An AI-powered follow-up projection sand table and its projection method

Abstract

This application discloses an artificial intelligence-guided projection sand table and its projection method, belonging to the field of digital sand table technology. The sand table includes a data acquisition device, a data processing terminal, and a projector. The data processing terminal is connected to the data acquisition device, and the projector is connected to the data processing terminal. The data acquisition device is used to collect data information about the projection area of ​​the sand table, the observer, and the distance between the observer and the sand table. The data processing terminal is used to process the observer's data and distance information, calculate the projection data using a projection display prediction model, and send it to the projector. The projector is used to adjust the projection parameters of the device according to the projection data to project images. This application's solution uses sensors in the sand table in conjunction with corresponding analysis algorithms to process data, enabling dynamic adjustment of the direction and effect of the sand table's projection display based on changes in the observer's position.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of digital sand table, and particularly relates to an artificial intelligence following type projection sand table and a projection method thereof. BACKGROUND

[0002] The existing geographical information projection mode of the sand table projects geographical information into the sand table through a projector.

[0003] At present, the projection effect of the sand table projector and the position of the viewer have strong correlation. The viewer can only identify the "correct" content when being at the set position. If the position of the viewer moves, the observed content will be offset, and the understanding of the displayed content will be greatly discounted. Figure 2 As shown in the figure, the observer can understand the "displayed text" when being at position A. However, when being at positions B, C and D, the observed content is offset to a certain extent. For example, the content observed at position B is offset by 90 degrees, which greatly affects the user experience. SUMMARY

[0004] Therefore, the application provides an artificial intelligence following type projection sand table and a projection method thereof, which can help solve the problem that the existing projection sand table is limited by the position of the viewer and cannot dynamically adjust the projection data, thereby causing poor viewing effect.

[0005] To achieve the above purpose, the application adopts the following technical solutions:

[0006] In a first aspect, the application provides an artificial intelligence following type projection sand table, which comprises:

[0007] A data acquisition device is configured to acquire data information of a projection area of a sand table, an observer and a distance between the observer and the sand table.

[0008] A data processing terminal is connected to the data acquisition device and configured to process the data information of the observer and the distance information, calculate projection data by using a projection display prediction model and deliver the projection data to a projector.

[0009] The projector is connected to the data processing terminal and configured to adjust projection parameters of the device according to the projection data to project an image.

[0010] Further, the data acquisition device comprises a binocular camera, a depth camera and an eye tracker sensor arranged at a preset data acquisition position around the sand table, and the data processing terminal is connected with the binocular camera, the depth camera and the eye tracker sensor respectively; the binocular camera is used to acquire picture images within a preset data acquisition position range according to a first preset acquisition frequency, and a contour image information of an observer within the preset data acquisition position range is recognized by using a character detection and recognition algorithm, and a projection area of the sand table is acquired at the same time; the depth camera is used to acquire distance information between the observer and the sand table within the preset data acquisition position range according to the first preset acquisition frequency; and the eye tracker sensor is used to capture eye data of the observer within the preset data acquisition position range.

[0011] Further, the data processing terminal specifically comprises an observer data processing module, a projection display prediction module and a projection output module.

[0012] The observer data processing module is used to perform character modeling processing on the data information of the observer and the distance information between the observer and the sand table, to obtain actual observer information, and to analyze a preset data acquisition position with the largest number of observers.

[0013] The projection display prediction module is used to process the actual observer information and the projection area of the sand table by using a projection display prediction model, to obtain a new projection direction and a new projection area after changing the projection direction, and to calculate optimal projection data in combination with the distance information between the observer and the sand table.

[0014] The projection output module is used to transmit the projection data to the projector.

[0015] Further, the preset data acquisition position comprises four preset data acquisition positions with the same azimuth angle, which are a first preset data acquisition position, a second preset data acquisition position, a third preset data acquisition position and a fourth preset data acquisition position.

[0016] In a second aspect, the present application provides a projection method of an artificial intelligence following type projection sand table, which is applied to the artificial intelligence following type projection sand table provided in the first aspect, and the projection method comprises the following steps:

[0017] S1: recognizing observer data, acquiring a projection area of a sand table, data information of an observer and distance information between the observer and the sand table by using a data acquisition device;

[0018] S2: observer data processing, performing data processing on the data information of the observer and the distance information, calculating projection data by using a projection display prediction model and delivering the projection data to a projector;

[0019] S3: projection display control, adjusting projection parameters of a projection device to project an image.

