AI technology-based non-abandoned content pushing method and system

Abstract

The invention provides an AI technology-based non-missing content pushing method and system, relates to a data processing technology, and aims at enabling a user to experience a ceramic firing process by wearing virtual-real combined VR glasses in a process of experiencing non-missing, such as ceramic firing, of the user. In the operation process of the user, image acquisition can be performed on the user, and the current process of the user can be recognized, so that corresponding operation instructions are recommended for the user, for example, operation gestures are recommended for the user, the manufacturing process of the user is guided to better meet the standard, and automatic recommendation is realized.

Description

Technical Field

[0001] The present invention relates to data processing technology, and in particular to a method and system for pushing intangible cultural heritage content based on AI technology. Background Art

[0002] In today's society, the inheritance and development of intangible cultural heritage has received widespread attention. Among them, traditional ceramic firing techniques, as an important intangible cultural heritage project, contain profound cultural heritage and exquisite craftsmanship.

[0003] At present, when users experience the production process of intangible cultural heritage, they generally rely on manual guidance or combine it with pre-recorded film and television videos. For example, when users try to make a ceramic embryo into a specific shape (such as a bowl), they need to be manually told the specific operation method, including the position and angle of the hand and the corresponding movements. This method has many disadvantages. On the one hand, the labor cost is high, and professional personnel are required to be on site at all times for guidance; on the other hand, due to the subjectivity and differences of manual guidance, the guidance effects obtained by different users may be uneven, and it is difficult to ensure the consistency and standardization of user experience.

[0004] Therefore, how to combine the user's operational data during the intangible cultural heritage experience to provide users with customized guidance has become a problem that needs to be solved urgently. Summary of the invention

[0005] The present invention provides an AI-based intangible cultural heritage content push method and system, which can provide customized guidance for users in combination with the user's operation data during the intangible cultural heritage experience process.

[0006] A first aspect of the present invention provides a method for pushing intangible cultural heritage content based on AI technology, comprising: Pushing basic display content according to the display device, the basic display content includes a benchmark identifier; Obtain visual feature parameters of the intangible cultural heritage products based on the acquisition equipment, and determine the process nodes of the intangible cultural heritage products based on the comparison results of the visual feature parameters and the standard feature parameters of each process stage; When the process node meets the judgment condition, a guide identifier corresponding to the process node is generated according to the difference between the visual feature parameters and the corresponding standard feature parameters, and the basic identifier is updated to the guide identifier.

[0007] Optionally, in a possible implementation of the first aspect, obtaining visual feature parameters of the intangible cultural heritage product based on a collection device, and determining the process node of the intangible cultural heritage product according to a comparison result between the visual feature parameters and standard feature parameters of each process stage include: Identify the outline of the intangible cultural heritage product according to the acquisition device, and the visual feature parameters include the outline of the item; Selecting process stages in the process sequence as target stages in sequence, retrieving standard profiles of the target stages, wherein standard feature parameters include standard profiles; The total similarity values ​​of the object outline and the standard outline under each viewing angle are calculated by fusion, and the progress value corresponding to the total similarity value is determined; The process axis is retrieved to determine the position of the progress value on the process axis, a process node is generated at the position, and when the progress value is greater than the stage threshold, the next target stage is entered.

[0008] Optionally, in a possible implementation of the first aspect, the visual feature parameters of the intangible cultural heritage product are obtained through the following steps, including: Identify the breakpoints in the outline of the intangible cultural heritage artifact and calculate the point distances between adjacent breakpoints; The arc adjustment parameters are obtained according to the ratio of the preset constant and the point distance, and the padding lines are generated based on the arc adjustment parameters to connect the adjacent breakpoints, and the visual feature parameters are obtained according to the outline of the object after padding.

[0009] Optionally, in a possible implementation manner of the first aspect, in the process of acquiring visual feature parameters of the intangible cultural heritage product based on a collection device, and determining the process node of the intangible cultural heritage product according to a comparison result between the visual feature parameters and standard feature parameters of each process stage, the process further includes: Obtain a similarity value sequence of the visual feature parameter and the corresponding standard feature parameter according to the time sequence, and obtain a similarity difference value of adjacent total similarity values ​​in the similarity value sequence; When a preset number of consecutive similarity differences are all smaller than the similarity difference threshold, the collection time interval of the collection device is increased according to the extension multiple; When a preset number of consecutive similarity difference values ​​are all greater than the similarity difference threshold, the collection time interval of the collection device is reduced according to the shortening factor.

[0010] Optionally, in a possible implementation manner of the first aspect, when the process node meets the determination condition, generating a guidance mark corresponding to the process node according to a difference between a visual feature parameter and a corresponding standard feature parameter includes: When the progress threshold of the process node is greater than or equal to the determination threshold, it is determined that the preset determination condition is met; Compare the feature parameter differences of the contour points corresponding to the visual feature parameters and the corresponding standard feature parameters, and determine the contour points whose feature parameter differences in the visual feature parameters are not within the standard difference value interval as difference points; Determine the feature type of the difference point, traverse the guide actions corresponding to the feature type in the gesture action library, locate the gesture identifier to the actual position corresponding to the difference point, and configure the guide action for the gesture identifier to obtain the guide identifier.

