Display driving method and display device

Abstract

This application discloses a display driving method and a display device. The display driving method includes: acquiring an original image to be displayed; performing error diffusion processing on the image data of pixels in the original image to obtain a first image, wherein at least two pixels have different error diffusion weights in the same direction; splitting the first image into a first sub-image and a second sub-image, wherein the first sub-image and the second sub-image have the same number of pixels, and the number of pixels in the first sub-image is less than the number of pixels in the original image, and the pixels in the first sub-image and the pixels in the second sub-image do not overlap in position; and sequentially displaying the first sub-image and the second sub-image. According to the embodiments of this application, the resolution of the image presented by the display driving is higher than the resolution of the display device itself, and a realistic high-resolution image can be displayed.

Description

Technical Field

[0001] This application relates to the field of display technology, specifically to a display driving method and a display device. Background Technology

[0002] In a projection device, a projection component can be used to project images onto a projection screen, causing the screen to display a corresponding image. To present users with more vivid and realistic images, higher resolution projected images are sought after by those skilled in the art. However, due to the limitations of the projection system's hardware structure, the pixel density of the projection system itself is fixed. For example, a projection system can project images at 2K resolution. To display images at 4K resolution, the pixel size of the projection system itself can be reduced to increase the resolution, but this method has drawbacks such as high manufacturing difficulty, low aperture ratio, and poor luminous efficiency.

[0003] How to enable a projection device to display higher resolution images based on the existing hardware of the projection device is a challenge faced by those skilled in the art. Summary of the Invention

[0004] This application provides a display driving method and a display device that enable the displayed image resolution to be higher than the resolution of the display device itself, and to display realistic high-resolution images.

[0005] In a first aspect, embodiments of this application provide a display driving method, comprising: acquiring an original image to be displayed; performing error diffusion processing on the image data of pixels in the original image to obtain a first image, wherein at least two pixels have different error diffusion weights in the same direction; splitting the first image into a first sub-image and a second sub-image, wherein the first sub-image and the second sub-image have the same number of pixels, and the number of pixels in the first sub-image is less than the number of pixels in the original image, and the pixels of the first sub-image and the pixels of the second sub-image do not overlap in position; and sequentially displaying the first sub-image and the second sub-image.

[0006] In one possible implementation of the first aspect, error diffusion processing is performed on the image data of pixels in the original image to obtain a first image, including: The quantization error of the target pixel is determined based on the difference between the target image data of the target pixel in the original image and the preset quantization data. The target pixel is any pixel in the original image. The quantization error is assigned to the pixels adjacent to the target pixel according to the preset weight information to obtain the first image, wherein the scanning time of the pixels assigned the quantization error and adjacent to the target pixel is later than the scanning time of the target pixel.

[0007] In one possible implementation of the first aspect, the image data includes luminance data and chrominance data. Error diffusion processing is performed on the image data of pixels in the original image to obtain a first image, including: A first error diffusion process is performed on the brightness data of pixels in the original image to obtain a third sub-image. The first error diffusion process includes at least two pixels having different error diffusion weights in the same direction. A second error diffusion process is performed on the chromaticity data of the pixels in the original image to obtain a fourth sub-image; The first image is obtained based on the third and fourth sub-images.

[0008] In one possible implementation of the first aspect, the image data includes brightness data, the gray level of the target pixel in the original image is any one of gray levels 1 to 244, and the quantization data is 128 gray levels; Alternatively, the target pixel in the original image has a gray level of 255, and the quantized data has a gray level of 255.

[0009] In one possible implementation of the first aspect, the target pixel is the pixel located in the first row and first column, and the target image data of the target pixel is the original image data of the target pixel in the original image; Alternatively, the target pixel is a pixel located outside the first row and first column, and the target image data of the target pixel is the sum of the original image data of the target pixel in the original image and the quantization error assigned to it.

[0010] In one possible implementation of the first aspect, the image data includes brightness data, and the weight information corresponding to the brightness data is first weight information. Before distributing the quantization error to pixels adjacent to the target pixel according to the preset weight information to obtain the first image, the method further includes: Based on the grayscale difference between the target pixel and its neighboring pixels, determine the gradient magnitude of the pixels adjacent to the target pixel. Based on the gradient magnitude, determine the first weight information corresponding to the target pixel and its neighboring pixels.

[0011] In one possible implementation of the first aspect, determining the first weight information of the target pixel and its neighboring pixels based on the gradient magnitude includes: Based on the relationship between the gradient magnitude and the preset gradient threshold, the region type to which the target pixel and its neighboring pixels belong is determined. The region type includes high-frequency regions and low-frequency regions. If the gradient magnitude is greater than the gradient threshold, the region type to which the target pixel and its neighboring pixels belong is a high-frequency region. If the gradient magnitude is less than or equal to the gradient threshold, the region type to which the target pixel and its neighboring pixels belong is a low-frequency region. The first weight information of the target pixel and its neighboring pixels is determined based on the region type to which the target pixel and its neighboring pixels belong.

[0012] In one possible implementation of the first aspect, the error diffusion weight of pixels in the high-frequency region is greater than that of pixels in the low-frequency region.

[0013] In one possible implementation of the first aspect, the pixels adjacent to the target pixel include a first pixel, a second pixel, and a third pixel, wherein the first pixel is adjacent to the target pixel in the row direction, the second pixel is adjacent to the target pixel in the column direction, and the third pixel is adjacent to the target pixel in the diagonal direction. The first pixel, the second pixel, and the third pixel all belong to the low-frequency region type, and the error diffusion weights of the first pixel, the second pixel, and the third pixel are equal. Alternatively, the region to which the first pixel belongs is a high-frequency region, and the gradient magnitude of the first pixel is the largest, and the error diffusion weight of the first pixel is the largest. Alternatively, the region to which the second pixel belongs is a high-frequency region, and the gradient magnitude of the second pixel is the largest, and the error diffusion weight of the second pixel is the largest. Alternatively, the region to which the third pixel belongs is a high-frequency region, and the gradient magnitude of the third pixel is the largest, and the error diffusion weight of the third pixel is the largest.

[0014] In one possible implementation of the first aspect, determining the first weight information of the target pixel and its neighboring pixels based on the gradient magnitude includes: Based on the gradient magnitude, determine the second weight information of the pixels adjacent to the target pixel; The second weight information is corrected based on visual sensitivity, and the first weight information of the pixels adjacent to the target pixel is obtained.

[0015] In one possible implementation of the first aspect, the second weight information is corrected based on visual sensitivity to obtain the first weight information of pixels adjacent to the target pixel, including: If the region type of the pixels adjacent to the target pixel is a high-frequency region, the error diffusion weight of the pixels adjacent to the target pixel in the second weight information is reduced; or, if the region type of the pixels adjacent to the target pixel is a low-frequency region, the error diffusion weight of the pixels adjacent to the target pixel in the second weight information is increased, thus obtaining the first weight information of the pixels adjacent to the target pixel.

[0016] In one possible implementation of the first aspect, the second error diffusion process includes at least two pixels having the same error diffusion weight in the same direction.

[0017] In one possible implementation of the first aspect, if the chromaticity data of a pixel in the original image is greater than a preset chromaticity threshold, a second error diffusion process is performed on the chromaticity data of the pixel in the original image to obtain a fourth sub-image. Alternatively, if the chromaticity data of pixels in the original image is less than or equal to the chromaticity threshold, error propagation of the chromaticity data of pixels in the original image is not performed.

[0018] In one possible implementation of the first aspect, error diffusion processing is performed on the image data of pixels in the original image to obtain a first image, including: Based on the image type of the original image and the pre-defined mapping relationship between multiple image types and multiple error diffusion algorithms, the target error diffusion algorithm corresponding to the original image is determined; Based on the target error diffusion algorithm, the image data of pixels in the original image are processed by error diffusion to obtain the first image.

[0019] In one possible implementation of the first aspect, different error diffusion algorithms cover different error diffusion windows, and / or, the weight distribution of individual error diffusion windows of different error diffusion algorithms is different.

[0020] In one possible implementation of the first aspect, the number of pixels in the original image is four times the number of pixels in the first sub-image.

[0021] In one possible implementation of the first aspect, the pixels of the second sub-image are offset diagonally relative to the pixels of the first sub-image.

[0022] In one possible implementation of the first aspect, the pixels of the second sub-image are offset by half a pixel in the row direction relative to the pixels of the first sub-image, and the pixels of the second sub-image are offset by half a pixel in the column direction relative to the pixels of the first sub-image.

[0023] Secondly, embodiments of this application provide a display device, comprising: An image processing component is configured to: acquire an original image to be displayed; perform error diffusion processing on the image data of pixels in the original image to obtain a first image, wherein at least two pixels have different error diffusion weights in the same direction; split the first image into a first sub-image and a second sub-image, wherein the number of pixels in the first sub-image and the second sub-image are the same, and the number of pixels in the first sub-image is less than the number of pixels in the original image, and the pixels in the first sub-image and the pixels in the second sub-image do not overlap in position; The display module is configured to display the first sub-image and the second sub-image sequentially.

[0024] In one possible implementation of the first aspect, the display device further includes a projection screen configured to receive a first sub-image and a second sub-image from the display module.

