Red education sharing method based on cloud computing

Abstract

The invention discloses a red education sharing method based on cloud computing. The invention relates to the technical field of education, and solves the technical problems that in the prior art, cultural relics are damaged or missing due to time and only two-dimensional pictures are reserved, so that the cultural relics are difficult to directly convert into three-dimensional images, and learners in shared red education cannot feel vividly and learn comprehensively. In particular to a red education sharing method based on cloud computing, and the method comprises the following steps: S1, obtaining a to-be-detected cultural relic image, and preprocessing the to-be-detected cultural relic image to obtain a two-dimensional image. According to the method, the point cloud data of the residual cultural relics are restored according to the features in the image, the three-dimensional image of the cultural relics is finally restored, the problem that the three-dimensional image of the red related cultural relics is difficult to directly obtain in the prior art is solved, meanwhile, the missing part of the red related objects is restored, and the feeling and learning interest of learners can be further improved.

Description

Technical Field

[0001] The present invention relates to the field of red education technology, and in particular to a red education sharing method based on cloud computing. Background Art

[0002] Cloud computing is a type of distributed computing, which refers to breaking down huge data computing programs into countless small programs through the network "cloud", and processing and analyzing these small programs through a system composed of multiple servers to obtain results and return them to users. The existing methods of using cloud computing for distance education generally use traditional indoctrination education methods. Especially in red education, most of the education methods revolve around a time or a person to instill content, and it is difficult for learners to learn in boring and monotonous indoctrination.

[0003] The existing technology has made improvements to this problem, adding picture and text matching learning to the shared red education, and then vividly displaying the objects or characters of the red stories. Due to the advancement of science and technology and people's cognitive progress, two-dimensional pictures are difficult to meet the needs of current learners, and objects need to be transformed into three-dimensional images. However, due to the passage of time, the main bodies of some red figures' related cultural relics have been damaged or missing, and only two-dimensional pictures remain, making it difficult to directly transform these objects into three-dimensional images, and thus learners in the shared red education cannot get vivid feelings and comprehensive learning. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a red education sharing method based on cloud computing, which solves the technical problem that damaged or missing cultural relics are difficult to directly convert into three-dimensional images, and achieves the purpose of using damaged or missing cultural relics as templates to extract features according to two-dimensional pictures, repairing the templates into three-dimensional images according to the features, and repairing the missing parts of red-related items, so that they can be presented to learners in a more complete and realistic manner, thereby improving learners' educational experience and learning interest.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: a red education sharing method based on cloud computing, the method comprising the following steps:

[0006] S1, obtaining an image of a cultural relic to be detected, and preprocessing the image of the cultural relic to be detected to obtain a two-dimensional image;

[0007] S2, using transition operator S a Perform convolution operation on the two-dimensional image to obtain the transition matrix H of the two-dimensional image b ;

[0008] S3, according to the transition matrix H b Calculate the maximum principal curvature k of a 2D image max and the minimum principal curvature kmin ;

[0009] S4, according to the maximum principal curvature k max and the minimum principal curvature k min Calculate the curvature index value CV and shape index value SL, and get the initial feature set C d ;

[0010] S5. Obtain the cultural relic point cloud dataset D of the cultural relic to be restored through VR equipment c ;

[0011] S6. Based on the initial feature set C d Calculate the curvature index value CV2 and shape index value SL2 of the local surface z formed by any point B;

[0012] S7, calculate the deformation variable D according to the curvature index value CV2 and the shape index value SL2, and determine the cultural relic point cloud data set D c whether correction is needed;

[0013] If D>0, reduce the curvature index value CV2 and the shape index value SL2 and repeat S7;

[0014] If D<0, increase the curvature index value CV2 and the shape index value SL2 and repeat S7;

[0015] If D=0, the process ends and returns to step S6 to detect the next point until all point cloud data are judged and a three-dimensional point set of cultural relic restoration is obtained;

[0016] S8. Convert the three-dimensional point set of cultural relic restoration into a three-dimensional cultural relic image and store it in the education sharing database.