[0011] Optionally, in a possible implementation manner of the first aspect, comparing the visual feature parameters with the feature parameter differences of the contour points corresponding to the corresponding standard feature parameters, and determining the contour points whose feature parameter differences in the visual feature parameters are not within the standard difference value interval as difference points, includes: Obtaining height difference and curvature difference of contour points corresponding to the visual feature parameters and the corresponding standard feature parameters, wherein the feature parameter difference includes the height difference and the curvature difference; Determine the contour points whose height difference values ​​in the visual feature parameters are not within the corresponding standard deviation value interval and / or whose curvature difference values ​​are not within the corresponding standard deviation value interval as difference points.

[0012] Optionally, in a possible implementation of the first aspect, determining a feature type of a difference point, traversing guide actions corresponding to the feature type in a gesture action library, locating a gesture identifier to an actual position corresponding to the difference point, and configuring a guide action for the gesture identifier to obtain a guide identifier includes: Determine the difference point corresponding to the height difference as a height adjustment type, determine the difference point corresponding to the curvature difference as a shape adjustment type, determine the difference point corresponding to the height difference and the curvature difference as a comprehensive adjustment type, and the feature type includes a height adjustment type, a shape adjustment type, and a comprehensive adjustment type; The operation direction of the corresponding guiding action is determined according to the positive and negative deviations of the feature type, the gesture marker is positioned to the actual position, and while configuring the guiding action for the gesture marker, its action direction is determined as the operation direction to obtain an updated guiding marker.

[0013] Optionally, in a possible implementation manner of the first aspect, in the process of locating the gesture identifier to the actual position corresponding to the difference point and configuring a guiding action for the gesture identifier to obtain the guiding identifier, the process further includes: Based on the current moment, the moving speed of the user's hand motion in the dynamic time period is obtained, and the moving speed is adjusted to the movement speed of the guide mark.

[0014] Optionally, in a possible implementation manner of the first aspect, determining the dynamic time period by the following steps includes: Taking the current time as the benchmark, trace back the benchmark time to get the benchmark time period; Obtain the displacement distance of the user's hand in adjacent image frames within a reference time period, and obtain the distance difference between the displacement distances corresponding to a preset number of consecutive image frames; When the distance differences are all less than the threshold, the preset backtracking time is added to the base time period to obtain the dynamic time period; When there is a distance difference greater than or equal to the threshold, the dynamic time period is obtained by subtracting the preset backtracking time length from the reference time period.

[0015] A second aspect of the present invention provides an intangible cultural heritage content push system based on AI technology, comprising: A push module, used for pushing basic display content according to a display device, where the basic display content includes a reference identifier; The process module is used to obtain the visual feature parameters of the intangible cultural heritage products based on the acquisition equipment, and determine the process nodes of the intangible cultural heritage products according to the comparison results between the visual feature parameters and the standard feature parameters of each process stage; The guidance module is used to generate a guidance mark corresponding to the process node according to the difference between the visual feature parameters and the corresponding standard feature parameters when the process node meets the judgment conditions, and update the basic mark to the guidance mark.

[0016] The beneficial effects of the present invention are as follows: The present invention generates targeted guidance marks based on the difference points between the visual feature parameters and the standard feature parameters. By comparing the feature parameter differences between the visual feature parameters and the standard feature parameters corresponding to the contour points, the difference points and feature types are determined, including height adjustment type, shape adjustment type and comprehensive adjustment type. The operation direction of the corresponding guidance action is determined according to the positive and negative deviations of the feature type, the gesture mark is positioned to the actual position corresponding to the difference point, and the guidance action is configured for it. This personalized guidance method can help users accurately understand the gap between their own production process and the standard, and make adjustments according to the guidance marks, thereby improving the accuracy and flexibility of the user's operation process.

[0017] The present invention can dynamically adjust the movement speed of the guide mark according to the changes in the user's hand movements. Taking the current moment as a reference, the reference time period is obtained by tracing back the reference time length, and the distance difference between the displacement distance of the user's hand in adjacent image frames within the time period and the displacement distance corresponding to a preset number of consecutive image frames is obtained. According to the comparison result of the distance difference and the threshold, the reference time period is dynamically adjusted to obtain the dynamic time period, and then the moving speed of the user's hand movement within the time period is obtained, and it is adjusted to the movement speed of the guide mark. When the user's hand movement changes smoothly, the dynamic time period is extended to obtain more comprehensive speed data; when the hand movement changes greatly, the dynamic time period is shortened to capture the speed change more accurately. In this way, the movement speed of the guide mark is consistent with the user's operation rhythm, providing users with more natural and smooth operation guidance, and improving the flexibility of the guidance data. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a schematic diagram of an application scenario provided by an embodiment of the present invention; Figure 2 It is a flowchart of a method for pushing intangible cultural heritage content based on AI technology provided by an embodiment of the present invention; Figure 3It is a structural diagram of an intangible cultural heritage content push system based on AI technology provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] See also Figure 1 , is a schematic diagram of an application scenario provided by an embodiment of the present invention. In this embodiment, when users experience intangible cultural heritage, such as in the process of firing ceramics, users can wear virtual and real VR glasses to experience the ceramic firing process. During the user's operation, images can be collected to identify the user's current process progress, so as to recommend corresponding operation instructions, such as recommending operation gestures, guiding the user's production process to be more in line with the standard, and realizing automated recommendation. Among them, Figure 1 The display device may be a VR device, and the acquisition device may be a camera. Cameras may be installed above and around the operation area. The server may analyze the images taken by the acquisition devices at various viewing angles to determine the user's current process. The server then provides corresponding operation instructions to the user in combination with the display device, thereby improving the accuracy and efficiency of the user's operation.

