Three-dimensional chart generation method and system

Abstract

The invention relates to the technical field of remote sensing surveying and mapping, in particular to a three-dimensional chart generation method and system, and the method comprises the steps: obtaining satellite images at different shooting angles, processing the satellite images to generate engineering data, and opening the engineering data as a three-dimensional engineering image; performing screenshot operation on the three-dimensional engineering image, and splicing the multimedia sub-images based on the edge pixel information of the multimedia sub-images to obtain a multimedia three-dimensional image; acquiring RGB (Red, Green and Blue) data of each pixel point in the multimedia stereo image and integrating the RGB data into a control group; performing printing color mode conversion on the multimedia stereo image to generate a to-be-printed image in a CMYK format, and integrating CMYK data corresponding to each pixel point into a processing group; adjusting the CMYK data of each pixel point in the processing group to obtain a visual action evaluation reward; and generating an adjustment plan corresponding to the highest reward, and printing the to-be-printed image. The method and the device have the effects of exporting the three-dimensional engineering file and reducing the influence on the three-dimensional effect during printing.

Description

Technical Field

[0001] The present application relates to the technical field of remote sensing mapping, and in particular to a method and system for generating a stereoscopic plate. Background Art

[0002] Remote sensing refers to all non-contact long-distance detection technologies. Remote sensing is a technology that combines inductive telemetry of the earth's surface with monitoring of resource management (such as trees, grass, soil, water, minerals, crops, fish and wildlife, etc.) through telemetry instruments on artificial satellites, aviation and other platforms.

[0003] At present, in order to improve the image display effect and facilitate the viewing of surveying and mapping results by industry personnel, many industry personnel will convert the remote sensing images taken by remote sensing satellites from 2D display mode to 3D display mode. At the same time, in some publicity, project activities and other occasions, in order to realize and popularize remote sensing visualization, they often cooperate with various cities and scenic spots to display 3D remote sensing images in front of tourists, so as to achieve the effect of remote sensing viewing of the world.

[0004] In the current stereoscopic image conversion technology, the obtained stereoscopic images are displayed through a computer or mobile display screen. When used in publicity and event occasions, the stereoscopic images are often required to be displayed in the form of paper brochures, paper posters and other paper plates. Although the existing surveying and mapping software can convert satellite images into stereoscopic images, it can only display the stereoscopic images on the display screen based on its engineering files. It does not have the function of exporting stereoscopic images based on engineering files, and it is also impossible to directly print the stereoscopic images in multimedia media as paper plates. Summary of the invention

[0005] In order to realize the export of three-dimensional engineering files and reduce the impact on the three-dimensional effect during printing, the present application provides a three-dimensional plate generation method and system.

[0006] In the first aspect, the present application provides a method for generating a three-dimensional plate, which adopts the following technical solution: A method for generating a three-dimensional plate comprises the following steps: Obtain satellite images from different shooting angles, perform preprocessing, color channel processing, and feature point geometric relationship matching on the satellite images to generate engineering data; Opening the engineering data to generate a three-dimensional engineering image; Performing a screenshot operation on the stereoscopic engineering image to derive a plurality of multimedia sub-images, and splicing the multimedia sub-images based on edge pixel information of the multimedia sub-images to obtain a multimedia stereoscopic image; Obtaining RGB data of each pixel in the multimedia stereoscopic image and integrating them into a control group; Converting the multimedia 3D image into a printing color mode to generate a to-be-printed image in CMYK format, and integrating the CMYK data corresponding to each pixel into a processing group; Performing pixel mapping between the treatment group and the control group and adjusting the CMYK data of each pixel in the treatment group to obtain a visual action evaluation reward, wherein the visual action evaluation reward is characterized by a score corresponding to the color of the pixel approaching or moving away from the color of the RGB data after the adjustment action; An adjustment plan corresponding to the highest reward is generated, and after the adjustment of each CMYK data of the processing group is completed, the image to be printed is saved and printed.

[0007] In some embodiments, the satellite image is preprocessed, the color channel is processed, and the geometric relationship of the feature points is matched to generate engineering data, which includes the following steps: Acquire a first panchromatic image and a first multispectral image of a first satellite, and fuse and mosaic them after radiation correction and atmospheric correction respectively to obtain a first satellite image at a first surveying and mapping position; Acquire a second panchromatic image and a second multispectral image of a second satellite, and fuse and mosaic them after radiation correction and atmospheric correction respectively to obtain a second satellite image at a second surveying and mapping position; Respectively obtain pixel RGB data corresponding to the first satellite image and the second satellite image, select one of them to remove the red channel value and select the other to remove the blue channel value and the green channel value; Aerial triangulation is performed based on the pixel feature information on the first satellite image and the pixel feature information on the second satellite image to determine the geometric relationship of the feature points and merge them to obtain the engineering data.

[0008] In some of the embodiments, before selecting one of the images to remove the red channel value and selecting another image to remove the blue channel value and the green channel value, the following steps are also included: Black and white RGB data are added to each of the pixel points on the first satellite image and the second satellite image as a gain parameter.

[0009] In some embodiments, taking a screenshot of the stereo engineering image to derive a plurality of multimedia sub-images, and splicing the multimedia sub-images based on edge pixel information of the multimedia sub-images to obtain a multimedia stereo image comprises the following steps: Obtaining the number of horizontal pixels and the number of vertical pixels of the three-dimensional engineering image to calculate the total number of pixels; Performing geometric segmentation on the horizontal pixel points of the stereo engineering image, performing geometric segmentation on the vertical pixel points of the stereo engineering image to obtain a plurality of segmentation calibration images, and labeling the plurality of segmentation calibration images; Taking screenshots based on the edges of the segmented and calibrated image and exporting them to obtain a plurality of the multimedia sub-images; sorting the plurality of multimedia sub-images based on the labels, and verifying the splicing matching degree between adjacent multimedia sub-images; When the stitching matching degree is higher than a threshold, adjacent multimedia sub-images are stitched together to obtain the multimedia stereoscopic image.

