AI-driven sand table picture emotion and psychological state analysis and recognition data processing method and system

Abstract

The invention provides an AI-driven sand table picture emotional psychological state analysis and recognition data processing method and system, and the method comprises the steps: receiving a prototype construction graph collected by a user side for a customized sand table based on an auxiliary collection interface, recognizing virtual elements corresponding to all entity elements in the prototype construction graph, and constructing a sand base model; determining virtual positions of the virtual elements according to a mapping positioning strategy and the prototype construction graph, and updating the virtual elements to the sand base model based on the virtual positions to obtain an element layout model; determining a virtual orientation of a corresponding virtual element in the element layout model based on an orientation positioning strategy and the prototype construction graph, and updating the corresponding virtual element according to the virtual orientation to obtain an orientation conversion model; and obtaining a sand table type of the customized sand table, calling a plurality of analysis identification angles based on the sand table type, and intercepting the directional conversion model according to the plurality of analysis identification angles to obtain a plurality of analysis training graphs.

Description

Technical Field

[0001] The present invention relates to data processing technology, and in particular to an AI-driven sandbox image emotional and psychological state analysis and recognition data processing method and system. Background Art

[0002] In many fields such as psychology, education, business management and urban planning, sandbox simulation is widely used as an important tool to simulate and predict psychological, behavioral and economic changes in various scenarios. With the rapid development of artificial intelligence technology, especially the advancement of image processing and deep learning technology, new solutions have been provided for the emotional and psychological state analysis of sandbox images.

[0003] In the prior art, personnel are usually required to shoot the placed sand table model from multiple angles, which places high technical requirements on the photographers, so as to obtain multiple pictures and upload them to the sand table picture feature recognition model for model training, so as to realize sand table analysis and obtain the psychological analysis results of the personnel. However, when personnel shoot the sand table from multiple angles and upload the sand table pictures, the shooting angles are prone to deviations and the efficiency is low, which seriously affects the analysis results.

[0004] Therefore, how to collect accurate sand table pictures according to needs, save manual shooting time, and improve the accuracy and collection efficiency of sand table pictures has become an urgent problem to be solved. Summary of the invention

[0005] The present invention provides an AI-driven sand table picture emotional psychological state analysis and recognition data processing method and system, which can collect accurate sand table pictures according to needs, save manual shooting time, and improve the accuracy and collection efficiency of sand table pictures.

[0006] A first aspect of the present invention provides an AI-driven sandbox image emotional psychological state analysis and recognition data processing method, comprising: Receiving a prototype construction diagram collected by the user end based on the auxiliary collection interface for the customized sand table, identifying virtual elements corresponding to each physical element in the prototype construction diagram, and constructing a sand base model; Determine the virtual position of the virtual element according to the mapping positioning strategy and the prototype construction diagram, and update the virtual element to the sand base model based on the virtual position to obtain an element layout model; Determine the virtual orientation of the corresponding virtual element in the element layout model based on the orientation positioning strategy and the prototype construction diagram, and update the corresponding virtual element according to the virtual orientation to obtain an orientation conversion model; The sandbox type of the customized sandbox is obtained, a plurality of analysis and identification angles are retrieved based on the sandbox type, and the directional conversion model is intercepted according to the plurality of analysis and identification angles to obtain a plurality of analysis training diagrams.

[0007] Optionally, in a possible implementation manner of the first aspect, the receiving user end collects a prototype construction diagram of the customized sandbox based on the auxiliary collection interface, and identifying virtual elements corresponding to each physical element in the prototype construction diagram, including: Receiving a prototype construction diagram collected by the user end based on the auxiliary collection interface for the customized sandbox, and identifying the entity outlines of the entity elements in the prototype construction diagram based on OpenCV; Extracting pixel values ​​within the entity contour as identification pixel values, and selecting preset elements with the identification pixel values ​​in the preset element library as candidate elements; Determine the identification pixel points within the entity outline based on the identification pixel values, and count the number of identification pixel points corresponding to each identification pixel value to obtain the identification number; Counting the identified numbers to obtain a total number, and obtaining the actual pixel value ratio of each identified pixel value according to the ratio of the identified number to the total number; Obtaining a preset pixel value ratio corresponding to the identified pixel value in the candidate element, and obtaining a sub-difference corresponding to each identified pixel value according to an absolute value of a difference between the preset pixel value ratio corresponding to each identified pixel value and an actual pixel value ratio; The sub-differences are counted to obtain a total difference corresponding to each of the candidate elements, and the candidate element corresponding to the smallest total difference is selected as the virtual element corresponding to the corresponding physical element.

[0008] Optionally, in a possible implementation of the first aspect, the method further includes: When determining that the prototype construction diagram has intersecting entity contours, taking the corresponding entity contours as overlapping contours; The virtual elements corresponding to the overlapping contours with gaps in the overlapping contours are acquired as active adjustment elements, and the remaining virtual elements are acquired as automatic adjustment elements.

[0009] Optionally, in a possible implementation of the first aspect, determining the virtual position of the virtual element according to the mapping positioning strategy and the prototype construction diagram, and updating the virtual element to the sand base model based on the virtual position to obtain the element layout model includes: Connect the midpoints of the upper and lower boundary lines in the prototype construction diagram to obtain a first center line, and connect the midpoints of the left and right boundary lines in the prototype construction diagram to obtain a second center line; Based on the intersection of the first center line and the second center line, obtaining the center point of the image corresponding to the prototype construction diagram; A coordinate system is constructed with the center point of the image as the coordinate origin, the center point of the base corresponding to each entity element in the prototype construction diagram is obtained as the entity midpoint, and the entity coordinates corresponding to the entity midpoint are determined; Determine the left side proportion, right side proportion, top side proportion and bottom side proportion corresponding to the midpoint of the entity according to the entity coordinates; Based on the left side proportion and the right side proportion corresponding to the midpoint of the entity, a virtual horizontal coordinate in the sand base model is determined, and according to the upper side proportion and the lower side proportion, a virtual vertical coordinate in the sand base model is determined; The virtual position of the virtual element is determined according to the virtual horizontal coordinate and the virtual vertical coordinate, and the virtual element is updated to the sand base model based on the virtual position to obtain an element layout model.

[0010] Optionally, in a possible implementation manner of the first aspect, determining, according to the entity coordinates, a left side proportion, a right side proportion, an upper side proportion, and a lower side proportion corresponding to the midpoint of the entity, includes: Determine a first ordinate corresponding to the upper boundary line and a second ordinate corresponding to the lower boundary line in the prototype construction diagram; Construct a transverse vertical line perpendicular to the longitudinal axis based on the entity coordinates, and obtain a first intersection point of the transverse vertical line with the left boundary line and a second intersection point of the right boundary line in the prototype construction diagram; Obtain a first horizontal coordinate corresponding to the first intersection point and a second horizontal coordinate corresponding to the second intersection point; Extracting the entity horizontal coordinate and the entity vertical coordinate corresponding to the entity coordinate, calculating based on the entity horizontal coordinate and the first horizontal coordinate to obtain the left distance, and calculating based on the entity horizontal coordinate and the second horizontal coordinate to obtain the right distance; According to the sum of the left distance and the right distance, a horizontal distance is obtained, according to the left distance and the horizontal distance, a left proportion is obtained, and based on the right distance and the horizontal distance, a right proportion is obtained; Calculate based on the entity ordinate and the first ordinate to obtain an upper distance, and calculate based on the entity ordinate and the second ordinate to obtain a lower distance; The longitudinal distance is obtained according to the sum of the upper distance and the lower distance, the upper proportion is obtained according to the upper distance and the longitudinal distance, and the lower proportion is obtained according to the lower distance and the longitudinal distance.

[0011] Optionally, in a possible implementation of the first aspect, determining a virtual orientation of a corresponding virtual element in the element layout model based on the orientation positioning strategy and the prototype construction diagram, and updating the corresponding virtual element according to the virtual orientation to obtain a directional conversion model includes: Retrieving the auxiliary angle corresponding to the auxiliary acquisition interface, adjusting the element layout model based on the auxiliary angle, and obtaining a direction positioning model; Acquire element attributes of the automatically adjusted element in the virtual element in the orientation positioning model, wherein the element attributes include a rotation attribute and a fixed attribute; When it is determined that the element attribute is a rotation attribute, the corresponding automatically adjusted element is used as a rotation element; Taking the virtual position of the rotating element as the center point, rotating the rotating element in the orientation positioning model in real time, obtaining the image center point of the intercepted image corresponding to the orientation positioning model, and constructing a coordinate system based on the image center point of the intercepted image; Based on OpenCV, a rotation contour of a rotating element in the orientation positioning model is acquired in real time, and the rotation contour coordinates of contour pixels in the rotation contour are extracted; The virtual orientation of each virtual element in the orientation positioning model is determined according to the rotation contour coordinates, and the corresponding virtual element is updated based on the virtual orientation to obtain an orientation conversion model.