[0021] See also Figure 2 , is a flow chart of a method for pushing intangible cultural heritage content based on AI technology provided by an embodiment of the present invention, Figure 2 The execution subject of the method shown may be software and / or hardware devices. The execution subject of the present application may include but is not limited to at least one of the following: user equipment, network equipment, etc. Among them, user equipment may include but is not limited to computers, smart phones, personal digital assistants (PDA) and the electronic devices mentioned above. Network equipment may include but is not limited to a single network server, a server group consisting of multiple network servers, or a cloud based on cloud computing consisting of a large number of computers or network servers, wherein cloud computing is a type of distributed computing, a super virtual computer composed of a group of loosely coupled computers. This embodiment does not limit this. It includes steps S101 to S103, as follows: S101, pushing basic display content according to a display device, where the basic display content includes a reference identifier.

[0022] When the user is ready to start the intangible cultural heritage experience of ceramic production, he puts on the VR device. The VR device can push basic display content at this time. The basic display content contains the benchmark mark that can be a standard hand posture image, so as to guide the user's operation process. Among them, the basic display content refers to the pre-configured guidance data used to guide the user to perform the initial production operation, and the benchmark mark refers to the mark for guidance.

[0023] For example, after the user puts on the VR device, a translucent, suspended standard hand posture image can be displayed on the operating table directly in front of the user, showing the hand posture for the initial action of ceramic throwing, with both hands in a ring, simulating the state of holding the clay. This benchmark mark is an important guide for users to start the ceramic production experience. It not only allows users to intuitively see the basic hand posture at the current production stage, providing the initial imitation object for subsequent actual operations, but also throughout the ceramic production process, users can review this benchmark mark at any time to calibrate whether their hand posture deviates from the standard to ensure the standardization and accuracy of the operation.

[0024] S102, obtaining visual feature parameters of the intangible cultural heritage product based on the acquisition device, and determining the process node of the intangible cultural heritage product according to the comparison results of the visual feature parameters and the standard feature parameters of each process stage.

[0025] When users are making ceramics, cameras placed in different locations in the workshop start working. These cameras can capture the state of the ceramic body in the user's hand in all directions and in real time. Visual feature parameters refer to the shape change parameters of intangible cultural heritage products, such as the outline of intangible cultural heritage products. Standard feature parameters refer to the shape parameters of standard products at each process stage. Through their comparison results, the user's current production progress can be determined, so that corresponding production guidance can be provided for them in the future. The process node refers to the node corresponding to the user's process progress. Intangible cultural heritage products can be ceramic products.

[0026] Detailed and precise standard characteristic parameters can be pre-set for each process stage of ceramic production. For example, in the throwing stage, the standard body contour can be a cylinder with uniform top and bottom and flat bottom.

[0027] The server compares the visual feature parameters of the ceramic body collected by the camera with the standard feature parameters of each process stage. The corresponding process node is determined by carefully analyzing the body contour and the standard contour from multiple perspectives (such as top view and side view).

[0028] Based on the above embodiment, the specific implementation of step S102 may be: The object outline of the intangible cultural heritage product is identified according to the acquisition equipment, and the visual feature parameters include the object outline.

[0029] The acquisition devices at different viewing angles can obtain the contours of the intangible cultural heritage products at different angles, so that more comprehensive data can be obtained from the contours of the items at different viewing angles. For example, in the initial stage of throwing, the clay may be an irregular block, and the camera can recognize its approximate external shape; with the user's operation, the clay is gradually shaped into a roughly cylindrical shape, and the camera can capture this changed contour; in the stage of trimming, the contour of the body will be more refined and regular, and the camera can also accurately identify and record it. The identified object contour is an important part of the visual feature parameters. It records the shape information of the ceramic body at every moment in the production process.

[0030] It is understandable that, in the actual ceramic production process, various factors may cause the collected object contour to not be a continuous and complete line, such as the user's hand blocking, so that the extracted object contour may be incomplete. Therefore, in order to improve the accuracy of the process judgment, in some embodiments, the visual feature parameters can be obtained by the following steps: Identify the breakpoints in the object contour of the intangible cultural heritage product, calculate the point distance between adjacent breakpoints, obtain the arc adjustment parameter according to the ratio of the preset constant and the point distance, generate the filling line to connect the adjacent breakpoints based on the arc adjustment parameter, and obtain the visual feature parameters according to the object contour after filling.

[0031] Among them, breakpoints refer to discontinuous contour points in the object contour. For example, during the throwing process, due to uneven hand operation or differences in clay texture, some bumps may appear on the surface of the blank, resulting in interruptions in the object contour in the image. When calculating the point distance between adjacent breakpoints, the coordinate position of the breakpoint in the image can be determined, and the actual distance between them can be calculated using mathematical formulas. This distance reflects the degree of contour interruption.

[0032] The arc adjustment parameter refers to the coefficient for adjusting the arc of the line segment connecting adjacent breakpoints. The fill line is the arc connecting adjacent breakpoints. According to the determined arc parameters, an arc can be drawn to connect two breakpoints, and the arc can be adjusted by the parameter.

[0033] It is understandable that when the two breakpoints are very close, if a smaller arc is set, the arc will be smoother. In this way, when connecting the two breakpoints, the arc may not be able to closely fit the contour trend near the breakpoints, resulting in a large deviation between the contour after completion and the contour of the actual object, resulting in distortion. Setting a larger arc can make the arc better fit the contour near the breakpoint, maintain the continuity and authenticity of the contour, and make the contour after completion closer to the actual shape of the ceramic body. On the contrary, if the distance between the points is large, a smaller arc can be used to make the arc relatively smooth.