[0010] In some embodiments, sorting the plurality of multimedia sub-images based on the labels and verifying the splicing matching degree between adjacent multimedia sub-images includes the following steps: Obtaining the number of horizontal pixels and the number of vertical pixels of each multimedia sub-image to calculate the number of sub-pixels, and determining whether the number of sub-pixels of each multimedia sub-image matches a reference value, wherein the reference value is represented by the total number of pixels divided by the total number of labels; The matched multimedia sub-images are defined as valid screenshots, the unmatched multimedia sub-images whose number is greater than the reference value are defined as screenshots to be optimized, and the unmatched multimedia sub-images whose number is less than the reference value are defined as invalid screenshots; Placing the valid screenshot and the screenshot to be optimized in corresponding areas based on the labels, obtaining the labels of the invalid screenshots and re-screaming the corresponding areas on the stereo engineering image until the invalid screenshots are changed into the valid screenshots or the screenshots to be optimized; The edge pixel information of each of the valid screenshots and each of the screenshots to be optimized is obtained, wherein the edge pixel information of the valid screenshots includes a layer of pixels extending inward from the edge of the image, and the edge pixel information of the screenshots to be optimized includes several layers of pixels extending inward from the edge of the image.

[0011] In some of the embodiments, sorting the plurality of multimedia sub-images based on the labels and verifying the splicing matching degree between adjacent multimedia sub-images further includes the following steps: Performing splicing optimization to compare whether there is overlapping data in the edge pixel information of adjacent multimedia sub-images; If so, eliminating the pixel points in one of the multimedia sub-images that are identical to the overlapped data; Pre-stitching the plurality of multimedia sub-images after the splicing optimization to obtain a pre-image, obtaining the number of horizontal pixels and the number of vertical pixels of the pre-image and comparing them with the total number of pixels, and generating a first matching coefficient based on the comparison result; Determine the number of overlapping edges and the number of overlapping points of each of the multimedia sub-layers having the overlapping data, and calculate a second matching coefficient based on the number of overlapping edges and the number of overlapping points, wherein the number of overlapping edges is represented by the number of edges having overlapping data among the four sides of the multimedia sub-layer, and the number of overlapping points is represented by the number of pixel points in the multimedia sub-layer corresponding to the overlapping data; The splicing matching degree of each of the multimedia sub-layers is calculated based on the first matching coefficient and the second matching coefficient.

[0012] In some embodiments, pixel mapping is performed between the treatment group and the control group and the CMYK data of each pixel in the treatment group is adjusted to obtain a visual action evaluation reward, comprising the following steps: Exporting the full-size multimedia stereoscopic image and the image to be printed respectively, and overlapping and fusing them based on pixel positions to generate a fused image; Acquire RGB data of each pixel in the fused image and integrate them into an evaluation group; Using the RGB data in the control group as a target state, and retrieving a color difference list based on the target state to generate a reward range; An adjustment action is performed on the data in the processing group to obtain whether the new state of the RGB value of each pixel point in the evaluation group after the adjustment action is within the reward range to generate a reward or penalty.

[0013] In some embodiments, respectively exporting the full-size multimedia stereoscopic image and the image to be printed, and overlapping and fusing them based on pixel positions to generate a fused image, comprises the following steps: Setting the multimedia stereoscopic image as the background and the image to be printed as the foreground; respectively setting an Alpha blending coefficient and allocating it to the background and the foreground, wherein the Alpha blending coefficient of the background is equal to 1, and the Alpha blending coefficient of the foreground is less than 1; In the foreground, corresponding reflectivity and refraction index are generated based on the alpha blending coefficient.

[0014] In some embodiments, generating an adjustment plan corresponding to the highest reward, and saving the image to be printed and printing it after the CMYK data of the processing group are adjusted completely, includes the following steps: After the adjustment plan is generated, the total color difference between the RGB values ​​of each pixel in the evaluation group and the RGB values ​​of each pixel in the control group is not greater than a preset value; After the adjustment of each CMYK data of the processing group is completed, the current image to be printed is saved and exported for printing.

[0015] In a second aspect, the present application provides a three-dimensional plate generation system, which adopts the following technical solution: A three-dimensional plate generation system is used to implement the above method.

[0016] The technical solution provided by the embodiment of the present application has the following technical effects: Stereoscopic engineering images are generated by remote sensing satellites combined with industry software. In order to solve the problem that engineering images cannot be directly exported, the engineering images are indirectly converted and exported into stereoscopic images in multimedia picture formats through screenshots and splicing. In order to print the stereoscopic images displayed on the monitor into paper-based stereoscopic plates, while ensuring the impact of the stereoscopic presentation effect due to the color format conversion, RGB data with known stereoscopic effects is used as the control group, and the parameters of the data converted to CMYK are used as the processing group for callback to approach the color display effect of the RGB image before printing. The stereoscopic images can be printed as paper-based stereoscopic plates in turn, which can be used as a medium for publicity and explanation in enterprises, scenic spots, and attractions. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic diagram of the steps of a method for generating a three-dimensional plate provided in this embodiment.

[0018] Figure 2 It is a flow chart of the method for generating a three-dimensional plate provided in an embodiment of the present application. DETAILED DESCRIPTION

[0019] To more clearly understand the purpose, technical solutions and advantages of the present application, the present application is described and illustrated below in conjunction with the accompanying drawings and embodiments. However, it should be understood by those of ordinary skill in the art that the present application can be implemented without these details. In some cases, in order to avoid unnecessary descriptions that make various aspects of the present application obscure, well-known methods, processes, systems, components and / or circuits that have been described at a higher level will not be described in detail. For those of ordinary skill in the art, it is obvious that various changes can be made to the embodiments disclosed in the present application, and without departing from the principles and scope of the present application, the general principles defined in the present application can be applied to other embodiments and application scenarios. Therefore, the present application is not limited to the embodiments shown, but conforms to the broadest scope consistent with the scope claimed for protection of the present application.

[0020] It should be noted that the description of these embodiments is used to help understand the present invention, but does not constitute a limitation of the present invention. In addition, the technical features involved in each embodiment of the present invention described below can be combined with each other as long as there is no conflict between them.

[0021] In the description of this application, "several" means one or more, "more" means more than two, "greater than", "less than", "exceed", etc. are understood to exclude the number itself, and "above", "below", "within", etc. are understood to include the number itself. If there is a description of "first" or "second", it is only used to distinguish the technical features, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the order of the indicated technical features.

[0022] In the description of the present application, the description with reference to the terms "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples.

[0023] like Figure 1 and Figure 2 As shown, the embodiment of the present application discloses a method for generating a three-dimensional plate, comprising the following steps: S100, obtaining satellite images at different shooting angles, performing pre-processing, color channel processing and feature point geometric relationship matching on the satellite images to generate engineering data.

[0024] First, satellite images of the same object taken at different angles and in the same orbit are obtained, and the satellite images at different shooting angles are used as basic data.