[0025] According to the display driving method and display device provided in the embodiments of this application, a high-resolution original image is acquired, nonlinear error diffusion processing is performed on the original image, and the image after non-error diffusion processing is split into a low-resolution first sub-image and a second sub-image, with a pixel offset between the first sub-image and the second sub-image. The first sub-image and the second sub-image are displayed alternately, so that the first sub-image and the second sub-image are arranged in an interlaced complementary manner in two-dimensional space. The pixels of the two sets of sub-images uniformly cover all pixel sampling grids corresponding to the high-resolution original image. Relying on the persistence of vision of the human eye, two consecutive frames are visually fused. The sparse two low-resolution pixel samples are reconstructed to form complete high-density image information, ultimately achieving an equivalent high-resolution display effect. In addition, nonlinear error diffusion processing can optimize the distribution of complementary sampling points, so that the effective pixel density after the first sub-image and the second sub-image are superimposed is closer to the real high-resolution image. For example, when processing certain image content (such as sharp edges and fine textures), nonlinear error diffusion processing can better utilize the information gain brought by displacement, making the final displayed image more realistic. In addition, this application does not require modification of the pixel resolution of existing display devices; it only needs to embed the error diffusion processing function into the display device, which can reduce costs. Attached Figure Description

[0026] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings, in which the same or similar reference numerals denote the same or similar features, and the drawings are not drawn to scale.

[0027] Figure 1 This illustration shows a first flowchart of a display driving method provided in an embodiment of this application; Figure 2 This illustration shows a schematic diagram of a display driving method provided in an embodiment of this application. Figure 3 This illustration shows a second flowchart of the display driving method provided in an embodiment of this application; Figure 4 This illustration shows a schematic diagram of the error diffusion processing principle provided in an embodiment of this application; Figure 5 This illustration shows a third flowchart of the display driving method provided in an embodiment of this application; Figure 6 This illustration shows a fourth flowchart of the display driving method provided in an embodiment of this application; Figure 7This illustration shows a fifth flowchart of the display driving method provided in an embodiment of this application; Figure 8 This illustration shows a schematic diagram of an error diffusion weight allocation provided in an embodiment of this application; Figure 9 This illustration shows a sixth flowchart of the display driving method provided in an embodiment of this application; Figure 10 This illustration shows a seventh flowchart of the display driving method provided in an embodiment of this application; Figure 11 This diagram illustrates the coverage and weight distribution of various error propagation algorithms provided in embodiments of this application. Figure 12 This illustration shows a schematic diagram of a display device provided in an embodiment of this application; Figure 13 This invention provides another schematic diagram of the structure of a display device according to an embodiment of the present application. Figure 14 This illustration shows a structural diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0028] The features and exemplary embodiments of various aspects of this application will now be described in detail. To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only configured to explain this application and are not configured to limit this application. For those skilled in the art, this application can be implemented without some of these specific details. The following description of the embodiments is merely to provide a better understanding of this application by illustrating examples of this application.

[0029] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

[0030] It should be understood that when describing the structure of a component, when referring to a layer or region as being "above" or "on top of" another layer or region, it can mean that it is directly above the other layer or region, or that it contains other layers or regions between it and the other layer or region. Furthermore, if the component is flipped over, that layer or region will be located "below" or "under" the other layer or region.

[0031] It should be understood that the term "and / or" used in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0032] In the embodiments of this application, the term "electrical connection" can refer to a direct electrical connection between two components, or it can refer to an electrical connection between two components via one or more other components.

[0033] The term "connection" can refer to "electrical connection" or "electrical connection without an intermediate transistor." The term "insulation" can refer to "electrical insulation" or "electrical isolation." The term "drive" can refer to "control" or "operation." The term "part" can refer to "section." The term "pattern" can refer to "component." The term "end" can refer to "end segment" or "end edge." A display module can be a display device or a module / part of a display device.

[0034] Various modifications and variations can be made to this application without departing from its spirit or scope, which will be apparent to those skilled in the art. Therefore, this application is intended to cover modifications and variations falling within the scope of the corresponding claims (the claimed technical solutions) and their equivalents. It should be noted that the implementation methods provided in the embodiments of this application can be combined with each other without contradiction.

[0035] As described in the background section, how to enable a display device to display higher resolution images based on existing display device hardware is a challenge faced by those skilled in the art.

[0036] For example, pixel shifting technology can be used to improve the resolution of the displayed image without changing the pixel size of the existing display device. In some implementations, pixel shifting technology includes: first, removing some pixels from the original image and performing error diffusion processing on the removed pixels; then, directly dividing or copying the original image data of the removed pixels onto the surrounding unremoved pixels; then, splitting the image into multiple sub-frame images, with the pixels of different sub-frame images spatially shifted relative to each other; and then displaying different sub-frame images alternately. These sub-frame images are superimposed on the screen, thus ultimately forming an image with a higher resolution than a single sub-frame image.

[0037] However, directly splitting or copying the original image data in error diffusion processing results in limitations in detail representation, edge sharpness, and temporal stability (such as potential flicker or motion artifacts) in the final high-resolution image, making the final image less realistic.

[0038] To address the aforementioned technical problems, this application provides a display driving method and a display driving device. The embodiments of this application will be described below with reference to the accompanying drawings.

[0039] Figure 1 This diagram illustrates a flowchart of a display driving method provided in an embodiment of this application. Figure 1 As shown, the display driving method provided in this application embodiment includes S110~S140.

[0040] S110, Obtain the original image to be displayed.

[0041] S120: Error diffusion processing is performed on the image data of pixels in the original image to obtain the first image. At least two pixels have different error diffusion weights in the same direction.

[0042] S130, the first image is split into a first sub-image and a second sub-image. The first sub-image and the second sub-image have the same number of pixels, and the number of pixels in the first sub-image is less than the number of pixels in the original image. The pixels of the first sub-image and the pixels of the second sub-image do not overlap in position.

[0043] S140, the first sub-image and the second sub-image are displayed in sequence.

[0044] Figure 2 This image illustrates a display driving method provided in an embodiment of this application. The following is a schematic diagram in conjunction with... Figure 1 and Figure 2 This paper introduces the processing flow of the display driving method provided in the embodiments of this application.

[0045] The original image 2.1 to be displayed can be obtained. The original image 2.1 can be a 4K resolution image.

[0046] Image 2.2 shows the first image obtained after error diffusion processing of the pixel image data in the original image. The arrows in Image 2.2 indicate the direction of error diffusion. The pixel image data may include luminance data and / or chrominance data. Luminance data may include the grayscale values ​​of the pixels. Chromaticity data may include data in the CIELAB uniform color space, or chrominance data may include tristimulus values, color coordinates, etc.

[0047] In pixel displacement technology, error diffusion processing can refer to: according to a certain error diffusion weight, distributing and diffusing the image data of the target pixel (or the current pixel) to the surrounding unscanned neighboring pixels (neighboring pixels are those adjacent to the target pixel and whose scanning time is later than that of the target pixel). By fine-tuning the brightness of neighboring pixels, the brightness loss of a single point is compensated. Relying on the low-pass visual fusion characteristics of the human eye, grid artifacts, color banding, moiré patterns, and screen door effects caused by pixel displacement are eliminated. Under the premise that the physical pixels remain unchanged, smooth grayscale transitions and continuous color details are restored, ensuring that the local average brightness and color of the entire image remain unchanged.

[0048] In the original image, multiple pixels are arranged in an array along the row and column directions. The error propagation algorithm is a serial, pixel-by-pixel algorithm that processes pixels sequentially in a fixed scanning order (e.g., traversing the image row by row from left to right and top to bottom). For pixels that have already been processed, the hardware has already completed the output and cannot modify their image data. Therefore, the error is allocated to unprocessed adjacent pixels that have not yet been scanned or calculated.

[0049] For example, when traversing to the target pixel in row i and column j, in the row direction, the image data (e.g., quantization error) of the target pixel can be distributed to the pixel to its right; in the column direction, the image data of the target pixel can be distributed to the pixel below it; and in the diagonal direction, the image data of the target pixel can be distributed to the pixel to its lower left or lower right.

[0050] When traversing to the target pixel in row i and column j, the pixels above and to the left of it have already been quantized and output, and the image data is already determined. There is no way to go back and modify their image data to compensate for the quantization error of the current pixel. Only the pixels on the right, below, lower right, and lower left have not yet been traversed and calculated, and have not yet been quantized and output. The error can be added to the original input value of these pixels in advance, and compensated together when processing these pixels later.

[0051] As shown in Figure 2.2, any pixel in the original image except for the single pixel in the last row and last column can be used as the target pixel. The quantization error of the target pixel can be determined based on the image data of the target pixel.

[0052] In related technologies, when performing error diffusion processing on any pixel, the error diffusion weight along any direction remains constant. This method can be called linear error diffusion processing. However, when processing certain image content (such as sharp edges and fine textures), this linear error diffusion processing method fails to fully utilize the information gain brought by displacement, limiting the potential for improving perceptual resolution and making the final displayed image less realistic.