[0034] Through the above method, the breakpoints in the outline of the ceramic body object can be effectively filled in and more accurate visual feature parameters can be obtained.

[0035] The process stages in the process sequence are selected in sequence as the target stage, and the standard profile of the target stage is retrieved. The standard feature parameters include the standard profile.

[0036] It is understandable that ceramic production is a process with strict order and stage division. There is a pre-set process sequence, which usually includes multiple stages such as throwing, trimming, decoration, firing, etc. Each process stage can be set as the target stage in turn according to the preset process sequence.

[0037] For each target stage, the database stores the corresponding standard contours. These standard contours can be standard shapes determined by the staff based on actual conditions. For example, in the throwing stage, the standard contour is a perfect cylinder with a flat bottom, vertical side walls and uniform thickness. When the system sets a process stage as the target stage, it will retrieve the standard contour corresponding to that stage from the database. This standard contour becomes an important reference for measuring whether the current ceramic body production meets the standards and is a key content of the standard feature parameters.

[0038] The total similarity values ​​of the object outline and the standard outline at each viewing angle are calculated and integrated to determine the progress value corresponding to the total similarity value.

[0039] Specifically, the similarity of the object outline under each viewing angle (such as top, side, etc.) and the standard outline under that viewing angle can be compared, and then the similarities calculated from multiple viewing angles can be added to obtain the total similarity. The total similarity value reflects the degree of completion of the current ceramic production at the target stage. According to the pre-set correspondence, the total similarity value can be converted into the corresponding progress value. For example, if the total similarity value is 80%, the corresponding progress value may show that the current ceramic production has been completed by 80% at this process stage. The progress value provides a quantitative indicator for users and systems, clearly indicating the progress of ceramic production at the current stage.

[0040] The process axis is retrieved to determine the position of the progress value on the process axis, a process node is generated at the position, and when the progress value is greater than the stage threshold, the next target stage is entered.

[0041] The process axis is a data axis used to represent the user's progress. On the process axis, each process stage is arranged in chronological order, and each process stage has its corresponding range and threshold. Each process stage on the process axis has its corresponding progress value at each position. Therefore, after determining the corresponding progress value of the user, the corresponding process node can be generated at the corresponding position.

[0042] Each process stage has a stage threshold set, which is the standard for judging whether to enter the next process stage. When the progress value is greater than this stage threshold, it means that the user has basically met the production requirements in the current process stage and can enter the next process stage. For example, the stage threshold of the throwing stage is set to 90%. When the progress value reaches 92%, the system will set the next process stage (such as the trimming stage) as the target stage, and then repeat the above steps to continue analyzing and guiding the new target stage, thereby helping users to gradually complete the entire ceramic production process.

[0043] In addition, in step S102 (obtaining visual feature parameters of the intangible cultural heritage product based on the acquisition device, and determining the process node of the intangible cultural heritage product according to the comparison results of the visual feature parameters and the standard feature parameters of each process stage), the following embodiments may also be included: A similarity value sequence of visual feature parameters and corresponding standard feature parameters is obtained in chronological order, and a similarity difference value of adjacent total similarity values ​​in the similarity value sequence is obtained; when a continuous preset number of similarity difference values ​​are all smaller than a similarity difference threshold, the collection time interval of the collection device is increased according to an extension multiple; when a continuous preset number of similarity difference values ​​are all larger than the similarity difference threshold, the collection time interval of the collection device is reduced according to a shortening multiple.

[0044] It can be understood that in order to reduce the amount of data collected by the acquisition device and to timely capture changes in the user's production process, so as to quickly determine whether there are deviations in the process and provide users with more accurate operation guidance, the frequency of image acquisition by the acquisition device can be dynamically adjusted.

[0045] Specifically, in the process of users making ceramics, the camera continuously collects images of the ceramic body, arranges and records the total similarity values ​​at adjacent moments to obtain a similarity value sequence, and calculates the similarity difference of adjacent total similarity values ​​in real time. It can be understood that the similarity difference at adjacent moments can reflect the rate of change of the body state, which is crucial for accurately grasping the dynamic changes in the ceramic production process. The similarity threshold refers to a preset value used to determine the magnitude of the change in the similarity difference.

[0046] If the similarity differences at consecutive moments are large and on an upward trend, this indicates that the state of the ceramic body is changing rapidly. This means that during this period, the difference between the shape and other features of the ceramic body and similar samples in the feature library is increasing rapidly. This may be because the user is making significant shape adjustments, such as quickly shaping the rough shape of the body at the beginning of throwing.

[0047] When the state of the ceramic body changes rapidly, it is necessary to capture these changes in time so that the system can quickly determine whether there is a deviation in the process and provide users with accurate operation guidance. Therefore, in this case, the interval of the acquisition equipment can be shortened, the acquisition frequency can be increased, and the changes in the body can be recorded more comprehensively, so that abnormal changes in the body state can be detected in time. The shortening factor refers to the preset shortening factor of the acquisition interval.

[0048] For example, when the similarity differences of a preset number of consecutive adjacent moments are large and show an increasing trend, the collection interval can be shortened to half of the original one.

[0049] If the difference in similarity between adjacent moments is small, or even shows a downward trend, it means that the state of the ceramic body changes slowly or tends to be stable. This indicates that the ceramic body is gradually approaching a stable state similar to a sample in the feature library, and may be in the stage of fine-tuning the body in the later stage of repair.

