A waveguide type HUD image brightness uniformity improvement method

Abstract

This invention relates to a method for improving the uniformity of image brightness in a waveguide-type HUD. The method is implemented without altering the optical structure. First, a preset uniform test image is displayed under normal HUD operation to obtain the virtual image brightness distribution within the eye box area, resulting in a brightness matrix representing the brightness response at each location. Second, based on the matrix, the deviation between the target brightness and the measured brightness is calculated, generating a position-dependent brightness compensation gain map or lookup table. The gain map uses single-channel encoding to carry the compensation coefficients for each pixel or sub-region. Third, the gain map is deployed in an image processing unit, and during the pixel coloring stage, the RGB values ​​of the input image are adjusted pixel-by-pixel in the linear brightness domain, so that areas with low brightness receive gain and areas with high brightness are moderately attenuated, thereby making the virtual image brightness within the eye box area more uniform. Finally, the compensated frame image is output and sent to the display end to form a far-field virtual image. This method can significantly improve the brightness uniformity within the field of view of a waveguide HUD.

Description

Technical Field

[0001] This invention belongs to the field of head-up display (HUD) technology, and specifically relates to a method for improving image brightness uniformity suitable for waveguide-type HUD systems. Background Technology

[0002] Waveguide-type HUDs utilize optical waveguides to couple and transmit images emitted from an image source to the driver's field of vision (eyebox). Their display quality is easily affected by the waveguide's pupil expansion transmission characteristics, resulting in uneven brightness distribution. Maintaining consistent brightness across the entire image within the waveguide HUD's field of view (FOV) and eyebox is often difficult, becoming a key issue in AR display technology. The causes of uneven brightness include gradual light loss during multiple pupil expansion transmissions within the waveguide and uneven efficiency of the coupling devices. Therefore, precise control of the coupling efficiency of the waveguide's input and output coupling structures is necessary to ensure consistent brightness as much as possible. Even with complex optical design and optimized coupler parameters, the HUD virtual image may still exhibit localized areas that are either too dark or too bright. Thus, improving HUD brightness uniformity without significantly reducing optical efficiency is currently the main technical challenge in this field. Summary of the Invention

[0003] Purpose of the invention: The purpose of this invention is to address the problem of uneven brightness within the field of view caused by waveguide transmission and pupil expansion in waveguide-type HUDs, and to provide an image processing method that can improve the uniformity of virtual image brightness without modifying the optical structure.

[0004] Technical solution: To solve the above technical problems, the following technical solution is adopted:

[0005] A method for improving the brightness uniformity of waveguide-type HUD images includes the following steps:

[0006] (1) Static brightness calibration: Under the normal display screen of the HUD system or the predetermined test image, the brightness distribution of the virtual image within the eye box is obtained by using an imaging brightness meter to obtain a brightness matrix representing the brightness response at each position;

[0007] (2) Compensation data generation: Calculate the deviation between the target brightness and the measured brightness based on the brightness matrix, and generate a position-related brightness compensation gain map or lookup table;

[0008] (3) Real-time image compensation: The gain map or lookup table is deployed in the image processing unit. During the pixel coloring stage, the RGB values ​​of the input image are adjusted pixel by pixel so that the low brightness area gains gain and the high brightness area is moderately attenuated, thereby improving the uniformity of the virtual image brightness within the eye box range.

[0009] Preferably, the static brightness calibration includes: displaying a preset uniform brightness test image, acquiring the brightness of different positions of the HUD virtual image using an imaging luminance meter, and obtaining a brightness matrix or brightness map.

[0010] Preferably, each element of the brightness matrix corresponds to the brightness value at a predetermined position in the HUD display field of view, and a correction coefficient or offset value relative to a reference brightness is calculated for each position based on the matrix.

[0011] Preferably, the compensation data is a grayscale compensation lookup table, which stores the grayscale value mapping relationship for each pixel of the HUD display screen, and is used to map the original grayscale value of the input image to the compensated grayscale value.

[0012] Preferably, the compensation data is a set of partitioned brightness compensation functions, which divides the HUD display area into multiple sub-regions, each sub-region corresponding to a brightness compensation function, and calculates the compensation coefficient and adjusts the pixel value based on the pixel coordinates.

[0013] Preferably, the compensation data is deployed in the image processing unit, and the RGB values ​​of each pixel in each frame of the image are adjusted in real time through pixel coloring or pixel-level lookup operations.

[0014] Preferably, the static brightness calibration and compensation processing are performed independently for the three color channels of red, green and blue, generating compensation data for each channel and simultaneously correcting brightness and color uniformity.

[0015] Preferably, the method is implemented entirely in the image processing unit through software algorithms, without requiring any modification to the optical components or waveguide structure of the HUD.