[0017] Preferably, in step S1, the specific implementation steps are as follows:

[0018] S11, filtering the image of the cultural relics to be detected by using a median filter to obtain a filtered image;

[0019] S12, performing expansion and corrosion processing on the filtered image to obtain a corrosion image;

[0020] S13, performing linear transformation on the eroded image through a linear transformation algorithm to obtain a two-dimensional image.

[0021] Preferably, in step S2, the specific implementation steps are as follows:

[0022] S21, converting the two-dimensional image into a two-dimensional grayscale image I(x,y) using an RGB weighted method;

[0023] S22, transition operator S using Sobel algorithm a The core S in the x direction axConvolution operation with the two-dimensional image I(x,y) obtains the gradient matrix set G e , the calculation formula is as follows:

[0024] G e =S ax ×I(x,y)

[0025] Among them, G e Represents the gradient matrix set after convolution in the x direction, including the matrix obtained by each convolution operation, S ax represents the transition operator S a The core in the x direction;

[0026] S23, according to the transition operator S a The core S in the y direction ay Convolution operation with the two-dimensional image I(x,y) obtains the gradient matrix set G f , the calculation formula is as follows:

[0027] G f =S ay ×I(x,y)

[0028] Among them, G f Represents the gradient matrix set after convolution in the y direction, including the matrix obtained by each convolution operation, S ay represents the transition operator S a The core in the x direction;

[0029] S24, calculate the gradient matrix set G respectively e and the gradient matrix set G f Transition Elements I xx ,I xy ,I yx and I yy , the calculation formula is as follows:

[0030] I xx =G e ×S ax

[0031] I yy =G f ×S ay

[0032]

[0033] Among them, I xx ,I xy ,I yx and I yy Represents the second-order partial derivatives of the two-dimensional image I(x,y) in the x and y directions;

[0034] S25, the transition element I xx,I xy ,I yx and I yy Combined into transition matrix H b , where H b =(I xx I xy I yx I yy ).

[0035] Preferably, in step S3, the specific implementation steps are as follows:

[0036] S31. According to I xx ,I xy ,I yx and I yy Calculate the transition matrix H b The matrix trace tr(H) and matrix shape det(H) are calculated as follows:

[0037] tr(H)=I xx +I yy

[0038] det(H)=I xx I yy -I xy I yx

[0039] Among them, tr(H) represents the transition matrix H corresponding to a point in the two-dimensional image I(x,y) b The matrix trace of det(H) represents the transition matrix H corresponding to a point in the two-dimensional image I(x,y). b Matrix form;

[0040] S32, calculating the maximum principal curvature k of the two-dimensional image according to the matrix trace tr(H) and the matrix shape det(H) max and the minimum principal curvature k min , the calculation formula is as follows:

[0041]

[0042] Among them, k max represents the maximum principal curvature of a point in a two-dimensional image, k min Represents the minimum principal curvature of a two-dimensional image.

[0043] Preferably, in step S4, the specific implementation steps are as follows:

[0044] S41, select a point A in the two-dimensional image, and calculate the maximum principal curvature k of point A. max and the minimum principal curvature k min Calculate the curvature index value CV and the shape index value SL;

[0045] S42, repeat step S41 to obtain the curvature index value CV and shape index value SL of all points in the two-dimensional image, and form an initial feature set C d .

[0046] Preferably, the calculation formula of the curvature index value CV is as follows:

[0047]

[0048] Among them, CV represents the curvature index value of point A in the two-dimensional image, which is used to measure the curvature of the local surface.

[0049] Preferably, the calculation formula of the shape index value SL is as follows:

[0050]

[0051] Wherein, SL represents the shape index value of point A in the two-dimensional image, which is used to measure the shape of the point.