[0012] Optionally, in a possible implementation manner of the first aspect, determining the virtual orientation of each virtual element in the orientation positioning model according to the rotation contour coordinates, and updating the corresponding virtual element based on the virtual orientation to obtain the orientation conversion model includes: Counting the rotation contour coordinates to obtain a real-time coordinate set corresponding to the rotation element; Based on OpenCV, the entity contour corresponding to the entity element in the prototype construction diagram is obtained, and the entity contour coordinates of the contour pixel points corresponding to the entity contour are counted to obtain an entity coordinate set; When it is determined that the real-time coordinate set is equal to the physical coordinate set, the current orientation corresponding to the corresponding rotation element is used as a virtual orientation; The automatic adjustment element is updated based on the virtual orientation to obtain a directional conversion model.

[0013] Optionally, in a possible implementation manner of the first aspect, determining the virtual orientation of each virtual element in the orientation positioning model according to the rotation contour coordinates, and updating the corresponding virtual element based on the virtual orientation to obtain the orientation conversion model includes: Counting the rotation contour coordinates to obtain a real-time coordinate set corresponding to the rotation element, and obtaining a maximum horizontal coordinate, a minimum horizontal coordinate, a maximum vertical coordinate, and a minimum vertical coordinate in the real-time coordinate set; Obtaining a virtual horizontal distance corresponding to the rotation element according to a difference between a maximum horizontal coordinate and a minimum horizontal coordinate in the real-time coordinate set; Obtaining a virtual longitudinal distance corresponding to the rotation element based on a difference between a maximum longitudinal coordinate and a minimum longitudinal coordinate in the real-time coordinate set; Obtaining the entity contour corresponding to the corresponding entity element in the prototype construction diagram according to OpenCV, and counting the entity contour coordinates corresponding to the entity contour to obtain an entity coordinate set; Based on the difference between the maximum horizontal coordinate and the minimum horizontal coordinate in the entity coordinate set, obtaining the entity horizontal distance corresponding to the corresponding entity element; Obtaining a physical longitudinal distance corresponding to the corresponding physical element according to a difference between a maximum vertical coordinate and a minimum vertical coordinate in the physical coordinate set; Based on the absolute value of the difference between the physical horizontal distance and the virtual horizontal distance, a horizontal judgment value is obtained, and based on the absolute value of the difference between the physical vertical distance and the virtual vertical distance, a vertical judgment value is obtained; When it is determined that both the horizontal judgment value and the vertical judgment value are smaller than a preset difference, the current orientation corresponding to the corresponding rotation element is used as a virtual orientation; The corresponding automatically adjusted element is updated based on the virtual orientation to obtain a directional conversion model.

[0014] Optionally, in a possible implementation of the first aspect, the method further includes: When it is determined that the rotating element does not have the virtual orientation, the corresponding rotating element is used as a manual orientation element; The manual adjustment element and the active adjustment element are sent to an analysis end.

[0015] A second aspect of the present invention provides an AI-driven sandbox image emotional and psychological state analysis and recognition data processing system, comprising: The recognition module is used to receive the prototype construction diagram collected by the user end based on the auxiliary collection interface for the customized sand table, identify the virtual elements corresponding to the physical elements in the prototype construction diagram, and build a sand base model; A determination module, used to determine the virtual position of the virtual element according to the mapping positioning strategy and the prototype construction diagram, and update the virtual element to the sand base model based on the virtual position to obtain an element layout model; An updating module, used for determining the virtual orientation of the corresponding virtual element in the element layout model based on the orientation positioning strategy and the prototype construction diagram, and updating the corresponding virtual element according to the virtual orientation to obtain an orientation conversion model; The interception module is used to obtain the sandbox type of the customized sandbox, call multiple analysis and identification angles based on the sandbox type, intercept the directional conversion model according to the multiple analysis and identification angles, and obtain multiple analysis training diagrams.

[0016] According to a third aspect of the present invention, there is provided an electronic device, comprising: a memory, a processor and a computer program, wherein the computer program is stored in the memory, and the processor runs the computer program to execute the first aspect of the present invention and various methods that may be involved in the first aspect.

[0017] According to a fourth aspect of the present invention, a storage medium is provided, in which a computer program is stored. When the computer program is executed by a processor, it is used to implement the first aspect of the present invention and various methods that may be involved in the first aspect.

[0018] The beneficial effects of the present invention are as follows: 1. The present invention can collect accurate sand table pictures according to the needs, save manual shooting time, and improve the accuracy and collection efficiency of sand table pictures. First, the present invention can determine the virtual elements corresponding to the physical elements in the prototype construction diagram according to the received prototype construction diagram of the customized sand table, and at the same time construct the sand base model of the corresponding sand table, so as to facilitate the subsequent update of the corresponding virtual elements in the sand base model. Secondly, the present invention can determine the virtual position of the virtual element, so as to update the virtual element at the corresponding virtual position in the sand base model, so as to obtain the element layout model, so as to adjust the orientation of the virtual element in the element layout model, so as to obtain the directional conversion model corresponding to the customized sand table, so that the display image in the intercepted analysis training diagram corresponds to the actual customized sand table, improve the accuracy of the collected image, and can realize the automatic interception of multiple analysis and recognition angles of the directional conversion model, and improve the image acquisition efficiency.

[0019] 2. The present invention can determine the virtual position of the corresponding virtual element by the position ratio of the physical element in the prototype construction diagram, improve the accuracy of determining the virtual position, and further enhance the accuracy of the element layout model. Among them, the present invention can establish a coordinate system in the prototype construction diagram, so as to determine the horizontal left and right proportions of the physical coordinates corresponding to the physical element and the vertical upper and lower proportions, so as to obtain the virtual position of the physical element corresponding to the virtual element, avoid the deviation of the coordinates of the physical element caused by the shooting angle, thereby improving the accuracy of the determined virtual position, and further improving the accuracy of the subsequent analysis training diagram corresponding to the sand table image.

[0020] 3. The present invention can adjust the orientation of the virtual elements in the element layout model, thereby obtaining a directional conversion model corresponding to the customized sandbox, so as to obtain multiple accurate analysis training diagrams corresponding to multiple analysis and recognition angles. Among them, the present invention can adjust the element layout model according to the auxiliary angle to obtain the orientation positioning model, so as to facilitate the determination of the virtual orientation of the virtual elements in the element layout model according to the orientation of the physical elements in the prototype construction diagram, and the present invention can distinguish the automatically adjusted elements according to the element attributes, so as to reduce the subsequent recognition of the images of the physical elements corresponding to the fixed attributes of the element attributes in the prototype construction diagram, thereby reducing the amount of data processing, and finally, the present invention can compare the real-time coordinate set of the rotating element with the physical outline of the physical element in the prototype construction diagram, so as to determine the virtual orientation, and then update the element layout model to obtain the directional conversion model, so as to automatically intercept the directional conversion model, reduce the acquisition time of personnel shooting, and at the same time improve the accuracy of the analysis training diagram. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 A flowchart of an AI-driven sandbox image emotional and psychological state analysis and recognition data processing method provided by the present invention; Figure 2 A schematic diagram of a prototype construction diagram provided by the present invention; Figure 3 A schematic diagram of an image center point corresponding to a prototype construction diagram provided by the present invention; Figure 4 This is a structural schematic diagram of an AI-driven sandbox image emotional and psychological state analysis and recognition data processing system provided by the present invention; Figure 5 A schematic diagram of the hardware structure of an electronic device provided by the present invention. DETAILED DESCRIPTION

[0022] The technical solution of the present invention is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0023] like Figure 1 As shown, the present invention provides an AI-driven sand table picture emotional psychological state analysis and recognition data processing method, the AI-driven sand table picture emotional psychological state analysis and recognition data processing method comprises: S1, receiving a prototype construction diagram collected by the user end based on the auxiliary collection interface for the customized sand table, identifying virtual elements corresponding to each physical element in the prototype construction diagram, and constructing a sand base model.