[0020] Further, the step S1 specifically comprises: firstly, the binocular camera collects picture images in a preset data collection orientation range according to a first preset collection frequency, and uses a person detection and recognition algorithm to recognize the contour image information of the observer in the preset data collection orientation range and simultaneously obtain the projection area of the sand table; then, the depth camera collects the distance information between the observer and the sand table in the preset data collection orientation range according to the first preset collection frequency; and finally, the eye tracker sensor captures the eye data of the observer in the preset data collection orientation range.

[0021] Further, the step S2 specifically comprises the following sub-steps:

[0022] S201: performing person modeling processing on the data information of the observer and the distance information between the observer and the sand table to obtain actual observer information, and analyzing a preset data collection orientation with the largest number of observers;

[0023] S202: using a projection display prediction model to process the actual observer information and the projection area of the sand table respectively, obtaining a new projection direction and a new projection area after changing the projection direction, and combining the distance information between the observer and the sand table to calculate optimal projection data;

[0024] S203: transmitting the optimal projection data to the projector.

[0025] Further, the sub-step S201 specifically comprises:

[0026] S2011: firstly, combining the contour image information and the distance information of the observer in each preset data collection orientation range, drawing all observer images around the sand table, and each observer image contains the recognized observer contour information and the position thereof;

[0027] S2022: based on the eye data of the observer in each preset data collection orientation range, performing eye data analysis on the observer who does not gaze at the sand table as a false observer to obtain the actual number of observers in each preset data collection orientation range;

[0028] S2023: selecting an arbitrary preset data collection orientation as a starting point, numbering the actual observers in each preset data collection orientation range in a clockwise or counterclockwise direction, comparing the actual number of observers in each preset data collection orientation range, and obtaining a preset data collection orientation with the largest number of observers;

[0029] S2024: sending the data in steps S2021-S2023 as observer information to the projection display prediction model.

[0030] Further, the sub-step S202 specifically comprises:

[0031] S2021: Based on the projection area of the sand table, the actual projection area data of the projector is calculated by using the projection area algorithm and sent to the projection display prediction model;

[0032] S2022: The projection display prediction model imports the observer information and the projection area data, and performs data cleaning on the observer information and the projection area data according to the preset data cleaning rule;

[0033] S2023: The projection display prediction model determines a new projection direction according to the preset data collection direction with the largest number of people in the observer information, and calculates a new projection area based on the new projection direction and the projection area data;

[0034] S2024: The projection display prediction model calculates the best projection parameters based on the new projection direction, the actual number of observers and the positions of the observers in the observer information, wherein the projection parameters include the offset angle of projection, the projection picture resolution and the projection size.

[0035] The above technical solution has at least the following beneficial effects:

[0036] The artificial intelligence following projection sand table provided by the present application includes a data acquisition device, a data processing terminal and a projector. The data processing terminal is connected to the data acquisition device, and the projector is connected to the data processing terminal. The data acquisition device is used to collect the projection area of the sand table, the data information of the observer and the distance information between the observer and the sand table. The data processing terminal is used to process the observer data information and the distance information, calculate the projection data by using the projection display prediction model, and send the projection data to the projector. The projector is used to adjust the projection parameters of the equipment according to the projection data to project the image. In this setting, the data acquisition device identifies the observer and the position parameters thereof by using the sensor, and then the data processing terminal takes the position of the observer as the reference surface of the display, calculates the offset angle (projection direction) and the projection display area of the projection picture by using the existing "projection display prediction model", and transmits the projection data to the projector. Finally, the projector is controlled to project and display. The present application can dynamically adjust the projection display direction and effect of the sand table based on the position change of the observer by using the sensor in the sand table and the corresponding analysis algorithm for data processing, which helps to solve the problem that the existing projection sand table is limited by the position of the viewer and cannot dynamically adjust the projection data, thereby causing poor viewing effect.

