High-definition naked-eye 3D display structure and method based on lab color space

Abstract

This invention discloses a high-definition naked-eye 3D display structure and method based on the Lab color space, comprising a K-image, a color image, a backlight source, and a beam splitter in front of it. The K-image and color image are stereoscopic images synthesized by sampling multi-viewpoint sequence images with horizontal parallax. After channel extraction, they represent brightness and color information, respectively. The K-image is a grayscale image, which is converted into a 1-bit image by RIP and then output on a transparent film by laser exposure, having a high horizontal resolution. The color image can be a low-resolution image output by inkjet printing or printing, or it can be directly displayed on a display screen. The beam splitter has a horizontal parallax stereoscopic display function. The beam splitter can be configured vertically or tilted. After the K-image and color image are registered, a high-definition stereoscopic dynamic or static image is displayed under the combined action of the beam splitter and the backlight source. It is particularly suitable for stereoscopic image or video displays designed for fixed themes such as product advertising, landscapes, and portraits.

Description

Technical fields:

[0001] This invention relates to the field of glasses-free 3D display technology, and in particular to a high-definition glasses-free 3D display structure and method based on the Lab color space, which can display static and dynamic 3D images. Background technology:

[0002] In multi-view naked-eye 3D display or output methods used for public demonstrations, existing 2D display screens or inkjet printing and printing methods have low resolution and cannot obtain high-definition stereoscopic images because the number of pixels is divided into dozens of parts.

[0003] Laser-based typesetting or lithography equipment can output tiny 1-bit black and white dots. After vertical ripping, the resulting exposed film has a horizontal resolution of 2400–9600 PPI and a size of several square meters. It is currently the main method for large-format, high-precision naked-eye 3D output, but it is only suitable for displaying black and white images.

[0004] Patent application 200710079378.5 discloses a high-resolution lenticular stereoscopic image based on the RGB color space. It combines laser typesetting and additive color mixing techniques with filters to output the RGB three-color components of the ripped stereoscopic image onto the same film, which is then registered with an RGB filter to form a color stereoscopic image. This method has two drawbacks: first, the cost of RGB filters is too high compared to lenticular stereoscopic images; second, it requires high registration accuracy, and registration errors can lead to a reduction in saturation and color accuracy, thus lowering image quality.

[0005] Patent application CN202311676310 discloses a high-color-accuracy static light field display system and encoding algorithm based on the CMYK color space. It decomposes a stereoscopic image into two parts: a CMY color channel and a K channel, making it suitable only for displaying static stereoscopic images. The CMY color channel is printed onto a film substrate with ink, forming low-PPI printed pixels of different colors, with a printing accuracy greater than or equal to 400 PPI. The K channel is a black-and-white image where light intensity in different areas is controlled by laser exposure, with an exposure accuracy greater than or equal to 2400 DPI. The high-precision K image is then overlaid with the low-precision CMY image to create a high-definition color stereoscopic image. However, it is clear that no matter how a CMYK image is decomposed into K and CMY channels, the K channel cannot fully display the light intensity of the image. In addition to color information, the CMY channel also contains a considerable amount of light intensity components. In other words, many details of the image will not appear in the K channel and cannot be output with high precision; these details will appear in the CMY channel and can only be output with low precision. Another drawback is that, since both the K and CMY channels contain a lot of image details, high-precision registration is required. Registration deviations can cause the black and white and color contours of the stereoscopic image to separate, resulting in ghosting.

[0006] Therefore, it is necessary to find a simple and effective structure and method to solve the problem of missing details in high-definition image display, and to reduce the detail content of the image in the color channel while maintaining color accuracy, thereby reducing the requirements for registration accuracy. Summary of the Invention:

[0007] The purpose of this invention is to provide a high-definition glasses-free 3D display structure and method based on the Lab color space, to solve the problems of high cost and difficult registration in existing high-definition color glasses-free 3D displays. To achieve the above objective, the specific technical solution of the high-definition glasses-free 3D display structure and method based on the Lab color space of this invention is as follows:

[0008] A high-definition naked-eye 3D display method based on the Lab color space comprises basic components such as a K-image, a color image, a backlight, and a beam splitter grating located in front of it. The K-image is a horizontally high-resolution image output by laser exposure. The color image can be inkjet-painted, printed, output by color exposure, or displayed directly on a display screen. The beam splitter grating refers to various gratings with horizontal parallax stereoscopic display capabilities, primarily lenticular gratings, which are configured vertically or at an angle. The beam splitter grating, K-image, and color image are registered together, and the backlight is combined to display a high-definition stereoscopic image. When the color image is displayed on the screen, the backlight is already contained within the screen. In this invention, the K-image and color image refer to the stereoscopic image obtained after channel extraction or modulation from a multi-viewpoint sequence image with horizontal parallax sampling and synthesis. Furthermore, separating the K-image and color image from the parallax sequence image and then sampling and synthesizing them separately to form a stereoscopic K-image and a stereoscopic color image is equivalent to sampling and synthesizing a stereoscopic image from the parallax sequence image and then separating the stereoscopic K-image and stereoscopic color image.

[0009] Furthermore, in this section, sequential images or stereo images are simply referred to as images. The images are first converted to Lab mode and decomposed into L channel, a channel, and b channel. The L channel displays brightness information, and the details of the image are mainly contained in the L channel. The a and b channels display color information, and the image details are less, which reduces the difficulty of registration. The brightness data of the L channel is converted into a K image, which is a single-channel grayscale image. The a color channel and the b color channel are merged into a color image.

[0010] Furthermore, the color image can be obtained using the following methods: Set all data of the L channel of the Lab image to the median value of its range, save it as a Lab mode color image, or convert it to an RGB mode color image, or convert it to a CMYK mode color image; perform blurring processing on the color image to further reduce the registration accuracy requirements. Blur processing includes, but is not limited to, Gaussian blur, horizontal blur, average blur, aperture blur, and out-of-focus blur. The median value of the L channel range is set differently in different image processing software; for example, in Photoshop, the median value is 127.

[0011] Furthermore, the K image is a stereoscopic grayscale image. After RIP processing and laser exposure, the K image is output onto a transparent film with extremely high horizontal resolution. The color image is in Lab or RGB mode. After being appropriately sized, it is displayed on a screen. The K image film is overlaid on the outer surface of the screen to form a color stereoscopic image. The beam splitter is located on the outermost layer, and together with the K image and the screen, it displays a high-definition color stereoscopic image. Preferably, the K image is located at the focal plane of the beam splitter, and the color image is located slightly off-focus behind the focal plane. The screen can be an LCD or OLED screen. To reduce the possibility of screen burn-in, the color image needs to be turned off, replaced, or have its hue changed after being displayed in a fixed position for a certain period of time. Color changing effects can be added during the image display process.

[0012] Furthermore, dynamic images can be overlaid on color images on the display screen. The dynamic images can be 2D or 3D images. The dynamic images are not separated into K images and color images. The color images and dynamic images can be displayed as layers overlapping each other or blended together. If the dynamic image is a 3D image, it must be a stereoscopic image sampled and synthesized according to the beam splitter pitch and tilt angle parameters. The lower resolution 2D and 3D images on the screen are placed in a slightly out-of-focus position, which helps to smooth the texture and reduce the sense of jaggedness.

[0013] The shape of a pixel after ripping in a K image is an extremely thin line segment. Considering the characteristics of laser exposure and development, extremely small laser points need to be arranged in a clustered state for exposure to reduce the expansion of the exposed area or the loss of isolated points. Therefore, the K image after ripping will show a periodic line structure, referred to as RIP lines. RIP lines, raster lines, and the pixel structure of the display screen are all periodic, and when stacked, they can easily produce multiple levels of moiré patterns.

[0014] First, determine the matching of the tilt direction of the beam splitter with the angle of the display screen. The angle between the beam splitter line and the vertical direction of the display screen is θ. Observe the image on the display screen through the beam splitter. Start from 0 and adjust the size of θ clockwise or counterclockwise. When the moiré pattern superimposed on the image disappears or becomes less obvious, determine the value of θ. If there are multiple values ​​of θ, the smallest value is preferred.

