Method and system for correcting three-dimensional space data of virtual reality

Abstract

The invention discloses a three-dimensional space data correction method and system for virtual reality, and relates to the technical field of virtual reality, and the method comprises the steps: obtaining basic three-dimensional space information and camera lens information, matching a real image feature point with a corresponding preset image feature point, and obtaining normal positioning identification information and abnormal positioning identification information. According to the method, the basic space anchor points are divided, the virtual scene construction efficiency is improved, the position of the real image in the three-dimensional space is predicted, the accuracy of image superposition is conveniently evaluated, the actual position of the real image in the three-dimensional space is compared with the predicted position, and the real scene construction efficiency is improved. According to the method, whether the real image meets the virtual scene requirement or not is judged, and the abnormal image feature points are adjusted, so that the positioning deviation is reduced, the trust degree and immersion of the user to the virtual environment are improved, and richer and more real experience is provided for the user.

Description

Technical Field

[0001] The present invention relates to the field of virtual reality technology, and in particular to a method and system for correcting three-dimensional space data of virtual reality. Background Art

[0002] Virtual reality technology is a computer simulation technology that can create and experience a virtual reality world. Based on virtual reality technology, in various fields such as manufacturing, medicine, and entertainment, people can get an experience that is increasingly similar to the real scene in terms of vision, hearing, touch, and interaction with virtual objects, and thus complete many tasks that cannot be completed in the real scene. The realization of virtual reality technology needs to be based on the construction of virtual scenes. Therefore, the construction result of a complete and comprehensive virtual scene is the basis for ensuring the virtual reality experience. In addition, the effect of virtual scene construction can also enhance the authenticity of virtual tasks in the virtual scene.

[0003] At present, there are still problems with virtual reality, such as the inability to fully, completely and highly restore the real scene, the inability to accurately analyze the three-dimensional space of virtual reality, the inability to accurately evaluate the image superposition state of real images and three-dimensional space, and the inability to accurately identify abnormal positioning during the image superposition process, which easily causes distortion of the virtual scene and the inability to correct the three-dimensional space according to the abnormal positioning. Summary of the invention

[0004] In order to solve the above technical problems, a method and system for correcting three-dimensional space data of virtual reality are provided. The technical solution solves the problems raised in the above background technology that the real scene cannot be fully, completely and highly restored, the three-dimensional space of virtual reality cannot be accurately analyzed, the image superposition state of the real image and the three-dimensional space cannot be accurately evaluated, the abnormal positioning in the image superposition process cannot be accurately identified, which easily causes distortion of the virtual scene, and the three-dimensional space cannot be corrected according to the abnormal positioning.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is:

[0006] A method for correcting three-dimensional space data of virtual reality, comprising:

[0007] Acquire basic three-dimensional space information and camera lens information, wherein the basic three-dimensional space information includes a basic space anchor point, and the camera lens information represents lens position information for framing a three-dimensional space to generate a virtual display image;

[0008] Divide the basic spatial anchor points to obtain the spatial core anchor point information and spatial boundary anchor point information;

[0009] Collecting real image information, wherein the real image information represents a real image used to superimpose images in a three-dimensional space to form a virtual scene;

[0010] According to the real image information, obtain the real image feature points and the real image space anchor points;

[0011] Taking the real image feature points and the real image space anchor points as references and based on the space boundary anchor point information, obtaining the preset image feature points, wherein the preset image feature points are the predicted positions of each real image feature point in the three-dimensional space, and the preset image feature points correspond to the real image feature points;

[0012] Match the actual image feature points with the corresponding preset image feature points to obtain normal positioning identification information and abnormal positioning identification information;

[0013] According to the abnormal location identification information, determine whether it is necessary to re-capture the real image. If so, re-capture the real image based on the abnormal location identification position information. If not, based on the real image information, overlay the three-dimensional space to form a virtual scene.

[0014] Preferably, the dividing of the space basic anchor points to obtain the space core anchor point information and the space boundary anchor point information specifically includes:

[0015] According to the camera lens information, obtain the lens movement trajectory information;

[0016] Based on the lens movement trajectory information, obtain the closest distance and the farthest distance between the spatial basic anchor point and the lens;

[0017] The square difference between the farthest distance and the closest distance between the spatial basic anchor point and the lens is used as the virtual image influence characteristic coefficient;

[0018] The spatial basic anchor point with the smallest virtual image influence characteristic coefficient is used as the spatial core anchor point;

[0019] Based on the three-dimensional space coordinate system, the lens movement trajectory is analyzed to obtain the movement trajectory pole information, where the movement trajectory pole information is the extreme point information of the lens movement trajectory on the three-dimensional space coordinate axis;

[0020] The spatial basic anchor point closest to the moving trajectory pole is selected as the spatial boundary anchor point.