[0025] After obtaining the satellite image corresponding to the basic data, the pixel point data of the satellite image is first preprocessed, the color data is processed, and the geometric relationship of different spatial parameters is processed and matched based on the remote sensing satellite image processing software corresponding to the industry to generate engineering data. The engineering data is represented by the data of the engineering software itself corresponding to the fusion of two basic data to obtain three-dimensional data.

[0026] S200, opening engineering data to generate a three-dimensional engineering image.

[0027] The engineering data is combined with engineering software to open it to obtain an engineering image with a three-dimensional effect. The three-dimensional engineering image here is used by industry surveying and mapping personnel to perform surveying and mapping related work through remote sensing images displayed in three-dimensional form. Subsequently, they can only export the engineering data. If they want to continue to see the three-dimensional effect, they need to import the engineering data into the corresponding engineering software for display. They cannot directly export and display the three-dimensional engineering image.

[0028] Therefore, a series of subsequent processing is required to convert the engineering images that cannot be directly exported into conventional multimedia image formats that can be directly opened, such as PNG, JPG and other formats. Only through images in this format can the images be printed as plates with visual three-dimensional effects.

[0029] S300, taking a screenshot of the stereoscopic engineering image to derive a plurality of multimedia sub-images, and splicing the multimedia sub-images based on edge pixel information of the multimedia sub-images to obtain a multimedia stereoscopic image.

[0030] First, take a screenshot of the 3D engineering image displayed in the engineering software, and cut out several sub-images in normal picture format from a 3D engineering image. The screenshot method can use the screenshot command provided by the computer or the screenshot function in the engineering software. The cut multimedia sub-images are in PNG, JPG and other formats.

[0031] Then, the cropped multimedia sub-images are matched and spliced ​​based on the pixel data at the edge of each sub-image, and the sub-images in multiple picture formats are stitched into a multimedia image in the entire image format. In this way, the engineering images that cannot be exported can be indirectly converted into multimedia stereo images that can be opened by any picture software.

[0032] S400, obtaining RGB data of each pixel in the multimedia stereo image and integrating them into a control group.

[0033] The RGB data of each point of the multimedia stereoscopic image is extracted and integrated into a data group, and the data group is defined to obtain a control group.

[0034] S500, converting the multimedia 3D image into a printing color mode to generate a CMYK format image to be printed, and integrating CMYK data corresponding to each pixel into a processing group.

[0035] The multimedia stereoscopic image is imported through PS software, and the color mode of the multimedia stereoscopic image in RGB format is converted into a printed image in CMYK format for printing. At the same time, the data in CMYK format corresponding to each pixel point is integrated to obtain a data group, and finally the data group is defined as a processing group.

[0036] S600, pixel mapping is performed between the treatment group and the control group, and the CMYK data of each pixel in the treatment group is adjusted to obtain a visual action evaluation reward. The visual action evaluation reward is represented by the score corresponding to the color of the pixel point approaching or moving away from the color of the RGB data after the adjustment action.

[0037] Because RGB is the color format of each pixel corresponding to when the image is displayed on an electronic display, and when the image is printed, it is printed by the ink of the printer, and the printed product corresponds to the CMYK format when printed.

[0038] RGB is represented by red channel, green channel and blue channel respectively, which obtain different color presentations by different combinations of three colors, while CMYK is represented by cyan, magenta, yellow and black respectively, which obtain different color presentations by combining four different colors.

[0039] RGB has a wider color gamut, while CMYK has a limited color gamut compared to RGB, so some colors in RGB cannot be displayed when printed. These colors not included in the CMYK color gamut will be lost when printed, resulting in "color difference". Because the change of color in a 3D image will seriously affect the 3D presentation effect, in order to reduce the impact of color conversion on the 3D effect, on the premise that the RGB format image has a 3D effect when displayed, it is necessary to re-adjust the parameters of the CMYK image after the image is converted to CMYK format to adjust the CMYK image color performance back to the effect close to the RGB format as much as possible.

[0040] Therefore, it is necessary to adjust the processing group and determine whether the color performance of the picture is close to or away from the RGB image after the color data of each point is changed according to the point mapping method, and generate different adjustment behavior rewards according to the visual actions.

[0041] S700, generating an adjustment plan corresponding to the highest reward, and after completing the adjustment of each CMYK data of the processing group, saving the image to be printed and printing it.

[0042] After simulating various adjustment actions, an adjustment plan is generated according to the highest reward corresponding to all the actions. The adjustment plan is characterized by adjusting the CMYK parameters to minimize the color difference between the CMYK image and the RGB image. The CMYK data is adjusted through this plan, and the adjusted image to be printed is printed. The generated paper plate can achieve a three-dimensional display effect.

[0043] Through the above steps, a three-dimensional engineering image is generated by combining remote sensing satellites with industry software. In order to solve the problem that the engineering image cannot be directly exported, the engineering image is indirectly converted and exported into a three-dimensional image in a multimedia picture format through screenshots and splicing. In order to print the three-dimensional image displayed on the monitor as a paper version of a three-dimensional plate, while ensuring the influence of the color format conversion on the three-dimensional presentation effect, the RGB data with a known three-dimensional effect is used as a control group, and the parameters of the data converted to CMYK are used as a processing group for callback to approach the color display effect of the RGB image before printing. The three-dimensional image can be printed as a paper three-dimensional plate in turn, which can be used as a medium for publicity and explanation in enterprises, scenic spots, and attractions.

[0044] In other embodiments, the satellite image is preprocessed, the color channel is processed, and the geometric relationship of the feature points is matched to generate engineering data, which includes the following steps: S110, acquiring a first panchromatic image and a first multispectral image of a first satellite, and fusing and mosaicking the images after radiation correction and atmospheric correction, respectively, to obtain a first satellite image at a first surveying and mapping position.

[0045] S120, acquiring a second panchromatic image and a second multispectral image of a second satellite, and fusing and mosaicking them after radiation correction and atmospheric correction respectively to obtain a second satellite image at a second surveying and mapping position.

[0046] When remote sensing satellites take images, they take two images at each angle. Panchromatic images are images that only contain black and white grayscale information, and multispectral images are images that contain color information in multiple bands. By fusing the two images, images with high spatial resolution and rich spectral information can be obtained.

[0047] It first needs to perform radiation correction and atmospheric correction to perform preprocessing steps such as denoising to eliminate the errors caused by radiation and atmospheric refraction of light data.

[0048] Resolution matching is then performed to match the two images with different spatial resolutions. For example, the multispectral image is sampled using interpolation techniques so that it has the same resolution as the panchromatic image.