[0015] The second step is to determine the alignment of the RIP line direction with the display angle. The direction of the next pixel segment after the RIP should be consistent with the direction of the raster lines, such as... Figure 6 As shown. The angle between the RIP line and the vertical direction of the display screen is α. Observe the display screen image through the K-image exposed by the laser, and adjust the value of α clockwise or counterclockwise starting from 0. Determine the value of α when the superimposed moiré patterns on the image disappear or become indistinct. If there are multiple α values, the α value with an absolute value of |α-θ| close to 90° is preferred. Alternatively, the RIP method of adjusting the K-image can be modified by generating irregular curves or broken lines to disrupt the periodic structure of the RIP lines while ensuring that the laser points are exposed in a concentrated state. This eliminates the moiré patterns between the RIP lines and the display screen pixels. Irregular RIP lines superimposed on the display screen may cause the image to appear jagged.

[0016] In the static display scheme, the K image is a stereo grayscale image. After RIP processing and laser exposure, the K image is output onto film with extremely high horizontal resolution. The color image is in Lab, RGB, or CMYK mode and is printed or inkjet at a relatively low resolution. If the color image is in Lab or RGB mode, the output device will internally convert it to CMYK mode for output. The K image film is overlaid on the color image to form a high-definition color stereoscopic image under backlight illumination.

[0017] Furthermore, the K image is obtained through hybrid modulation as follows: ① The image is converted into Lab mode and CMYK mode respectively; ② The Lab mode is decomposed into three channels: L, a, and b. The L channel is converted into a grayscale image and labeled as K. L； ③ The CMYK mode is decomposed into four channels: C, M, Y, and K, labeled CMYK. K ④ The color image is a CMYK mode image composed of the CMY three channels, and the K channel data in the original CMYK image has been set to zero; ⑤ The K image is composed of K L and K K According to the formula K=f1(i)*K L +f2(i)*K K Mixed and modulated, K K The missing details can be found in K L This is partially supplemented. f1(i) and f2(i) are modulation coefficients, which can be constants or functions related to grayscale.

[0018] Furthermore, using the undecomposed Lab image as a reference, the mixed-modulated K image data is implanted into the K channel of the color image to form a new CMYK image. The brightness curve and saturation are adjusted, or color matching is used to make the CMYK image similar in color to the reference image. The adjusted CMYK image is then decomposed back into two parts: the K channel and the CMY channel. The K channel is used for high-precision output of the grayscale image via laser exposure, and the CMY channel is used for the color image in inkjet printing output.

[0019] The present invention provides a high-definition naked-eye 3D display structure and method based on the Lab color space, which has the following advantages:

[0020] It utilizes existing black-and-white stereoscopic image and color inkjet stereoscopic image output equipment and production processes, as well as existing flat panel displays. The structure and production method are relatively simple, and the stereoscopic image has high clarity. Based on the Lab color space, it separates the luminance image and color image for high and low resolution display, preserving the details and color accuracy of the stereoscopic image relatively completely. By reducing the image details contained in the color image and further blurring or bokeh processing, the registration requirements of the luminance image and color image are significantly reduced. It can display high-definition naked-eye 3D dynamic images, which is particularly suitable for video displays designed for fixed themes such as product advertisements, landscapes, and portraits. Attached image description:

[0021] Figure 1 This is a schematic diagram of a high-definition naked-eye 3D dynamic display structure based on the Lab color space;

[0022] Figure 2 This is a schematic diagram of a high-definition naked-eye 3D static display structure based on the Lab color space;

[0023] Figure 3 A block diagram for separating the K-image and color image from a Lab mode image;

[0024] Figure 4 A block diagram of K-image mixing modulation;

[0025] Figure 5 A block diagram for high-definition naked-eye 3D display based on the Lab color space;

[0026] Figure 6 This is a schematic diagram showing the angular relationship between the beam splitter, RIP lines, and display screen.