[0021] Preferably, the step of obtaining preset image feature points based on the spatial boundary anchor point information and taking the real image feature points and the real image spatial anchor points as references specifically includes:

[0022] According to the real image information, obtain the real image space center information;

[0023] According to the spatial core anchor point and the spatial boundary anchor point, spatial vector information is obtained, where the spatial vector information is a direction vector with the spatial core anchor point as the starting point and each spatial boundary anchor point as the end point;

[0024] The center of the real image space is matched with the core anchor point of the space, and the space vector is proportionally enlarged based on the space vector until the space vector reaches the boundary of the real image space, so as to obtain the space vector information of the real image;

[0025] The end point of the real image space vector is used as the anchor point of the real image space;

[0026] According to the positional relationship between the real image feature points and the real image space anchor points, the preset image feature points are obtained with the space boundary anchor points as the reference. The preset image feature points are used to predict the spatial position of each real image feature point after the real image is superimposed on the three-dimensional space.

[0027] Preferably, matching the actual image feature points with the corresponding preset image feature points to obtain normal positioning identification information and abnormal positioning identification information specifically includes:

[0028] According to the feature points of the real image, two of the feature points of the real image are selected as the image positioning feature points;

[0029] Using the preset image feature points corresponding to the image positioning feature points as the spatial positioning feature points;

[0030] According to the positional relationship between the feature points of the real image and the feature points of the image positioning, the positional association information of the feature points of the real image is obtained;

[0031] Overlap the image positioning feature points with the space positioning feature points, use the space positioning feature points as reference points, and obtain the real image space position information according to the real image feature point position association information, wherein the real image space position information indicates the specific position of the real image feature points in the three-dimensional space when the images are superimposed;

[0032] According to the real image spatial position information and the preset image feature points, image position difference information is obtained, where the image position difference information is the distance between the real image spatial position information and the corresponding preset image feature points;

[0033] Based on the requirements of virtual scene production, the image position difference distance threshold is obtained;

[0034] According to the image position difference information and the image position difference distance threshold, it is judged whether the preset image feature point is abnormal. If the image position difference information exceeds the image position difference distance threshold, the preset image feature point is a normal positioning mark; if the image position difference information does not exceed the image position difference distance threshold, the preset image feature point is an abnormal positioning mark.

[0035] Preferably, judging whether it is necessary to re-collect the real image according to the abnormal location identification information specifically includes:

[0036] According to the abnormal positioning identification information and the preset image feature points, a positioning abnormality coefficient is obtained, where the positioning abnormality coefficient is the ratio of the number of abnormal positioning identifications to the total number of preset image feature points;

[0037] Based on the virtual scene production requirements, obtain the positioning anomaly coefficient threshold;

[0038] According to the positioning anomaly coefficient, determine whether it is necessary to re-collect the real image;

[0039] If the positioning anomaly coefficient exceeds the positioning anomaly coefficient threshold, the real image needs to be re-collected, and the real image collection position is adjusted based on the abnormal positioning mark position information, and the real image is re-collected;

[0040] If the positioning anomaly coefficient does not exceed the positioning anomaly coefficient threshold, there is no need to re-collect the real image, and the three-dimensional space is superimposed on the image based on the real image information to form a virtual scene.

[0041] Preferably, the step of superimposing images in three-dimensional space to form a virtual scene based on real image information specifically includes:

[0042] According to the abnormality positioning mark information, a real image correction feature point is obtained, wherein the real image correction feature point is a real image feature point corresponding to the abnormality positioning mark;

[0043] Acquire correction feature point distance information, wherein the correction feature point distance information is the distance between the correction feature point of the real image and the feature point of the real image;

[0044] Sort the corrected feature point distances in descending order to obtain corrected feature point distance sorting information;

[0045] Based on the distance sorting information of the corrected feature points, the first two feature points of the real image are selected as the first corrected influencing feature point and the second corrected influencing feature point;

[0046] Obtain image positioning feature points and space positioning feature points;

[0047] Obtaining a distance correction coefficient according to the first correction affecting feature points, the second correction affecting feature points, the image positioning feature points, the space positioning feature points, the space core anchor point information and the space boundary anchor point information;

[0048] Adjusting the positions of the preset image feature points corresponding to the first correction-affecting feature points, the second correction-affecting feature points and the actual image correction feature points through the distance correction coefficient;

[0049] The first correction-affected feature point, the second correction-affected feature point and the actual image correction feature point are directly overlapped with the corresponding preset image feature points after adjustment to complete image superposition and obtain virtual scene information;

[0050] The calculation formula of the distance correction coefficient is:

[0051]

[0052] Where Q is the distance correction coefficient, is the position influence coefficient of the first correction influencing feature point on the second correction influencing feature point, is the position influence coefficient of the second correction influencing feature point on the real image correction feature point, S1 is the moving distance of the preset image feature point corresponding to the first correction influencing feature point, S2 is the moving distance of the preset image feature point corresponding to the second correction influencing feature point, S3 is the moving distance of the preset image feature point corresponding to the real image correction feature point, σ0 is the virtual image influence feature coefficient of the spatial core anchor point, σ i is the virtual image influence feature coefficient of the i-th spatial boundary anchor point.