[0024] It should be noted that a corresponding twin model can be constructed for the sand table where the personnel are placed, so that the twin model can be automatically intercepted from multiple angles to capture the sand table pictures, ensuring the accuracy of the angles of the collected sand table pictures. At the same time, the efficiency of collecting multi-angle sand table pictures can be improved. Therefore, personnel are required to take a sand table picture first so that the twin model can be constructed later.

[0025] It can be understood that the user end is the information terminal of the user who uploads the sand table image, for example, it can be a mobile phone, and the auxiliary acquisition interface is an interface for assisting personnel to collect sand table images, which can be artificially pre-constructed. Since the shape of the border outline in the corresponding sand table picture will also change with different shooting angles, in order to obtain the sand table picture of the corresponding angle, so as to subsequently build the twin model of the sand table, the sand table can be imaged with the help of the auxiliary acquisition interface. For example, when the sand table is imaged at a front view angle or a top view angle, the models placed in the sand table may block each other, resulting in missing elements and affecting the analysis results. In order to more completely identify the models and positions placed in the sand table in the future, a shooting posture can be pre-adjusted so that it can maximize the clear display of the element models placed in the sand table. Therefore, in order for personnel to accurately collect sand table pictures of the corresponding angles, image acquisition can be performed through the auxiliary interface, for example, the boundary outline in the auxiliary acquisition interface is aligned with the border of the custom sand table, so as to shoot and obtain a prototype construction diagram.

[0026] Among them, the customized sand table is the sand table for corresponding personnel to conduct testing, the prototype construction diagram is the prototype image for reference in building the virtual model, the physical elements are the element models placed in the prototype construction diagram, the virtual elements are the virtual element models corresponding to the physical elements, and the sand base model is the model corresponding to the base in the sand table.

[0027] It is not difficult to understand that OpenCV can be used to identify the pixel values ​​corresponding to the sand in the prototype component diagram, thereby identifying the corresponding sand stacking situation, such as a sand pit or a sand pile. When occlusion occurs, the prototype construction diagram can be manually observed to autonomously adjust the shape of the corresponding position, thereby constructing a sand base model.

[0028] It is worth mentioning that in order to obtain accurate psychological analysis of personnel through analysis of customized sand tables, it is necessary to collect images of the sand table from multiple angles. At present, it is necessary to manually judge the shooting angle to take pictures of the sand table, which places high technical requirements on the photographers. Due to the inaccurate shooting angle, the accuracy of the collected images is seriously affected. Therefore, the present invention can construct a corresponding model through a collected prototype construction diagram, so as to collect images from multiple angles according to needs, reduce the collection time, improve the collection efficiency and ensure the accuracy of the collected images.

[0029] Through the above implementation, the present invention can obtain a prototype construction diagram and construct a sand base model, so as to facilitate the subsequent construction of a corresponding customized sand table model according to the sand base model, thereby facilitating the collection of corresponding accurate sand table pictures.

[0030] In some embodiments, the specific implementation of step S1 (receiving the prototype construction diagram collected by the user end based on the auxiliary collection interface for the customized sandbox, and identifying the virtual elements corresponding to the physical elements in the prototype construction diagram) includes: S11, receiving a prototype construction diagram collected by the user end based on the auxiliary collection interface from the customized sandbox, and identifying the entity outlines of the entity elements in the prototype construction diagram based on OpenCV.

[0031] It is understandable that if Figure 2 As shown, after receiving the prototype construction drawing, the OpenCV recognition technology can be used to recognize the model contours of the physical elements in the prototype construction drawing, so as to subsequently select the corresponding virtual elements and update the sand base model.

[0032] Among them, OpenCV provides a large number of image processing functions and algorithms, which are mainly used for real-time image processing and computer vision tasks. It can recognize images to facilitate the subsequent determination of virtual elements for model construction, and the entity contour is the model contour of the entity element.

[0033] S12, extracting pixel values ​​within the entity contour as identification pixel values, and selecting preset elements with the identification pixel values ​​in a preset element library as candidate elements.

[0034] It should be noted that the physical element models that can be placed in the sandbox are limited. Therefore, the corresponding preset elements in the preset element library are also limited, and there are many types of corresponding models, such as people, vehicles, natural landscapes, etc. After classifying the models, the number of element models of each type will become smaller. Therefore, the preset elements can be quickly selected by the outline of the corresponding element, the pixel values ​​and the corresponding pixel value ratio.

[0035] It can be understood that the identified pixel value is the pixel value within the entity contour. For example, when the entity elements in the prototype construction diagram are identified to obtain the corresponding entity contour, when the pixel values ​​extracted within the corresponding entity contour are red and white, the corresponding red and white can be used as the identified pixel values.

[0036] Among them, the preset element library is a pre-set database with element models corresponding to all sandbox models, the preset element is the element model in the preset element library, and the candidate element is the preset element in the preset element library with the identification pixel value. For example, when the identification pixel values ​​corresponding to the current entity element are extracted as red and white, the preset element model with red and white colors in the preset element library will be used as a candidate model.

[0037] Through the above implementation, the present invention can obtain the selected elements, so as to further determine the virtual elements corresponding to the physical elements in the subsequent process.

[0038] S13, determining the identification pixel points within the entity contour based on the identification pixel values, and counting the number of identification pixel points corresponding to each identification pixel value to obtain the identification number.

[0039] It should be noted that different preset elements in the preset element library may have the same identification pixel value, but since the sizes of the image areas corresponding to the various identification pixel values ​​are different, the number of pixels will also be inconsistent. Therefore, the candidate elements can be further selected according to the identification number to select the accurate virtual elements.

[0040] It can be understood that the identified pixel points are the pixel points corresponding to the identified pixel values ​​within the entity contour, and the identified quantity is the quantity of the identified pixel points corresponding to each identified pixel value.

[0041] S14, counting the identified numbers to obtain a total number, and obtaining an actual pixel value ratio of each identified pixel value according to a ratio of the identified number to the total number.

[0042] It can be understood that the total number is the sum of the identification numbers corresponding to multiple identification pixel values ​​in the entity contour corresponding to each entity element. For example, when there are 2 identification pixel values ​​in the entity contour, which are red and white, and the corresponding identification numbers are 2000 and 500 respectively, the total number is 2000+500=2500.

[0043] Among them, the actual pixel value ratio is the ratio of the pixel points corresponding to each identification pixel value. For example, when the total number is 2500, and the number of identification pixels with red identification value is 2000, the actual pixel value ratio of red is 2000 / 2500=0.8, and when the number of identification pixels with white identification value is 500, the corresponding actual pixel value ratio of white is 500 / 2500=0.2.

[0044] Through the above implementation, the present invention can obtain the actual pixel value ratio of each identified pixel value, so as to compare it with the candidate elements in the subsequent process, thereby selecting the corresponding virtual element.

[0045] S15, obtaining the preset pixel value ratios corresponding to the identified pixel values ​​in the candidate elements, and obtaining the sub-differences corresponding to the identified pixel values ​​according to the absolute values ​​of the differences between the preset pixel value ratios corresponding to the identified pixel values ​​and the actual pixel value ratios.

[0046] It should be noted that the actual pixel value ratio calculated from the pixel values ​​identified in the prototype component diagram may deviate from the preset pixel value ratio of the corresponding preset element. Therefore, the absolute value of the difference between the actual pixel value ratio and the preset pixel value ratio can be calculated to facilitate the subsequent selection of the selected element with the smallest difference as the virtual element, thereby improving the accuracy of element selection.

[0047] It can be understood that the preset pixel value ratio is the ratio of each pixel value determined by calculating each identified pixel value in the candidate element in advance, and the sub-difference is the absolute value of the difference between the preset pixel value ratio and the actual pixel value ratio corresponding to each identified pixel value.

[0048] Through the above implementation, the present invention can obtain the sub-difference corresponding to each identified pixel value, so as to determine the corresponding virtual element later.

[0049] S16, counting the sub-differences to obtain a total difference corresponding to each of the candidate elements, and selecting a candidate element corresponding to the smallest total difference as a virtual element corresponding to the corresponding physical element.

[0050] It can be understood that the total difference is the sum of the sub-differences corresponding to the identified pixel values ​​corresponding to each candidate element. For example, when the sub-difference of the red pixel value in the first candidate element is 0.1, and the sub-difference of the white pixel value is 0.05, the corresponding total difference of the first candidate element is 0.1+0.05=0.15. When the sub-difference of the red pixel value in the second candidate element is 0.3, and the sub-difference of the white pixel value is 0.1, the corresponding total difference of the first candidate element is 0.1+0.3=0.4. Since 0.15<0.4, the first candidate element can be used as a virtual element corresponding to the corresponding physical element.