[0052] Preferably, in step S6, the specific implementation steps are as follows:

[0053] S61. Based on the cultural relics point cloud dataset D c Establish a three-dimensional coordinate system and obtain the coordinates of each point;

[0054] S62. Select the cultural relics point cloud dataset D c Take any point B in the image, and perform local surface fitting on the points around point B whose distance is less than or equal to b by the least square method, and obtain the local surface z = r (p, q);

[0055] S63, calculate the first invariant (E, F, G) of the local surface z, the calculation formula is as follows:

[0056] E=r p ·r p

[0057] F=r p ·r q

[0058] G=r q ·r q

[0059] Among them, r p represents the partial derivative of the local surface z = r (p, q) in the p direction, which is calculated using the surface partial derivative, r q represents the partial derivative of the local surface z=r(p,q) in the q direction;

[0060] S64. Calculate the second invariant (L, M, N) of the local surface z. The calculation formula is as follows:

[0061]

[0062] Among them, r pp represents the second-order partial derivative of the local surface z=r(p,q) in the r direction, which is calculated using the second-order partial derivative formula of the surface. pq represents the mixed partial derivative of the local surface z, r qq represents the second-order partial derivative of the local surface z in the q direction, Represents the normal vector of the surface at that point, which can be calculated using the partial derivative formula;

[0063] S65. Calculate the average curvature J and Gaussian curvature K of the local surface z according to the first invariant (E, F, G) and the second invariant (L, M, N). The calculation formula is as follows:

[0064]

[0065] Among them, E, F, G represent the first invariant, L, M, N represent the second invariant;

[0066] S67. Calculate the maximum principal curvature m of the local surface z according to the mean curvature J and the Gaussian curvature K max and the minimum principal curvature m min , the calculation formula is as follows:

[0067]

[0068] Where J represents the average curvature of the local surface z, and K represents the Gaussian curvature of the local surface z;

[0069] S68, according to the maximum principal curvature m max and the minimum principal curvature m min Calculate the curvature index value CV2 and the shape index value SL2, the calculation formula is as follows:

[0070]

[0071] Among them, CV2 represents the curvature index value of the local surface z, which is used to measure the curvature of the local surface, and SL2 represents the shape index value of the local surface z, which is used to measure the shape of the point.

[0072] Preferably, in step S7, specifically: calculate the deformation amount D between the curvature index value CV2 and the curvature index value CV and the shape index value SL2 and the shape index value SL, and the calculation formula is as follows:

[0073] D=w1CV2+w2SL2-w1CV-w2SL

[0074] Among them, w1 and w2 are weight coefficients, which are calculated using the hierarchical analysis method.

[0075] By means of the above technical solution, the present invention provides a red education sharing method based on cloud computing, which has at least the following beneficial effects:

[0076] 1. The present invention performs convolution processing on the two-dimensional image of the red cultural relics, takes the curvature index value and shape index value of the object in the image as features, and then collects point cloud of the existing status of the remaining cultural relics, and uses this as a template to repair and restore the template according to the features of the curvature index value and shape index value of the image, and finally restores the cultural relics into a three-dimensional image, thereby solving the problem that it is difficult to directly obtain the three-dimensional image of the damaged and missing red-related cultural relics in the prior art. At the same time, the missing parts of the red-related objects can be repaired, so that they can be presented to learners in a more complete and realistic manner, thereby improving the learners' educational experience and learning interest.

[0077] 2. The present invention calculates the brightness change rate in the two-dimensional image through convolution processing, obtains the surface features of the object in the two-dimensional image based on the calculation, and quickly compares and repairs the three-dimensional image based on these features, which not only improves the accuracy of cultural relic restoration, but also improves the efficiency of three-dimensional image restoration.

[0078] 3. The present invention obtains the surface features of damaged or missing cultural relics through processing and analysis of point cloud data, and then accurately and quickly modifies the point cloud data and restores the three-dimensional image. The steps are simple and the feature restoration accuracy is high, which greatly improves work efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0080] Figure 1 It is a flow chart of the red education sharing method of the present invention. DETAILED DESCRIPTION

[0081] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific implementation methods, so that the implementation process of how the present application uses technical means to solve technical problems and achieve technical effects can be fully understood and implemented accordingly.