[0053] Furthermore, a virtual reality three-dimensional space data correction system is proposed, which is used to implement the above-mentioned data correction method, including:

[0054] A main control module, wherein the main control module is used to determine whether a preset image feature point is abnormal based on image position difference information and an image position difference distance threshold, determine whether it is necessary to re-collect the real image based on a positioning abnormality coefficient, adjust the real image collection position based on the abnormal positioning identifier position information, and re-collect the real image, obtain a distance correction coefficient based on the first correction influencing feature point, the second correction influencing feature point, the image positioning feature point, the space positioning feature point, the space core anchor point information, and the space boundary anchor point information, directly overlap the first correction influencing feature point, the second correction influencing feature point, and the real image correction feature point with the corresponding preset image feature point after adjustment, and complete image superposition;

[0055] An information acquisition module, the information acquisition module is used to acquire basic three-dimensional space information, camera lens information and space basic anchor points, divide the space basic anchor points, acquire space core anchor point information and space boundary anchor point information, collect real image information, and acquire real image feature points and real image space anchor points based on the real image information;

[0056] An image preset module, wherein the image preset module is used to obtain preset image feature points based on the positional relationship between the real image feature points and the real image spatial anchor points, taking the spatial boundary anchor points as a reference, superimposing the image positioning feature points and the spatial positioning feature points, taking the spatial positioning feature points as reference points, and obtaining the real image spatial position information based on the real image feature point position association information, and obtaining the image position difference information based on the real image spatial position information and the preset image feature points;

[0057] The display module interacts with the main control module and is used to display real image information, basic three-dimensional space information and virtual scene information.

[0058] Optionally, the main control module specifically includes:

[0059] A control unit, the control unit is used to adjust the real image acquisition position based on the abnormal positioning mark position information, and re-acquire the real image, obtain the distance correction coefficient according to the first correction influence feature point, the second correction influence feature point, the image positioning feature point, the space positioning feature point, the space core anchor point information and the space boundary anchor point information, and directly overlap the first correction influence feature point, the second correction influence feature point and the real image correction feature point with the corresponding preset image feature point after adjustment to complete image superposition;

[0060] An information receiving unit, which interacts with the information acquisition module and the image preset module to receive data and transmit it to the judgment unit;

[0061] The judgment unit is used to judge whether the preset image feature point is abnormal according to the image position difference information and the image position difference distance threshold, and judge whether the real image needs to be re-collected according to the positioning abnormality coefficient.

[0062] Optionally, the information acquisition module specifically includes:

[0063] A first acquisition unit, the first acquisition unit is used to acquire basic three-dimensional space information, camera lens information and space basic anchor points, divide the space basic anchor points, and acquire space core anchor point information and space boundary anchor point information;

[0064] The second acquisition unit is used to collect real image information, and acquire real image feature points and real image space anchor points based on the real image information.

[0065] Optionally, the image preset module specifically includes:

[0066] A preset image unit, wherein the preset image unit is used to obtain preset image feature points based on the positional relationship between the real image feature points and the real image space anchor points and taking the space boundary anchor points as a reference;

[0067] An image positioning evaluation unit is used to overlap the image positioning feature points and the spatial positioning feature points, use the spatial positioning feature points as reference points, obtain the real image spatial position information according to the real image feature point position association information, and obtain the image position difference information according to the real image spatial position information and the preset image feature points.

[0068] Compared with the prior art, the present invention has the following beneficial effects:

[0069] The present invention proposes a method and system for correcting three-dimensional space data of virtual reality. By dividing basic space anchor points, the efficiency of virtual scene construction is improved. By predicting the position of a real image in three-dimensional space, the accuracy of image superposition is evaluated. By comparing the actual position of a real image in three-dimensional space with the predicted position, it is determined whether the real image meets the requirements of the virtual scene. By adjusting abnormal image feature points, positioning deviation is reduced, the user's trust in and sense of immersion in the virtual environment are improved, and a richer and more realistic experience is provided to the user. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 This is a flow chart of a method for correcting three-dimensional space data of virtual reality proposed by the present invention;

[0071] Figure 2 A flowchart for obtaining the spatial core anchor point and the spatial boundary anchor point in the present invention;

[0072] Figure 3 A flowchart for obtaining preset image feature points in the present invention;

[0073] Figure 4 A flow chart for obtaining normal positioning identification information and abnormal positioning identification information in the present invention;

[0074] Figure 5 This is a flowchart for obtaining virtual scene information in the present invention;

[0075] Figure 6 This is a structural block diagram of a virtual reality three-dimensional space data correction system proposed by the present invention. DETAILED DESCRIPTION

[0076] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0077] Reference Figure 1 - Figure 5 As shown, a method for correcting three-dimensional space data of virtual reality in an embodiment of the present invention includes:

[0078] Acquire basic three-dimensional space information and camera lens information, wherein the basic three-dimensional space information includes a basic space anchor point, and the camera lens information represents lens position information for framing a three-dimensional space to generate a virtual display image;

[0079] Divide the basic spatial anchor points to obtain the spatial core anchor point information and spatial boundary anchor point information;

[0080] Specifically, the spatial basic anchor points are divided to obtain the spatial core anchor point information and spatial boundary anchor point information, including:

[0081] According to the camera lens information, obtain the lens movement trajectory information;

[0082] Based on the lens movement trajectory information, obtain the closest distance and the farthest distance between the spatial basic anchor point and the lens;

[0083] The square difference between the farthest distance and the closest distance between the spatial basic anchor point and the lens is used as the virtual image influence characteristic coefficient;

[0084] The spatial basic anchor point with the smallest virtual image influence characteristic coefficient is used as the spatial core anchor point;

[0085] Based on the three-dimensional space coordinate system, the lens movement trajectory is analyzed to obtain the movement trajectory pole information, where the movement trajectory pole information is the extreme point information of the lens movement trajectory on the three-dimensional space coordinate axis;

[0086] The spatial basic anchor point closest to the moving trajectory pole is selected as the spatial boundary anchor point.