[0051] Through the above implementation, the present invention can select virtual elements corresponding to corresponding physical elements, so as to facilitate subsequent updating in the sand base model to obtain an element layout model.

[0052] It should be noted that, when the custom sand table is photographed to obtain the prototype construction diagram, the corresponding shooting angle has a certain offset, so there is overlap of elements in the collected prototype construction diagram. For example, a human element model appears behind the corresponding tree model. However, due to the overlap of the images, the placement position of the corresponding physical element in the sand table cannot be intuitively determined. Therefore, the virtual elements can be classified according to the prototype component diagram, so as to facilitate the subsequent placement of the corresponding virtual elements at the corresponding model position to realize the construction of the model. Therefore, in some embodiments, it also includes: A1, when it is determined that the prototype construction diagram has intersecting entity contours, the corresponding entity contours are used as overlapping contours.

[0053] It is understandable that when multiple solid contours intersect in the prototype component drawing, the intersecting solid contours can be treated as overlapping contours to facilitate the subsequent classification of virtual elements corresponding to the solid contours, thereby determining the placement of the virtual elements in the model position.

[0054] A2, obtaining virtual elements corresponding to overlapping contours with gaps in the overlapping contours as active adjustment elements, and taking the remaining virtual elements as automatic adjustment elements.

[0055] It can be understood that the actively adjusted element is a virtual element corresponding to the overlapping contour with a gap in the overlapping contour, and the automatically adjusted element is a virtual element corresponding to the overlapping contour without a gap.

[0056] It is not difficult to understand that when there is a gap in the overlapping contour, it means that the corresponding physical element is blocked. Therefore, the position of the corresponding element model in the sandbox cannot be accurately determined, and thus, it can be actively adjusted manually.

[0057] S2, determining the virtual position of the virtual element according to the mapping positioning strategy and the prototype construction diagram, and updating the virtual element to the sand base model based on the virtual position to obtain an element layout model.

[0058] It can be understood that when the virtual position corresponding to the virtual element is determined, the virtual element can be updated at the corresponding virtual position in the sand base model, thereby obtaining an element layout model to facilitate the subsequent adjustment of the orientation of the virtual element, to achieve automatic capture of the sand table image, to enhance image accuracy, and to improve acquisition efficiency.

[0059] The virtual position is the position of the virtual element placed in the model, and the element layout model is the model for laying out the virtual elements.

[0060] In some embodiments, the specific implementation of step S2 (determining the virtual position of the virtual element according to the mapping positioning strategy and the prototype construction diagram, and updating the virtual element to the sand base model based on the virtual position to obtain the element layout model) includes: S21, connecting the midpoints of the upper and lower boundary lines in the prototype construction diagram to obtain a first center line, and connecting the midpoints of the left and right boundary lines in the prototype construction diagram to obtain a second center line.

[0061] It is understandable that in order to determine the corresponding position of each entity element, so as to construct a corresponding coordinate system in the prototype construction diagram, it is necessary to obtain the center point position of the prototype construction diagram through the first center line and the second center line to facilitate the subsequent establishment of the coordinate system.

[0062] Among them, the upper boundary line is the upper boundary line of the prototype component drawing, the lower boundary line is the lower boundary line of the prototype component drawing, the left boundary line is the left boundary line of the prototype component drawing, the right boundary line is the right boundary line of the prototype component drawing, the first center line is the line between the midpoints of the upper and lower boundary lines in the prototype construction diagram, and the second center line is the line between the midpoints of the left and right boundary lines in the prototype construction diagram.

[0063] S22: Based on the intersection of the first center line and the second center line, obtain an image center point corresponding to the prototype construction image.

[0064] It is understandable that if Figure 3 As shown, the center point of the image is the center point of the prototype construction diagram, that is, the intersection of the first center line and the second center line.

[0065] Through the above implementation, the present invention can obtain the center point of the image, so as to subsequently establish a coordinate system with the center point of the image as the origin.

[0066] S23, constructing a coordinate system with the center point of the image as the coordinate origin, obtaining the center point of the base corresponding to each entity element in the prototype construction diagram as the entity midpoint, and determining the entity coordinates corresponding to the entity midpoint.

[0067] It can be understood that when placing models in the sandbox, in order to enable the models to stand normally, the models corresponding to the physical elements all have corresponding bases to facilitate the placement of the models. Then, the center point of the base corresponding to each physical element can be obtained, thereby obtaining the corresponding physical coordinates, so as to subsequently determine the corresponding position ratio according to the coordinates, so as to subsequently determine the virtual coordinates of the corresponding virtual elements and realize the position update of the virtual elements.

[0068] Among them, the entity midpoint is the center point of the base corresponding to the entity element, and the entity coordinates are the coordinates corresponding to the entity midpoint.

[0069] S24, determining the left side proportion, right side proportion, top side proportion and bottom side proportion corresponding to the midpoint of the entity according to the entity coordinates.

[0070] It is understandable that the boundary lines of the sandbox in the prototype construction diagram will be affected by the shooting angle, resulting in image deviations in the upper and lower boundary lines of the image. For example, when the real sandbox is in the shape of a rectangle, the prototype component diagram photographed shows a trapezoidal shape. Therefore, there may be a large positional deviation in determining the coordinate position based on the coordinate system constructed in the prototype construction diagram. Therefore, the proportions of the midpoint of the entity in multiple directions can be obtained through the coordinates, and the accuracy of the virtual position can be improved based on the proportions.

[0071] Among them, the left side ratio is the ratio of the position of the entity midpoint in the left direction of the image, the right side ratio is the ratio of the position of the entity midpoint in the right direction of the image, the upper side ratio is the ratio of the position above the entity midpoint in the image, and the lower side ratio is the ratio of the position below the entity midpoint in the image.

[0072] Through the above implementation, the present invention can obtain the left side proportion, right side proportion, upper side proportion and lower side proportion corresponding to the midpoint of the entity, so as to obtain the accurate virtual position and realize the update of the virtual element, so as to improve the efficiency of capturing the sandbox picture.

[0073] In some embodiments, the specific implementation of step S24 (determining the left side proportion, right side proportion, top side proportion and bottom side proportion corresponding to the midpoint of the entity according to the entity coordinates) includes: S241, determining a first ordinate corresponding to the upper boundary line and a second ordinate corresponding to the lower boundary line in the prototype construction diagram.

[0074] It can be understood that the first ordinate is the ordinate corresponding to the upper boundary line in the coordinate system, and the second ordinate is the ordinate corresponding to the lower boundary line in the coordinate system.

[0075] Through the above implementation, the present invention can obtain the first ordinate and the second ordinate corresponding to the upper and lower boundary lines in the prototype construction diagram, so as to facilitate subsequent calculation with the physical coordinates, thereby obtaining the corresponding position ratio, so as to facilitate the subsequent determination of the accurate virtual position coordinates.

[0076] S242, constructing a transverse vertical line perpendicular to the longitudinal axis based on the entity coordinates, and obtaining a first intersection point of the transverse vertical line with a left boundary line and a second intersection point of the right boundary line in the prototype construction diagram.

[0077] It can be understood that in order to obtain the proportion of each entity coordinate in different horizontal directions on the left and right, the position of the corresponding entity element is determined, so that a horizontal vertical line perpendicular to the longitudinal axis can be constructed according to the entity coordinates, so as to obtain the first intersection and the second intersection of the horizontal vertical line and the left and right boundary lines, and then obtain the horizontal coordinates of the corresponding intersections, so as to determine the accurate coordinate position of the virtual element corresponding to the entity element of the corresponding entity coordinate in the model.

[0078] Among them, the horizontal vertical line is a straight line segment passing through the entity coordinates and perpendicular to the vertical axis. The first intersection point is the intersection point of the horizontal vertical line and the left boundary line in the prototype construction diagram, and the second intersection point is the intersection point of the horizontal vertical line and the right boundary line in the prototype construction diagram.

[0079] S243: Obtain a first horizontal coordinate corresponding to the first intersection point and a second horizontal coordinate corresponding to the second intersection point.

[0080] It can be understood that the first horizontal coordinate is the horizontal coordinate corresponding to the first intersection point, and the second horizontal coordinate is the horizontal coordinate corresponding to the second intersection point.

[0081] Through the above implementation, the present invention can obtain the first horizontal coordinate and the second horizontal coordinate corresponding to the entity coordinates, so as to obtain the corresponding left-side proportion and right-side proportion subsequently.