[0082] Due to the progress of science and technology and people's cognition, two-dimensional pictures can no longer meet the needs of current learners, and objects need to be converted into three-dimensional images. However, due to the passage of time, the bodies of some red-related objects have faded, deformed or disappeared, and only two-dimensional pictures remain, which makes it difficult to directly convert these objects into three-dimensional images, and thus learners in shared red education cannot get vivid feelings and comprehensive learning. The present invention performs convolution processing on the two-dimensional pictures of red objects, uses the curvature index value and shape index value of the objects in the image as features, and then collects point clouds based on the existing states of the residual objects, and then repairs and restores the templates in combination with the picture features, and finally restores the objects to form three-dimensional images, thereby solving the problem that the prior art is difficult to directly obtain three-dimensional images of red-related objects. At the same time, it can also repair the actual parts of the red-related objects, so that they can be presented to learners in a more complete and realistic manner, thereby improving the learners' feelings and learning interest.

[0083] Please refer to Figure 1 This embodiment provides a red education sharing method based on cloud computing to solve the problem that due to the advancement of science and technology and the advancement of people's cognition, two-dimensional pictures are difficult to meet the needs of current learners, and objects need to be converted into three-dimensional images. However, due to the passage of time, the main bodies of some red characters' related objects fade, deform or disappear, and only two-dimensional pictures remain, which makes it difficult to directly convert these objects into three-dimensional images, thereby preventing learners in the shared red education from getting vivid feelings and comprehensive learning. The method includes the following steps:

[0084] S1, obtaining an image of a cultural relic to be detected, and preprocessing the image of the cultural relic to be detected to obtain a two-dimensional image; in this step, the image of the cultural relic to be detected is filtered, corroded, expanded, and linearly transformed to convert the image into a two-dimensional image, which is convenient for convolution operation. The specific implementation steps are as follows:

[0085] S11, filtering the image of the cultural relics to be detected by using a median filter to obtain a filtered image;

[0086] S12, performing expansion and corrosion processing on the filtered image to obtain a corrosion image;

[0087] S13, performing linear transformation on the eroded image through a linear transformation algorithm to obtain a two-dimensional image.

[0088] It should be noted that the median filter is a commonly used method for image filtering. This method can complete image filtering without other operations, which will not be repeated here. Dilation and erosion are also a common method for image processing. Linear transformation requires the rate of change of grayscale in the image to be more linear. It is a commonly used method in the prior art, which is convenient for subsequent convolution and calculation of principal curvature, which will not be repeated here.

[0089] S2, using transition operator Sa Perform convolution operation on the two-dimensional image to obtain the transition matrix H of the two-dimensional image b ; In this step, the image is first converted into a grayscale image to facilitate convolution operation, and then the image is further processed by convolution operation to form a transition matrix. The transition matrix is ​​a Hessian matrix that can help calculate the curvature of a surface and can be used as an eigenvalue to facilitate comparison in the restoration of three-dimensional images. The specific implementation steps are as follows:

[0090] S21, converting the two-dimensional image into a two-dimensional grayscale image I(x,y) using an RGB weighted method;

[0091] S22, transition operator S using Sobel algorithm a The core S in the x direction ax Convolution operation with the two-dimensional image I(x,y) obtains the gradient matrix set G e , the calculation formula is as follows:

[0092] G e =S ax ×I(x,y)

[0093] Among them, G e Represents the gradient matrix set after convolution in the x direction, including the matrix obtained by each convolution operation, S ax represents the transition operator S a The core in the x direction;

[0094] S23, according to the transition operator S a The core S in the y direction ay Convolution operation with the two-dimensional image I(x,y) obtains the gradient matrix set G f , the calculation formula is as follows:

[0095] G f =S ay ×I(x,y)

[0096] Among them, G f Represents the gradient matrix set after convolution in the y direction, including the matrix obtained by each convolution operation, S ay represents the transition operator S a The core in the x direction;