[0053] Specifically, linear error diffusion processing can refer to: using a fixed neighborhood weight matrix to allocate quantization error, with the error diffusion weight remaining constant throughout the process and not adaptively changing with pixel chroma values, saturation, or image area, and employing low-intensity, narrow-range fixed weight diffusion.

[0054] In this application, pixels P1 and P2 have different error diffusion weights in the same direction, and pixels P1 and P2 can be any two pixels in the original image. For example, pixels P1 and P2 may have different error diffusion weights to their right, and / or, pixels P1 and P2 may have different error diffusion weights to their down, and / or, pixels P1 and P2 may have different error diffusion weights to their lower right. This method can be called nonlinear error diffusion processing. Nonlinear error diffusion can better utilize the information gain brought by displacement when processing certain image content (such as sharp edges and fine textures), making the final displayed image more realistic.

[0055] Specifically, nonlinear error diffusion processing refers to the dynamic adjustment of error diffusion weights based on the current pixel brightness and local image texture gradient, rather than a fixed weight matrix. For example, differentiated error allocation strategies are used for dark areas, bright areas, and edge areas.

[0056] For example, error diffusion processing can be performed using a preset error diffusion algorithm, which includes any one of Floyd-Steinberg, Jarvis-Judice-Ninke, Stucki, and Burkes.

[0057] For example, the original image is a 4K image. The display device's hardware architecture supports displaying a single frame as a 2K image. The physical resolution of the display device can be 2K. The display device's frame buffer, driving timing, and hardware clock operate according to the 2K resolution specification. Therefore, image 2.2 needs to undergo pixel downsampling processing (removing some pixels) to obtain the reduced image 2.3. Half the number of pixels in image 2.2 can be removed to obtain the lower-resolution image 2.3. For example, image 2.2 can be processed by removing pixels in alternating rows and columns, retaining one column every two columns horizontally and one row every two rows vertically, removing half the number of pixels. The total number of pixels after removal is half the total number of pixels before removal.

[0058] Furthermore, the deleted image 2.3 is split into a first sub-image 2.41 and a second sub-image 2.42. The first sub-image 2.41 and the second sub-image 2.42 have the same number of pixels, and the first sub-image 2.41 has fewer pixels than the image 2.3. In some embodiments, the number of pixels in the original image is four times the number of pixels in the first sub-image. Similarly, the number of pixels in the original image is four times the number of pixels in the second sub-image.

[0059] For example, the number of pixels in the first sub-image 2.41 is half the number of pixels in image 2.3, the number of pixels in the first sub-image 2.41 is one-quarter the number of pixels in image 2.2, and the number of pixels in the second sub-image 2.42 is one-quarter the number of pixels in image 2.2.

[0060] The pixels of the first sub-image 2.41 and the second sub-image 2.4 do not overlap in position. For example, the first sub-image 2.41 consists of pixels in odd-numbered rows and odd-numbered columns in the original image, and the second sub-image 2.4 consists of pixels in even-numbered rows and even-numbered columns in the original image.

[0061] In the images 2.3, 2.41 (first sub-image), and 2.4 (second sub-image), the pixels at the positions of the dashed boxes represent the pixels that were removed.

[0062] In terms of physical specifications, the number of images in the first sub-image 2.41 and the second sub-image 2.4 is 1 / 4 of the number of images in the original image. However, by displaying the first sub-image 2.41 and the second sub-image 2.4 in sequence, for example, the first sub-image 2.41 as a B-frame image FB and the second sub-image 2.4 as an A-frame image FA, the final image can be visually equivalent to a 4K resolution image by relying on the spatial misalignment of the first sub-image 2.41 and the persistence of vision of the human eye.

[0063] refer to Figure 2 The original 4K resolution image is equivalent to a 6×6 pixel grid, and the first sub-image 2.41 and the second sub-image 2.42 are each equivalent to a 3×3 pixel grid. Compared to the original image, the first sub-image 2.41 has one pixel in the upper left position within the 2×2 grid, and the second sub-image 2.42 has one pixel in the lower right position within the 2×2 grid.

[0064] In some embodiments, the pixels of the second sub-image are offset diagonally relative to the pixels of the first sub-image. As an example, the pixels of the second sub-image are offset by half a pixel in the row direction relative to the pixels of the first sub-image, and the pixels of the second sub-image are offset by half a pixel in the column direction relative to the pixels of the first sub-image.

[0065] As shown in Figure 2.5, the second sub-image 2.42 can be displaced by a distance L in the diagonal direction relative to the first sub-image 2.41. For example, the second sub-image 2.42 can be offset by 0.5 pixels in both the row and column directions relative to the first sub-image 2.41. The first sub-image 2.41 and the second sub-image 2.42 can be refreshed alternately at a frequency higher than that of human vision (e.g., greater than or equal to 120Hz). The first sub-image 2.41 and the second sub-image 2.42 have preset pixel offsets, so that the first sub-image 2.41 and the second sub-image 2.42 are arranged in an alternating and complementary manner in two-dimensional space. The pixels of the two sets of sub-images uniformly cover all pixel sampling grids corresponding to the original 4K resolution image. Relying on the persistence of vision of human eyes, the two temporally consecutive frames are visually fused. The sparse two low-resolution pixel samplings are reconstructed to form complete high-density image information, ultimately achieving an equivalent 4K resolution display effect.

[0066] According to the display driving method provided in this application, a high-resolution original image is acquired, nonlinear error diffusion processing is performed on the original image, and the image after non-error diffusion processing is split into a low-resolution first sub-image and a second sub-image. There is a pixel offset between the first sub-image and the second sub-image, and the first sub-image and the second sub-image are displayed alternately. This makes the first sub-image and the second sub-image arranged in an interlaced complementary manner in two-dimensional space. The pixels of the two sets of sub-images uniformly cover all pixel sampling grids corresponding to the high-resolution original image. Relying on the persistence of vision of the human eye, two consecutive frames are visually fused. The sparse two-way low-resolution pixel sampling reconstructs complete high-density image information, ultimately achieving an equivalent high-resolution display effect. In addition, nonlinear error diffusion processing can optimize the distribution of complementary sampling points, making the effective pixel density of the superimposed first sub-image and the second sub-image closer to the real high-resolution image. For example, when processing certain image content (such as sharp edges and fine textures), nonlinear error diffusion processing can better utilize the information gain brought by displacement, making the final displayed image more realistic. In addition, this application does not require modification of the pixel resolution of existing display devices; it only needs to embed the error diffusion processing function into the display device, which can reduce costs.

[0067] Figure 3 This illustration shows a second flowchart of a display driving method provided in an embodiment of this application. In some embodiments, such as Figure 3 As shown, error diffusion processing is performed on the image data of pixels in the original image to obtain the first image, which may include S310 and S320.

[0068] S310, determine the quantization error of the target pixel based on the difference between the target image data of the target pixel in the original image and the preset quantization data, where the target pixel is any pixel in the original image.

[0069] S320: The quantization error is assigned to pixels adjacent to the target pixel according to preset weight information to obtain a first image. The scanning time of the pixels assigned quantization error and adjacent to the target pixel is later than the scanning time of the target pixel.

[0070] The target image data for the target pixel may include the target brightness data and / or target chromaticity data of the target pixel.

[0071] The quantization data are preset values; for example, the specific values ​​of the quantization data can be set based on experience and are not limited here. If the target image data is luminance data, the corresponding quantization data can be a grayscale threshold. If the target image data is chrominance data, the corresponding quantization data can be a chrominance threshold.

[0072] For example, the difference between the target image data and the preset quantization data can be directly used as the quantization error of the target pixel.

[0073] The final image data of the pixels adjacent to the target pixel is: its original image data plus the quantization error allocated from the target image data.

[0074] Figure 4 This diagram illustrates a principle of error diffusion processing provided in an embodiment of this application. Figure 4 As shown, the target pixel Px can be any pixel in the original image. As an example, the target pixel Px is the red pixel R, and the target image data of the target pixel Px includes brightness data. For example, if the target gray level of the target pixel Px is 120 gray levels, and the quantization data corresponding to the target pixel Px is 128 gray levels, then the quantization error of the target pixel Px is E = 120 - 128 = -8.

[0075] The error diffusion weight of the right pixel Px1 of the target pixel Px is w1, the error diffusion weight of the bottom pixel Px2 of the target pixel Px is w2, the error diffusion weight of the bottom left pixel Px3 of the target pixel Px is w3, and the error diffusion weight of the bottom right pixel Px4 of the target pixel Px is w4.

[0076] The final image data of pixel Px1 is G1 + (-8) * w1, where G1 is the original gray level of pixel Px1.

[0077] The final image data of pixel Px2 is G2+(-8)*w2, where G2 is the original gray level of pixel Px2.

[0078] The final image data for pixel Px3 is G3 + (-8) * w3, where G3 is the original grayscale of pixel Px3.

[0079] The final image data for pixel Px4 is G4 + (-8) * w4, where G4 is the original grayscale of pixel Px4.

[0080] The above example, using target image data including luminance data, illustrates the process of determining quantization error and the final luminance data of neighboring pixels of the target pixel. When the target image data includes chrominance data, the corresponding process of determining quantization error and the final chrominance data of neighboring pixels of the target pixel is similar, and will not be repeated here.

