head-up display according to claim 1, characterized in that the head-up display comprises image decomposing means by which the at least one frame is separable into pixels, each of which are arranged at the same position within the frame. Head-up display according to claim 1 or 2, characterized in that the assignment made by the pixel assignment means of pixels of the individual image to luminous points of the display unit in dependence on a focusing state of the eye is variable. Head-up display according to claim 3, characterized in that for different image parts of the frame, an adaptation of the assignment to different focusing states of the eye is vornehmbar. Head-up display according to one of claims 3 or 4, characterized in that at the same time two or more different image parts, in particular displays of driver assistance systems with different meanings, in the head-up display can be displayed in the focusing or accommodation of the eye Inmates focused on different distances appear. Head-up display according to one of claims 3 to 5, characterized in that for at least two objects in the vehicle environment, the distance between the object and a reference point on the motor vehicle can be detected metrologically and that the at least two objects simultaneously by the head-up Display can be augmented in focusing on the detected distance dependent. Head-up display according to one of Claims 1 to 6, characterized in that the assignment of pixels of the individual image to luminous points of the display unit made by the pixel-allocating means is adaptable in dependence on a visual impairment of the occupant. Head-up display according to one of Claims 1 to 7, characterized in that the assignment of pixels of the individual image to luminous points of the display unit made by the pixel association means is adaptable as a function of a metrologically detected viewing direction of the occupant. Head-up display according to one of claims 1 to 8, characterized in that the head-up display has demand recognition means and / or is connected to such demand recognition means by which a need for changing the focus state of the occupant's eye can be recognized, in that the assignment made by the pixel allocation means of pixels of the single image to luminous points of the display unit is variable in response to the detection of such a need to effect a steering of the attention of the occupant. Head-up display according to one of claims 1 to 9, characterized in that the pixel allocation means apply an assignment rule for the assignment of pixels of the individual images to luminous points of the display unit, which has been determined in a calibration process, the control of selected luminous points of the display unit and includes the time-synchronized camera-based detection of the respective resulting reflected in the windshield image.

Images