[0097] S24, calculate the gradient matrix set G respectively e and the gradient matrix set G f Transition Elements I xx ,I xy ,I yx and I yy , the calculation formula is as follows:

[0098] Ixx =G e ×S ax

[0099] I yy =G f ×S ay

[0100]

[0101] Among them, I xx ,I xy ,I yx and I yy Represents the second-order partial derivative of the two-dimensional image I(x,y) in the x and y directions. The second-order partial derivative can be solved by the second-order partial derivative formula;

[0102] It should be noted that the solution I xx ,I xy ,I yx and I yy The method is the second-order partial derivative, and its formula is common knowledge. Only the second-order partial derivative formula is needed to complete the solution. No other steps are required, so I will not go into details here.

[0103] S25, the transition element I xx ,I xy ,I yx and I yy Combined into transition matrix H b , where H b =(I xx I xy I yx I yy ).

[0104] S3, according to the transition matrix H b Calculate the maximum principal curvature k of a 2D image max and the minimum principal curvature k min ; In this step, the maximum principal curvature k max and the minimum principal curvature k min The transition element I that needs to be obtained after convolution xx ,I xy ,I yx and I yy Perform matrix trace operations and matrix shape operations to facilitate the calculation of the curvature index value CV and the shape index value SL in the later stage. The specific implementation steps are as follows:

[0105] S31. According to I xx ,I xy ,I yx and I yy Calculate the transition matrix H bThe matrix trace tr(H) and matrix shape det(H) are calculated as follows:

[0106] tr(H)=I xx +I yy

[0107] det(H)=I xx I yy -I xy I yx

[0108] Among them, tr(H) represents the transition matrix H corresponding to a point in the two-dimensional image I(x,y) b The matrix trace of det(H) represents the transition matrix H corresponding to a point in the two-dimensional image I(x,y). b Matrix form;

[0109] S32, calculating the maximum principal curvature k of the two-dimensional image according to the matrix trace tr(H) and the matrix shape det(H) max and the minimum principal curvature k min , the calculation formula is as follows:

[0110]

[0111] Among them, k max represents the maximum principal curvature of a point in a two-dimensional image, k min Represents the minimum principal curvature of a two-dimensional image.

[0112] S4, according to the maximum principal curvature k max and the minimum principal curvature k max Calculate the curvature index value CV and shape index value SL, and get the initial feature set C d In this step, the curvature index value CV is the maximum principal curvature k max and the minimum principal curvature k min The value of the shape index SL is calculated based on the brightness change rate in the previous convolution result. This curvature can reflect the characteristics of a certain position of the cultural relic in the image. Similarly, the shape index value SL is the maximum principal curvature k max and the minimum principal curvature k min The shape of a certain position of the cultural relic in the image is calculated and used as a feature. The specific implementation steps are as follows:

[0113] S41, select a point A in the two-dimensional image, and calculate the maximum principal curvature k of point A. max and the minimum principal curvature k min Calculate the curvature index value CV and the shape index value SL. The calculation formula is as follows:

[0114]

[0115] Among them, CV represents the curvature index value of point A in the two-dimensional image, which is used to measure the curvature of the local surface, and SL represents the shape index value of point A in the two-dimensional image, which is used to measure the shape of the point;

[0116] S42, repeat step S41 to obtain the curvature index value CV and shape index value SL of all points in the two-dimensional image, and form an initial feature set C d .

[0117] S5. Obtain the cultural relic point cloud dataset D of the cultural relic to be restored through VR equipment c ; It should be noted in this step that the VR device includes a device for collecting point cloud images in the prior art. The present invention selects a low-cost ultrasonic device to collect point cloud images. Since the size of the cultural relics is moderate, the accuracy and economy are better than other methods and equipment.