[0016] The present invention also provides a waveguide-type HUD head-up display system, comprising:

[0017] Microdisplay imaging light source, optical waveguide assembly, and image processing unit;

[0018] The image processing unit includes an image brightness compensation module, which is used to execute the image brightness uniformity improvement method according to any one of claims 1 to 8, and to perform pixel-level brightness compensation on the image signal from the imaging light source.

[0019] Furthermore, the image processing unit is a graphics processor or a digital signal processor, and the image brightness compensation module stores a brightness compensation lookup table or compensation function generated based on static calibration, and is configured to adjust the pixel signals of each frame of the image in real time.

[0020] Beneficial effects:

[0021] This invention introduces an image compensation algorithm unit into the HUD system. First, it performs static brightness calibration on the HUD display in its initial state to obtain brightness uniformity data for each pixel observed at the eye position. In subsequent design, a set of brightness compensation maps is generated based on this calibration result. This brightness compensation map is deployed in the HUD's image compensation algorithm unit to adjust and compensate the pixel-level RGB values ​​of the real-time image signal stream. This enhances the image in areas with low brightness and moderately reduces the image in areas with excessive brightness, ultimately improving the uniformity of the brightness distribution across the entire virtual image.

[0022] The method of this invention does not rely on changes to external optical structures, but achieves brightness uniformity improvement entirely through software algorithms. All algorithms are implemented within the HUD image processing unit. Existing waveguide HUD optical structures can be directly used with the software compensation module of this invention without altering the waveguide HUD's optical design or redesigning the grating distribution, thus lowering the barrier to achieving improved brightness uniformity. By statically calibrating the brightness within the display area, brightness deviation data for each sub-region is obtained, and the RGB input value of each pixel is adjusted accordingly. This ensures uniform and consistent virtual image brightness across the entire HUD field of view; areas that were originally too dark are brightened, while overly bright areas are appropriately dimmed, significantly improving image uniformity. The compensation algorithm runs in real-time in image processing units such as GPUs, adjusting brightness values ​​pixel-by-pixel for each frame, ensuring that the correction process has no significant delay impact on the HUD display. Leveraging the parallel processing capabilities of the image processing unit, dynamic brightness uniformity compensation is achieved at high frame rates, ensuring that the driver always sees a uniform and clear HUD image.

[0023] Furthermore, the method of this invention has good scalability. The above calibration and compensation process can be performed independently for the three color channels: red (R), green (G), and blue (B), generating independent compensation data (LUT or function) for each channel. During real-time processing, a channel-specific compensation coefficient is applied to the R, G, and B components of each pixel, thereby improving brightness uniformity while simultaneously correcting color inhomogeneity or color shift that may be caused by waveguide dispersion or light source characteristics, thus comprehensively improving the image quality of the HUD virtual image.

[0024] In summary, this invention achieves uniform improvement in virtual image brightness without modifying the HUD optical components through static brightness calibration and pixel-level compensation. This method is applicable to waveguide-type HUD systems employing various micro-display light sources such as DLP, LCoS, and MicroLED, and has broad versatility and practical value. Attached Figure Description

[0025] Figure 1 This is a schematic diagram of the display and brightness adjustment method of the head-up display device of the present invention;

[0026] Figure 2 This is a schematic diagram of the head-up display device and its brightness adjustment method according to Embodiment 1 of the present invention;

[0027] Figure 3 This is a schematic diagram of the head-up display device and its brightness adjustment method according to Embodiment 2 of the present invention;

[0028] Annotation instructions:

[0029] 101 Image processing unit; 102 Waveguide HUD; 103 Gray-scale compensation LUT input; 104 Eye box; 105 Virtual image position; 106 Automotive windshield.

[0030] 201 GPU; 202 Waveguide HUD; 203 Luminance compensation function mapping input for zones; 204 Eyebox; 205 Virtual image position; 206 Car windshield. Detailed Implementation

[0031] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be noted that these embodiments are used to illustrate the technical solutions of the present invention and not to limit its scope. Those skilled in the art can adjust the specific structure and parameters without departing from the spirit of the present invention.

[0032] Example 1: Optimizing Brightness Uniformity Based on Gray-Scale Compensation LUT Using Static Brightness Map

[0033] This embodiment provides a scheme for generating a pixel grayscale compensation lookup table (LUT) using statically measured brightness distribution and applying it to real-time image signals in a HUD. Please refer to... Figure 1 This embodiment demonstrates the basic structure and brightness adjustment process of the head-up display device. The system includes an image processing unit 101 and a waveguide-type HUD 102. During the laboratory calibration phase, brightness measurements are performed on the waveguide HUD 102 display. The HUD projects a uniform standard grayscale image (pure white or equal grayscale), and a high-resolution imaging luminance meter is used to obtain the brightness distribution map of the HUD virtual image position 105 observed at the eye box 104 across the entire field of view. A brightness map with the same resolution as the HUD display is obtained, containing the actual brightness value of each pixel position. This brightness map can clearly reflect the non-uniformity of the brightness response at various points in the HUD virtual image, identifying overly bright or dark areas of the entire image.