[0087] In this scheme, the closest and farthest distances between the space basic anchor point and the lens are obtained through the lens movement trajectory information. The square difference between the farthest and closest distances between the space basic anchor point and the lens is used as the virtual image influence characteristic coefficient, and the space basic anchor point closest to the movement trajectory pole is selected as the space boundary anchor point.

[0088] It is understandable that for the three-dimensional space of virtual reality, the three-dimensional space is analyzed according to the spatial direction, and the spatial boundary anchor points are selected through the boundaries of the three-dimensional space in different spatial directions, thereby achieving accurate analysis of the three-dimensional space and facilitating the subsequent construction of virtual scenes.

[0089] The spatial basic anchor point with the smallest virtual image influence characteristic coefficient is taken as the spatial core anchor point

[0090] Collecting real image information, wherein the real image information represents a real image used to superimpose images in a three-dimensional space to form a virtual scene;

[0091] According to the real image information, obtain the real image feature points and the real image space anchor points;

[0092] In this solution, there are at least six feature points of the real image, and the feature points of the real image can be color blocks, pixel points, image contour points, markers, etc.

[0093] Taking the real image feature points and the real image space anchor points as references and based on the space boundary anchor point information, obtaining the preset image feature points, wherein the preset image feature points are the predicted positions of each real image feature point in the three-dimensional space, and the preset image feature points correspond to the real image feature points;

[0094] Specifically, based on the real image feature points and the real image space anchor points, the preset image feature points are obtained based on the space boundary anchor point information, including:

[0095] According to the real image information, obtain the real image space center information;

[0096] According to the spatial core anchor point and the spatial boundary anchor point, spatial vector information is obtained, where the spatial vector information is a direction vector with the spatial core anchor point as the starting point and each spatial boundary anchor point as the end point;

[0097] The center of the real image space is matched with the core anchor point of the space, and the space vector is proportionally enlarged based on the space vector until the space vector reaches the boundary of the real image space, so as to obtain the space vector information of the real image;

[0098] The end point of the real image space vector is used as the anchor point of the real image space;

[0099] According to the positional relationship between the real image feature points and the real image space anchor points, the preset image feature points are obtained with the space boundary anchor points as the reference. The preset image feature points are used to predict the spatial position of each real image feature point after the real image is superimposed on the three-dimensional space.

[0100] In this solution, the spatial vector information is obtained through the spatial core anchor point and the spatial boundary anchor point, the spatial center of the real image is matched with the spatial core anchor point, and the spatial vector is proportionally enlarged with the spatial vector as a reference until the spatial vector reaches the spatial boundary of the real image to obtain the spatial vector information of the real image, and the preset image feature points are obtained based on the positional relationship between the real image feature points and the real image spatial anchor points and with the spatial boundary anchor points as a reference;

[0101] It can be understood that when constructing a virtual scene, the spatial relationship of the real image is directly related to the realism of the virtual scene. By aligning the spatial center of the real image with the spatial core anchor point of the three-dimensional space, the real image spatial anchor point is selected in the real image space based on the direction of the spatial core anchor point pointing to the spatial boundary anchor point. The real image spatial anchor point corresponds to the spatial boundary anchor point, and the positional relationship between the real image spatial anchor point and the real image spatial center and the positional relationship between the spatial boundary anchor point and the spatial core anchor point remain the same.

[0102] On this basis, the positional relationship between the real image feature points and the real image space anchor points is used as a benchmark, and preset image feature points with the same positional relationship as the spatial boundary anchor points are selected in the three-dimensional space, that is, the spatial position prediction of each real image feature point is realized after the real image is superimposed on the three-dimensional space.

[0103] Match the actual image feature points with the corresponding preset image feature points to obtain normal positioning identification information and abnormal positioning identification information;

[0104] Specifically, matching the feature points of the actual image with the corresponding preset image feature points to obtain normal positioning identification information and abnormal positioning identification information includes:

[0105] According to the feature points of the real image, two of the feature points of the real image are selected as the image positioning feature points;

[0106] Using the preset image feature points corresponding to the image positioning feature points as the spatial positioning feature points;

[0107] According to the positional relationship between the feature points of the real image and the feature points of the image positioning, the positional association information of the feature points of the real image is obtained;

[0108] Overlap the image positioning feature points with the space positioning feature points, use the space positioning feature points as reference points, and obtain the real image space position information according to the real image feature point position association information, wherein the real image space position information indicates the specific position of the real image feature points in the three-dimensional space when the images are superimposed;

[0109] According to the real image spatial position information and the preset image feature points, image position difference information is obtained, where the image position difference information is the distance between the real image spatial position information and the corresponding preset image feature points;

[0110] Based on the requirements of virtual scene production, the image position difference distance threshold is obtained;

[0111] According to the image position difference information and the image position difference distance threshold, it is judged whether the preset image feature point is abnormal. If the image position difference information exceeds the image position difference distance threshold, the preset image feature point is a normal positioning mark; if the image position difference information does not exceed the image position difference distance threshold, the preset image feature point is an abnormal positioning mark.