[0118] S6. Based on the initial feature set C d Calculate the curvature index value CV2 and shape index value SL2 of the local surface z formed by any point B; in this step, use the point cloud data to perform a preliminary local surface fitting on the point cloud, and establish a three-dimensional coordinate system, which is convenient for the later calculation of the two characteristics of the curvature and shape of the local surface, and then calculate the curvature index value CV2 and shape index value SL2 of the local surface corresponding to the point one by one, which is convenient for the next step of comparison and repair. The specific implementation steps are as follows:

[0119] S61. Based on the cultural relics point cloud dataset D c Establish a three-dimensional coordinate system and obtain the coordinates of each point;

[0120] S62. Select the cultural relics point cloud dataset D c Take any point B in the image, and perform local surface fitting on the points around point B whose distance is less than or equal to b by the least square method, and obtain the local surface z = r (p, q);

[0121] S63, calculate the first invariant (E, F, G) of the local surface z, the calculation formula is as follows:

[0122] E=r p ·r p

[0123] F=r p ·r q

[0124] G=r q ·r q

[0125] Among them, r prepresents the partial derivative of the local surface z = r (p, q) in the p direction, which is calculated using the surface partial derivative, r q represents the partial derivative of the local surface z=r(p,q) in the q direction;

[0126] S64. Calculate the second invariant (L, M, N) of the local surface z. The calculation formula is as follows:

[0127]

[0128] Among them, r pp represents the second-order partial derivative of the local surface z=r(p,q) in the p direction, which is calculated using the second-order partial derivative formula of the surface. pq represents the mixed partial derivative of the local surface z, r qq represents the second-order partial derivative of the local surface z in the q direction, Represents the normal vector of the surface at that point, which can be calculated using the partial derivative formula;

[0129] S65. Calculate the average curvature J and Gaussian curvature K of the local surface z according to the first invariant (E, F, G) and the second invariant (L, M, N). The calculation formula is as follows:

[0130]

[0131] Among them, E, F, G represent the first invariant, L, M, N represent the second invariant;

[0132] S67. Calculate the maximum principal curvature m of the local surface z according to the mean curvature J and the Gaussian curvature K max and the minimum principal curvature m min , the calculation formula is as follows:

[0133]

[0134] Where J represents the average curvature of the local surface z, and K represents the Gaussian curvature of the local surface z;

[0135] S68, according to the maximum principal curvature m max and the minimum principal curvature m min Calculate the curvature index value CV2 and the shape index value SL2, the calculation formula is as follows:

[0136]

[0137] Among them, CV2 represents the curvature index value of the local surface z, which is used to measure the curvature of the local surface, and SL2 represents the shape index value of the local surface z, which is used to measure the shape of the point.

[0138] S7, calculate the deformation variable D according to the curvature index value CV2 and the shape index value SL2, and determine the cultural relic point cloud data set D c whether correction is needed;

[0139] If D>0, reduce the curvature index value CV2 and the shape index value SL2 and repeat S7;

[0140] If D<0, increase the curvature index value CV2 and the shape index value SL2 and repeat S7;

[0141] If D=0, the process ends and returns to step S6 to detect the next point until all point cloud data are judged and the three-dimensional point set of cultural relic restoration is obtained; it should be noted in this step that since the curvature index value CV, the shape index value SL, the curvature index value CV2 and the shape index value SL2 are all different eigenvalues, in order to facilitate comparison, the eigenvalues ​​are normalized using weight coefficients, so that their similarities and differences can be directly compared. Due to the passage of time, some cultural relics are prone to missing or damaged. Directly obtaining three-dimensional point cloud images will prevent learners from seeing complete cultural relics, so that they cannot vividly learn and understand the hardships of the revolution at that time. Therefore, the intact cultural relic images in the pictures are used for collection, and then the three-dimensional point clouds of existing cultural relics can be compared for restoration, and the missing of existing cultural relics can be corrected, and then the point cloud data can be converted into three-dimensional images. If the cultural relic is damaged and the original cultural relic cannot be found, an object with a similar appearance can be found instead, and the three-dimensional image of the cultural relic can also be restored through the two-dimensional image of the cultural relic. Specifically, the deformation variable D between the curvature index value CV2 and the curvature index value CV and the shape index value SL2 and the shape index value SL is calculated, and the calculation formula is as follows:

[0142] D=w1CV2+w2SL2-w1CV-w2SL

[0143] Among them, w1 and w2 are weight coefficients, which are calculated using the hierarchical analysis method.