[0034] Furthermore, based on the acquired luminance map data, a grayscale compensation LUT103 is calculated. Specifically, a reference area in the luminance map (e.g., the area with the highest brightness) can be selected as a baseline, and the luminance deviation coefficient of each pixel relative to this baseline is calculated. For example, if the actual brightness of a pixel in the calibrated image is only 80% of the baseline brightness, then the luminance compensation coefficient for that pixel needs to be increased to compensate for the 20% difference. The positions of all pixels are associated with their corresponding compensation coefficients to generate a two-dimensional lookup table, which stores the adjustment value of the output grayscale required for each pixel under different input grayscale levels. To simplify the calculation, it can be assumed that the luminance deviation is approximately linear with respect to the input grayscale, thus using a fixed gain / loss coefficient table for each pixel for compensation.

[0035] The generated grayscale compensation LUT 103 is burned into the image processing unit 101 of the HUD system for storage. When the waveguide HUD 102 is running, all image pixel signals output to the display screen undergo mapping adjustment by this LUT before imaging at the eye box position 104: for each pixel, an adjusted output grayscale value is calculated based on its original input grayscale value and the compensation mapping pre-stored in the LUT. When rendering each frame, the image processing unit applies this lookup table transformation to every pixel across the entire screen, thereby correcting the effects of uneven brightness in real time. Specifically, after processing in this embodiment, the brightness of the entire image displayed by the HUD virtual image 105 tends to be uniform, darker areas are significantly brightened, and overly bright areas are slightly darkened, greatly improving the brightness consistency of the HUD projection information observed by the driver through the car's windshield 106.

[0036] It should be noted that the LUT-based pixel compensation method described in this embodiment is suitable for scenarios where high-precision pixel-by-pixel brightness data is obtained during the calibration stage. Its advantages are that the compensation is fine and direct, and even subtle differences in brightness inhomogeneity can be corrected. However, in practical applications, the working environment and imaging characteristics of the HUD may change slowly (e.g., temperature changes cause light source brightness drift). Therefore, the static LUT can be updated periodically or irregularly through recalibration as needed to ensure that the compensation effect is effective in the long term.

[0037] Example 2: Frame-by-Frame Adjustment Method Based on Luminosity Matrix Partition Function Mapping

[0038] This embodiment provides another implementation method, namely, using a brightness matrix to construct a partition compensation function, which is then used by the GPU201 to adjust pixel brightness in real time according to the calculation formula during each frame of image rendering. Please refer to [link to relevant documentation]. Figure 2 This illustrates the system configuration of this embodiment. Unlike Embodiment 1, which uses a full-pixel lookup table, this solution reduces storage overhead and improves compensation flexibility by partitioning the calibration data for modeling.

[0039] First, static brightness calibration measurements are performed. In this embodiment, the waveguide HUD 202 at the virtual image position 205 of the eye box 204 is divided into several sub-regions according to the field of view, and the average brightness value of each region is recorded to obtain a lower-resolution brightness matrix. Specifically, the HUD virtual image can be divided into a 196×108 matrix grid according to coordinates, and an average brightness value is obtained for each grid region through measurement. The brightness matrix obtained in this way reflects the brightness distribution trend at each position. Compared with the pixel-by-pixel data volume in Embodiment 1, this matrix significantly reduces the data volume, but still retains the main characteristics of spatial brightness variation.

[0040] Next, a brightness compensation function mapping 203 for each partition is constructed based on the brightness matrix data. The deviation ratio of each grid point in the brightness matrix relative to the target brightness is calculated, and a continuous compensation coefficient function F(x,y) is formed in two-dimensional space through interpolation fitting. This function takes the pixel's coordinates (x, y) on the screen as the independent variable and outputs the brightness compensation coefficient at the corresponding position. Specifically, a bilinear interpolation method can be used: for any pixel within the brightness matrix grid, its compensation coefficient is obtained by weighting the compensation requirements of the four nearest calibration grid points by distance, thus ensuring a smooth transition of the compensation coefficient in space. The generated function F(x,y) can provide a pixel brightness adjustment coefficient distribution map for the entire display area.

[0041] During the operation of the waveguide HUD 202, the luminance compensation function 203 for each region is integrated into the frame-by-frame image processing flow of the GPU 201. When rendering each frame, the GPU 201 calculates the corresponding compensation coefficient F(x,y) for each pixel based on its screen coordinates and applies this coefficient to the pixel's original color value. Thus, pixels in low-brightness areas receive a gain with a coefficient greater than 1, increasing their brightness, while pixels in high-brightness areas receive a loss with a coefficient less than 1, decreasing their brightness. Since this calculation only involves weighted and multiplicative operations, the GPU 201 can efficiently perform this operation on all screen pixels in parallel, achieving real-time luminance uniformity correction frame by frame. Finally, an improved uniformity virtual image is formed in the eyebox 204 via the optical system and projected onto the car's windshield 206.