[0037] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present application. BRIEF DESCRIPTION OF DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the accompanying drawings needed to be used in the embodiments or prior art description will be briefly introduced. Obviously, the accompanying drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative labor.

[0039] Figure 1 It is a principle schematic diagram of an artificial intelligence following projection sand table according to an exemplary embodiment;

[0040] Figure 2 It is a sand table projection schematic diagram in the related art according to an embodiment;

[0041] Figure 3 It is a device layout schematic diagram of an artificial intelligence following projection sand table according to an exemplary embodiment;

[0042] Figure 4 It is a projection method flow chart according to an exemplary embodiment;

[0043] Figure 5 It is an observer image schematic diagram around a sand table according to an exemplary embodiment;

[0044] Figure 6 It is a true and false observer image schematic diagram around a sand table according to an exemplary embodiment;

[0045] Figure 7 It is an actual observer image schematic diagram around a sand table according to an exemplary embodiment;

[0046] Figure 8 It is an actual observer number schematic diagram around a sand table according to an exemplary embodiment;

[0047] Figure 9 It is a new projection direction schematic diagram of a sand table according to an exemplary embodiment;

[0048] Figure 10 It is a new projection area schematic diagram of a sand table according to an exemplary embodiment;

[0049] Appendix Figure 1 In the drawings: 1-data acquisition device, 2-data processing terminal, 3-projector. DETAILED DESCRIPTION

[0050] In order to make the purposes, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be described in detail below. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work belong to the scope of protection of the present application.

[0051] Please refer to Figure 1 , Figure 1 is a schematic diagram of a principle of an artificial intelligence following projection sand table, as shown in Figure 1 , the artificial intelligence following projection sand table comprises:

[0052] a data acquisition device 1 for acquiring data information of a projection area of a sand table, an observer and a distance between the observer and the sand table;

[0053] a data processing terminal 2 connected to the data acquisition device 1, for data processing of the data information of the observer and the distance, calculating projection data by using a projection display prediction model and delivering the projection data to a projector 3;

[0054] the projector 3 connected to the data processing terminal 2, for adjusting projection parameters of the equipment according to the projection data to project an image.

[0055] Further, in an embodiment, the data acquisition device 1 comprises a binocular camera, a depth camera and an eye tracker sensor arranged at a preset data acquisition position around the sand table, and the data processing terminal 2 is connected to the binocular camera, the depth camera and the eye tracker sensor respectively; the binocular camera is used to acquire picture images within a preset data acquisition position range according to a first preset acquisition frequency, to identify contour image information of the observer within the preset data acquisition position range by using a person detection and recognition algorithm, and to acquire the projection area of the sand table; the depth camera is used to acquire distance information between the observer and the sand table within the preset data acquisition position range according to the first preset acquisition frequency; and the eye tracker sensor is used to capture eye data of the observer within the preset data acquisition position range.

[0056] The binocular camera is used to identify the image of the observer and the distance information from the position of the sand table. The binocular camera can realize character recognition by capturing images and cooperating with corresponding algorithms, and can also judge the distance based on the principle of human eyes looking at things, that is, the smaller the parallax, the farther the distance, and the larger the parallax, the closer the distance. The role of the depth camera is to collect the distance information between the observer and the sand table according to the parallax of the binocular camera. The role of the eye tracker sensor is to capture eye data to identify how many people are currently observing the content of the sand table, so as to obtain the actual number of observers and avoid the situation that people around the sand table gather on one side for some reason, but do not observe the content of the sand table, resulting in misjudgment of the data processing terminal 2. Finally, the three sensors send the collected data to the observer data processing module of the data processing terminal 2 for data processing.

[0057] Referring to Figure 3 The sand table around the sand table is divided into four preset data collection directions with the same azimuth, that is, the first preset data collection direction, the second preset data collection direction, the third preset data collection direction and the fourth preset data collection direction, which correspond to the positions A, B, C and D of the observer respectively, and then the data collection device 1 is arranged on the four directions.

[0058] Further, in an embodiment, the data processing terminal 2 specifically includes an observer data processing module, a projection display prediction module and a projection output module.