[0027] The figure labels in the above figures are as follows:

[0028] 10. Diode grating, 20K image, 30 color image, 40 backlight, 50 display, 60 dynamic image, 70 pixel line segments after rip, 80 rip lines of K image, 90 laser exposure points, 100 diode grating line direction.

[0029] The high-definition naked-eye 3D display structure and method based on Lab color space disclosed in this invention has a relatively simple structure and manufacturing process, which reduces the accuracy requirements for stereoscopic image registration. It can use a display screen to replace the inkjet-painted color image, thereby displaying high-definition static and dynamic stereoscopic images. The structure and method can be applied to advertising displays, portrait photography, and virtual landscape displays. Detailed implementation method:

[0030] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0031] Implementation Case 1: A high-definition naked-eye 3D dynamic display structure and method based on the Lab color space, consisting of a beam splitter grating 10, a K image 20, and a display screen 50 displaying color images 30, such as... Figure 1 As shown. The K image 20 is a horizontally high-resolution image output by laser exposure. The color image 30 is displayed on the display screen 50, and the size and position of the color image 30 can be adjusted on the display screen 50 to align it with the K image 20. The beam splitter 10 is configured vertically or at an angle, with an angled configuration being preferred. The angle of the angle is such that no obvious moiré patterns or color separation lines are produced between the screen image and the beam splitter 10. The beam splitter 10, the K image 20, and the color image 30 are combined to display a high-definition color stereoscopic image.

[0032] Multi-viewpoint image sequences with horizontal parallax are acquired through stereoscopic photography, 3D scene rendering, and 2D-to-3D conversion. These images are then sampled and synthesized into a stereoscopic image. The stereoscopic image is converted to Lab mode, and the L-channel data is extracted and converted into a K-image 20. The a-channel and b-channel data are merged and converted into a color image 30. The K-image 20 is a grayscale image, converted to a 1-bit Tif image via RIP, and output onto a transparent film via laser exposure, achieving a horizontal output resolution of over 2400 PPI. The color image 30 is a color image in Lab or RGB mode, displayed on the display screen 50 after being reduced in resolution. The K-image 20 is positioned on the plane constructed by the focus lines of the beam splitter 10, allowing for a clear display of the stereoscopic image. The color image 30 on the display screen 50 is slightly out of focus behind the K-image 20, mitigating the image quality degradation caused by interference between the periodic pixel structure and the beam splitter 10.

[0033] Color image 30 can be obtained by the following method: In Photoshop, set all data of the L channel of the Lab image to 127 and save it as a Lab mode image, or convert it to an RGB mode image, reduce the image resolution and adjust the position and size so that color image 30 can be registered with K image 20, and blur or defocus the color image.

[0034] The display screen 50 can be an LCD or OLED screen. To reduce the possibility of screen burn-in, the color image 30 needs to be turned off, replaced, or have its hue changed after being displayed in a fixed position for a certain period of time. Color changing effects can be added during the image display process. Furthermore, a dynamic image 60 can be superimposed on the color image 30 on the display screen 50. The dynamic image can be a 2D image or a 3D image. The dynamic image 60 is not separated into a K image and a color image. If the dynamic image 60 is a 3D image, it must be a stereoscopic image synthesized by sampling according to the grating pitch and tilt angle θ of the beam splitter 10. Placing the lower resolution 2D or 3D images on the screen at a slightly out-of-focus position helps to smooth the texture and reduce the appearance of jagged edges.

[0035] Implementation Case 2: A high-definition naked-eye 3D static display structure and method based on the Lab color space, consisting of a beam splitter grating 10, a K image 20, a display color image 30, and a backlight 40, as shown below. Figure 2 As shown. The K image 20 is a horizontally high-resolution image output by laser exposure, while the color image 30 is printed or printed onto a light film at a lower resolution. The K image 20 is registered with the color image 20 to display brightness and color. The beam splitter 10 is configured vertically or obliquely, preferably vertically. The beam splitter 10, the K image 20, and the color image 30 are combined to display a high-definition color stereoscopic image under the illumination of the backlight 40.