[0042] The partitioned function mapping method in this embodiment features dynamic adjustment. When the optical output characteristics of the waveguide HUD202 change (e.g., light source aging leading to a decrease in overall brightness), the change can be adapted by adjusting the parameters of the function F(x,y) or recalibrating to generate a new brightness matrix, without requiring pixel-by-pixel measurements or updating the massive lookup table data. Furthermore, this method saves storage resources while ensuring improved uniformity, as only a small number of regions' compensation parameters and calculation formulas need to be stored. This method is well-suited for use in HUD image processing firmware, executed by the GPU201, thus fully utilizing the parallel computing power of the graphics hardware to achieve real-time correction.

[0043] It should be noted that in this embodiment, if the brightness non-uniformity of the waveguide HUD includes complex small-scale variations, the partitioning function approximation may not be as precise as the pixel-by-pixel LUT. However, the compensation accuracy can be improved by increasing the resolution of the brightness matrix and using a finer grid. The solution provided in this embodiment can be flexibly adjusted as needed in practical engineering to achieve a balance between compensation accuracy and system resources.

[0044] The two embodiments above illustrate the core software compensation approach of this invention. It is worth noting that the method of this invention is implemented entirely through software algorithms in an image processing unit (such as a GPU or dedicated DSP), without requiring any physical modifications to the optical components, waveguide, or grating structure of the waveguide HUD, thereby significantly reducing the technical threshold and cost for achieving improved brightness uniformity.

Claims

1. A method for improving the brightness uniformity of waveguide-type HUD images, characterized in that, Includes the following steps: (1) Static brightness calibration: Under the normal display screen of the HUD system or the predetermined test image, the brightness distribution of the virtual image within the eye box is obtained by using an imaging brightness meter to obtain a brightness matrix representing the brightness response at each position; (2) Compensation data generation: Calculate the deviation between the target brightness and the measured brightness based on the brightness matrix, and generate a position-related brightness compensation gain map or lookup table; (3) Real-time image compensation: The gain map or lookup table is deployed in the image processing unit. During the pixel coloring stage, the RGB values ​​of the input image are adjusted pixel by pixel so that the low brightness area gains gain and the high brightness area is moderately attenuated, thereby improving the uniformity of the virtual image brightness within the eye box range.

2. The method according to claim 1, characterized in that, The static brightness calibration includes: displaying a preset uniform brightness test image, collecting the brightness of different positions of the HUD virtual image using an imaging brightness meter, and obtaining a brightness matrix or brightness map.

3. The method according to claim 2, characterized in that, Each element of the brightness matrix corresponds to the brightness value at a predetermined position in the HUD display field of view. Based on the matrix, a correction coefficient or offset value relative to the reference brightness is calculated for each position.

4. The method according to claim 1, characterized in that, The compensation data is a grayscale compensation lookup table, which stores the grayscale value mapping relationship for each pixel of the HUD display screen, and is used to map the original grayscale value of the input image to the compensated grayscale value.

5. The method according to claim 1, characterized in that, The compensation data is a set of partitioned brightness compensation functions. The HUD display area is divided into multiple sub-regions, and each sub-region corresponds to a brightness compensation function. The compensation coefficient is calculated based on the pixel coordinates and the pixel value is adjusted.

6. The method according to claim 4, characterized in that, The compensation data is deployed in the image processing unit, and the RGB values ​​of each pixel in each frame of the image are adjusted in real time through pixel coloring or pixel-level lookup operations.

7. The method according to claim 1, characterized in that, The static brightness calibration and compensation processing are performed independently for the three color channels: red, green, and blue, generating compensation data for each channel and simultaneously correcting brightness and color uniformity.

8. The method according to claim 1, characterized in that, The method is implemented entirely in the image processing unit through software algorithms, without requiring any changes to the optical components or waveguide structure of the HUD.

9. A waveguide-type HUD head-up display system, characterized in that, include: Microdisplay imaging light source, optical waveguide assembly, and image processing unit; The image processing unit includes an image brightness compensation module, which is used to execute the image brightness uniformity improvement method according to any one of claims 1 to 8, and to perform pixel-level brightness compensation on the image signal from the imaging light source.

10. The HUD system according to claim 9, characterized in that, The image processing unit is a graphics processor or a digital signal processor. The image brightness compensation module stores a brightness compensation lookup table or compensation function generated based on static calibration and is configured to adjust the pixel signals of each frame of the image in real time.

Images