[0082] S244, extracting the entity horizontal coordinate and the entity vertical coordinate corresponding to the entity coordinate, calculating based on the entity horizontal coordinate and the first horizontal coordinate to obtain the left distance, and calculating based on the entity horizontal coordinate and the second horizontal coordinate to obtain the right distance.

[0083] It should be noted that the difference between the entity horizontal coordinate in the entity coordinate and the first horizontal coordinate of the first intersection of the corresponding horizontal vertical line is calculated, so as to obtain the distance of the entity coordinate from the left boundary line in the prototype construction diagram, so as to obtain the left side proportion corresponding to the entity coordinate in the subsequent process. At the same time, the difference between the entity horizontal coordinate in the entity coordinate and the second horizontal coordinate of the second intersection of the corresponding horizontal vertical line is calculated, so as to obtain the distance of the entity coordinate from the right boundary line in the prototype construction diagram, so as to obtain the right side proportion corresponding to the entity coordinate in the subsequent process.

[0084] It can be understood that the entity horizontal coordinate is the horizontal coordinate in the entity coordinate, the entity vertical coordinate is the vertical coordinate in the entity coordinate, the left distance is the distance between the entity horizontal coordinate and the first horizontal coordinate, and the right distance is the distance between the entity horizontal coordinate and the second horizontal coordinate.

[0085] S245, obtaining the horizontal distance according to the sum of the left distance and the right distance, obtaining the left proportion according to the left distance and the horizontal distance, and obtaining the right proportion based on the right distance and the horizontal distance.

[0086] It should be noted that in order to calculate the position proportion corresponding to the entity coordinates, it is necessary to obtain the total distance in the horizontal direction where the corresponding entity coordinates are located, so as to calculate the corresponding left proportion and right proportion.

[0087] It can be understood that the lateral distance is the sum of the left distance and the right distance, the left proportion is the ratio of the left distance to the lateral distance, and the right proportion is the ratio of the right distance to the lateral distance.

[0088] S246, calculating based on the entity ordinate and the first ordinate to obtain an upper distance, and calculating based on the entity ordinate and the second ordinate to obtain a lower distance.

[0089] It can be understood that the upper distance is the difference distance between the entity ordinate and the first ordinate, and the lower distance is the difference distance between the entity ordinate and the second ordinate.

[0090] S247, obtaining a longitudinal distance according to the sum of the upper distance and the lower distance, obtaining an upper proportion according to the upper distance and the longitudinal distance, and obtaining a lower proportion according to the lower distance and the longitudinal distance.

[0091] It can be understood that the longitudinal distance is the sum of the upper distance and the lower distance, the upper proportion is the ratio between the upper distance and the longitudinal distance, and the lower proportion is the ratio between the lower distance and the longitudinal distance.

[0092] S25, determining a virtual horizontal coordinate in the sand base model based on the left side proportion and the right side proportion corresponding to the midpoint of the entity, and determining a virtual vertical coordinate in the sand base model according to the upper side proportion and the lower side proportion.

[0093] It can be understood that the virtual horizontal coordinate is the horizontal coordinate of the virtual element corresponding to the midpoint of the entity in the sand base model, and the virtual vertical coordinate is the vertical coordinate of the virtual element corresponding to the midpoint of the entity in the sand base model.

[0094] For example: when the left side corresponding to the midpoint of the entity accounts for 1 / 3, the right side accounts for 2 / 3, the upper side accounts for 2 / 5, and the lower side accounts for 3 / 5, the virtual horizontal coordinate and virtual vertical coordinate of the corresponding midpoint of the entity in the sand base model can be determined, so as to facilitate the subsequent update of the corresponding virtual elements in the sand base model.

[0095] S26, determining a virtual position of the virtual element according to the virtual horizontal coordinate and the virtual vertical coordinate, and updating the virtual element to the sand base model based on the virtual position to obtain an element layout model.

[0096] It can be understood that the virtual position is the corresponding position of the virtual element in the sand base model, so that the corresponding virtual element can be updated to obtain an element layout model, so as to facilitate the subsequent adjustment of the orientation of the virtual element, thereby obtaining a directional conversion model and improving the efficiency of collecting sand table pictures.

[0097] S3, determining the virtual orientation of the corresponding virtual element in the element layout model based on the orientation positioning strategy and the prototype construction diagram, and updating the corresponding virtual element according to the virtual orientation to obtain a directional conversion model.

[0098] It should be noted that each entity element in the prototype construction diagram has a corresponding model orientation. For example, the human elements in the sandbox in the prototype construction diagram face forward, and the animal element models face left. In order to make the constructed model consistent with the actual sandbox information and ensure the accuracy of the information displayed in the captured sandbox picture, the orientation of the corresponding virtual elements in the element layout model can be adjusted according to the prototype component diagram to obtain a directional conversion model, thereby realizing automatic capture of sandbox pictures and improving acquisition efficiency.

[0099] It can be understood that the virtual orientation is the placement direction of the virtual element, and the orientation conversion model is a model obtained by updating the placement direction of the virtual element in the element layout model.

[0100] In some embodiments, the specific implementation of step S3 (determining the virtual orientation of the corresponding virtual element in the element layout model based on the orientation positioning strategy and the prototype construction diagram, and updating the corresponding virtual element according to the virtual orientation to obtain the orientation conversion model) includes: S31, retrieving the auxiliary angle corresponding to the auxiliary acquisition interface, adjusting the element layout model based on the auxiliary angle, and obtaining a direction positioning model.

[0101] It should be noted that in order to facilitate the determination of the virtual orientation corresponding to the virtual element, the element layout model can be adjusted to the same auxiliary angle. For example, with the horizontal plane as the reference 0°, when the auxiliary angle is 50° with the horizontal plane, the element layout model can also be adjusted by 50°, so that the display image of the layout element model seen by the personnel at a level angle is consistent with the collected prototype construction. Figure 1 Therefore, it is convenient to determine the virtual orientation of the virtual elements in the element layout model according to the orientation of the physical elements in the prototype construction diagram.

[0102] It can be understood that the auxiliary angle is the offset angle of the auxiliary acquisition interface for shooting the sandbox, and the orientation positioning model is the model after the element layout model is adjusted to the auxiliary angle.

[0103] S32, obtaining element attributes of the automatically adjusted element in the virtual element in the orientation positioning model, wherein the element attributes include a rotation attribute and a fixed attribute.

[0104] It should be noted that some of the models placed in the sandbox have a distinction between the front and back sides, while some models have consistent images on all sides and do not distinguish between the front and back sides. For example, the character models, animals, vehicles, etc. in the sandbox have a distinction between the front and back sides of the models. Among them, the tree model has a consistent image on all sides, and the image displayed after rotating to different angles is the same. Therefore, the attributes of entity elements can be divided into rotation attributes and fixed attributes.

[0105] It can be understood that the element attribute is the adjustment attribute corresponding to the virtual element, including rotation attribute and fixed attribute, wherein the rotation attribute is the attribute that can be rotated to adjust the displayed image, and the fixed attribute is the attribute that does not need to adjust the angle and the image is consistent.

[0106] Through the above implementation, the present invention can obtain the element attribute corresponding to the automatically adjusted element, so as to automatically adjust the orientation angle of the corresponding virtual element according to the element attribute later.

[0107] S33: When it is determined that the element attribute is a rotation attribute, the corresponding automatically adjusted element is used as a rotation element.

[0108] It is understandable that when the element attribute is a rotation attribute, it means that the corresponding automatic adjustment element is a model that distinguishes the front and back sides, and the corresponding automatic adjustment element can be used as a rotation element to facilitate subsequent adjustment of the corresponding automatic adjustment element.

[0109] It is worth mentioning that when the element attribute is a fixed attribute, it can be said that the corresponding automatic adjustment element does not distinguish between the front and back sides. Therefore, there is no need to adjust the orientation of the corresponding automatic adjustment element, thereby reducing the subsequent image recognition of the corresponding physical elements in the prototype construction diagram and reducing the amount of data processing.

[0110] S34, taking the virtual position of the rotating element as the center point, rotating the rotating element in the orientation positioning model in real time, obtaining the image center point of the intercepted image corresponding to the orientation positioning model, and constructing a coordinate system based on the image center point of the intercepted image.