[0081] Figure 4 This is merely an example and is not intended to limit the scope of this application. In other embodiments, the number of pixels assigned quantization error may be greater than... Figure 4 More in China.

[0082] In this embodiment, the quantization error of the target pixel is determined based on the difference between the target image data and the quantization data of the target pixel. This quantization error is then distributed to its neighboring pixels according to preset weighting information, thus completing the pixel quantization error diffusion process. The brightness and chromaticity of the original image are continuous floating-point values, but display devices can generally only output a finite number of discrete values. The closest value is taken from the target image data; the difference between the target image data and the approximate value (i.e., the quantization data) is the quantization error. If this quantization error is directly discarded, the error will be concentrated in the target pixel. Error diffusion can distribute the brightness difference of a single pixel to surrounding pixels, simulating a large number of intermediate gray levels using human visual color mixing, significantly weakening gradient breaks, and making the final display-driven image's brightness transition more natural.

[0083] Figure 5 This diagram illustrates a third flowchart of a display driving method provided in an embodiment of this application. In some embodiments, image data includes luminance data and chrominance data, such as... Figure 5 As shown, error diffusion processing is performed on the image data of pixels in the original image to obtain the first image, which may include steps S510 to S530.

[0084] S510, perform a first error diffusion process on the brightness data of pixels in the original image to obtain a third sub-image. The first error diffusion process includes at least two pixels having different error diffusion weights in the same direction.

[0085] S520, performs a second error diffusion process on the chromaticity data of the pixels in the original image to obtain the fourth sub-image.

[0086] S530, the first image is obtained based on the third sub-image and the fourth sub-image.

[0087] The original image is in the RGB color space. It can be converted to a perceptibly uniform color space (such as the YCbCr or CIE Lab* color space). After color space conversion, the original image is decomposed into independent luminance (Y) and chrominance channels (Cb, Cr or a*, b*). Then, independent error diffusion processing is performed on the luminance and chrominance channels respectively, resulting in a third and fourth sub-image, with at least the luminance channel undergoing non-linear error diffusion processing. The third sub-image of the luminance channel and the fourth sub-image of the chrominance channel, after independent error diffusion processing, are recombine, and the data in the YCbCr color space (or CIE Lab* color space) is inversely transformed back to the RGB color space to obtain the first image. Based on the first image, the final output frames are obtained, namely the sub-frames optimized by error diffusion for both luminance and chrominance (i.e., the first and second sub-images). These sub-frames are then sent to pixel displacement hardware for displacement display.

[0088] The human visual system perceives brightness and chromaticity very differently. The human eye is extremely sensitive to brightness, for example, to differences in light and dark areas, edges, textures, directional artifacts, and noise. Conversely, the human eye is very insensitive to chromaticity, but has a high tolerance for color shifts, color noise, and color distortion. Sharing the same error diffusion weights for both the brightness and chromaticity channels can lead to the following problems: brightness is prone to directional artifacts and noise in light and dark areas; and excessively fine processing of chromaticity wastes computational resources.

[0089] This embodiment separates the luminance and chrominance channels, performing independent error diffusion processing on each. This isolates and diffuses luminance and chrominance errors, improving issues of brightness distortion and color distortion. Furthermore, the luminance channel, which carries texture and brightness details, employs non-linear adaptive error diffusion processing. This dynamically adjusts the error allocation weights based on the image brightness gradient and perception threshold, effectively compensating for quantization distortion during pixel displacement, suppressing gradient banding and periodic moiré artifacts, and maximizing the preservation of edge details and high dynamic range.

[0090] In some embodiments, the image data of the target pixel includes brightness data. If the gray level of the target pixel in the original image is any one of gray levels 1 to 244, the quantization data is 128 gray levels; or, if the gray level of the target pixel in the original image is 255 gray levels, the quantization data is 255 gray levels.

[0091] If the grayscale of the target pixel in the original image is 0, since pure black pixels themselves have no effective brightness information, if the quantization error is spread to surrounding pixels, it is very easy to introduce random color noise and dark texture noise in the dark areas. Especially in displaying dark scenes and night scenes, patches of fine noise will appear, severely reducing the purity of black levels. Therefore, the quantization error corresponding to the 0 grayscale is directly discarded and not allocated to neighboring pixels. In addition, for 0 grayscale pixels, only the error propagation process is truncated; the pixel itself still outputs 0 grayscale normally for display, rather than blocking the 0 grayscale pixel.

[0092] If the grayscale of the target pixel in the original image is 255, the pure white highlight pixel is quantized using a saturation threshold of 255. Once the pixel is quantized to the maximum grayscale, it will no longer overflow upwards, avoiding overshoot and whitening of the highlight area. At the same time, the quantization error generated at this position can be normally diffused into non-linear error, distributing the brightness deviation at the highlight to adjacent pixels, preserving the highlight edge details, and preventing the bright edges of the image from becoming blurred and losing their sense of depth.

[0093] Using 128 as a unified quantization threshold for the intermediate range of 1 to 255, symmetrical quantization judgment is achieved in the intermediate brightness area. If the original grayscale is greater than 128, it is quantized upwards to obtain the nearest higher grayscale value. If the original grayscale is less than 128, it is quantized downwards to obtain the nearest lower grayscale value. In this way, it can better cooperate with the nonlinear error diffusion, so that the quantization distribution of the mid-gray stage (such as faces, scenery, sky and most other image areas) is balanced, effectively suppressing the periodic grid artifacts caused by gradient color banding and pixel displacement, and ensuring smooth grayscale transition in the intermediate brightness range that the human eye is most sensitive to.

[0094] In some embodiments, the target pixel is a pixel located in the first row and first column, and the target image data of the target pixel is the original image data of the target pixel in the original image; or, the target pixel is a pixel located outside the first row and first column, and the target image data of the target pixel is the sum of the original image data of the target pixel in the original image and its assigned quantization error.

[0095] During the pixel-by-pixel error diffusion process, the quantization error generated after the current pixel is quantized will be distributed to the unprocessed neighboring pixels on the right and below according to the preset weight. The pixel in the i-th row and j-th column is the target pixel.

[0096] When i=1 and j=1, the pixel in the first row and first column is the first pixel to be processed. It has no pre-allocated quantization error. Therefore, the target image data for the pixel in the first row and first column is the original grayscale value of the pixel in the first row and first column of the original image. The pixel in the first row and first column directly uses its own original grayscale value for quantization processing, without adding external errors.

[0097] If at least one of i and j is not equal to 1, the pixel in the i-th row and j-th column has already been assigned quantization error by other pixels. Therefore, the target grayscale value of the pixel in the i-th row and j-th column is: the original grayscale value of the pixel in the i-th row and j-th column of the original image plus the quantization error assigned to it.

[0098] This embodiment enables the quantization error to be propagated pixel by pixel and globally conserved throughout the entire image. This avoids local accumulation of errors that cause distortion in brightness and darkness, and also uses a nonlinear error diffusion strategy to smooth grayscale transitions and suppress jagged edges, color banding, and moiré artifacts caused by pixel displacement.

[0099] The following section uses pixel image data, including luminance data, as an example to introduce some implementation methods for error diffusion processing of the luminance channel. Nonlinear error diffusion processing can be performed on the luminance channel.

[0100] Figure 6 This diagram illustrates a fourth flowchart of the display driving method provided in this application embodiment. In some embodiments, the image data includes brightness data, and the weight information corresponding to the brightness data is first weight information. Before distributing the quantization error to pixels adjacent to the target pixel according to the preset weight information to obtain the first image, the display driving method provided in this application embodiment may further include steps S610 to S620.

[0101] S610, determine the gradient magnitude of the pixels adjacent to the target pixel based on the grayscale difference between the target pixel and its adjacent pixels; S620 determines the first weight information corresponding to the target pixel and its neighboring pixels based on the gradient magnitude.

[0102] Specifically, the process of nonlinear quantization error processing for the brightness channel may include: determining the quantization error of the target pixel based on the difference between the target brightness data of the target pixel in the original image and the preset quantization data; determining the gradient magnitude of the pixels adjacent to the target pixel based on the grayscale difference between the target pixel and its adjacent pixels; determining the first weight information corresponding to the target pixel based on the gradient magnitude; and allocating the quantization error to the pixels adjacent to the target pixel based on the first weight information to obtain the third sub-image.

[0103] refer to Figure 3 The quantization error of the target pixel is distributed to multiple pixels in its neighborhood (e.g., 4 pixels in the neighborhood). The first weight information corresponding to the target pixel can be a weight matrix, which includes the weight values ​​corresponding to the multiple pixels in the neighborhood of the target pixel.

[0104] The first weight information is dynamic, and the process of determining the first weight information can be as follows: Calculate the grayscale difference between the grayscale value of the target pixel and the grayscale values ​​of its neighboring pixels, and then obtain the gradient magnitude of the target pixel and its neighboring pixels based on the grayscale difference.

[0105] For example, the Sobel operator can be used to calculate the gradient magnitude of the target pixel and its neighboring pixels.

[0106] The Sobel operator uses two 3x3 convolution kernels to slide across the image, calculating the brightness changes along the horizontal and vertical directions, respectively.