[0059] The observer data processing module is used for character modeling processing of the data information of the observer and the distance information between the observer and the sand table, obtaining actual observer information, and analyzing the preset data collection direction with the most observers;

[0060] The projection display prediction module is used for processing the actual observer information and the projection area of the sand table by using the projection display prediction model, the new projection direction and the new projection area after changing the projection direction, and calculating the best projection data in combination with the distance information between the observer and the sand table.

[0061] The projection output module is used for transmitting the projection data to the projector 3.

[0062] Referring to Figure 4 , the present application provides a projection method of an artificial intelligence following type projection sand table, which is applied to the artificial intelligence following type projection sand table provided in the first aspect, and the projection method comprises:

[0063] S1: identifying observer data, collecting the projection area of the sand table, the data information of the observer and the distance information between the observer and the sand table by using the data collection device 1.

[0064] S2: Observer data processing, which processes the observer data and distance information, calculates the projection data using the projection display prediction model, and sends it to the projector 3;

[0065] S3: Projection display control, adjusting the projection parameters of the device to project images.

[0066] Further, in one embodiment, step S1 specifically includes: first, the binocular camera acquires image data within a preset data acquisition area at a first preset acquisition frequency, and uses a person detection and recognition algorithm to identify the outline image information of the observer within the preset data acquisition area, while simultaneously acquiring the projection area of ​​the sand table; then, the depth camera acquires the distance information between the observer and the sand table within the preset data acquisition area at a first preset acquisition frequency; finally, the eye tracker sensor captures the human eye data of the observer within the preset data acquisition area.

[0067] Furthermore, in one embodiment, step S2 specifically includes the following sub-steps:

[0068] S201: Perform character modeling on the data information of the observers and their distance information from the sand table to obtain the actual observer information and analyze the preset data collection location with the most observers.

[0069] S202: The projection display prediction model is used to process the actual observer information and the projection area of ​​the sand table, respectively, to obtain the new projection direction and the new projection area after changing the projection direction, and to calculate the optimal projection data by combining the distance information between the observer and the sand table.

[0070] S203: Transmit the optimal projection data to the projector 3.

[0071] The purpose of step S201 is to collect sensor information from four directions. Then, the observer data processing module in the data processing terminal 2 performs modeling processing based on this information. The specific processing procedure is as follows:

[0072] S2011: First, combining the contour image information and distance information of the observers within the preset data acquisition range collected by the binocular camera and depth camera, images of all observers around the sand table are drawn, such as... Figure 5 As shown, each observer image contains the outline information of the identified observer and its location. The circular points represent observers around the sand table identified by the binocular cameras.

[0073] S022: Based on the eye data of observers within various preset data acquisition ranges collected by eye tracker sensors, observers not fixating on the sand table are used as dummy observers for eye data analysis, such as... Figure 6 As shown, the rectangular points represent fake observers identified by eye-tracking sensors, i.e., people who are not actually looking at the sand table. Figure 7 As shown, by combining the data collected by the eye tracker sensor for correction and eliminating false observers, the actual number of observers within each preset data collection location range is obtained.

[0074] S2023: Select any preset data collection location as the starting point, number the actual observers within each preset data collection location range in a clockwise or counterclockwise direction, and compare the number of actual observers within each preset data collection location range to obtain the preset data collection location with the most observers. For example... Figure 8 As shown, data processing terminal 2 uses the lower left corner of the sand table as the starting point and proceeds clockwise to number the observers. Then, it combines the information of each observer to create a data table. The distance information of the observers is represented by D1 to D4, and the preset data collection location of the sand table is represented by the geographical directions East, West, South, and North. The final data table is as follows:

[0075] Table 1 Observer Information Table

[0076]

[0077]

[0078] By comparing the number of people in each direction, the direction with the most people around the sand table is determined.

[0079] S2024: Finally, the data from steps S2021 to S2023 are sent as observer information to the projection display prediction model for further processing.