[0144] S8. Convert the three-dimensional point set of cultural relic restoration into a three-dimensional cultural relic image and store it in the education sharing database.

[0145] It should be noted that in the red education sharing method, the education sharing database is used to store relevant information, and when in use, the relevant information can be directly obtained by searching and extracting it from the database, such as the three-dimensional image of the present invention.

[0146] Those skilled in the art can understand that all or part of the steps in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a program, so the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0147] Each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the above embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0148] The above implementation methods have been described in detail. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A red education sharing method based on cloud computing, characterized in that: The method comprises the following steps: S1, obtaining an image of a cultural relic to be detected, and preprocessing the image of the cultural relic to be detected to obtain a two-dimensional image; S2, using transition operator S a Perform convolution operation on the two-dimensional image to obtain the transition matrix H of the two-dimensional image b ; S3, according to the transition matrix H b Calculate the maximum principal curvature k of a 2D image max and the minimum principal curvature k min ; S4, according to the maximum principal curvature k max and the minimum principal curvature k min Calculate the curvature index value CV and shape index value SL, and get the initial feature set C d ; S5. Obtain the cultural relic point cloud dataset D of the cultural relic to be restored through VR equipment c ; S6. Based on the initial feature set C d Calculate the curvature index value CV2 and shape index value SL2 of the local surface z formed by any point B; S7, calculate the deformation variable D according to the curvature index value CV2 and the shape index value SL2, and determine the cultural relic point cloud data set D c whether correction is needed; If D>0, reduce the curvature index value CV2 and the shape index value SL2 and repeat S7; If D<0, increase the curvature index value CV2 and the shape index value SL2 and repeat S7; If D=0, the process ends and returns to step S6 to detect the next point until all point cloud data are judged and a three-dimensional point set of cultural relic restoration is obtained; S8. Convert the three-dimensional point set of cultural relic restoration into a three-dimensional cultural relic image and store it in the education sharing database.

2. The sharing method according to claim 1, characterized in that: In step S1, the specific implementation steps are as follows: S11, filtering the image of the cultural relics to be detected by using a median filter to obtain a filtered image; S12, performing expansion and corrosion processing on the filtered image to obtain a corrosion image; S13, performing linear transformation on the eroded image through a linear transformation algorithm to obtain a two-dimensional image.

3. The sharing method according to claim 1, characterized in that: In step S2, the specific implementation steps are as follows: S21, converting the two-dimensional image into a two-dimensional grayscale image I (x, y) using an RGB weighted method; S22, transition operator S using Sobel algorithm a The core S in the x direction ax Convolution operation with the two-dimensional image I(x, y) obtains the gradient matrix set G e , the calculation formula is as follows: G e =S ax ×I(x,y) Among them, G e Represents the gradient matrix set after convolution in the x direction, including the matrix obtained by each convolution operation, S ax represents the transition operator S a The core in the x direction; S23, according to the transition operator S a The core S in the y direction ay Convolution operation with the two-dimensional image I(x, y) obtains the gradient matrix set G f , the calculation formula is as follows: G f =S ay ×I(x,y) Among them, G f Represents the gradient matrix set after convolution in the y direction, including the matrix obtained by each convolution operation, S ay represents the transition operator S a The core in the x direction; S24, calculate the gradient matrix set G respectively e and the gradient matrix set G f Transition Elements I xx ,I xy ,I yx and I yy , the calculation formula is as follows: I xx =G e ×S ax I yy =G f ×S ay Among them, I xx ,I xy ,I yx and I yy Represents the second-order partial derivatives of the two-dimensional image I(x, y) in the x and y directions; S25, the transition element I xx ,I xy ,I yx and I yy Combined into transition matrix H b , where H b =(I xx I xy I yx I yy ).