[0112] In this solution, by selecting two real image feature points as image positioning feature points, according to the positional relationship between the real image feature points and the image positioning feature points, the position association information of the real image feature points, that is, the mapping relationship of the real image in the three-dimensional space, is obtained, and by overlapping the image positioning feature points and the spatial positioning feature points, the spatial positioning feature points are used as reference points, and according to the position association information of the real image feature points, the real image spatial position information is obtained, and the specific positions of the real image feature points in the three-dimensional space are analyzed, and according to the image position difference information and the image position difference distance threshold, it is determined whether the preset image feature points are abnormal, and the abnormal positioning mark is identified, so as to facilitate the subsequent correction of the three-dimensional space. In this embodiment, the image position difference distance threshold is 0.05m.

[0113] According to the abnormal location identification information, determine whether it is necessary to re-capture the real image. If so, re-capture the real image based on the abnormal location identification position information. If not, based on the real image information, overlay the three-dimensional space to form a virtual scene.

[0114] Specifically, based on the abnormal location identification information, it is determined whether the real image needs to be re-collected, including:

[0115] According to the abnormal positioning identification information and the preset image feature points, a positioning abnormality coefficient is obtained, where the positioning abnormality coefficient is the ratio of the number of abnormal positioning identifications to the total number of preset image feature points;

[0116] Based on the virtual scene production requirements, obtain the positioning anomaly coefficient threshold;

[0117] According to the positioning anomaly coefficient, determine whether it is necessary to re-collect the real image;

[0118] If the positioning anomaly coefficient exceeds the positioning anomaly coefficient threshold, the real image needs to be re-collected, and the real image collection position is adjusted based on the abnormal positioning mark position information, and the real image is re-collected;

[0119] If the positioning anomaly coefficient does not exceed the positioning anomaly coefficient threshold, there is no need to re-collect the real image, and the three-dimensional space is superimposed on the image based on the real image information to form a virtual scene.

[0120] In this solution, the positioning anomaly coefficient is obtained through abnormal positioning identification information and preset image feature points. According to the positioning anomaly coefficient, it is determined whether the real image needs to be recaptured, thereby ensuring the authenticity and stability of the virtual scene. The positioning anomaly coefficient threshold is 0.2.

[0121] Specifically, based on the real image information, the three-dimensional space is overlaid with images to form a virtual scene, which specifically includes:

[0122] According to the abnormality positioning mark information, a real image correction feature point is obtained, wherein the real image correction feature point is a real image feature point corresponding to the abnormality positioning mark;

[0123] Acquire correction feature point distance information, wherein the correction feature point distance information is the distance between the correction feature point of the real image and the feature point of the real image;

[0124] Sort the corrected feature point distances in descending order to obtain corrected feature point distance sorting information;

[0125] Based on the distance sorting information of the corrected feature points, the first two feature points of the real image are selected as the first corrected influencing feature point and the second corrected influencing feature point;

[0126] Obtain image positioning feature points and space positioning feature points;

[0127] Obtaining a distance correction coefficient according to the first correction affecting feature points, the second correction affecting feature points, the image positioning feature points, the space positioning feature points, the space core anchor point information and the space boundary anchor point information;

[0128] Adjusting the positions of the preset image feature points corresponding to the first correction-affecting feature points, the second correction-affecting feature points and the actual image correction feature points through the distance correction coefficient;

[0129] The first correction-affected feature point, the second correction-affected feature point and the actual image correction feature point are directly overlapped with the corresponding preset image feature points after adjustment to complete image superposition and obtain virtual scene information;

[0130] The calculation formula of the distance correction coefficient is:

[0131]

[0132] Where Q is the distance correction coefficient, is the position influence coefficient of the first correction influencing feature point on the second correction influencing feature point, that is, when the position relationship remains unchanged, if the preset image feature point corresponding to the first correction influencing feature point is moved by a moving distance of A, then the preset image feature point corresponding to the second correction influencing feature point is moved by a moving distance of B, then is the position influence coefficient of the second correction influencing feature point on the real image correction feature point, S1 is the moving distance of the preset image feature point corresponding to the first correction influencing feature point, S2 is the moving distance of the preset image feature point corresponding to the second correction influencing feature point, S3 is the moving distance of the preset image feature point corresponding to the real image correction feature point, σ0 is the virtual image influence feature coefficient of the spatial core anchor point, σ i is the virtual image influence feature coefficient of the i-th spatial boundary anchor point.