[0080] Furthermore, in one embodiment, in step S202, the data processing terminal 2 processes the data based on the observer information and the projection display prediction model. The purpose of this process is to calculate the projection parameters, including the projection direction, the projection size, and the projection resolution, based on the projection display area data and the observer data, combined with the projection display prediction model. The specific process is as follows:

[0081] S2021: Based on the projection area of ​​the sand table captured by the binocular camera, the actual projection area data of the projector 3 is calculated using a projection area algorithm and sent to the projection display prediction model. The projection display prediction model is implemented using patent application number CN201910239617.1, which discloses a method for creating a laser holographic projection sand table. This method provides a laser holographic projection sand table prediction model, which allows for real-time adjustment of the projection parameters of the laser holographic projector. The projection area algorithm is implemented using the existing method for calculating the projection area of ​​the projector 3. Since the actual size of the sand table and the parameters of the projector 3 are known, only the projection area of ​​the sand table needs to be captured; the corresponding projection area data can be calculated based on the actual size of the sand table and the size of the sand table in the captured image.

[0082] S2022: The projection display prediction model imports observer information and projection area data, and performs data cleaning on the observer information and projection area data according to preset data cleaning rules. It determines whether there are any anomalies in each feature value in the data, including missing values, duplicate values, and outliers. Then, it uses corresponding anomaly handling methods to deal with the anomalies in the data, such as using mean imputation, cluster filling, and multiple imputation prediction to handle missing values, deduplication of duplicate values, and deletion of outliers.

[0083] S2023: After data cleaning is completed, as follows Figure 9 As shown, the projection display prediction model determines the new projection direction based on the preset data collection location with the most people in the observer information. Then, as... Figure 10 As shown, the projection display prediction model calculates the new projection area based on the new projection direction and projection area data.

[0084] S2024: The projection display prediction model calculates the optimal projection parameters based on the new projection direction and the actual number of observers and their positions in the observer information. The projection parameters include the projection offset angle, the resolution of the projected image, and the projection size.

[0085] Furthermore, the projector 3 adjusts the projection data according to the projection parameters. First, the projector 3 receives the projection information output parameters provided by the data processing terminal 2; then the projector 3 performs parameter parsing and projects the image based on the parsed parameters.

[0086] The advantage of this application lies in overcoming the limitation imposed on the observer's viewing position by the projection direction. In scenarios where geographic information is projected onto a sand table, data processing is performed using sensors within the sand table in conjunction with corresponding analysis algorithms. This allows for the dynamic adjustment of the projection direction and effect of the sand table based on changes in the observer's position.

[0087] It is understood that the same or similar parts in the above embodiments can be referred to each other, and the contents not described in detail in some embodiments can be referred to the same or similar contents in other embodiments.

[0088] It should be noted that in the description of this application, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Furthermore, in the description of this application, unless otherwise stated, "multiple" or "more" means at least two.

[0089] It should be understood that when an element is referred to as “fixed to” or “set on” another element, it may be directly on the other element or may be interposed with an intervening element; when an element is referred to as “connected to” another element, it may be directly connected to the other element or may be interposed with an intervening element. Furthermore, the term “connected” as used herein may include wireless connections; the word “and / or” as used includes any and all combinations of one or more of the associated listed items.

[0090] Any process or method description in the flowchart or otherwise herein can be understood as: representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the function involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0091] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0092] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0093] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0094] The storage media mentioned above can be read-only memory, disk, or optical disk, etc.