[0050] When the ceramic body is stable, the body changes slightly in a short period of time. Frequent data collection will generate a large amount of redundant data, increasing the processing burden. Extending the collection interval can reduce unnecessary collection times and reduce data processing volume. The extension multiple refers to the preset multiple of the collection interval.

[0051] For example, when the similarity differences of a continuous preset number of adjacent moments are small and show a downward trend, the collection interval can be extended to 1.5 times of the original one.

[0052] S103, when the process node meets the determination condition, a guide identifier corresponding to the process node is generated according to the difference between the visual feature parameters and the corresponding standard feature parameters, and the basic identifier is updated to the guide identifier.

[0053] The judgment condition refers to the judgment condition for providing more targeted guidance to users. When the progress of the process node of ceramic production reaches or exceeds the threshold, it means that the user needs more targeted operation guidance. At this time, it can be determined that the judgment condition is met. For example, in the ceramic trimming process stage, if it is judged that the progress of the process node of the current ceramic work has reached 70% (assuming the judgment threshold is 60%), then it meets the condition.

[0054] The difference point refers to the contour point in the visual feature parameters that is different from the standard feature parameters. Through the difference point, the gap between the ceramic body currently made by the user and the standard state can be determined. For example, the ceramic body made by the user may be too high compared to the standard state. The guide mark refers to the guide mark used to adjust the error for the user. For example, the guide mark can be a dynamic virtual hand posture image that simulates the correct hand posture. When the height of the ceramic body made by the user is too high, it can be a hand posture of a downward scraping action. This guide mark will directly replace the previous reference mark. The user can see this guidance clearly and intuitively in the VR device, so as to adjust his hand operation according to the hand posture and action displayed by the guide mark, better complete the various stages of ceramic production, and gradually make the ceramic body he makes closer to the standard process effect.

[0055] Based on the above embodiment, the specific implementation of step S103 may be: When the progress threshold of the process node is greater than or equal to the determination threshold, it is determined that the preset determination condition is met.

[0056] Among them, the judgment threshold refers to the threshold for providing more targeted guidance to users. The threshold can be pre-configured. For example, in the ceramic trimming process stage, assuming that the judgment threshold is set to 70%, the server calculates that the progress value of the current process node is 75%, and the preset judgment condition is met. This means that the user has completed most of the work in the trimming operation, and then more precise guidance is needed to complete the remaining part to make the body closer to the standard state. The advantage of setting the judgment conditions in this way is that it can clearly indicate when more detailed guidance is needed for users, avoiding intervening in user operations too early or too late, and ensuring the timeliness and effectiveness of the guidance.

[0057] The feature parameter differences of the contour points corresponding to the visual feature parameters and the corresponding standard feature parameters are compared, and the contour points whose feature parameter differences in the visual feature parameters are not within the standard difference value interval are determined as difference points.

[0058] Specifically, the contour points of the visual feature parameters and the corresponding standard feature parameters can be matched one by one. In practical applications, the feature points of the visual feature parameters and the corresponding standard feature parameters can be extracted first, and then matched through the feature points. Then the height, curvature and other parameter differences of these contour points are calculated, and the contour points that are not in the normal range are taken as difference points. The feature parameter difference refers to the difference between the feature parameters of the corresponding contour points, such as height parameters and curvature parameters. The standard deviation interval refers to the interval corresponding to the normal error range set in advance. When the feature parameter difference exceeds the standard deviation interval, the corresponding contour point is identified as a difference point. These difference points intuitively reflect the position and degree of deviation between the current state of the ceramic body and the standard state.

[0059] For example, when judging whether the contour of a certain part of a ceramic body meets the standard, assuming that the standard requires the curvature of a certain contour to be between 0.8 and 1.2, and the curvature of the corresponding contour point in the actual calculated visual feature parameters is 0.6, this difference is not within the standard difference range, then the contour point is determined as a difference point. In this way, the specific location where the current ceramic body is inconsistent with the standard state can be accurately found.

[0060] In some embodiments, the difference point can be determined by the following steps: Obtain the height difference and curvature difference of the contour points corresponding to the visual feature parameters and the corresponding standard feature parameters, where the feature parameter difference includes the height difference and the curvature difference; determine the contour points in the visual feature parameters whose height difference is not in the corresponding standard difference value interval and / or whose curvature difference is not in the corresponding standard difference value interval as difference points.

[0061] For each set of corresponding contour points, obtain their height values ​​respectively. The height value can be the vertical distance relative to a certain reference surface (such as the table of a wheel throwing machine). Then calculate the difference between the height value of the contour point in the visual feature parameter and the height value of the corresponding contour point in the standard feature parameter. This difference is the height difference. For example, the height of a contour point in the standard feature parameter is 10 cm, and the height of the corresponding contour point in the visual feature parameter is 10.3 cm, then the height difference is 10.3 - 10 = 0.3 cm.

[0062] Curvature is used to describe the degree of curvature of a curve. For the contour curve of a ceramic body, the curvature at the contour point in the visual feature parameters and the curvature at the corresponding contour point in the standard feature parameters are calculated respectively. Mathematical methods (such as calculating the curvature through the derivative of the curve, etc.) can be used to obtain these two curvature values. Then the difference between them is calculated to obtain the curvature difference. For example, the curvature at a contour point in the standard feature parameters is 0.5, and the curvature at the corresponding contour point in the visual feature parameters is 0.4, then the curvature difference is 0.1.