[0107] The coordinates of the target pixel are (x,y), and the gray level value of the target pixel is f(x,y). The gradient component Gx in the horizontal direction can be calculated according to equation (1), the gradient component Gy in the vertical direction can be calculated according to equation (2), and the gradient magnitude M can be calculated according to equation (3).

[0108] (1) (2) (3) Where f(x-1,y) and f(x+1,y) represent the grayscale values ​​of the left and right adjacent pixels of the target pixel, respectively, and f(x,y-1) and f(x,y+1) represent the grayscale values ​​of the top and bottom adjacent pixels of the target pixel, respectively.

[0109] As an example, the convolution kernel for calculating the gradient component Gx in the horizontal direction is the convolution kernel shown in equation (4).

[0110] (4) The convolution kernel for calculating the gradient component Gy in the vertical direction is the convolution kernel shown in equation (5).

[0111] (5) Assuming the target pixel I 22 The 3x3 image block centered on the image is given by equation (6): (6) The expansion formula for the horizontal gradient component Gx is given by equation (7): (7) The expansion formula for the vertical gradient component Gy is given by equation (8): (8) For example, the grayscale values ​​of each pixel in equation (6) are as shown in equation (9): (6) Then Gx=0, Gy=80, M=80.

[0112] Gx=0 indicates that the image patch centered on the target pixel has no grayscale variation in the horizontal direction and no vertical edges. Gy=80 indicates that the image patch centered on the target pixel has significant grayscale variation in the vertical direction and has horizontal edges.

[0113] Gradient magnitude can be used to describe the degree of drastic change in the grayscale of a target pixel.

[0114] A larger gradient magnitude indicates a greater difference in grayscale between the target pixel and its neighboring pixels. In this case, the target pixel and its neighboring pixels may belong to high-detail areas such as object edges, text, and thin lines.

[0115] The smaller the gradient magnitude, the less variation there is in the grayscale of the target pixel and its neighboring pixels. In this case, the target pixel and its neighbors may belong to a flat gradient region such as the sky, walls, or a solid color background.

[0116] By relying on gradient magnitude, the image region can be objectively divided. The weight coefficient of nonlinear error diffusion in the brightness channel can be adaptively adjusted to achieve differentiated image processing that preserves edge sharpness and smooths noise reduction in flat areas, thus suppressing various image quality defects such as jagged edges, color banding, and periodic moiré artifacts caused by pixel displacement imaging.

[0117] Figure 7 This diagram illustrates a fifth flowchart of the display driving method provided in an embodiment of this application. In some embodiments, such as Figure 7 As shown, determining the first weight information of the target pixel and its neighboring pixels based on the gradient magnitude can include: S710: Based on the relationship between the gradient magnitude and a preset gradient threshold, determine the region type to which the target pixel and its neighboring pixels belong. The region type includes high-frequency regions and low-frequency regions. If the gradient magnitude is greater than the gradient threshold, the target pixel and its neighboring pixels belong to a high-frequency region. If the gradient magnitude is less than or equal to the gradient threshold, the target pixel and its neighboring pixels belong to a low-frequency region. S720 determines the first weight information of the target pixel and its neighboring pixels based on the region type to which the target pixel and its neighboring pixels belong.

[0118] The gradient magnitude is obtained by calculating a 3x3 local window centered on the target pixel. The gradient magnitude represents the degree of drastic change in grayscale.

[0119] In high-frequency regions, the gradient magnitude exceeds the gradient threshold. Pixel grayscale values ​​change drastically within the high-frequency region, and may contain high-frequency detail information such as object outlines, text, lines, and textures.

[0120] In low-frequency regions, the gradient magnitude is less than or equal to the gradient threshold. Pixel grayscale changes are smooth (or gradual) in low-frequency regions, including low-frequency content such as the sky, walls, and solid-color backgrounds, with almost no fine edge details.

[0121] For pixels in high-frequency and low-frequency regions, error diffusion weights can be designed differently. This can "preserve details" in high-frequency regions and "smooth out" low-frequency regions, allowing the error diffusion strategy to better align with the human eye's perception of visual defects and optimize image quality at the optimal position.

[0122] The gradient threshold can be adaptively adjusted based on the original image data or experimental results.

[0123] A fixed single gradient threshold cannot adapt to video footage with large differences in brightness and varying scene types. In this embodiment, the gradient threshold supports dynamic adjustment based on the global image average brightness, local image contrast, image content type, and offline experimental calibration results.

[0124] For example, for high-contrast images, the gradient threshold can be appropriately increased to prevent textures from being misidentified as edges. For low-brightness images with rich gradients, the threshold can be appropriately decreased to better identify flat areas and suppress color banding. Alternatively, the gradient threshold can be adaptively updated based on parameter ranges calibrated through multiple sets of experiments to improve the accuracy of region segmentation and avoid image quality distortion caused by mismatches in weighting strategies.

[0125] In some embodiments, the error diffusion weight of pixels in high-frequency regions is greater than that of pixels in low-frequency regions.

[0126] High-frequency regions exhibit significant grayscale abrupt changes, typically resulting in larger quantization error values. If an excessively small error diffusion weight is used, the error cannot be distributed promptly, accumulating locally at the edges and causing distortion in brightness and contour. In this embodiment, the error diffusion weight of pixels in high-frequency regions is larger, enabling the quantization error to be concentrated and distributed to adjacent pixels in the horizontal and vertical directions. This results in a higher overall error distribution concentration and a larger effective weight magnitude. This approach can quickly absorb the quantization error generated by the target pixel, accurately compensate for the jagged distortion caused by edge grayscale abrupt changes, and constrain the error to propagate only in the local neighborhood, preventing edge details from being blurred.

[0127] In low-frequency regions, grayscale smoothing is crucial because the human eye is extremely sensitive to minute brightness disturbances. If a large weight is used to concentrate the error distribution, it can lead to excessive brightness shifts between adjacent pixels, resulting in visible grain noise and uneven brightness. In this embodiment, the error diffusion weight for pixels in low-frequency regions is small, and the error disturbance received by a single pixel is weak, achieving grayscale smoothing without introducing significant visible noise.

[0128] In some embodiments, the pixels adjacent to the target pixel include a first pixel, a second pixel, and a third pixel. The first pixel is adjacent to the target pixel in the row direction, the second pixel is adjacent to the target pixel in the column direction, and the third pixel is adjacent to the target pixel in the diagonal direction. For example, refer to Figure 3 Pixel Px1 is the first pixel, pixel Px2 is the second pixel, and either pixel Px3 or Px4 is the third pixel.

[0129] When the regions to which the first, second, and third pixels belong are all low-frequency regions, the error diffusion weights for the first, second, and third pixels are equal. This "equal" weighting allows for a certain degree of error. In this case, the error diffusion weights of the target pixel are nearly uniform in all directions, allowing the quantization error of the target pixel to be evenly distributed among its neighboring pixels, thereby reducing overall noise.

[0130] Alternatively, if the region to which the first pixel belongs is a high-frequency region and the gradient magnitude of the first pixel is the largest, the error diffusion weight of the first pixel can be maximized. In this case, it can be assumed that there are horizontal edges, and the error diffusion weight of the target pixel along the left-right direction can be set to be larger, while the error diffusion weight along the up-down direction can be smaller, in order to avoid blurring of horizontal edge details.

[0131] Alternatively, if the region to which the second pixel belongs is a high-frequency region, and the gradient magnitude of the second pixel is the largest, then the error diffusion weight of the second pixel can be the largest. In this case, it can be assumed that there is a vertical edge, and the error diffusion weight of the target pixel along the vertical direction can be set to be larger, while the error diffusion weight along the horizontal direction can be smaller, in order to avoid blurring of vertical edge details.

[0132] Alternatively, if the region to which the third pixel belongs is a high-frequency region and the gradient magnitude of the third pixel is the largest, then the error diffusion weight of the third pixel can be the largest. In this case, it can be assumed that there are diagonal edges, and the error diffusion weight of the target pixel along the diagonal direction can be set to be larger, while the error diffusion weight along the vertical and horizontal directions can be smaller, in order to avoid blurring the details of the diagonal edges.

[0133] Figure 8 This diagram illustrates an example of error diffusion weight allocation provided in an embodiment of this application. As an example, such as... Figure 8 As shown, if the calculated gradient component Gy in the vertical direction is large and the gradient component Gx in the horizontal direction is small, it can be considered that a horizontal edge exists, and the image brightness and darkness change abruptly along the vertical direction, with the edge line running horizontally. The horizontal edge only extends along the horizontal direction, while the vertical direction is the boundary of the brightness and darkness transition. Therefore, the weight of the vertical edge position is reduced, and only the pixels below and to the right of the same edge side are assigned high weights to prevent the edge from being blurred by interpolation.

[0134] Specifically, for pixel Px2 directly below the target pixel, which is a pixel extending along the horizontal edge, a relatively large weight of 0.45 can be set to protect the edge sharpness. For pixels Px3 to the lower left and Px4 to the lower right of the target pixel, which are pixels crossing the edge, relatively small weights of 0.05 and 0.15 can be set to avoid blurring caused by pixels blending on both sides of the edge. For pixel Px1 to the right of the target pixel, which is a pixel in the same area as the horizontal edge, a medium weight of 0.35 can be set.