[0049] The two images are merged through a fusion algorithm. Common fusion methods include: transformation-based methods (such as wavelet transform, texture synthesis, etc.), statistics-based methods (such as principal component analysis, Laplace pyramid transform, etc.) and feature-based methods (such as IHS transform, HSV transform, etc.).

[0050] Secondly, when the target area exceeds the range covered by a single remote sensing image, two or more images need to be stitched together to form one or a series of larger images covering the entire area. At this time, image mosaicking is required.

[0051] When mosaicking, a reference image needs to be determined. The reference image will serve as the benchmark for the output mosaic image, determine the contrast matching of the mosaic image, and the pixel size and data type of the output image. The same or similar imaging time is selected for the two or more mosaicked images to keep the image tones consistent. However, when the tones differ too much, histogram equalization and color smoothing can be used to make the edges as consistent as possible. However, when used for variation information extraction, the tones of adjacent images are not allowed to be smoothed to avoid information variation.

[0052] After the above processing, the first satellite image and the second satellite image are obtained.

[0053] S130, respectively obtain RGB data of pixels corresponding to the first satellite image and the second satellite image, select one of them to remove red channel values ​​and select the other to remove blue channel values ​​and green channel values.

[0054] Secondly, the color channel selection process is performed. First, one of the satellite images is selected, and the red channel value (R value) of the satellite image is removed, and another satellite image is selected, and the blue channel (B value) and green channel (G value) of the satellite image are removed.

[0055] One satellite image is an image with only the cyan channel (blue + green), and the other satellite image is an image with only the red channel. After the two images are fused, the three-dimensional effect of the image can be seen through red-cyan glasses with a red light lens and a cyan light lens respectively.

[0056] S140, performing aerial triangulation based on the pixel feature information on the first satellite image and the pixel feature information on the second satellite image to determine the geometric relationship of the feature points and merge them to obtain engineering data.

[0057] Aerial triangulation refers to the use of photogrammetric analysis to determine the exterior orientation elements of all images in the area. In traditional photogrammetry, this is achieved by measuring the point positions, that is, according to the image point coordinates measured on the image and the geodetic coordinates of a small number of control points, the geodetic coordinates of the unknown points are calculated, so that the number of known points in each model is increased to no less than 4, and then these known points are used to solve the exterior orientation elements of the image.

[0058] The collinear equation of the central projection is used as the basic equation of adjustment. Through the rotation and translation of each light beam in space, the light rays at the common points between the models can achieve the best intersection, and the entire area can be optimally incorporated into the known control point coordinate system. Under the conditions that the coordinates of the common intersection points of adjacent photos are equal and the internal coordinates of the control points are equal to the known external coordinates, the error equations of the control points and the encrypted points are listed, and a unified adjustment calculation is performed for the entire area to solve the exterior orientation elements of each photo and the ground coordinates of the encrypted points.

[0059] Through aerial triangulation, the positional relationship between related pixel points on the two images, such as the foot of the mountain, the peak of the mountain, etc., is determined, and based on the stereo conversion relationship, the superposition difference between the pixel points of two different color channels under different positional relationships is determined, so that the two images are fused into a stereo image.

[0060] After preprocessing, color processing and geometric relationship matching, the two satellite images can be superimposed into an image with a three-dimensional display effect, which corresponds to the corresponding engineering data in the engineering software.

[0061] In some other embodiments, before selecting one of the images to remove the red channel value and selecting another image to remove the blue channel value and the green channel value, the following steps are also included: S121, adding black and white RGB data to each pixel on the first satellite image and the second satellite image as a gain parameter.

[0062] Traditional red-cyan 3D may result in reduced image resolution for each eye due to the segmentation of color channels, because each color channel may only use part of the pixel information, so the resolution of each channel may be limited to the bandwidth of the monochrome channel (usually one-third of the total pixels).

[0063] The human eye tends to be more sensitive to brightness details, and brightness data is often provided by the black and white channel values ​​of the image. The black and white channels can provide clear contours and textures of the image, while the red, green and blue color channels retain the three-dimensional clues required for color separation by superimposing color information.

[0064] Therefore, in an embodiment of the present application, before eliminating the different channel data of the two satellite images, the black and white channel data are first added to the two satellite images to be merged with other color channels. In this way, after the other channel data are subsequently eliminated, one image can receive red and black and white information through a red filter, and the other image can receive green, blue and black and white information through a cyan filter.

[0065] The black and white communication data provides high-resolution brightness, ensuring that each image has pixels corresponding to the color channels and complete pixels of full-resolution black and white information. Experimental results show that the clarity is 4 times higher than traditional red and blue 3D.

[0066] In some other embodiments, a screenshot operation is performed on a stereoscopic engineering image to derive a plurality of multimedia sub-images, and each multimedia sub-image is spliced ​​based on edge pixel information of the multimedia sub-images to obtain a multimedia stereoscopic image, including the following steps: S310, obtaining the number of horizontal pixels and the number of vertical pixels of the stereo engineering image to calculate the total number of pixels.

[0067] First, before taking a screenshot, first obtain the number of horizontal and vertical pixels of the entire stereo engineering image, and combine them to calculate the total number of pixels of the entire image, which represents the overall resolution of the image.

[0068] S320, performing geometric segmentation on the horizontal pixel points of the stereo engineering image, performing geometric segmentation on the vertical pixel points of the stereo engineering image to obtain a plurality of segmented calibration images, and labeling the plurality of segmented calibration images.

[0069] Firstly, based on the screenshot tool, equidistant annotations are performed in the stereo engineering image, so that the image is divided into several parts in a geometric ratio in the horizontal direction and in the vertical direction. The entire stereo engineering image is divided into several equal-sized segmented calibration images by combining the horizontal and vertical geometric segmentation points. At the same time, each segmented calibration image is labeled based on its position in the entire image.

[0070] S330, taking screenshots based on the edge of the segmented and calibrated image and exporting the screenshots to obtain a plurality of multimedia sub-images.

[0071] Based on the edge positions of each segmented and calibrated image, corresponding screenshots are taken, and each screenshot is exported to obtain a plurality of multimedia sub-images in a multimedia image format.

[0072] When taking screenshots, you can use the built-in screenshot tool on your calculator, a third-party screenshot tool, or the screenshot tool on your engineering software.

[0073] S340, sorting the plurality of multimedia sub-images based on the labels, and verifying the splicing matching degree between adjacent multimedia sub-images.