[0111] It can be understood that by using the virtual position of the rotating element as the center point of rotation, the rotating element can be rotated in real time, and then the image displayed after rotation is captured in real time to obtain a captured image, and the center point of the captured image is obtained, so as to establish a coordinate system with the center point of the captured image as the origin, so as to facilitate the subsequent real-time acquisition of the outline of the corresponding rotating element, thereby determining the virtual orientation of the corresponding virtual element.

[0112] The captured image is an image captured by capturing a display image of a rotating element that is rotated in real time.

[0113] It is not difficult to understand that the corresponding rotation contour is obtained while the rotating element is rotated in real time. When the rotation contour of the corresponding rotating element after being rotated to a certain angle is consistent with the contour of the corresponding element in the prototype construction diagram, it can be said that the orientation of the current rotating element is consistent with the orientation of the corresponding physical element in the prototype construction diagram.

[0114] S35, acquiring the rotation contour of the rotating element in the orientation positioning model in real time based on OpenCV, and extracting the rotation contour coordinates of the contour pixel points in the rotation contour.

[0115] It can be understood that the rotation contour is the image contour of the rotation element, the contour pixel points are the pixel points of the rotation contour, and the rotation contour coordinates are the coordinates of the contour pixel points in the rotation contour.

[0116] Through the above implementation, the present invention can obtain the rotation contour coordinates of the contour pixel points in the rotation contour, so as to determine the virtual orientation of the corresponding virtual element.

[0117] S36, determining the virtual orientation of each virtual element in the orientation positioning model according to the rotation contour coordinates, and updating the corresponding virtual element based on the virtual orientation to obtain an orientation conversion model.

[0118] It can be understood that the virtual orientation is a model orientation corresponding to a virtual element in the orientation positioning model.

[0119] Through the above implementation, the present invention can obtain a directional conversion model, so as to facilitate the subsequent customized interception of sand table pictures for the directional conversion model, thereby improving the accuracy of intercepting and displaying information and improving the image acquisition efficiency.

[0120] In some embodiments, the specific implementation of step S36 (determining the virtual orientation of each virtual element in the orientation positioning model according to the rotation contour coordinates, and updating the corresponding virtual element based on the virtual orientation to obtain the orientation conversion model) includes: S361, counting the rotation contour coordinates to obtain a real-time coordinate set corresponding to the rotation element.

[0121] It can be understood that the real-time coordinate set is a set of rotation profile coordinates corresponding to the rotation element in real time.

[0122] Through the above implementation, the present invention can obtain the real-time coordinate set corresponding to the rotating element, so as to compare it with the physical outline of the physical element in the prototype construction diagram, so as to determine the virtual orientation.

[0123] S362, based on OpenCV, obtain the entity contour corresponding to the corresponding entity element in the prototype construction diagram, and count the entity contour coordinates of the contour pixel points corresponding to the entity contour to obtain a set of entity coordinates.

[0124] It can be understood that the entity contour is the contour of the corresponding entity element in the prototype construction diagram, and the entity contour set is the set of coordinates of the contour pixel points corresponding to the entity contour.

[0125] Through the above implementation, the present invention can obtain the physical coordinate set corresponding to each physical element, so as to compare it with the real-time coordinate set of the rotating element later. When the physical coordinate set is consistent with the real-time coordinate set, the virtual orientation is determined.

[0126] S363: When it is determined that the real-time coordinate set is equal to the physical coordinate set, the current orientation corresponding to the corresponding rotation element is used as a virtual orientation.

[0127] It should be noted that the angle specifications of the captured image and the prototype construction drawing are the same, so the coordinates in the corresponding coordinate system are also the same. Therefore, the real-time coordinate set can be compared with the physical coordinate set. When the real-time coordinate set is equal to the physical coordinate set, it can be said that the contours of the corresponding elements are the same, and the current orientation of the rotating element can be determined.

[0128] It can be understood that the virtual orientation is the current orientation corresponding to the corresponding rotation element when the real-time coordinate set is equal to the physical coordinate set.

[0129] S364: Update the automatically adjusted element based on the virtual orientation to obtain a directional conversion model.

[0130] It is understandable that after the virtual orientation is determined, the orientation of the automatically adjusted element may be updated, thereby obtaining a directional conversion model after the orientation is updated.

[0131] In some other embodiments, the specific implementation of the step (determining the virtual orientation of each virtual element in the orientation positioning model according to the rotation contour coordinates, and updating the corresponding virtual element based on the virtual orientation to obtain the orientation conversion model) includes: B1, counting the rotation contour coordinates to obtain a real-time coordinate set corresponding to the rotation element, and obtaining a maximum horizontal coordinate, a minimum horizontal coordinate, a maximum vertical coordinate and a minimum vertical coordinate in the real-time coordinate set.

[0132] It should be noted that in order to determine the virtual orientation corresponding to the virtual element, it can be judged not only by the element's contour coordinate set, but also by comparing the maximum and minimum values ​​of the virtual element's horizontal coordinates and the maximum and minimum values ​​of the vertical coordinates with the contours of the corresponding physical elements in the prototype construction diagram.

[0133] It can be understood that the real-time coordinate set is a set of rotation contour coordinates, the maximum horizontal coordinate is the maximum value of the horizontal coordinate in the real-time coordinate set, the minimum horizontal coordinate is the minimum value of the horizontal coordinate in the real-time coordinate set, the maximum vertical coordinate is the maximum value of the vertical coordinate in the real-time coordinate set, and the maximum vertical coordinate is the minimum value of the vertical coordinate in the real-time coordinate set.

[0134] B2: Obtain a virtual horizontal distance corresponding to the rotation element according to a difference between a maximum horizontal coordinate and a minimum horizontal coordinate in the real-time coordinate set.

[0135] It can be understood that the virtual horizontal distance is the difference between the maximum horizontal coordinate and the minimum horizontal coordinate in the real-time coordinate set.

[0136] B3, obtaining a virtual longitudinal distance corresponding to the rotation element based on a difference between a maximum longitudinal coordinate and a minimum longitudinal coordinate in the real-time coordinate set.

[0137] It can be understood that the virtual longitudinal distance is the difference between the maximum longitudinal coordinate and the minimum longitudinal coordinate in the real-time coordinate set.

[0138] B4, obtaining the entity contours corresponding to the corresponding entity elements in the prototype construction diagram according to OpenCV, and counting the entity contour coordinates corresponding to the entity contours to obtain an entity coordinate set.

[0139] It can be understood that the entity contour is the image contour corresponding to the corresponding entity element in the prototype construction diagram, the entity contour coordinates are the coordinates of the contour pixel points corresponding to the entity contour, and the entity coordinate set is the set of entity contour coordinates.

[0140] Through the above implementation, the present invention can obtain the physical coordinate set of the physical element, so as to determine the physical horizontal distance of the corresponding physical element according to the physical coordinate set, thereby comparing it with the virtual horizontal distance to determine the corresponding virtual orientation.

[0141] B5, obtaining a physical horizontal distance corresponding to the corresponding physical element based on a difference between a maximum horizontal coordinate and a minimum horizontal coordinate in the physical coordinate set.

[0142] It can be understood that the entity lateral distance is the difference between the maximum horizontal coordinate and the minimum horizontal coordinate in the entity coordinate set.

[0143] B6, obtaining the entity longitudinal distance corresponding to the corresponding entity element according to the difference between the maximum longitudinal coordinate and the minimum longitudinal coordinate in the entity coordinate set.

[0144] It can be understood that the entity longitudinal distance is the difference between the maximum longitudinal coordinate and the minimum longitudinal coordinate in the entity coordinate set.

[0145] B7, obtaining a horizontal judgment value based on the absolute value of the difference between the physical horizontal distance and the virtual horizontal distance, and obtaining a vertical judgment value based on the absolute value of the difference between the physical vertical distance and the virtual vertical distance.

[0146] It can be understood that the horizontal judgment value is the distance difference between the physical element and the virtual element in the horizontal direction, that is, the absolute value of the difference between the physical horizontal distance and the virtual horizontal distance, and the vertical judgment value is the distance difference between the physical element and the virtual element in the vertical direction, that is, the absolute value of the difference between the physical vertical distance and the virtual vertical distance.

[0147] B8. When it is determined that both the horizontal judgment value and the vertical judgment value are smaller than a preset difference, the current orientation corresponding to the corresponding rotation element is used as a virtual orientation.

[0148] It is understandable that since the element layout model is a three-dimensional model, the elements therein have a certain height, so there may be certain differences when extracting the outlines of the corresponding elements for coordinate comparison. Therefore, a benchmark deviation value can be set in advance to facilitate determining the corresponding virtual orientation.