[0135] If the calculated vertical gradient component Gx is large and the horizontal gradient component Gy is small, it can be considered that a vertical edge exists, and the image brightness and darkness change abruptly along the horizontal direction, with the edge line running vertically. The vertical edge extends along the vertical direction, and the horizontal direction is the boundary of the brightness and darkness transition. Therefore, the weight of the right edge position is reduced, and the lower left and lower right pixels extending vertically downwards are assigned high weights. Interpolation is performed along the edge direction to preserve the sharpness of the vertical edge.

[0136] Specifically, for pixel Px2 directly below the target pixel, a medium weight of 0.20 can be set. For pixels Px3 to the lower left and Px4 to the lower right of the target pixel, which are pixels extending downwards from the vertical edge, relatively large weights of 0.25 and 0.45 can be set respectively. For pixel Px1 to the right of the target pixel, which is a pixel crossing the edge, a relatively small weight of 0.10 can be set.

[0137] It should be noted that, Figure 8 The values ​​shown are merely examples and are not intended to limit this application.

[0138] In summary, the error diffusion weight setting in this embodiment is linked to the region type (horizontal edge / vertical edge / diagonal). Based on the gradient magnitude, the error diffusion weight is dynamically allocated, thereby dynamically allocating the quantization error of the target pixel. This enables more accurate differentiation between high-frequency regions (such as texture, contour, and detail regions) and low-frequency regions (such as solid color and background regions). It ensures that errors in high-frequency regions are preferentially diffused to complementary sampling points, while errors in low-frequency regions are distributed more smoothly, avoiding large-area noise and grayscale distortion; improving high-resolution reproduction and suppressing directional artifacts.

[0139] Figure 9 This diagram illustrates a sixth flowchart of a display driving method provided in an embodiment of this application. In some embodiments, first weight information of a target pixel and its neighboring pixels is determined based on the gradient magnitude, including steps S910 to S920.

[0140] S910, based on the gradient magnitude, determines the second weight information of the pixels adjacent to the target pixel; S920 obtains the first weight information of pixels adjacent to the target pixel by correcting the second weight information based on visual sensitivity.

[0141] Visual sensitivity refers to the sensitivity of the human visual system (HVS) to images. A core characteristic of HVS is spatial frequency sensitivity. The human eye is extremely sensitive to low-frequency regions (such as smooth, solid colors), perceiving even minute errors. Conversely, the human eye is less sensitive to high-frequency regions (such as edges, textures, and details), tolerating larger errors.

[0142] For example, calculating the gradient magnitude using a 3x3 window yields the magnitudes of individual local gradients. These local gradient magnitudes only consider the target pixel and its neighboring pixels, resulting in a relatively small window that represents the local spatial frequency of the image.

[0143] The gradient magnitude can be mapped to the human visual sensitivity coefficient S based on the contrast sensitivity function (CSF) of the human visual system. HVS The second weight information obtained based on the gradient magnitude is the basic weight information, and the human visual sensitivity coefficient S is used. HVS The second weighting information can be corrected, for example, based on the human visual sensitivity coefficient S. HVS The second weight information is scaled to obtain the final first weight information.

[0144] The window for HVS is relatively large compared to the calculation window for gradient magnitude. For example, if most of the original image has a white background and only a small local window contains light-colored dots, without correcting the error diffusion weights, these light-colored dots would be identified as high-frequency areas, making them more noticeable. In this embodiment, the second weight information is corrected based on the visual sensitivity of HVS to obtain the first weight information. This no longer only performs edge protection for local areas but also constrains noise from a visual perception perspective. For example, it can eliminate obvious light-colored dots. The corrected first weight information makes the graininess of the final displayed image more comfortable.

[0145] In some embodiments, the second weight information is corrected based on visual sensitivity to obtain the first weight information of pixels adjacent to the target pixel, including: If the region type of the pixels adjacent to the target pixel is a high-frequency region, the error diffusion weight of the pixels adjacent to the target pixel in the second weight information is reduced; or, if the region type of the pixels adjacent to the target pixel is a low-frequency region, the error diffusion weight of the pixels adjacent to the target pixel in the second weight information is increased, thus obtaining the first weight information of the pixels adjacent to the target pixel.

[0146] If the target pixel's neighboring pixels belong to a high-frequency region, but the large global window to which the pixel belongs is a low-frequency region, then the error diffusion weight of the pixel at that location is reduced to suppress noise propagation.

[0147] If the target pixel's neighboring pixels belong to a low-frequency region, but the large global window to which the pixel belongs is a high-frequency region, then the error diffusion weight of the pixel at that location is increased to suppress noise propagation outward.

[0148] This embodiment captures localized noise details using a small gradient window and constrains the global background using a large HVS window, avoiding misclassifying isolated noise points on a pure white background as valid texture edges. Furthermore, isolated bright spots are not amplified and diffused due to high-frequency detection; the large window background constraint compresses weights, eliminating background noise.

[0149] The above example, using pixel image data including luminance data, illustrates some implementation methods for error diffusion processing of the luminance channel.

[0150] The following section uses pixel-level image data, including chroma data, as an example to introduce some implementation methods for error diffusion processing of the chroma channel. Linear error diffusion processing can be performed on the chroma channel.

[0151] In some embodiments, the image data of a pixel includes chroma data. A second error diffusion process is performed on the chroma data of the pixels in the original image to obtain a fourth sub-image. The second error diffusion process includes at least two pixels having the same error diffusion weight in the same direction.

[0152] For example, with Figure 2 Taking image 2,2 as an example, for the chromaticity data of pixels P1 and P2, the error diffusion weights of pixels P1 and P2 in the same direction can be the same.

[0153] Error diffusion processing of chroma data can be performed on the target pixel to its right, lower left, lower right, and lower right. The error diffusion weights of pixels P1 and P2 to their right are equal, the error diffusion weights of pixels P1 and P2 to their lower left are equal, the error diffusion weights of pixels P1 and P2 to their lower right are equal, and the error diffusion weights of pixels P1 and P2 to their lower right are equal.

[0154] The human eye is less sensitive to the chroma channel than to the luminance channel, and has lower requirements for the spatial resolution of chroma. By adopting a linear error diffusion processing method with fixed weights, we can avoid directional color shift and color noise introduced by dynamic adjustment of the weights of the chroma channel, ensure uniform and stable color transition, reduce the amount of calculation, and improve efficiency.

[0155] The luminance channel employs nonlinear error diffusion processing, and the error diffusion weights can be adaptively and dynamically adjusted. Combined with HVS-corrected error diffusion weights, it can suppress noise in flat areas, protect edge sharpness, and eliminate isolated light-colored bright spots on a white background.

[0156] In some embodiments, if the chromaticity data of a pixel in the original image is greater than a preset chromaticity threshold, a second error diffusion process is performed on the chromaticity data of the pixel in the original image to obtain a fourth sub-image; or, if the chromaticity data of a pixel in the original image is less than or equal to the chromaticity threshold, no error diffusion is performed on the chromaticity data of the pixel in the original image.

[0157] In the chroma channel, since the human eye is less sensitive to changes in chroma than to changes in luminance, the spatial resolution requirement for chroma is relatively low. In this embodiment, error diffusion processing is performed only when the original chroma value of a pixel is greater than the chroma threshold; otherwise, error diffusion processing is not performed, which can save computational resources and improve efficiency.

[0158] Figure 10 This diagram illustrates a seventh flowchart of a display driving method provided in an embodiment of this application. In some embodiments, such as Figure 10 As shown, error diffusion processing is performed on the image data of pixels in the original image to obtain the first image, which may include S1011 and S1012: S1011, Based on the image type of the original image and the pre-defined mapping relationship between multiple image types and multiple error diffusion algorithms, determine the target error diffusion algorithm corresponding to the original image.

[0159] S1012, based on the target error diffusion algorithm, performs error diffusion processing on the image data of pixels in the original image to obtain the first image.

[0160] Image types include, but are not limited to, images of people, landscapes, night scenes, and architecture.

[0161] Multiple error propagation algorithms include Floyd-Steinberg, Jarvis-Judice-Ninke, Stucki, Burkes, and others.

[0162] The mapping relationship between multiple image types and multiple error diffusion algorithms includes: the optimal error diffusion algorithm corresponding to each image type.

[0163] For example, the optimal error diffusion algorithm for images of people is the Floyd-Steinberg error diffusion algorithm, the optimal error diffusion algorithm for images of landscapes is the Jarvis-Judice-Ninke error diffusion algorithm, the optimal error diffusion algorithm for images of buildings is the Stucki error diffusion algorithm, and the optimal error diffusion algorithm for images of night scenes is the Burkes error diffusion algorithm.

[0164] If the original image is a portrait, the Floyd-Steinberg error diffusion algorithm can be selected from the mapping relationship to perform error diffusion processing. If the original image is a landscape, the Jarvis-Judice-Ninke error diffusion algorithm can be selected from the mapping relationship to perform error diffusion processing. If the original image is an architectural image, the Stucki error diffusion algorithm can be selected from the mapping relationship to perform error diffusion processing. If the original image is a night scene, the Burkes error diffusion algorithm can be selected from the mapping relationship to perform error diffusion processing.