[0074] The plurality of cut multimedia sub-images are sorted according to their corresponding labels to rearrange the plurality of sub-images into a whole image. At the same time, adjacent multimedia sub-images need to be verified based on the splicing matching degree before splicing.

[0075] Because pixels are extremely small points in an image, when taking a screenshot, although there is edge information of the screenshot mark, there may still be more or fewer pixels in the screenshot. When there are errors in the captured pixels between several sub-images, it may cause pixel overlap or loss in some locations. The stitching matching degree is used to verify and ensure that the adjacent sub-images have complete stitching conditions before stitching.

[0076] S350: When the stitching matching degree is higher than a threshold, adjacent multimedia sub-images are stitched together to obtain a multimedia stereoscopic image.

[0077] When the corresponding splicing matching degrees between adjacent multimedia sub-images are all higher than a preset value, it is considered that the sub-images meet the splicing conditions, and the multimedia sub-images are spliced ​​based on edge pixel information.

[0078] In some other embodiments, sorting a plurality of multimedia sub-images based on labels and verifying the splicing matching degree between adjacent multimedia sub-images includes the following steps: S341, obtaining the horizontal pixel number and the vertical pixel number of each multimedia sub-image to calculate the sub-pixel number, and determining whether the sub-pixel number of each multimedia sub-image matches a reference value, wherein the reference value is represented by the total number of pixels divided by the total number of labels.

[0079] For calculation and generation of the stitching matching degree, first determine the number of horizontal and vertical pixels of several multimedia sub-images obtained after the screenshot operation, and use this to calculate the number of sub-pixels of each sub-image, each sub-pixel number representing the resolution of each sub-image.

[0080] At the same time, it is determined whether the number of sub-pixels corresponding to each sub-image is equal to the total pixel value corresponding to the entire project image divided by the number of labels. The reference value represents the number of pixels that should exist in each sub-image theoretically after segmentation. If they match, it means that the corresponding sub-image has passed the verification at the quantitative level. If they do not match, it means that the sub-image has captured more or less pixels when taking the screenshot.

[0081] Among them, it should be noted that, whether it is the above-mentioned engineering stereoscopic image or the multimedia stereoscopic image, it is displayed through a display screen. When an image is displayed in different sizes on the display screen, the number of pixels it occupies on the display is different. Therefore, in order to achieve the low error in the above-mentioned pixel number verification, no enlargement or reduction operation will be performed on the image before and after the screenshot.

[0082] For example, when a project image is screenshotted at a display ratio of 100%, the number of screenshots is 5*2=10, and the multimedia sub-images generated after each screenshot are spliced ​​at a display ratio of 10%.

[0083] S342, defining the matched multimedia sub-images as valid screenshots, defining the unmatched multimedia sub-images whose number is greater than a reference value as screenshots to be optimized, and defining the unmatched multimedia sub-images whose number is less than a reference value as invalid screenshots.

[0084] When performing the matching verification of the above number of pixels, there are three situations: When the number of pixels in the sub-image matches the reference value, it is considered that the screenshot verification of the sub-image has passed from the quantitative level; When the number of pixels of a sub-image does not match the reference value, and when the number of pixels is greater than the reference value, it is considered that some pixels are captured more than others during the screenshot. In this case, the captured pixels can be optimized based on the edge pixel information of other sub-images adjacent to the sub-image. Therefore, the multimedia sub-image at this time still has room for remedy, so it is defined as a screenshot to be optimized. When the number of pixels in a sub-image does not match the reference value and when the number of pixels is less than the reference value, it is considered that some pixels are missed when taking the screenshot. In this case, since it is uncertain where the missed part is located in the multimedia sub-image, it is impossible to directly supplement or predict the exact pixels of the sub-image through other adjacent multimedia sub-images. Therefore, when faced with such an image that cannot be optimized and remedied, it is defined as an invalid screenshot.

[0085] S343, placing the valid screenshots and the screenshots to be optimized in corresponding areas based on the labels, obtaining the labels of the invalid screenshots and re-screwing the corresponding areas on the stereo engineering image until the invalid screenshots are changed into valid screenshots or screenshots to be optimized.

[0086] Different processing methods correspond to screenshots in different situations. For valid screenshots and screenshots to be optimized, because they match the requirements or can be optimized, they can be directly arranged and placed in the corresponding positions according to the labels and wait for subsequent stitching. Invalid screenshots do not match the requirements and cannot be remedied, so it is necessary to reselect the corresponding part on the three-dimensional engineering image according to the label of the invalid screenshot and re-screenshot it until the invalid screenshot becomes a valid screenshot or a screenshot to be optimized.

[0087] S344, obtaining edge pixel information of each valid screenshot and each screenshot to be optimized, wherein the edge pixel information of the valid screenshot includes a layer of pixels extending inward from the edge of the image, and the edge pixel information of the screenshot to be optimized includes several layers of pixels extending inward from the edge of the image.

[0088] When all sub-images are valid screenshots or screenshots to be optimized, for the subsequent stitching work, it is necessary to re-acquire the edge pixel information of each multimedia sub-image. At the same time, because the number of pixels of the valid screenshots matches the reference value, the edge pixel information of the valid screenshots only includes a layer of pixels extending inward on the four sides of the image. For the image to be optimized, because the number of pixels is more than the reference value, the edge pixel information that needs to be selected includes several layers of pixels extending inward on the four sides.

[0089] In some other embodiments, sorting the plurality of multimedia sub-images based on the labels and verifying the splicing matching degree between adjacent multimedia sub-images further includes the following steps: S345 , performing splicing optimization to compare whether there is overlapping data in edge pixel information of adjacent multimedia sub-images.

[0090] For the calculation of the splicing matching degree, the edge pixel information of each multimedia sub-image is first optimized so that the edge pixels of one multimedia sub-image match the edge pixels of its adjacent multimedia sub-image. The specific matching is represented by the fact that there is no overlapping data on the edges between two adjacent multimedia sub-images.

[0091] S346: If yes, eliminate the pixel points in one of the multimedia sub-images that are the same as the overlapped data.

[0092] If there is overlapping data in the edge pixel information of adjacent multimedia sub-images, in order to avoid the superposition of the same pixels and cause color changes, it is necessary to eliminate the pixels corresponding to the overlapping data in any multimedia sub-image. Once the overlapping data is eliminated, it can be guaranteed that the two adjacent multimedia sub-images meet the splicing matching requirements.

[0093] S347, pre-stitching the plurality of multimedia sub-images after the stitching optimization to obtain a pre-image, obtaining the number of horizontal pixels and the number of vertical pixels of the pre-image and comparing them with the total number of pixels, and generating a first matching coefficient based on the comparison result.