[0063] For height difference and curvature difference, corresponding standard deviation intervals can be pre-set. These intervals can be determined according to the precision requirements of the ceramic manufacturing process. For example, for height difference, the standard deviation interval that may be set is ±0.2 cm, which means that the height difference within this interval is considered to be within the normal manufacturing error range; for curvature difference, the standard deviation interval that may be set is ±0.1, that is, the curvature difference within this range is acceptable.

[0064] The calculated height difference and curvature difference are compared with the corresponding standard deviation interval. If the height difference of a contour point in the visual feature parameter is not within the corresponding standard deviation interval, or the curvature difference is not within the corresponding standard deviation interval, or both are not within the corresponding interval, then this contour point is determined as a difference point.

[0065] It is understandable that the height difference can directly reflect the degree of deviation of the blank from the standard shape in the vertical direction. For example, when making bowls and plates, if the height difference at a certain position exceeds the standard range, it may mean that the depth of the bowls and plates does not meet the requirements. The curvature difference is used to measure the difference between the curvature of the blank contour curve and the standard. Taking a vase as an example, the body of the vase usually has a specific curve shape. The curvature difference can help determine whether the curve of the body meets the design requirements and whether there is a local over-bending or over-straightening situation. Therefore, when determining the difference point, the parameters of these two dimensions can be combined to determine it.

[0066] Determine the feature type of the difference point, traverse the guide actions corresponding to the feature type in the gesture action library, locate the gesture identifier to the actual position corresponding to the difference point, and configure the guide action for the gesture identifier to obtain the guide identifier.

[0067] After determining the difference point, it is necessary to clarify the feature type of the difference point in order to find the appropriate guiding action. The feature type is the adjustment type corresponding to the difference point, which can be classified according to different feature parameters, such as the type related to height, the type related to curvature, etc. The gesture action library is a pre-built database that stores various guiding action information for different feature type difference points.

[0068] Taking the curvature adjustment type as an example, the server will traverse the gesture action library, which pre-stores various guide actions corresponding to different feature types. Assume that there is a guide action for curvature adjustment in the gesture action library, "Use a trimming knife to scrape in a specific direction to adjust the curvature." The system will locate the gesture identifier representing this guide action to the actual position of the difference point through VR technology. For example, if the difference point is located at a certain position on the side of the ceramic body where the curvature does not meet the standard, the gesture identifier will appear precisely near the position and be displayed to the user in an intuitive way. At the same time, the corresponding guide action is configured for the gesture identifier, that is, the animation process of scraping with a trimming knife is simulated, so that the user can clearly understand how to operate to adjust the state of the body at the difference point, and finally generate a complete guide identifier. The guide identifier generated in this way can guide the user's operation in an intuitive and easy-to-understand way, and use the immersion and interactivity of VR technology to help users better complete the fine adjustment in the ceramic production process and improve the production quality and efficiency.

[0069] In some embodiments, the guide identifier may be obtained by the following steps: Determine the difference point corresponding to the height difference as the height adjustment type, determine the difference point corresponding to the curvature difference as the shape adjustment type, determine the difference point corresponding to the height difference and the curvature difference as the comprehensive adjustment type, and the feature types include height adjustment type, shape adjustment type and comprehensive adjustment type; determine the operation direction of the corresponding guidance action according to the positive and negative bias of the feature type, locate the gesture marker to the actual position, and while configuring the guidance action for the gesture marker, determine its action direction as the operation direction to obtain an updated guidance marker.

[0070] Among them, the height adjustment type is a type for adjusting the height, the shape adjustment type is a type for adjusting the shape, and the comprehensive adjustment type is a type for comprehensive adjustment of the height and shape.

[0071] When determining the operation direction of the corresponding guiding action according to the positive and negative bias of the feature type, for the height adjustment type, if the height difference is positive (that is, the actual height is greater than the standard height), the operation direction may be downward adjustment, such as pressing down or scraping; if the height difference is negative (the actual height is less than the standard height), the operation direction may be adding material upward, etc. For the shape adjustment type, if the curvature difference is positive (the actual curvature is greater than the standard curvature), the operation direction may be to make the curve flatter, such as by grinding, etc.; if the curvature difference is negative (the actual curvature is less than the standard curvature), the operation direction may be to increase the curvature of the curve, for example, by pressing the surface of the mud pile with a finger locally to form a depression on the surface of the mud pile, thereby increasing the curvature of the area. For the comprehensive adjustment type, it is necessary to comprehensively consider the positive and negative biases of the height and curvature to determine a suitable operation direction to solve the problems of height and shape at the same time. For example, a gesture action to reduce the height, such as pressing down or scraping, can be performed first to make the height close to the standard value. After the height is adjusted, a gesture operation to reduce the curvature is performed, such as sliding and grinding with the palm of the hand.

[0072] After determining the direction, VR technology can be used to accurately locate the gesture mark representing the guiding action to the actual position corresponding to the difference point. For example, in a VR scene, when a contour point is determined to be a difference point of shape adjustment type and the operation direction is to make the curve smoother, a gesture mark simulating a polishing action will appear at the actual position, and a guiding action consistent with the operation direction will be configured for the gesture mark, intuitively informing the user of the location and action to be operated.

[0073] Through the above method, targeted and operational guidance signs can be generated according to the actual differences in ceramic blanks, providing users with effective guidance and helping them to better complete the ceramic production experience.