[0036] Multi-viewpoint image sequences with horizontal parallax are acquired through methods such as stereoscopic photography, 3D scene rendering, and 2D-to-3D conversion. These multi-viewpoint image sequences are then sampled and synthesized into stereoscopic images. The stereoscopic images are converted to Lab mode, and the L channel data is extracted and converted into K image 20. The a and b channel data are then merged and converted into color image 30. K image 20 is a grayscale image, which is converted into a 1-bit Tif image via RIP and output onto a transparent film via laser exposure, achieving a horizontal output resolution of over 2400 PPI. The Lab mode color image 30 is converted into a CMYK mode color image, reduced in resolution, and printed onto a light film. In this embodiment, it is also feasible to replace the front and rear positions of K image 20 and color image 30 for registration.

[0037] Color image 30 can be obtained as follows: In Photoshop, set all data of the L channel of the Lab image to 127, convert it to CMYK mode, reduce the image resolution to match the printing output device, and then register it with the K image 20 after printing. Blurring or defocusing the color image before printing can further reduce the registration accuracy requirements and avoid the separation of black and white and color image outlines caused by registration errors.

[0038] Implementation Case 3: A high-definition naked-eye 3D static display structure and K-image modulation method based on the Lab color space, consisting of a beam splitter grating 10, a K-image 20, a display color image 30, and a backlight 40, as shown below. Figure 2 As shown. K image 20 is a horizontally high-resolution image output by laser exposure, and color image 30 is printed or printed on a light film at a lower resolution, and the brightness and color are registered with K image 20.

[0039] Multi-viewpoint image sequences with horizontal parallax were acquired through stereoscopic photography, 3D scene rendering, and 2D-to-3D conversion. These multi-viewpoint image sequences were then sampled and synthesized into stereoscopic images, which were then converted into Lab and CMYK modes respectively. The Lab mode was decomposed into three channels: L, a, and b. The L channel was converted into a grayscale image and labeled KL. The CMYK mode was decomposed into four channels: C, M, Y, and K, labeled CMYK. KK Color image 30 is a CMYK mode image composed of CMY three-channel data after setting all K channel data to zero; K image 20 is composed of K... L and K K According to the formula K=f1(i)*K L +f2(i)*K K Mixed and modulated, K K The missing details can be found in K L This is partially supplemented. f1(i) and f2(i) are modulation coefficients, which can be constants or functions related to gray level i.

[0040] Using the Lab mode stereo image as a reference, the mixed-modulated K image 20 data is implanted into the K channel of the color image 30 to form a new CMYK image. The brightness curve and saturation of the CMYK image are adjusted to make the CMYK image colors similar to the reference image. The adjusted CMYK image is then decomposed into two parts: the K channel and the CMY channel. The K channel is used for the high-precision laser exposure output of the K image 20, and the CMY channel is used for the printing output of the color image 30.

[0041] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A high-definition naked-eye 3D display structure and method based on the Lab color space, characterized in that, It consists of a K-image, a color image, a backlight, and a beam splitter in front of it. The K-image and color image are stereoscopic images synthesized by sampling multi-viewpoint sequence images with horizontal parallax. After channel extraction, they represent brightness and color information, respectively. The K-image is a grayscale image, which is converted into a 1-bit image by RIP and then output on a transparent film by laser exposure, with high horizontal resolution. The color image can be a low-resolution image output by inkjet printing or printing, or it can be displayed directly on a display screen. When the color image is displayed on the display screen, the backlight is already included in the display screen. The beam splitter has horizontal parallax stereoscopic display function. The beam splitter can be configured vertically or tilted. After the K-image and color image are registered, a high-definition stereoscopic dynamic or static image is displayed under the combined action of the beam splitter and the backlight.

2. The structure and method as described in claim 1, further characterized in that, The stereo image is first converted to Lab mode and decomposed into L channel, a channel, and b channel. The L channel displays brightness information. The brightness data of the L channel is converted into a K image. The K image is a grayscale image. The main details of the stereo image are contained in the K image. The a channel and b channel display color information. The a color channel and b color channel are merged into a color image. The color image contains less detail, which reduces the difficulty of registration between the K image and the color image.