[0133] In this solution, the values ​​of S1, S2 and S3 are adjusted until Q·H≤0.05m and When it is the minimum value, S1≤0.02 and S2≤0.03, the data correction of the three-dimensional space is realized by adjusting the abnormal image feature points, which reduces the positioning deviation, improves the user's trust in the virtual environment and immersion, and provides users with a richer and more realistic experience;

[0134] It should be noted that, in this embodiment, for the real image feature points remaining after removing the first correction affecting feature points, the second correction affecting feature points and the real image correction feature points, the image positioning feature points and the space positioning feature points are overlapped, and the space positioning feature points are used as reference points. Image superposition is performed based on the positional relationship between the real image feature points and the image positioning feature points.

[0135] Reference Figure 6 As shown, further, in combination with the above-mentioned virtual reality three-dimensional space data correction method, a virtual reality three-dimensional space data correction system is proposed, including:

[0136] A main control module, wherein the main control module is used to determine whether a preset image feature point is abnormal based on image position difference information and an image position difference distance threshold, determine whether it is necessary to re-collect the real image based on a positioning abnormality coefficient, adjust the real image collection position based on the abnormal positioning identifier position information, and re-collect the real image, obtain a distance correction coefficient based on the first correction influencing feature point, the second correction influencing feature point, the image positioning feature point, the space positioning feature point, the space core anchor point information, and the space boundary anchor point information, directly overlap the first correction influencing feature point, the second correction influencing feature point, and the real image correction feature point with the corresponding preset image feature point after adjustment, and complete image superposition;

[0137] An information acquisition module, the information acquisition module is used to acquire basic three-dimensional space information, camera lens information and space basic anchor points, divide the space basic anchor points, acquire space core anchor point information and space boundary anchor point information, collect real image information, and acquire real image feature points and real image space anchor points based on the real image information;

[0138] An image preset module, wherein the image preset module is used to obtain preset image feature points based on the positional relationship between the real image feature points and the real image spatial anchor points, taking the spatial boundary anchor points as a reference, superimposing the image positioning feature points and the spatial positioning feature points, taking the spatial positioning feature points as reference points, and obtaining the real image spatial position information based on the real image feature point position association information, and obtaining the image position difference information based on the real image spatial position information and the preset image feature points;

[0139] The display module interacts with the main control module and is used to display real image information, basic three-dimensional space information and virtual scene information.

[0140] Main control module, specifically including:

[0141] A control unit, the control unit is used to adjust the real image acquisition position based on the abnormal positioning mark position information, and re-acquire the real image, obtain the distance correction coefficient according to the first correction influence feature point, the second correction influence feature point, the image positioning feature point, the space positioning feature point, the space core anchor point information and the space boundary anchor point information, and directly overlap the first correction influence feature point, the second correction influence feature point and the real image correction feature point with the corresponding preset image feature point after adjustment to complete image superposition;

[0142] An information receiving unit, which interacts with the information acquisition module and the image preset module to receive data and transmit it to the judgment unit;

[0143] The judgment unit is used to judge whether the preset image feature point is abnormal according to the image position difference information and the image position difference distance threshold, and judge whether the real image needs to be re-collected according to the positioning abnormality coefficient.

[0144] Information acquisition module, specifically including:

[0145] A first acquisition unit, the first acquisition unit is used to acquire basic three-dimensional space information, camera lens information and space basic anchor points, divide the space basic anchor points, and acquire space core anchor point information and space boundary anchor point information;

[0146] The second acquisition unit is used to collect real image information, and acquire real image feature points and real image space anchor points based on the real image information.

[0147] Image preset module, including:

[0148] A preset image unit, wherein the preset image unit is used to obtain preset image feature points based on the positional relationship between the real image feature points and the real image space anchor points and taking the space boundary anchor points as a reference;

[0149] An image positioning evaluation unit is used to overlap the image positioning feature points and the spatial positioning feature points, use the spatial positioning feature points as reference points, obtain the real image spatial position information according to the real image feature point position association information, and obtain the image position difference information according to the real image spatial position information and the preset image feature points.

[0150] In summary, the advantages of the present invention are: by setting basic space anchor points for virtual reality, it is convenient to analyze virtual scenes, by dividing the basic space anchor points, the efficiency of virtual scene construction is improved, by analyzing real images, the feature points of real images are obtained, by comparing the real image space with the three-dimensional space, the position of the real image in the three-dimensional space is predicted, which is convenient for evaluating the accuracy of image superposition, by comparing the actual position of the real image in the three-dimensional space with the predicted position, it is determined whether the real image meets the requirements of the virtual scene, by adjusting the abnormal image feature points, the data correction of the three-dimensional space is realized, the positioning deviation is reduced, the user's trust in the virtual environment and the sense of immersion are improved, and a richer and more realistic experience is provided to the user.