[0165] Of course, the correspondence between the various image types and algorithms mentioned above are merely examples and are not intended to limit this application.

[0166] For example, the optimal error propagation algorithm for each image type distribution can be determined in the following manner: Prepare test images of multiple image types (such as images with different content features, such as portraits and landscapes); For any type of test image, different error diffusion algorithms are used for processing; The outputs of each error diffusion algorithm are compared with the reference standard image to calculate objective quality indicators such as PSNR (Peak Signal-to-Noise Ratio) and MSE (Mean Square Error). Subjective evaluations (such as directional accuracy, geometric consistency, and visual naturalness) can also be performed. The optimal error diffusion algorithm for this type of image is selected based on a combination of objective quality indicators (such as selecting the highest PSNR and the lowest MSE) and subjective evaluation. Establish a mapping relationship between various image types and the optimal error diffusion algorithm, and support adaptive selection of the optimal error diffusion algorithm for different image contents.

[0167] In this embodiment, the optimal error diffusion algorithm corresponding to the original image is selected from multiple error diffusion algorithms based on the image type of the original image, which can further improve the final display effect.

[0168] In some embodiments, different error diffusion algorithms cover different error diffusion windows, and / or, the underlying weight distribution of a single error diffusion window differs among different error diffusion algorithms.

[0169] Figure 11 This diagram illustrates the coverage and weight distribution of various error propagation algorithms provided in embodiments of this application. Figure 11 In the middle, x represents the current pixel, and the other values ​​represent the base weights.

[0170] like Figure 11 As shown, the Floyd-Steinberg algorithm covers the error diffusion window of the current pixel and the three pixels in the next row, with basic weight distributions of 7 / 16, 3 / 16, 5 / 16, and 1 / 16, respectively. The characteristics of the Floyd-Steinberg algorithm include: suitability for portrait images, small diffusion range, smooth and delicate skin transitions, no large-area grainy noise, natural skin tone, and moderate detail preservation.

[0171] The Jarvis-Judice-Ninke algorithm covers the error diffusion window of the current pixel, specifically the two pixels to its right and four pixels in each of the next two rows. The basic weight distributions are: 7 / 48, 5 / 48, 3 / 48, 5 / 48, 7 / 48, 5 / 48, 3 / 48, 1 / 48, 3 / 48, 5 / 48, 3 / 48, 1 / 48. The characteristics of the Jarvis-Judice-Ninke algorithm include: suitability for landscape images, wide diffusion range, uniform distribution of large-area gradient noise, and good suppression of moiré patterns.

[0172] The Stucki algorithm covers the error diffusion window of the two pixels to the right of the previous pixel and four pixels in each of the next two rows, with the following basic weight distributions: 8 / 42, 4 / 42, 2 / 42, 4 / 42, 8 / 42, 4 / 42, 2 / 42, 1 / 42, 2 / 42, 4 / 42, 2 / 42, 1 / 42. The Stucki algorithm's characteristics include: suitability for architectural images, strengthening edge direction weights to produce sharp lines in close-up architectural images, while smoothing gradient areas.

[0173] The Burkes algorithm covers an error diffusion window that extends to the two pixels to the right of the current pixel and the five pixels in the next row, with basic weight distributions of 8 / 32, 4 / 32, 2 / 32, 4 / 32, 8 / 32, 4 / 32, and 2 / 32. The Burkes algorithm is characterized by its suitability for night scene images and excellent masking of highlight and shadow noise.

[0174] It should be noted that the display driving method provided in this application embodiment can be applied to projection devices.

[0175] Based on the same technical concept, embodiments of this application also provide a display device. Figure 12 This diagram illustrates a structural schematic of a display device provided in an embodiment of this application. Figure 12 As shown, the display device 100 provided in this application embodiment includes an image processing component 10, a pixel displacement component 21, and a display driving component 22.

[0176] Image processing component 10 is configured to: acquire the original image to be displayed; and perform error diffusion processing on the image data of pixels in the original image to obtain a first image. Specifically, at least two pixels have different error diffusion weights in the same direction. Image processing component 10 essentially embeds the error diffusion processing function.

[0177] The image processing component 10 is further configured to: split the first image into a first sub-image and a second sub-image, wherein the first sub-image and the second sub-image have the same number of pixels, and the number of pixels in the first sub-image is less than the number of pixels in the original image, and the pixels of the first sub-image and the pixels of the second sub-image do not overlap in position.

[0178] Display module 20 is configured to display the first sub-image and the second sub-image sequentially.

[0179] According to the display device provided in this application embodiment, a high-resolution original image is acquired, nonlinear error diffusion processing is performed on the original image, and the image after non-error diffusion processing is split into a low-resolution first sub-image and a second sub-image. There is a pixel offset between the first sub-image and the second sub-image, and the first sub-image and the second sub-image are displayed alternately. This makes the first sub-image and the second sub-image arranged in an interlaced complementary manner in two-dimensional space. The pixels of the two sets of sub-images uniformly cover all pixel sampling grids corresponding to the high-resolution original image. Relying on the persistence of vision of the human eye, two temporally consecutive frames are visually fused. The sparse two-way low-resolution pixel sampling reconstructs complete high-density image information, ultimately achieving an equivalent high-resolution display effect. In addition, nonlinear error diffusion processing can optimize the distribution of complementary sampling points, making the effective pixel density of the superimposed first sub-image and the second sub-image closer to the real high-resolution image. For example, when processing certain image content (such as sharp edges and fine textures), nonlinear error diffusion processing can better utilize the information gain brought by displacement, making the final displayed image more realistic. In addition, this application does not require modification of the pixel resolution of existing display devices; it only needs to embed the error diffusion processing function into the display device, which can reduce costs.

[0180] In some embodiments, such as Figure 13 As shown, the display device also includes: The projection screen 30 is configured to receive a first sub-image and a second sub-image from the display module 20.

[0181] Figure 13 This illustration shows another structural diagram of the display device provided in an embodiment of this application. As an example, such as... Figure 13As shown, the image processing component 10 may include a central processing unit (CPU), a graphics processing unit (GPU), and a display driver integrated circuit (DDIC).

[0182] The central processing unit (CPU) acquires the original image to be displayed; it performs error diffusion processing on the pixel data of the original image to obtain a first image. At least two pixels have different error diffusion weights in the same direction. The first image is then divided into a first sub-image and a second sub-image. The first sub-image and the second sub-image have the same number of pixels, but the number of pixels in the first sub-image is less than the number of pixels in the original image. The pixels in the first sub-image and the pixels in the second sub-image do not overlap in position.

[0183] The central processing unit (CPU) sends the processed first and second sub-images to the image processor (GPU). The GPU performs relevant processing (such as rendering) on ​​the first and second sub-images respectively. The GPU then sends the processed first and second sub-images to the display driver chip (DDIC). The DDIC drives the display module to display the first and second sub-images sequentially.

[0184] The display module 20 may include a display panel 21, a polarization modulator 22, and an optical anisotropy layer 23.

[0185] The display panel 21 can be a low-pixel-density polarized light display panel, such as a regular LCD display panel or an OLED display panel equipped with a circular polarizer. The image processing component 10 incorporates an error diffusion processing function. The display panel 21 displays a first sub-image in one frame and a second sub-image in another frame. The image light emitted by the display panel 21 first passes through the polarization modulator 22, and the polarization state is rotated by 90° between adjacent frames. The polarization modulator can be a fast-response liquid crystal cell, operating in electrically controlled birefringence (ECB) or twisted nematic (TN) mode. Subsequently, the time-division modulated image light passes through the optical anisotropy layer 23 (the optical axis is tilted relative to the surface), and the anomalous light (e-ray) is deflected and shifted relative to the ordinary light (o-ray); finally, half a frame (e.g., Figure 2 The B-frame image (FB) is displayed in its original position, and half a frame (e.g.) is displayed in its original position. Figure 2 The A-frame image (FA) is displayed at the shifted position, and a new image is synthesized with double the spatial resolution and halved the temporal resolution.

[0186] This application also provides an electronic device. Figure 14 A schematic diagram of the hardware structure of the electronic device provided in an embodiment of this application is shown.

[0187] The electronic device 800 may include a processor 801 and a memory 802 storing computer program instructions.

[0188] Specifically, the processor 801 may include a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of the present invention.

[0189] Memory 802 may include mass storage for data or instructions. For example, and not limitingly, memory 802 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 802 may include removable or non-removable (or fixed) media. Where appropriate, memory 802 may be internal or external to the integrated gateway disaster recovery device. In a particular embodiment, memory 802 is non-volatile solid-state memory.

[0190] In a particular embodiment, memory 802 includes read-only memory (ROM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically rewritable ROM (EAROM), or flash memory, or a combination of two or more of these. Exemplarily, the memory may include non-volatile transient memory.

[0191] The processor 801 implements any of the display driving methods described in the above embodiments by reading and executing computer program instructions stored in the memory 802.