[0074] In addition, based on the above embodiment, in the process of step "locating the gesture identifier to the actual position corresponding to the difference point, and configuring a guiding action for the gesture identifier to obtain a guiding identifier", the following embodiments are also included: Based on the current moment, the moving speed of the user's hand motion in the dynamic time period is obtained, and the moving speed is adjusted to the movement speed of the guide mark.

[0075] It can be understood that, in the process of users experiencing ceramic production with the help of VR devices, this step is mainly to match the movement speed of the guide signs with the speed of the user's hand movements, thereby providing guidance that is more in line with the user's operating rhythm.

[0076] Specifically, when the speed of the guide sign needs to be adjusted, a current moment can be first determined as a reference point, and then a dynamic time period is defined forward, which is used to obtain the user's historical speed. Generally speaking, this time period may be between a few seconds and more than ten seconds.

[0077] When obtaining the moving speed of the user's hand motion, the distance the hand moves in unit time can be calculated based on the position coordinates of the hand at different times, thereby obtaining the moving speed, and the moving speed is set as the motion speed of the guide mark.

[0078] In some embodiments, the dynamic time period may be determined by the following steps: Taking the current moment as the benchmark, trace back the benchmark time length to obtain the benchmark time period; obtain the displacement distance of the user's hand in adjacent image frames within the benchmark time period, and obtain the distance difference between the displacement distances corresponding to a preset number of consecutive image frames; when the distance differences are all less than the threshold, add the preset backtracking time length to the benchmark time period to obtain the dynamic time period; when there is a distance difference greater than or equal to the threshold, subtract the preset backtracking time length from the benchmark time period to obtain the dynamic time period.

[0079] Among them, the reference duration is a pre-set duration. The reference time period can be traced back to the corresponding time period through the current moment and the reference moment. Within the reference time period, the displacement distance of the user's hand in adjacent image frames can be obtained. The user's hand movement information can be collected in real time and recorded in the form of image frames. By analyzing the position changes of the user's hand in these image frames, the distance moved by the hand between adjacent image frames is calculated. Then, the distance difference between the displacement distances corresponding to a preset number of consecutive image frames can be obtained. The preset number is a pre-set value. By calculating these distance differences, the changes in the user's hand displacement distance can be understood.

[0080] When the distance difference corresponding to a preset number of consecutive image frames is less than the threshold, it means that the change of the user's hand movement is relatively stable within the reference time period, and the change of the displacement distance is not large. In this case, in order to more comprehensively analyze the speed trend of the user's hand movement, a preset backtracking time can be added to the reference time period to obtain a longer dynamic time period. For example, if the preset backtracking time is 2 seconds, then the original 5-second reference time period will be increased to 7 seconds, and the 7-second backtracking time period is the new dynamic time period. In this way, more data can be obtained to accurately calculate the hand movement speed when the user's hand movement is stable.

[0081] When the distance difference is greater than or equal to the threshold, it indicates that the user's hand motion has changed significantly within the reference time period, and the displacement distance has obvious fluctuations. At this time, in order to focus more on the key part of the user's hand motion change, the preset backtracking time can be subtracted from the reference time period to obtain a shorter dynamic time period. For example, after subtracting the preset backtracking time of 2 seconds, the original 5-second reference time period becomes 3 seconds, and this 3-second time period is the new dynamic time period. By shortening the time period, the speed information of the user's hand motion changes can be captured more accurately, so as to adjust the movement speed of the guide sign in time.

[0082] Through the above steps, the time period used to analyze the hand movement speed can be dynamically adjusted according to the changes in the user's hand movements, so as to more accurately calculate the hand movement speed, provide a more reliable basis for adjusting the movement speed of the guide sign, and enable the movement speed of the guide sign to better match the user's operating rhythm.

[0083] See also Figure 3 , is a schematic diagram of the structure of an intangible cultural heritage content push system based on AI technology provided by an embodiment of the present invention, and the intangible cultural heritage content push system based on AI technology includes: A push module, used for pushing basic display content according to a display device, where the basic display content includes a reference identifier; The process module is used to obtain the visual feature parameters of the intangible cultural heritage products based on the acquisition equipment, and determine the process nodes of the intangible cultural heritage products according to the comparison results between the visual feature parameters and the standard feature parameters of each process stage; The guidance module is used to generate a guidance mark corresponding to the process node according to the difference between the visual feature parameters and the corresponding standard feature parameters when the process node meets the judgment conditions, and update the basic mark to the guidance mark.

[0084] Figure 3 The apparatus of the embodiment shown can be used to perform Figure 2 The implementation principles and technical effects of the steps in the method embodiment shown are similar and will not be repeated here.

[0085] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for pushing intangible cultural heritage content based on AI technology, characterized in that: include: Pushing basic display content according to the display device, the basic display content includes a benchmark identifier; Obtain visual feature parameters of the intangible cultural heritage products based on the acquisition equipment, and determine the process nodes of the intangible cultural heritage products based on the comparison results of the visual feature parameters and the standard feature parameters of each process stage; When the process node meets the judgment condition, a guide identifier corresponding to the process node is generated according to the difference between the visual feature parameters and the corresponding standard feature parameters, and the basic identifier is updated to the guide identifier.