[0151] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for correcting three-dimensional space data of virtual reality, characterized in that: include: Acquire basic three-dimensional space information and camera lens information, wherein the basic three-dimensional space information includes a basic space anchor point, and the camera lens information represents lens position information for framing a three-dimensional space to generate a virtual display image; Divide the basic spatial anchor points to obtain the spatial core anchor point information and spatial boundary anchor point information; Collecting real image information, wherein the real image information represents a real image used to superimpose images in a three-dimensional space to form a virtual scene; According to the real image information, obtain the real image feature points and the real image space anchor points; Taking the real image feature points and the real image space anchor points as references and based on the space boundary anchor point information, obtaining the preset image feature points, wherein the preset image feature points are the predicted positions of each real image feature point in the three-dimensional space, and the preset image feature points correspond to the real image feature points; Match the actual image feature points with the corresponding preset image feature points to obtain normal positioning identification information and abnormal positioning identification information; According to the abnormal location identification information, determine whether it is necessary to re-capture the real image. If so, re-capture the real image based on the abnormal location identification position information. If not, based on the real image information, overlay the three-dimensional space to form a virtual scene.

2. A method for correcting three-dimensional space data of virtual reality according to claim 1, characterized in that: The dividing of the space basic anchor points to obtain the space core anchor point information and the space boundary anchor point information specifically includes: According to the camera lens information, obtain the lens movement trajectory information; Based on the lens movement trajectory information, obtain the closest distance and the farthest distance between the spatial basic anchor point and the lens; The square difference between the farthest distance and the closest distance between the spatial basic anchor point and the lens is used as the virtual image influence characteristic coefficient; The spatial basic anchor point with the smallest virtual image influence characteristic coefficient is used as the spatial core anchor point; Based on the three-dimensional space coordinate system, the lens movement trajectory is analyzed to obtain the movement trajectory pole information, where the movement trajectory pole information is the extreme point information of the lens movement trajectory on the three-dimensional space coordinate axis; The spatial basic anchor point closest to the moving trajectory pole is selected as the spatial boundary anchor point.

3. The method for correcting three-dimensional space data of virtual reality according to claim 1, characterized in that: The method of obtaining the preset image feature points based on the spatial boundary anchor point information and the real image feature points and the real image spatial anchor points as a reference specifically includes: According to the real image information, obtain the real image space center information; According to the spatial core anchor point and the spatial boundary anchor point, spatial vector information is obtained, where the spatial vector information is a direction vector with the spatial core anchor point as the starting point and each spatial boundary anchor point as the end point; The center of the real image space is matched with the core anchor point of the space, and the space vector is proportionally enlarged based on the space vector until the space vector reaches the boundary of the real image space, so as to obtain the space vector information of the real image; The end point of the real image space vector is used as the anchor point of the real image space; According to the positional relationship between the real image feature points and the real image space anchor points, the preset image feature points are obtained with the space boundary anchor points as the reference. The preset image feature points are used to predict the spatial position of each real image feature point after the real image is superimposed on the three-dimensional space.

4. The method for correcting three-dimensional space data of virtual reality according to claim 1, characterized in that: The matching of the actual image feature points with the corresponding preset image feature points to obtain normal positioning identification information and abnormal positioning identification information specifically includes: According to the feature points of the real image, two of the feature points of the real image are selected as the image positioning feature points; Using the preset image feature points corresponding to the image positioning feature points as the spatial positioning feature points; According to the positional relationship between the feature points of the real image and the feature points of the image positioning, the positional association information of the feature points of the real image is obtained; Overlap the image positioning feature points with the space positioning feature points, use the space positioning feature points as reference points, and obtain the real image space position information according to the real image feature point position association information, wherein the real image space position information indicates the specific position of the real image feature points in the three-dimensional space when the images are superimposed; According to the real image spatial position information and the preset image feature points, image position difference information is obtained, where the image position difference information is the distance between the real image spatial position information and the corresponding preset image feature points; Based on the requirements of virtual scene production, the image position difference distance threshold is obtained; According to the image position difference information and the image position difference distance threshold, it is judged whether the preset image feature point is abnormal. If the image position difference information exceeds the image position difference distance threshold, the preset image feature point is a normal positioning mark; if the image position difference information does not exceed the image position difference distance threshold, the preset image feature point is an abnormal positioning mark.

5. The method for correcting three-dimensional space data of virtual reality according to claim 1, characterized in that: The determining whether it is necessary to re-collect the real image according to the abnormal location identification information specifically includes: According to the abnormal positioning identification information and the preset image feature points, a positioning abnormality coefficient is obtained, where the positioning abnormality coefficient is the ratio of the number of abnormal positioning identifications to the total number of preset image feature points; Based on the virtual scene production requirements, obtain the positioning anomaly coefficient threshold; According to the positioning anomaly coefficient, determine whether it is necessary to re-collect the real image; If the positioning anomaly coefficient exceeds the positioning anomaly coefficient threshold, the real image needs to be re-collected, and the real image collection position is adjusted based on the abnormal positioning mark position information, and the real image is re-collected; If the positioning anomaly coefficient does not exceed the positioning anomaly coefficient threshold, there is no need to re-collect the real image, and the three-dimensional space is superimposed on the image based on the real image information to form a virtual scene.