[0192] In one example, the electronic device 800 may further include a communication interface 803 and a bus 810. For example, Figure 8 As shown, the processor 801, memory 802, and communication interface 803 are connected through bus 810 and complete communication with each other.

[0193] The communication interface 803 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of the present invention.

[0194] Bus 810 includes hardware, software, or both, that couples components of an electronic device together. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 810 may include one or more buses. While specific buses are described and illustrated in embodiments of the invention, the invention contemplates any suitable bus or interconnect.

[0195] For example, electronic device 800 may be a display device, etc.

[0196] This application also provides a computer-readable storage medium storing a computer program. When executed by a processor, the computer program can implement the display driving method described in the above embodiments and achieve the same technical effect. To avoid repetition, it will not be described again here. The computer-readable storage medium may include read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk, etc., and is not limited thereto.

[0197] This application also provides a computer program product, which includes computer program instructions. When the computer program instructions are executed by a processor, they implement the display driving method as described in the above embodiments.

[0198] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0199] The functional blocks shown in the above-described structural diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Computer-readable medium" can include any medium capable of storing or transmitting information. Examples of computer-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0200] According to embodiments of this application, the computer-readable storage medium may be a non-transitory computer-readable storage medium.

[0201] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0202] The aspects of this application have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by dedicated hardware performing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0203] The embodiments described above are not exhaustive, nor do they limit the application to the specific embodiments described herein. Clearly, many modifications and variations can be made based on the above description. These embodiments are selected and specifically described in this specification to better explain the principles and practical applications of this application, thereby enabling those skilled in the art to effectively utilize this application and its modifications. This application is limited only by the claims and their full scope and equivalents.

Claims

1. A display driving method, characterized in that, include: Get the original image to be displayed; Error diffusion processing is performed on the image data of pixels in the original image to obtain a first image, wherein at least two pixels have different error diffusion weights in the same direction; The first image is split into a first sub-image and a second sub-image. The first sub-image and the second sub-image have the same number of pixels, and the number of pixels in the first sub-image is less than the number of pixels in the original image. The pixels of the first sub-image and the pixels of the second sub-image do not overlap in position. The first sub-image and the second sub-image are displayed sequentially.

2. The method according to claim 1, characterized in that, The step of performing error diffusion processing on the image data of pixels in the original image to obtain a first image includes: The quantization error of the target pixel is determined based on the difference between the target image data of the target pixel in the original image and the preset quantization data, wherein the target pixel is any pixel in the original image; The quantization error is assigned to pixels adjacent to the target pixel according to preset weight information to obtain a first image, wherein the scanning time of the pixels assigned the quantization error and adjacent to the target pixel is later than the scanning time of the target pixel.

3. The method according to claim 1, characterized in that, The image data includes luminance data and chrominance data. The step of performing error diffusion processing on the image data of pixels in the original image to obtain a first image includes: A first error diffusion process is performed on the brightness data of pixels in the original image to obtain a third sub-image, wherein the first error diffusion process includes at least two pixels having different error diffusion weights in the same direction; A second error diffusion process is performed on the chromaticity data of the pixels in the original image to obtain a fourth sub-image; The first image is obtained based on the third sub-image and the fourth sub-image.

4. The method according to claim 2, characterized in that, The image data includes brightness data, the gray level of the target pixel in the original image is any one of gray levels from 1 to 244, and the quantization data is 128 gray levels; Alternatively, the target pixel in the original image has a gray level of 255, and the quantization data has a gray level of 255.

5. The method according to claim 2, characterized in that, The target pixel is the pixel located in the first row and first column, and the target image data of the target pixel is the original image data of the target pixel in the original image; Alternatively, the target pixel may be a pixel located outside the first row and first column, and the target image data of the target pixel may be the sum of the original image data of the target pixel in the original image and the quantization error assigned to it.

6. The method according to claim 2, characterized in that, The image data includes brightness data, and the weight information corresponding to the brightness data is first weight information. Before distributing the quantization error to pixels adjacent to the target pixel according to the preset weight information to obtain the first image, the method further includes: Based on the grayscale difference between the target pixel and its neighboring pixels, determine the gradient magnitude of the pixels adjacent to the target pixel; Based on the gradient magnitude, the first weight information corresponding to the target pixel and its neighboring pixels is determined.

7. The method according to claim 6, characterized in that, Determining the first weight information of the target pixel and its neighboring pixels based on the gradient magnitude includes: Based on the relationship between the gradient magnitude and a preset gradient threshold, the region type to which the target pixel and its neighboring pixels belong is determined; wherein, the region type includes high-frequency regions and low-frequency regions, if the gradient magnitude is greater than the gradient threshold, then the region type to which the target pixel and its neighboring pixels belong is the high-frequency region; if the gradient magnitude is less than or equal to the gradient threshold, then the region type to which the target pixel and its neighboring pixels belong is the low-frequency region. The first weight information of the target pixel and its neighboring pixels is determined based on the region type to which the target pixel and its neighboring pixels belong.

8. The method according to claim 7, characterized in that, The error diffusion weight of pixels in the high-frequency region is greater than that of pixels in the low-frequency region.

9. The method according to claim 7, characterized in that, The pixels adjacent to the target pixel include a first pixel, a second pixel, and a third pixel. The first pixel is adjacent to the target pixel in the row direction, the second pixel is adjacent to the target pixel in the column direction, and the third pixel is adjacent to the target pixel in the diagonal direction. The first pixel, the second pixel, and the third pixel all belong to the low-frequency region, and the error diffusion weights of the first pixel, the second pixel, and the third pixel are equal. Alternatively, the region type to which the first pixel belongs is the high-frequency region, and the gradient magnitude of the first pixel is the largest, and the error diffusion weight of the first pixel is the largest. Alternatively, the region type to which the second pixel belongs is the high-frequency region, and the gradient magnitude of the second pixel is the largest, and the error diffusion weight of the second pixel is the largest. Alternatively, the region type to which the third pixel belongs is the high-frequency region, and the gradient magnitude of the third pixel is the largest, and the error diffusion weight of the third pixel is the largest.

10. The method according to claim 7, characterized in that, Determining the first weight information of the target pixel and its neighboring pixels based on the gradient magnitude includes: The second weight information of the pixels adjacent to the target pixel is determined based on the gradient magnitude. The second weight information is corrected based on visual sensitivity to obtain the first weight information of the pixels adjacent to the target pixel.

11. The method according to claim 10, characterized in that, The step of correcting the second weight information based on visual sensitivity to obtain the first weight information of pixels adjacent to the target pixel includes: If the region type of the pixels adjacent to the target pixel is the high-frequency region, the error diffusion weight of the pixels adjacent to the target pixel in the second weight information is reduced; or, if the region type of the pixels adjacent to the target pixel is the low-frequency region, the error diffusion weight of the pixels adjacent to the target pixel in the second weight information is increased, thereby obtaining the first weight information of the pixels adjacent to the target pixel.

12. The method according to claim 3, characterized in that, The second error diffusion process includes at least two pixels having the same error diffusion weight in the same direction.

13. The method according to claim 3, characterized in that, If the chromaticity data of a pixel in the original image is greater than a preset chromaticity threshold, a second error diffusion process is performed on the chromaticity data of the pixel in the original image to obtain a fourth sub-image. Alternatively, if the chromaticity data of a pixel in the original image is less than or equal to the chromaticity threshold, error diffusion is not performed on the chromaticity data of the pixel in the original image.

14. The method according to claim 1, characterized in that, The step of performing error diffusion processing on the image data of pixels in the original image to obtain a first image includes: Based on the image type of the original image and the preset mapping relationship between multiple image types and multiple error diffusion algorithms, the target error diffusion algorithm corresponding to the original image is determined; Based on the target error diffusion algorithm, the image data of pixels in the original image are subjected to error diffusion processing to obtain the first image.

15. The method according to claim 14, characterized in that, Different error diffusion algorithms cover different error diffusion windows, and / or the weight distribution of a single error diffusion window differs among different error diffusion algorithms.

16. The method according to claim 1, characterized in that, The number of pixels in the original image is four times the number of pixels in the first sub-image.

17. The method according to claim 1, characterized in that, The pixels of the second sub-image are offset diagonally relative to the pixels of the first sub-image.

18. The method according to claim 17, characterized in that, The pixels of the second sub-image are offset by half a pixel in the row direction relative to the pixels of the first sub-image, and the pixels of the second sub-image are offset by half a pixel in the column direction relative to the pixels of the first sub-image.

19. A display device, characterized in that, include: The image processing component is configured to: acquire the original image to be displayed; Error diffusion processing is performed on the image data of pixels in the original image to obtain a first image, wherein at least two pixels have different error diffusion weights in the same direction; the first image is split into a first sub-image and a second sub-image, the first sub-image and the second sub-image have the same number of pixels, and the number of pixels in the first sub-image is less than the number of pixels in the original image, and the pixels in the first sub-image and the pixels in the second sub-image do not overlap in position; The display module is configured to display the first sub-image and the second sub-image sequentially.

20. The display device according to claim 19, characterized in that, Also includes: The projection screen is configured to receive the first sub-image and the second sub-image from the display module.

Images