[0094] After eliminating the overlapping pixels between all adjacent sub-images, several multimedia sub-images are pre-stitched to obtain a pre-image, and the number of horizontal and vertical pixels of the pre-image is determined again and compared with the total number of pixels of the previous engineering stereo image. A first matching coefficient is generated according to different comparison results.

[0095] In the optimal case, the total number of pixels of the pre-image obtained after screenshot and stitching should be the same as the total number of pixels of the previous engineering stereo image, and the corresponding first matching coefficient is the highest; conversely, the greater the difference between the two total pixel numbers, the smaller the corresponding first matching coefficient.

[0096] S348, determining the number of overlapping edges and overlapping points of each multimedia sub-layer having overlapping data, and calculating a second matching coefficient based on the number of overlapping edges and overlapping points.

[0097] Secondly, the number of edges and points of the overlapping data in each multimedia sub-image is determined, and a corresponding second matching coefficient is generated.

[0098] Among them, the number of overlapping edges is represented by the number of edges with overlapping data among the four sides of the multimedia sub-layer, such as overlapping data on the top and left sides of an image; the number of overlapping points is represented by the number of pixels in the multimedia sub-layer corresponding to overlapping data, such as the number of overlapping pixels on the left side of an image is 270 pixels.

[0099] The more overlapping edges there are, the greater the overall deviation of multiple screenshots. For example, if there is only one overlapping edge, it means that only one side has multiple screenshots when taking the screenshot. When there are two overlapping edges on the left and right, it means that multiple screenshots have occurred in both the left and right directions. This means that there is a large error in the screenshot operation, and the corresponding second matching coefficient is lower.

[0100] The generation logic of the second matching coefficient corresponding to the number of overlapping points is the same as that of the number of overlapping edges.

[0101] S349: Calculate the splicing matching degree of each multimedia sub-layer based on the first matching coefficient and the second matching coefficient.

[0102] The stitching matching degree is calculated by the first matching coefficient and the second matching coefficient, and compared with the preset stitching requirement value. When it is equal to or higher than the stitching requirement value, it means that the current multimedia sub-images have a higher stitching matching effect. When it is less than the stitching requirement value, it means that the current multimedia sub-images do not have the conditions for successful stitching. At this time, the above-mentioned screenshot operation needs to be performed again.

[0103] The problem of not meeting the stitching conditions may be that the number of pixels found in several multimedia sub-images after processing is far less than the total number of pixels in the engineering stereo image, which will cause a large number of pixels to be lost during the screenshot process, which will have a greater impact on the final stereo effect; for example, because the errors in screenshot operations are generally large, pixel optimization of screenshots needs to be performed frequently, which will also cause more pixels to be lost during the optimization process.

[0104] After stitching out a complete multimedia 3D image, the last operation to be performed is to convert the multimedia 3D image into a CMYK image in a printed format. In order to ensure that the 3D effect is not affected by the loss caused by the change in color gamut, the corresponding parameters of the CMYK image need to be further adjusted to make the color display effect of the CMYK image close to the color display effect of the multimedia 3D image. Specifically: In other embodiments, pixel mapping is performed between the treatment group and the control group and the CMYK data of each pixel in the treatment group is adjusted to obtain a visual action evaluation reward, including the following steps: S610, respectively exporting the full-size multimedia stereoscopic image and the image to be printed, and overlapping and fusing them based on pixel positions to generate a fused image.

[0105] First, export the 100% scale multimedia stereoscopic image and the image to be printed, and overlap and fuse them based on the position and pixel information of each pixel point, so that the fused image of RGB image + CMYK image is displayed in RGB mode. At this time, because the two images with color difference are overlapped, the color difference between the fused image and the RGB multimedia stereoscopic image in the control group will change again.

[0106] S620, obtaining RGB data of each pixel in the fused image and integrating them into an evaluation group.

[0107] The RGB data of the pixels corresponding to the fused image are integrated and defined as the evaluation group.

[0108] S630, taking the RGB data in the control group as the target state, and retrieving the color difference list based on the target state to generate a reward range.

[0109] The RGB data of each pixel in the control group is the target state, which represents the target result of the overall color change of the CMYK image after a series of parameter adjustments.

[0110] At the same time, a reward range is generated through a preset color difference list. The color difference list represents the experimental results obtained by the operator in a large number of previous experiments. It represents the color difference between the color change of the picture and the data of the control group when the CMYK parameters are adjusted, while ensuring that the three-dimensional effect is not affected. When the CMYK parameters change, if the difference between the final color change and the color in the control group is within the reward range, the intelligent agent will reward the adjustment behavior.

[0111] S640, performing an adjustment action on the data in the processing group and obtaining whether the new state of the RGB value of each pixel point in the evaluation group after the adjustment action is within the reward range to generate a reward or penalty.

[0112] When the CMYK value changes, in order to make the CMYK color change visualized and intuitive, after adjusting the processing of the processing group, the new state of the RGB value of each pixel in the evaluation group can be used to directly determine whether the color difference change is within the reward range.

[0113] If the CMYK and RGB images are not overlapped, then when the CMYK parameters are adjusted normally, the color values ​​of the image to be printed are still displayed based on the CMYK format, and the user still cannot directly compare and judge the specific data with the RGB data in the control group. Therefore, after overlapping the CMYK and RGB images, the changes in RGB parameters of the fused image caused by the changes in CMYK parameters can be compared intuitively with the RGB data of the control group.

[0114] Among them, the parameter adjustment of CMYK includes the adjustment of the values ​​of each channel of C, M, Y, and K, as well as the changes in parameters such as contrast, brightness, and exposure.

[0115] In some other embodiments, the full-size multimedia stereoscopic image and the image to be printed are respectively exported, and overlapped and fused based on the pixel positions to generate a fused image, including the following steps: S611, setting the multimedia stereoscopic image as the background and the image to be printed as the foreground.

[0116] When the multimedia 3D image and the image to be printed are to be fused, the background and foreground during fusion must first be determined. The background and foreground represent the positional relationship between the two images when they are fused, that is, in the fused image, the multimedia 3D image is at the back and the image to be printed is at the front.

[0117] S612, respectively set the Alpha blending coefficients and allocate them to the background and foreground, wherein the Alpha blending coefficient of the background is equal to 1, and the Alpha blending coefficient of the foreground is less than 1.