4. The sharing method according to claim 1, characterized in that: In step S3, the specific implementation steps are as follows: S31. According to I xx ,I xy ,I yx and I yy Calculate the transition matrix H b The matrix trace tr(H) and matrix shape det(H) are calculated as follows: tr(H)=I xx +I yy it(H)=I xx IN yy -IN xy IN yx Among them, tr(H) represents the transition matrix H corresponding to a point in the two-dimensional image I(x, y) b The matrix trace of det(H) represents the transition matrix H corresponding to a point in the two-dimensional image I(x, y). b Matrix form; S32, calculating the maximum principal curvature k of the two-dimensional image according to the matrix trace tr(H) and the matrix shape det(H) max and the minimum principal curvature k min , the calculation formula is as follows: Among them, k max represents the maximum principal curvature of a point in a two-dimensional image, k min Represents the minimum principal curvature of a two-dimensional image.

5. The sharing method according to claim 1, characterized in that: In step S4, the specific implementation steps are as follows: S41, select a point A in the two-dimensional image, and calculate the maximum principal curvature k of point A. max and the minimum principal curvature k min Calculate the curvature index value CV and the shape index value SL; S42, repeat step S41 to obtain the curvature index value CV and shape index value SL of all points in the two-dimensional image, and form an initial feature set C d .

6. The sharing method according to claim 5, characterized in that: The calculation formula of the curvature index value CV is as follows: Among them, CV represents the curvature index value of point A in the two-dimensional image, which is used to measure the curvature of the local surface.

7. The sharing method according to claim 5, characterized in that: The calculation formula of the shape index value SL is as follows: Wherein, SL represents the shape index value of point A in the two-dimensional image, which is used to measure the shape of the point.

8. The sharing method according to claim 1, characterized in that: In step S6, the specific implementation steps are as follows: S61. Based on the cultural relics point cloud dataset D c Establish a three-dimensional coordinate system and obtain the coordinates of each point; S62. Select the cultural relics point cloud dataset D c Take any point B in the image, and perform local surface fitting on the points around point B whose distance is less than or equal to b by the least square method, and obtain the local surface z = r (p, q); S63, calculate the first invariant (E, F, G) of the local surface z, the calculation formula is as follows: E＝r p ·r p F＝r p ·r q G＝r q ·r q Among them, r p represents the partial derivative of the local surface z = r (p, q) in the p direction, which is calculated using the surface partial derivative, r q represents the partial derivative of the local surface z=r(p,q) in the q direction; S64. Calculate the second invariant (L, M, N) of the local surface z. The calculation formula is as follows: Among them, r pp represents the second-order partial derivative of the local surface z=r(p,q) in the p direction, r pq represents the mixed partial derivative of the local surface z, r qq represents the second-order partial derivative of the local surface z in the q direction, Represents the normal vector of the surface at that point; S65. Calculate the average curvature J and Gaussian curvature K of the local surface z according to the first invariant (E, F, G) and the second invariant (L, M, N). The calculation formula is as follows: Among them, E, F, G represent the first invariant, L, M, N represent the second invariant; S67. Calculate the maximum principal curvature m of the local surface z according to the mean curvature J and the Gaussian curvature K max and the minimum principal curvature m min , the calculation formula is as follows: Where J represents the average curvature of the local surface z, and K represents the Gaussian curvature of the local surface z; S68, according to the maximum principal curvature m max and the minimum principal curvature m min Calculate the curvature index value CV2 and the shape index value SL2, the calculation formula is as follows: Among them, CV2 represents the curvature index value of the local surface z, which is used to measure the curvature of the local surface, and SL2 represents the shape index value of the local surface z, which is used to measure the shape of the point.

9. The sharing method according to claim 1, characterized in that: In step S7, specifically, the deformation amount D between the curvature index value CV2 and the curvature index value CV and the shape index value SL2 and the shape index value SL is calculated, and the calculation formula is as follows: D=w1CV2+w2SL2-w1CV-w2SL Among them, w1 and w2 are weight coefficients, which are calculated using the hierarchical analysis method.

Images