3. The structure and method as described in claim 2, further characterized in that, The color image is in Lab or RGB mode and is displayed on the screen after being reduced in resolution. The resolution, size, and position of the color image can be adjusted to align it with the K image. The K image is overlaid on the outer surface of the screen with a transparent film. The beam splitter is located on the outermost layer, and the K image is positioned at the focal plane of the beam splitter, which can display a high-definition stereoscopic image. The color image is positioned slightly out of focus behind the focal plane, which can weaken the image quality degradation caused by the periodic structure of pixels on the screen and the interference of the beam splitter to produce moiré patterns.

4. The structure and method as described in claim 3, further characterized in that, The display screen can be an LCD or OLED screen. To reduce the possibility of screen burn-in, the color image needs to be turned off, replaced, or have its hue changed after being displayed in a fixed position for a certain period of time. Color changing effects are added during the image display process. Dynamic images can also be superimposed on the color image on the display screen. The dynamic image can be a 2D image or a 3D image. The dynamic image is not separated into K image and color image. The color image and dynamic image can be displayed as upper and lower layers, in a way that covers or blends with each other. If the dynamic image is a 3D image, it must be a stereoscopic image synthesized by sampling according to the grating pitch and tilt angle θ of the beam splitter. The lower resolution 2D and 3D images on the screen are placed in a slightly out-of-focus position to reduce moiré patterns and reduce image jaggedness.

5. The structure and method as described in claim 2, further characterized in that, The color image is in Lab, RGB, or CMYK mode and is printed or inkjet at a relatively low resolution. The K image is registered on the color image, forming a high-definition color stereoscopic image under backlighting. Alternatively, to further reduce the registration accuracy requirements and avoid the separation of black and white and color image outlines, the color image is blurred before printing or inkjet.

6. The structure and method as described in claim 5, further characterized in that, The stereo image was converted to Lab and CMYK modes respectively. The Lab mode was decomposed into three channels: L, a, and b. The L channel was converted into a grayscale image and labeled as K. L The CMYK mode is decomposed into four channels: C, M, Y, and K, labeled CMYK. K A color image is a CMYK mode image composed of the CMY three-channel data after setting all K-channel data to zero. A K-image is composed of K... L and K K According to the formula K=f1(i)*K L +f2(i)*K K Mixed and modulated, K K The missing details can be found in K L The following is a partial supplement: f1(i) and f2(i) are modulation coefficients, which can be constants or functions related to gray level i.

7. The structure and method as described in claim 6, further characterized in that, Using the Lab mode image as a reference, the mixed-modulated K image data is implanted into the K channel of the color image to form a new CMYK image. The brightness curve and saturation are adjusted, or color matching is used to make the CMYK image similar to the reference image. The adjusted CMYK image is then decomposed into two parts: the K channel and the CMY channel. The K channel is used for high-precision output of grayscale image laser exposure, and the CMY channel is used for color image printing output.

8. The structure and method as described in claim 4, further characterized in that, First, determine the alignment of the beam splitter tilt direction with the display screen angle. The angle between the grating lines and the vertical direction of the display screen is θ. Observe the display screen image through the beam splitter and adjust θ clockwise or counterclockwise, starting from 0. When the moiré patterns superimposed on the image disappear or become less noticeable, determine the θ value. If there are multiple θ values, the smallest value is preferred. Second, determine the alignment of the RIP line direction with the display screen angle. The direction of the next pixel segment after the RIP line is consistent with the direction of the grating lines. The angle between the RIP line and the vertical direction of the display screen is α. Observe the display screen image through the K-image exposed by the laser and adjust α clockwise or counterclockwise, starting from 0. When the moiré patterns superimposed on the image disappear or become less noticeable, determine the α value. If there are multiple α values, the α value with an absolute value of |α-θ| close to 90° is preferred. Alternatively, adjust the RIP method of the K-image. While ensuring that the laser points are exposed in a focused state, generate the RIP lines as irregular curves or broken lines to disrupt the periodic structure of the RIP lines and eliminate the moiré patterns between the RIP lines and the display screen pixels.

Images