[0118] When the two images are configured as foreground and background respectively, in order to reduce the color impact of the front CMYK image on the original RGB image, the image in the foreground needs to have a certain degree of transparency. If it does not have transparency, the color of the background image will be completely invisible, and the fused image cannot be used as an evaluation reference for color adjustment.

[0119] Therefore, it is necessary to generate a blending coefficient through the Alpha algorithm. The blending coefficient represents the transparency setting of the image. When the blending coefficient is 1, it represents an image that is completely opaque. When the blending coefficient is less than 1, it represents that the corresponding image has a certain degree of transparency. The smaller the blending coefficient, the greater the transparency.

[0120] For example, when the mixing coefficient is 0.6, the image is characterized as glass with a transmittance of 40%, that is, 40% of the light can pass through the image, while the remaining 60% of the light cannot pass through and is reflected back to the user's eyes through color.

[0121] Then in the present application, the mixing coefficient of the multimedia stereoscopic image in the background image is 1, and the mixing coefficient of the CMYK image in the foreground image is 0.5.

[0122] S613, in the foreground, generating corresponding reflectivity and refraction index based on the Alpha blending coefficient.

[0123] Since transparency is the light transmittance of the corresponding layer of an image, the foreground and background can be regarded as glass. The glass in the foreground (mixing coefficient 0.5) means that 50% of the light can enter the background, while 50% of the light cannot pass through and is reflected back to the user's eyes, so its corresponding reflectivity is 0.5 and refractive index is 0.5. The mixing coefficient of the background image is 1, which means that no light can pass through, and all the light it receives will be reflected back to the user's eyes, so its reflectivity is 1 and its refractive index is 0.

[0124] Then, when actually calculating the RGB value of the mixed image, the above characteristics can be used to calculate RGB.

[0125] For example, the RGB values ​​of the RGB image of the known background are a1 (R), a2 (G), a3 (B), and the RGB values ​​of the CMYK image of the foreground are assumed to be b1 (R), b2 (G), b3 (B). The reflectivity of the background is 1, and the refractive index is 0, while the reflectivity of the foreground is 0.5, and the refractive index is 0.5.

[0126] When calculating, first for the background, its reflectivity is 1, so for the R channel, all the red of a1 is reflected, and then it passes through the foreground with a refractive index of 0.5 to enter the user's eyes, so the R channel value of the background after passing through the foreground is a1*0.5.

[0127] For the foreground, its reflection value is 0.5, so for the R channel, 50% of the red color of b1 will be reflected into the user's eyes, so the value of the R channel of the foreground is b1*0.5.

[0128] After the two images are fused, the total value of the R channel that enters the user's eyes is a1*0.5+b1*0.5=(a1+b1)*0.5.

[0129] According to the above calculation, it can be found that under the premise that the corresponding specific RGB value of the CMYK format image cannot be directly obtained, if the CMYK value is continuously adjusted so that the unknown RGB value of the CMYK in the foreground is the same as the RGB value of the multimedia stereo image in the background, the RGB value of the entire fused image (evaluation group) can be guaranteed to be the same as the RGB value of the original multimedia stereo image (control group).

[0130] Through the above calculation, when the CMYK values ​​in the processing group are adjusted, the color difference change between the image to be printed and the multimedia 3D image can be intelligently and quickly determined without manual visual color change comparison.

[0131] In other embodiments, generating an adjustment plan corresponding to the highest reward, and saving the image to be printed and printing it after the CMYK data of the processing group are adjusted completely, includes the following steps: S710: After the adjustment plan is generated, the total color difference between the RGB values ​​of each pixel in the evaluation group and the RGB values ​​of each pixel in the control group is not greater than a preset value.

[0132] After adjusting the CMYK data in the treatment group through the adjustment plan, it is necessary to ensure that the total color difference between the RGB values ​​of each pixel in the evaluation group and the RGB values ​​of each pixel in the control group is not greater than the preset value. This can ensure that after the parameters of the CMYK image are adjusted, it is adjusted to the same or similar color as the RGB image so that the three-dimensional display effect can be retained after printing.

[0133] S720, after the adjustment of each CMYK data of the processing group is completed, the current image to be printed is saved and exported for printing.

[0134] Finally, after the processing group is adjusted, the adjusted image to be printed in CMYK format is printed.

[0135] The embodiment of the present application also discloses a three-dimensional plate generation system for implementing the above method.

[0136] The implementation principle is: Stereoscopic engineering images are generated by remote sensing satellites combined with industry software. In order to solve the problem that engineering images cannot be directly exported, the engineering images are indirectly converted and exported into stereoscopic images in multimedia picture formats through screenshots and splicing. In order to print the stereoscopic images displayed on the monitor into paper-based stereoscopic plates, while ensuring the impact of the stereoscopic presentation effect due to the color format conversion, RGB data with known stereoscopic effects is used as the control group, and the parameters of the data converted to CMYK are used as the processing group for callback to approach the color display effect of the RGB image before printing. The stereoscopic images can be printed as paper-based stereoscopic plates in turn, which can be used as a medium for publicity and explanation in enterprises, scenic spots, and attractions.

[0137] It should be understood that although the steps in the flowchart of the accompanying drawings are shown in sequence according to the instructions of the arrows, these steps are not necessarily performed in the order indicated by the arrows. Unless otherwise clearly stated in this document, the execution of these steps is not strictly limited in order and can be performed in other orders.

[0138] The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereto. Therefore, any equivalent changes made according to the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. A method for generating a three-dimensional plate, characterized in that: The following steps are involved: Obtain satellite images from different shooting angles, perform preprocessing, color channel processing, and feature point geometric relationship matching on the satellite images to generate engineering data; Opening the engineering data to generate a three-dimensional engineering image; Performing a screenshot operation on the stereoscopic engineering image to derive a plurality of multimedia sub-images, and splicing the multimedia sub-images based on edge pixel information of the multimedia sub-images to obtain a multimedia stereoscopic image; Obtaining RGB data of each pixel in the multimedia stereoscopic image and integrating them into a control group; Converting the multimedia 3D image into a printing color mode to generate a to-be-printed image in CMYK format, and integrating the CMYK data corresponding to each pixel into a processing group; Performing pixel mapping between the treatment group and the control group and adjusting the CMYK data of each pixel in the treatment group to obtain a visual action evaluation reward, wherein the visual action evaluation reward is characterized by a score corresponding to the color of the pixel approaching or moving away from the color of the RGB data after the adjustment action; An adjustment plan corresponding to the highest reward is generated, and after the adjustment of each CMYK data of the processing group is completed, the image to be printed is saved and printed.