[0095] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0096] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. An artificial intelligence follow-up projection sand table, characterized in that, The application relates to a sand table projection method and device. The application comprises the following: a data acquisition device for acquiring the projection area of a sand table, data information of an observer and distance information between the observer and the sand table; a data processing terminal connected to the data acquisition device for processing the data information of the observer and the distance information, calculating projection data by using a projection display prediction model and transmitting the projection data to a projector; the projector connected to the data processing terminal for adjusting the projection parameters of the equipment according to the projection data to project an image; the data acquisition device comprises binocular cameras, depth cameras and eye movement sensor sensors arranged at preset data acquisition positions around the sand table, and the data processing terminal is connected to the binocular cameras, the depth cameras and the eye movement sensor sensors; the binocular cameras are used for acquiring picture images in a preset data acquisition position range at a first preset acquisition frequency, recognizing the contour image information of the observer in the preset data acquisition position range by using a character detection and recognition algorithm and simultaneously acquiring the projection area of the sand table; the depth cameras are used for acquiring the distance information between the observer and the sand table in the preset data acquisition position range at the first preset acquisition frequency; the eye movement sensor sensors are used for capturing the eye data of the observer in the preset data acquisition position range; the data processing terminal specifically comprises an observer data processing module, a projection display prediction module and a projection output module; the observer data processing module is used for character modeling processing of the data information of the observer and the distance information between the observer and the sand table, obtaining actual observer information and analyzing a preset data acquisition position with the largest number of observers; the projection display prediction module is used for processing the actual observer information and the projection area of the sand table by using a projection display prediction model, acquiring a new projection direction and a new projection area after changing the projection direction and combining the distance information between the observer and the sand table to calculate optimal projection data; the projection output module is used for transmitting the projection data to the projector; 2. A projection method of an artificial intelligence following projection sand table, applied to the artificial intelligence following projection sand table of claim 1, characterized in that, the preset data acquisition positions comprise four preset data acquisition positions with the same azimuth, namely a first preset data acquisition position, a second preset data acquisition position, a third preset data acquisition position and a fourth preset data acquisition position. The projection method comprises the following steps: S1: recognizing observer data, acquiring the projection area of a sand table, data information of an observer and distance information between the observer and the sand table by using a data acquisition device; 3. The projection method of the artificial intelligence following type projection sand table according to claim 2, characterized in that, S2: processing observer data, processing the data information of the observer and the distance information, calculating projection data by using a projection display prediction model and transmitting the projection data to a projector; S3: projection display control, adjusting the projection parameters of the equipment according to the projection data to project an image. The step S1 specifically comprises the following steps: firstly, acquiring picture images in a preset data acquisition position range at a first preset acquisition frequency by using binocular cameras, recognizing the contour image information of the observer in the preset data acquisition position range by using a character detection and recognition algorithm and simultaneously acquiring the projection area of the sand table; secondly, acquiring the distance information between the observer and the sand table in the preset data acquisition position range at the first preset acquisition frequency by using depth cameras; and finally, capturing the eye data of the observer in the preset data acquisition position range by using eye movement sensor sensors.

4. The projection method of the artificial intelligence following type projection sand table according to claim 2, characterized in that, The step S2 specifically comprises the following sub-steps: S201: performing character modeling processing on the data information of the observers and the distance information between the observers and the sand table, obtaining actual observer information, and analyzing a preset data acquisition orientation with the most number of observers; S202: processing the actual observer information and the projection area of the sand table respectively by using a projection display prediction model, obtaining a new projection direction and a new projection area after changing the projection direction, and calculating optimal projection data in combination with the distance information between the observers and the sand table; S203: transmitting the optimal projection data to the projector.

5. The projection method of the artificial intelligence following type projection sand table according to claim 4, characterized in that, The sub-step S201 specifically comprises: S2011: first, combining the contour image information and the distance information of the observers in each preset data acquisition orientation range, drawing all observer images around the sand table, each observer image containing recognized observer contour information and a position thereof; S2012: based on the eye data of the observers in each preset data acquisition orientation range, performing eye data analysis on the observers not gazing at the sand table as false observers, and obtaining the actual number of observers in each preset data acquisition orientation range; S2013: selecting an arbitrary preset data acquisition orientation as a starting point, numbering the actual observers in each preset data acquisition orientation range in a clockwise or counterclockwise direction, and comparing the actual number of observers in each preset data acquisition orientation range, to obtain a preset data acquisition orientation with the most number of observers; S2014: sending the data in steps S2011-S2013 as observer information to the projection display prediction model.

6. The projection method of the artificial intelligence following type projection sand table according to claim 4, characterized in that, The sub-step S202 specifically comprises: S2021: based on the projection area of the sand table, calculating actual projection area data of the projector by using a projection area algorithm and sending the projection area data to the projection display prediction model; S2022: the projection display prediction model imports the observer information and the projection area data, and performs data cleaning on the observer information and the projection area data according to a preset data cleaning rule; S2023: the projection display prediction model judges a new projection direction according to the preset data acquisition orientation with the most number of observers in the observer information, and calculates a new projection area based on the new projection direction and the projection area data; S2024: the projection display prediction model calculates optimal projection parameters in combination with the actual number of observers, the positions of the observers, and the new projection direction in the observer information, wherein the projection parameters include a projection offset angle, a projection picture resolution, and a projection size.

Images