6. The method for correcting three-dimensional space data of virtual reality according to claim 1, characterized in that: The method of forming a virtual scene by superimposing images in the three-dimensional space according to the real image information specifically includes: According to the abnormality positioning mark information, a real image correction feature point is obtained, wherein the real image correction feature point is a real image feature point corresponding to the abnormality positioning mark; Acquire correction feature point distance information, wherein the correction feature point distance information is the distance between the correction feature point of the real image and the feature point of the real image; Sort the corrected feature point distances in descending order to obtain corrected feature point distance sorting information; Based on the distance sorting information of the corrected feature points, the first two feature points of the real image are selected as the first corrected influencing feature point and the second corrected influencing feature point; Obtain image positioning feature points and space positioning feature points; Obtaining a distance correction coefficient according to the first correction affecting feature points, the second correction affecting feature points, the image positioning feature points, the space positioning feature points, the space core anchor point information and the space boundary anchor point information; Adjusting the positions of the preset image feature points corresponding to the first correction-affecting feature points, the second correction-affecting feature points and the actual image correction feature points through the distance correction coefficient; The first correction-affected feature point, the second correction-affected feature point and the actual image correction feature point are directly overlapped with the corresponding preset image feature points after adjustment to complete image superposition and obtain virtual scene information; The calculation formula of the distance correction coefficient is: Where Q is the distance correction coefficient, is the position influence coefficient of the first correction influencing feature point on the second correction influencing feature point, is the position influence coefficient of the second correction influencing feature point on the real image correction feature point, S1 is the moving distance of the preset image feature point corresponding to the first correction influencing feature point, S2 is the moving distance of the preset image feature point corresponding to the second correction influencing feature point, S3 is the moving distance of the preset image feature point corresponding to the real image correction feature point, σ0 is the virtual image influence feature coefficient of the spatial core anchor point, σ i is the virtual image influence feature coefficient of the i-th spatial boundary anchor point.

7. A virtual reality three-dimensional space data correction system, used to implement the data correction method according to any one of claims 1 to 6, characterized in that: include: A main control module, wherein the main control module is used to determine whether a preset image feature point is abnormal based on image position difference information and an image position difference distance threshold, determine whether it is necessary to re-collect the real image based on a positioning abnormality coefficient, adjust the real image collection position based on the abnormal positioning identifier position information, and re-collect the real image, obtain a distance correction coefficient based on the first correction influencing feature point, the second correction influencing feature point, the image positioning feature point, the space positioning feature point, the space core anchor point information, and the space boundary anchor point information, directly overlap the first correction influencing feature point, the second correction influencing feature point, and the real image correction feature point with the corresponding preset image feature point after adjustment, and complete image superposition; An information acquisition module, the information acquisition module is used to acquire basic three-dimensional space information, camera lens information and space basic anchor points, divide the space basic anchor points, acquire space core anchor point information and space boundary anchor point information, collect real image information, and acquire real image feature points and real image space anchor points based on the real image information; An image preset module, wherein the image preset module is used to obtain preset image feature points based on the positional relationship between the real image feature points and the real image spatial anchor points, taking the spatial boundary anchor points as a reference, superimposing the image positioning feature points and the spatial positioning feature points, taking the spatial positioning feature points as reference points, and obtaining the real image spatial position information based on the real image feature point position association information, and obtaining the image position difference information based on the real image spatial position information and the preset image feature points; The display module interacts with the main control module and is used to display real image information, basic three-dimensional space information and virtual scene information.

8. A virtual reality three-dimensional space data correction system according to claim 7, characterized in that: The main control module specifically includes: A control unit, the control unit is used to adjust the real image acquisition position based on the abnormal positioning mark position information, and re-acquire the real image, obtain the distance correction coefficient according to the first correction influence feature point, the second correction influence feature point, the image positioning feature point, the space positioning feature point, the space core anchor point information and the space boundary anchor point information, and directly overlap the first correction influence feature point, the second correction influence feature point and the real image correction feature point with the corresponding preset image feature point after adjustment to complete image superposition; An information receiving unit, which interacts with the information acquisition module and the image preset module to receive data and transmit it to the judgment unit; The judgment unit is used to judge whether the preset image feature point is abnormal according to the image position difference information and the image position difference distance threshold, and judge whether the real image needs to be re-collected according to the positioning abnormality coefficient.

9. The virtual reality three-dimensional space data correction system according to claim 7, characterized in that: The information acquisition module specifically includes: A first acquisition unit, the first acquisition unit is used to acquire basic three-dimensional space information, camera lens information and space basic anchor points, divide the space basic anchor points, and acquire space core anchor point information and space boundary anchor point information; The second acquisition unit is used to collect real image information, and acquire real image feature points and real image space anchor points based on the real image information.

10. The virtual reality three-dimensional space data correction system according to claim 7, characterized in that: The image preset module specifically includes: A preset image unit, wherein the preset image unit is used to obtain preset image feature points based on the positional relationship between the real image feature points and the real image space anchor points and taking the space boundary anchor points as a reference; An image positioning evaluation unit is used to overlap the image positioning feature points and the spatial positioning feature points, use the spatial positioning feature points as reference points, obtain the real image spatial position information according to the real image feature point position association information, and obtain the image position difference information according to the real image spatial position information and the preset image feature points.

Images