2. The method for generating a three-dimensional plate according to claim 1, characterized in that: The satellite image is preprocessed, color channel processed, and feature point geometric relationship matched to generate engineering data, including the following steps: Acquire a first panchromatic image and a first multispectral image of a first satellite, and fuse and mosaic them after radiation correction and atmospheric correction respectively to obtain a first satellite image at a first surveying and mapping position; Acquire a second panchromatic image and a second multispectral image of a second satellite, and fuse and mosaic them after radiation correction and atmospheric correction respectively to obtain a second satellite image at a second surveying and mapping position; Respectively obtain pixel RGB data corresponding to the first satellite image and the second satellite image, select one of them to remove the red channel value and select the other to remove the blue channel value and the green channel value; Aerial triangulation is performed based on the pixel feature information on the first satellite image and the pixel feature information on the second satellite image to determine the geometric relationship of the feature points and merge them to obtain the engineering data.

3. The method for generating a three-dimensional plate according to claim 2, characterized in that: Before selecting one of the images to remove the red channel value and another to remove the blue and green channel values, the following steps are also included: Black and white RGB data are added to each of the pixel points on the first satellite image and the second satellite image as a gain parameter.

4. The method for generating a three-dimensional plate according to claim 1, characterized in that: Taking a screenshot of the stereo engineering image to derive a plurality of multimedia sub-images, and splicing the multimedia sub-images based on edge pixel information of the multimedia sub-images to obtain a multimedia stereo image, comprises the following steps: Obtaining the number of horizontal pixels and the number of vertical pixels of the three-dimensional engineering image to calculate the total number of pixels; Performing geometric segmentation on the horizontal pixel points of the stereo engineering image, performing geometric segmentation on the vertical pixel points of the stereo engineering image to obtain a plurality of segmentation calibration images, and labeling the plurality of segmentation calibration images; Taking screenshots based on the edges of the segmented and calibrated image and exporting them to obtain a plurality of the multimedia sub-images; sorting the plurality of multimedia sub-images based on the labels, and verifying the splicing matching degree between adjacent multimedia sub-images; When the stitching matching degree is higher than a threshold, adjacent multimedia sub-images are stitched together to obtain the multimedia stereoscopic image.

5. The method for generating a three-dimensional plate according to claim 4, characterized in that: Sorting the plurality of multimedia sub-images based on the labels and verifying the splicing matching degree between adjacent multimedia sub-images comprises the following steps: Obtaining the number of horizontal pixels and the number of vertical pixels of each multimedia sub-image to calculate the number of sub-pixels, and determining whether the number of sub-pixels of each multimedia sub-image matches a reference value, wherein the reference value is represented by the total number of pixels divided by the total number of labels; The matched multimedia sub-images are defined as valid screenshots, the unmatched multimedia sub-images whose number is greater than the reference value are defined as screenshots to be optimized, and the unmatched multimedia sub-images whose number is less than the reference value are defined as invalid screenshots; Placing the valid screenshot and the screenshot to be optimized in corresponding areas based on the labels, obtaining the labels of the invalid screenshots and re-screaming the corresponding areas on the stereo engineering image until the invalid screenshots are changed into the valid screenshots or the screenshots to be optimized; The edge pixel information of each of the valid screenshots and each of the screenshots to be optimized is obtained, wherein the edge pixel information of the valid screenshots includes a layer of pixels extending inward from the edge of the image, and the edge pixel information of the screenshots to be optimized includes several layers of pixels extending inward from the edge of the image.

6. The method for generating a three-dimensional plate according to claim 5, characterized in that: The method further comprises the following steps: sorting the plurality of multimedia sub-images based on the labels, and verifying the splicing matching degree between adjacent multimedia sub-images; and: Performing splicing optimization to compare whether there is overlapping data in the edge pixel information of adjacent multimedia sub-images; If so, eliminating the pixel points in one of the multimedia sub-images that are identical to the overlapped data; Pre-stitching the plurality of multimedia sub-images after the splicing optimization to obtain a pre-image, obtaining the number of horizontal pixels and the number of vertical pixels of the pre-image and comparing them with the total number of pixels, and generating a first matching coefficient based on the comparison result; Determine the number of overlapping edges and the number of overlapping points of each of the multimedia sub-layers having the overlapping data, and calculate a second matching coefficient based on the number of overlapping edges and the number of overlapping points, wherein the number of overlapping edges is represented by the number of edges having overlapping data among the four sides of the multimedia sub-layer, and the number of overlapping points is represented by the number of pixel points in the multimedia sub-layer corresponding to the overlapping data; The splicing matching degree of each of the multimedia sub-layers is calculated based on the first matching coefficient and the second matching coefficient.

7. The method for generating a three-dimensional plate according to claim 1, characterized in that: The processing group and the control group are pixel mapped and the CMYK data of each pixel in the processing group is adjusted to obtain a visual action evaluation reward, comprising the following steps: Exporting the full-size multimedia stereoscopic image and the image to be printed respectively, and overlapping and fusing them based on pixel positions to generate a fused image; Acquire RGB data of each pixel in the fused image and integrate them into an evaluation group; Using the RGB data in the control group as a target state, and retrieving a color difference list based on the target state to generate a reward range; An adjustment action is performed on the data in the processing group to obtain whether the new state of the RGB value of each pixel point in the evaluation group after the adjustment action is within the reward range to generate a reward or penalty.

8. The method for generating a three-dimensional plate according to claim 7, characterized in that: The full-size multimedia stereoscopic image and the image to be printed are respectively exported, and overlapped and fused based on pixel positions to generate a fused image, including the following steps: Setting the multimedia stereoscopic image as the background and the image to be printed as the foreground; respectively setting an Alpha blending coefficient and allocating it to the background and the foreground, wherein the Alpha blending coefficient of the background is equal to 1, and the Alpha blending coefficient of the foreground is less than 1; In the foreground, corresponding reflectivity and refraction index are generated based on the alpha blending coefficient.

9. The method for generating a three-dimensional plate according to claim 7, characterized in that: Generate an adjustment plan corresponding to the highest reward, and save the image to be printed and print it after the CMYK data of the processing group are adjusted completely, including the following steps: After the adjustment plan is generated, the total color difference between the RGB values ​​of each pixel in the evaluation group and the RGB values ​​of each pixel in the control group is not greater than a preset value; After the adjustment of each CMYK data of the processing group is completed, the current image to be printed is saved and exported for printing.

10. A three-dimensional plate generation system, characterized in that: Used to implement the method according to any one of claims 1 to 9.

Images