[0149] The preset difference is a reference value of the pre-allowed deviation degree, which can be preset manually.

[0150] It is not difficult to understand that when the server determines that both the horizontal judgment value and the vertical judgment value are less than the preset difference, it can be explained that the direction of the automatically adjusted element after the corresponding rotation is the same as the direction of the corresponding physical element in the prototype construction diagram taken, and the current orientation corresponding to the corresponding rotated element can be used as the virtual orientation.

[0151] Through the above implementation, the present invention can obtain the virtual orientation for subsequent updating, thereby obtaining a directional conversion model.

[0152] B9, updating the corresponding automatically adjusted element based on the virtual orientation to obtain a directional conversion model.

[0153] It can be understood that after the virtual orientation is determined, the orientation of the corresponding automatic adjustment element is automatically updated to obtain a directional conversion model, thereby improving the accuracy of the actual sandbox information corresponding to the directional conversion model and the prototype construction diagram.

[0154] It should be noted that in actual shooting, different physical elements are blocked to a certain extent. When the deviation between the rotation of the corresponding rotating element and the actual position or orientation is large and the corresponding orientation cannot be determined, a person may observe and actively adjust the corresponding rotating element. Thus, in some embodiments, the following is also included: C1: When it is determined that the rotation element does not have the virtual orientation, the corresponding rotation element is used as a manual orientation element.

[0155] It can be understood that when the rotating element does not have a virtual orientation, that is, the corresponding element has a large deviation from the orientation in the prototype construction diagram, in order to make the information in the obtained directional conversion model more accurate, the corresponding rotating element can be used as a manual adjustment element to facilitate the subsequent manual adjustment of the corresponding manual adjustment element to ensure that the directional conversion model is consistent with the actual customized sand table model placement information.

[0156] The manual orientation element is a corresponding rotation element when the rotation element does not have a virtual orientation.

[0157] C2, sending the manual adjustment element and the active adjustment element to the analysis end.

[0158] It can be understood that the analysis end is the information terminal of the person who receives the sandbox information for analysis, which can be a mobile phone, computer, etc.

[0159] It is not difficult to understand that the manually adjusted elements and the actively adjusted elements are sent to the analysis end for personnel to view, thereby manually and actively adjusting the orientation of the corresponding virtual elements.

[0160] S4, obtaining the sandbox type of the customized sandbox, retrieving multiple analysis and identification angles based on the sandbox type, intercepting the directional conversion model according to the multiple analysis and identification angles, and obtaining multiple analysis training diagrams.

[0161] It should be noted that there are many types of sand tables, such as group sand tables, individual sand tables, sand tables of different sizes, etc. The corresponding collection angles and quantities of different types of sand tables may also be different. Therefore, the corresponding multiple analysis and recognition angles can be retrieved by the sand table type, so as to automatically capture the multi-angle images of the directional conversion model corresponding to the customized sand table, and obtain multiple analysis training graphs, so as to send them to the corresponding sand table image feature recognition model for data training.

[0162] It can be understood that the sand table type is the type of the customized sand table, and the analysis and recognition angle is the image acquisition angle when analyzing the sand table display information. The sand table type and the analysis and recognition angle can be corresponding. For example, when the customized sand table is an individual sand table, it may be necessary to collect 20 analysis training pictures with different analysis and recognition angles. When the customized sand table is a group sand table, it is necessary to collect 30 analysis training pictures with different analysis and recognition angles to facilitate transmission to the sand table picture feature recognition model for model training, thereby improving the accuracy of the analysis results of subsequent psychological analysis by the sand table picture feature recognition model.

[0163] Among them, the sand table image feature recognition model is an identification model that autonomously analyzes the sand table and generates psychological analysis results of the personnel. After the sand table image feature recognition model is trained, the sand table picture of the corresponding personnel can be input to obtain the psychological state of the personnel. This is the existing technology and will not be elaborated here.

[0164] See also Figure 4 , is a schematic diagram of the structure of an AI-driven sandbox image emotional and psychological state analysis and recognition data processing system provided by an embodiment of the present invention, the AI-driven sandbox image emotional and psychological state analysis and recognition data processing system comprising: The recognition module is used to receive the prototype construction diagram collected by the user end based on the auxiliary collection interface for the customized sand table, identify the virtual elements corresponding to each physical element in the prototype construction diagram, and build a sand base model.

[0165] A determination module is used to determine the virtual position of the virtual element according to the mapping positioning strategy and the prototype construction diagram, and update the virtual element to the sand base model based on the virtual position to obtain an element layout model.

[0166] An updating module is used to determine the virtual orientation of the corresponding virtual element in the element layout model based on the orientation positioning strategy and the prototype construction diagram, and to update the corresponding virtual element according to the virtual orientation to obtain a directional conversion model.

[0167] The interception module is used to obtain the sandbox type of the customized sandbox, call multiple analysis and identification angles based on the sandbox type, intercept the directional conversion model according to the multiple analysis and identification angles, and obtain multiple analysis training diagrams.

[0168] See also Figure 5 , is a schematic diagram of the hardware structure of an electronic device provided by an embodiment of the present invention, the electronic device 50 includes: a processor 51, a memory 52 and a computer program; wherein The memory 52 is used to store the computer program, which may also be a flash memory. The computer program is, for example, an application program, a functional module, etc. for implementing the above method.

[0169] The processor 51 is used to execute the computer program stored in the memory to implement each step performed by the device in the above method. For details, please refer to the relevant description in the above method embodiment.

[0170] Optionally, the memory 52 may be independent or integrated with the processor 51 .

[0171] When the memory 52 is a device independent of the processor 51, the device may further include: The bus 53 is used to connect the memory 52 and the processor 51 .

[0172] The present invention also provides a readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, it is used to implement the methods provided by the various embodiments described above.

[0173] Among them, the readable storage medium can be a computer storage medium or a communication medium. The communication medium includes any medium that facilitates the transmission of a computer program from one place to another. The computer storage medium can be any available medium that can be accessed by a general-purpose or special-purpose computer. For example, a readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an application-specific integrated circuit (Application Specific Integrated Circuits, referred to as: ASIC). In addition, the ASIC can be located in a user device. Of course, the processor and the readable storage medium can also exist in a communication device as discrete components. The readable storage medium can be a read-only memory (ROM), a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, and an optical data storage device, etc.

[0174] The present invention also provides a program product, which includes an execution instruction, which is stored in a readable storage medium. At least one processor of a device can read the execution instruction from the readable storage medium, and at least one processor executes the execution instruction so that the device implements the methods provided in the above various embodiments.

[0175] In the embodiments of the above-mentioned devices, it should be understood that the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the present invention may be directly implemented as being executed by a hardware processor, or may be executed by a combination of hardware and software modules in the processor.

[0176] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An AI-driven sandbox image emotional and psychological state analysis and recognition data processing method, characterized in that: include: Receiving a prototype construction diagram collected by the user end based on the auxiliary collection interface for the customized sand table, identifying virtual elements corresponding to each physical element in the prototype construction diagram, and constructing a sand base model; Determine the virtual position of the virtual element according to the mapping positioning strategy and the prototype construction diagram, and update the virtual element to the sand base model based on the virtual position to obtain an element layout model; Determine the virtual orientation of the corresponding virtual element in the element layout model based on the orientation positioning strategy and the prototype construction diagram, and update the corresponding virtual element according to the virtual orientation to obtain an orientation conversion model; The sandbox type of the customized sandbox is obtained, a plurality of analysis and identification angles are retrieved based on the sandbox type, and the directional conversion model is intercepted according to the plurality of analysis and identification angles to obtain a plurality of analysis training diagrams.

2. The method according to claim 1, characterized in that The receiving end collects a prototype construction diagram of the customized sandbox based on the auxiliary collection interface, and identifies virtual elements corresponding to each physical element in the prototype construction diagram, including: Receiving a prototype construction diagram collected by the user end based on the auxiliary collection interface for the customized sandbox, and identifying the entity outlines of the entity elements in the prototype construction diagram based on OpenCV; Extracting pixel values ​​within the entity contour as identification pixel values, and selecting preset elements with the identification pixel values ​​in the preset element library as candidate elements; Determine the identification pixel points within the entity outline based on the identification pixel values, and count the number of identification pixel points corresponding to each identification pixel value to obtain the identification number; Counting the identified numbers to obtain a total number, and obtaining the actual pixel value ratio of each identified pixel value according to the ratio of the identified number to the total number; Obtaining a preset pixel value ratio corresponding to the identified pixel value in the candidate element, and obtaining a sub-difference corresponding to each identified pixel value according to an absolute value of a difference between the preset pixel value ratio corresponding to each identified pixel value and an actual pixel value ratio; The sub-differences are counted to obtain a total difference corresponding to each of the candidate elements, and the candidate element corresponding to the smallest total difference is selected as the virtual element corresponding to the corresponding physical element.