2. The method according to claim 1, characterized in that: The visual feature parameters of the intangible cultural heritage products are obtained based on the acquisition equipment, and the process nodes of the intangible cultural heritage products are determined according to the comparison results between the visual feature parameters and the standard feature parameters of each process stage, including: Identify the outline of the intangible cultural heritage product according to the acquisition device, and the visual feature parameters include the outline of the item; Selecting process stages in the process sequence as target stages in sequence, retrieving standard profiles of the target stages, wherein standard feature parameters include standard profiles; The total similarity values ​​of the object outline and the standard outline under each viewing angle are calculated by fusion, and the progress value corresponding to the total similarity value is determined; The process axis is retrieved to determine the position of the progress value on the process axis, a process node is generated at the position, and when the progress value is greater than the stage threshold, the next target stage is entered.

3. The method according to claim 1, characterized in that The visual feature parameters of the intangible cultural heritage products are obtained through the following steps, including: Identify the breakpoints in the outline of the intangible cultural heritage artifact and calculate the point distances between adjacent breakpoints; The arc adjustment parameters are obtained according to the ratio of the preset constant and the point distance, and the padding lines are generated based on the arc adjustment parameters to connect the adjacent breakpoints, and the visual feature parameters are obtained according to the outline of the object after padding.

4. The method according to claim 1, characterized in that: In the process of obtaining the visual feature parameters of the intangible cultural heritage product based on the acquisition device and determining the process node of the intangible cultural heritage product according to the comparison results of the visual feature parameters and the standard feature parameters of each process stage, the following is also included: Obtain a similarity value sequence of the visual feature parameter and the corresponding standard feature parameter according to the time sequence, and obtain a similarity difference value of adjacent total similarity values ​​in the similarity value sequence; When a preset number of consecutive similarity differences are all less than the similarity difference threshold, the collection time interval of the collection device is increased according to the extension multiple; When a preset number of consecutive similarity difference values ​​are all greater than the similarity difference threshold, the collection time interval of the collection device is reduced according to the shortening factor.

5. The method according to claim 1, characterized in that When the process node meets the judgment condition, a guide mark corresponding to the process node is generated according to the difference between the visual feature parameters and the corresponding standard feature parameters, including: When the progress threshold of the process node is greater than or equal to the determination threshold, it is determined that the preset determination condition is met; Compare the feature parameter differences of the contour points corresponding to the visual feature parameters and the corresponding standard feature parameters, and determine the contour points whose feature parameter differences in the visual feature parameters are not within the standard difference value interval as difference points; Determine the feature type of the difference point, traverse the guide actions corresponding to the feature type in the gesture action library, locate the gesture identifier to the actual position corresponding to the difference point, and configure the guide action for the gesture identifier to obtain the guide identifier.

6. The method according to claim 5, characterized in that Compare the feature parameter difference of the corresponding contour point of the visual feature parameter and the corresponding standard feature parameter, and determine the contour point whose feature parameter difference in the visual feature parameter is not within the standard difference value interval as the difference point, including: Obtaining height difference and curvature difference of contour points corresponding to the visual feature parameters and the corresponding standard feature parameters, wherein the feature parameter difference includes the height difference and the curvature difference; Determine the contour points whose height difference values ​​in the visual feature parameters are not within the corresponding standard deviation value interval and / or whose curvature difference values ​​are not within the corresponding standard deviation value interval as difference points.

7. The method according to claim 6, characterized in that Determine the feature type of the difference point, traverse the guide action corresponding to the feature type in the gesture action library, locate the gesture identifier to the actual position corresponding to the difference point, and configure the guide action for the gesture identifier to obtain the guide identifier, including: Determine the difference point corresponding to the height difference as a height adjustment type, determine the difference point corresponding to the curvature difference as a shape adjustment type, determine the difference point corresponding to the height difference and the curvature difference as a comprehensive adjustment type, and the feature type includes a height adjustment type, a shape adjustment type, and a comprehensive adjustment type; The operation direction of the corresponding guiding action is determined according to the positive and negative deviations of the feature type, the gesture marker is positioned to the actual position, and while configuring the guiding action for the gesture marker, its action direction is determined as the operation direction to obtain an updated guiding marker.

8. The method according to claim 1, characterized in that The process of locating the gesture identifier to the actual position corresponding to the difference point and configuring the guiding action for the gesture identifier to obtain the guiding identifier also includes: Based on the current moment, the moving speed of the user's hand motion in the dynamic time period is obtained, and the moving speed is adjusted to the movement speed of the guide mark.

9. The method according to claim 8, characterized in that The following steps are used to determine the dynamic time period, including: Taking the current time as the benchmark, trace back the benchmark time to get the benchmark time period; Obtain the displacement distance of the user's hand in adjacent image frames within a reference time period, and obtain the distance difference between the displacement distances corresponding to a preset number of consecutive image frames; When the distance differences are all less than the threshold, the preset backtracking time is added to the base time period to obtain the dynamic time period; When there is a distance difference greater than or equal to the threshold, the dynamic time period is obtained by subtracting the preset backtracking time length from the reference time period.

10. An AI-based intangible cultural heritage content push system, characterized in that: include: A push module, used for pushing basic display content according to a display device, where the basic display content includes a reference identifier; The process module is used to obtain the visual feature parameters of the intangible cultural heritage products based on the acquisition equipment, and determine the process nodes of the intangible cultural heritage products according to the comparison results between the visual feature parameters and the standard feature parameters of each process stage; The guidance module is used to generate a guidance mark corresponding to the process node according to the difference between the visual feature parameters and the corresponding standard feature parameters when the process node meets the judgment conditions, and update the basic mark to the guidance mark.

Images