3. The method according to claim 1, characterized in that: Also includes: When determining that the prototype construction diagram has intersecting entity contours, taking the corresponding entity contours as overlapping contours; The virtual elements corresponding to the overlapping contours with gaps in the overlapping contours are acquired as active adjustment elements, and the remaining virtual elements are acquired as automatic adjustment elements.

4. The method according to claim 2, characterized in that: The step of determining the virtual position of the virtual element according to the mapping positioning strategy and the prototype construction diagram, and updating the virtual element to the sand base model based on the virtual position to obtain the element layout model includes: Connect the midpoints of the upper and lower boundary lines in the prototype construction diagram to obtain a first center line, and connect the midpoints of the left and right boundary lines in the prototype construction diagram to obtain a second center line; Based on the intersection of the first center line and the second center line, obtaining the center point of the image corresponding to the prototype construction image; A coordinate system is constructed with the center point of the image as the coordinate origin, the center point of the base corresponding to each entity element in the prototype construction diagram is obtained as the entity midpoint, and the entity coordinates corresponding to the entity midpoint are determined; Determine the left side proportion, right side proportion, top side proportion and bottom side proportion corresponding to the midpoint of the entity according to the entity coordinates; Based on the left side proportion and the right side proportion corresponding to the midpoint of the entity, a virtual horizontal coordinate in the sand base model is determined, and according to the upper side proportion and the lower side proportion, a virtual vertical coordinate in the sand base model is determined; The virtual position of the virtual element is determined according to the virtual horizontal coordinate and the virtual vertical coordinate, and the virtual element is updated to the sand base model based on the virtual position to obtain an element layout model.

5. The method according to claim 4, characterized in that The determining, according to the entity coordinates, the left side proportion, the right side proportion, the upper side proportion and the lower side proportion corresponding to the midpoint of the entity includes: Determine a first ordinate corresponding to the upper boundary line and a second ordinate corresponding to the lower boundary line in the prototype construction diagram; Construct a transverse vertical line perpendicular to the longitudinal axis based on the entity coordinates, and obtain a first intersection point of the transverse vertical line with the left boundary line and a second intersection point of the right boundary line in the prototype construction diagram; Obtain a first horizontal coordinate corresponding to the first intersection point and a second horizontal coordinate corresponding to the second intersection point; Extracting the entity horizontal coordinate and the entity vertical coordinate corresponding to the entity coordinate, calculating based on the entity horizontal coordinate and the first horizontal coordinate to obtain the left distance, and calculating based on the entity horizontal coordinate and the second horizontal coordinate to obtain the right distance; According to the sum of the left distance and the right distance, a horizontal distance is obtained, according to the left distance and the horizontal distance, a left proportion is obtained, and based on the right distance and the horizontal distance, a right proportion is obtained; Calculate based on the entity ordinate and the first ordinate to obtain an upper distance, and calculate based on the entity ordinate and the second ordinate to obtain a lower distance; The longitudinal distance is obtained according to the sum of the upper distance and the lower distance, the upper proportion is obtained according to the upper distance and the longitudinal distance, and the lower proportion is obtained according to the lower distance and the longitudinal distance.

6. The method according to claim 5, characterized in that The determining of the virtual orientation of the corresponding virtual element in the element layout model based on the orientation positioning strategy and the prototype construction diagram, and updating the corresponding virtual element according to the virtual orientation to obtain the orientation conversion model includes: Retrieving the auxiliary angle corresponding to the auxiliary acquisition interface, adjusting the element layout model based on the auxiliary angle, and obtaining a direction positioning model; Acquire element attributes of the automatically adjusted element in the virtual element in the orientation positioning model, wherein the element attributes include a rotation attribute and a fixed attribute; When it is determined that the element attribute is a rotation attribute, the corresponding automatically adjusted element is used as a rotation element; Taking the virtual position of the rotating element as the center point, rotating the rotating element in the orientation positioning model in real time, obtaining the image center point of the intercepted image corresponding to the orientation positioning model, and constructing a coordinate system based on the image center point of the intercepted image; Based on OpenCV, a rotation contour of a rotating element in the orientation positioning model is acquired in real time, and the rotation contour coordinates of contour pixels in the rotation contour are extracted; The virtual orientation of each virtual element in the orientation positioning model is determined according to the rotation contour coordinates, and the corresponding virtual element is updated based on the virtual orientation to obtain an orientation conversion model.

7. The method according to claim 6, characterized in that The step of determining the virtual orientation of each virtual element in the orientation positioning model according to the rotation contour coordinates, and updating the corresponding virtual element based on the virtual orientation to obtain the orientation conversion model includes: Counting the rotation contour coordinates to obtain a real-time coordinate set corresponding to the rotation element; Based on OpenCV, the entity contour corresponding to the entity element in the prototype construction diagram is obtained, and the entity contour coordinates of the contour pixel points corresponding to the entity contour are counted to obtain an entity coordinate set; When it is determined that the real-time coordinate set is equal to the physical coordinate set, the current orientation corresponding to the corresponding rotation element is used as a virtual orientation; The automatic adjustment element is updated based on the virtual orientation to obtain a directional conversion model.

8. The method according to claim 6, characterized in that The step of determining the virtual orientation of each virtual element in the orientation positioning model according to the rotation contour coordinates, and updating the corresponding virtual element based on the virtual orientation to obtain the orientation conversion model includes: Counting the rotation contour coordinates to obtain a real-time coordinate set corresponding to the rotation element, and obtaining a maximum horizontal coordinate, a minimum horizontal coordinate, a maximum vertical coordinate, and a minimum vertical coordinate in the real-time coordinate set; Obtaining a virtual horizontal distance corresponding to the rotation element according to a difference between a maximum horizontal coordinate and a minimum horizontal coordinate in the real-time coordinate set; Obtaining a virtual longitudinal distance corresponding to the rotation element based on a difference between a maximum longitudinal coordinate and a minimum longitudinal coordinate in the real-time coordinate set; Obtaining the entity contour corresponding to the corresponding entity element in the prototype construction diagram according to OpenCV, and counting the entity contour coordinates corresponding to the entity contour to obtain an entity coordinate set; Based on the difference between the maximum horizontal coordinate and the minimum horizontal coordinate in the entity coordinate set, obtaining the entity horizontal distance corresponding to the corresponding entity element; Obtaining a physical longitudinal distance corresponding to the corresponding physical element according to a difference between a maximum vertical coordinate and a minimum vertical coordinate in the physical coordinate set; Based on the absolute value of the difference between the physical horizontal distance and the virtual horizontal distance, a horizontal judgment value is obtained, and based on the absolute value of the difference between the physical vertical distance and the virtual vertical distance, a vertical judgment value is obtained; When it is determined that both the horizontal judgment value and the vertical judgment value are smaller than a preset difference, the current orientation corresponding to the corresponding rotation element is used as a virtual orientation; The corresponding automatically adjusted element is updated based on the virtual orientation to obtain a directional conversion model.

9. The method according to claim 7 or 8, characterized in that: Also includes: When it is determined that the rotating element does not have the virtual orientation, the corresponding rotating element is used as a manual orientation element; The manual adjustment element and the active adjustment element are sent to an analysis end.

10. An AI-driven sand table image emotional and psychological state analysis and recognition data processing system, characterized in that: include: The recognition module is used to receive the prototype construction diagram collected by the user end based on the auxiliary collection interface for the customized sand table, identify the virtual elements corresponding to the physical elements in the prototype construction diagram, and build a sand base model; A determination module, used to determine the virtual position of the virtual element according to the mapping positioning strategy and the prototype construction diagram, and update the virtual element to the sand base model based on the virtual position to obtain an element layout model; An updating module, used for determining the virtual orientation of the corresponding virtual element in the element layout model based on the orientation positioning strategy and the prototype construction diagram, and updating the corresponding virtual element according to the virtual orientation to obtain an orientation conversion model; The interception module is used to obtain the sandbox type of the customized sandbox, call multiple analysis and identification angles based on the sandbox type, intercept the directional conversion model according to the multiple analysis and identification angles, and obtain multiple analysis training diagrams.

Images