An efficient collection and storage method for cultural relics data

Abstract

The invention relates to the technical field of data storage, and specifically discloses an efficient acquisition and storage method for cultural relic data. The method comprises the following steps: obtaining an image of a cultural relic containing a shooting angle and a description text thereof, converting the description text into a layer, inserting an image as a feature image of the cultural relic with the shooting angle as an index; downsampling each feature image based on the shooting angle and the cultural relic level to obtain an image group as a feature set of the cultural relic with the shooting angle as an index; receiving a backup request sent by a display end, obtaining display end information, selecting an image from the feature set according to the display end information, and sending the image to the display end; the invention obtains images of the cultural relic at different shooting angles, downsamples the images, obtains images of multiple different precisions, and sends images of different precisions according to the number of tourists at the display end, thereby effectively reducing resource consumption, displaying more images under the premise of fixed costs, and improving display completeness.

Description

Technical Field

[0001] The invention relates to the technical field of data storage, and in particular to an efficient collection and storage method for cultural relics data. Background Art

[0002] With the improvement of social level, people's spiritual needs are increasing. Viewing cultural relics is one of the ways to meet spiritual needs. The existing cultural relic viewing process is generally offline. With the popularization of the Internet, online viewing has gradually become a mainstream method.

[0003] The existing online viewing architecture does not actually make any distinction. For some independent small display terminals, if the server performance of the small display terminal is not enough, the display content will be reduced, which will affect the completeness of the display of the small display terminal. In fact, this lack of completeness of the display greatly affects the user's viewing experience. Summary of the invention

[0004] The purpose of the present invention is to provide an efficient collection and storage method for cultural relics data to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] An efficient collection and storage method for cultural relics data, the method comprising:

[0007] Obtain an image of a cultural relic containing a shooting angle and its description text, convert the description text into a layer, insert the image as a feature image of the cultural relic with the shooting angle as an index; the shooting angle is represented by a vector;

[0008] Down-sampling each feature image based on the shooting angle and the level of the cultural relic to obtain an image group as a feature set of the cultural relic indexed by the shooting angle; different images in the same feature set have different clarity;

[0009] Receive a backup request sent by the display end, obtain display end information, select an image from a feature set according to the display end information, and send the image to the display end; wherein at least one image is selected from each feature set;

[0010] Receive demand information fed back by the display end, and update the image selection process according to the demand information; wherein the demand information is used to characterize the demand of visitors at the display end for cultural relics, and the demand includes shooting angle and clarity.

[0011] As a further solution of the present invention, the steps of obtaining an image of a cultural relic containing a shooting angle and a description text thereof, converting the description text into a layer, and inserting the image as a feature image of the cultural relic indexed by the shooting angle include:

[0012] Obtain images of cultural relics with shooting angles and their description texts;

[0013] Remove noise from the description text, and perform word segmentation on the description text after noise removal;

[0014] Count the word segmentation results of all description texts and sort the words according to the preset arrangement rules;

[0015] Determine the assignment step length according to the total number of words counted, and determine the value of each word after sorting based on the assignment step length; wherein the value of each word is determined by three parameters, and the value range of the three parameters is 0 to 255;

[0016] Each description text is transformed based on the value of each word to obtain a layer, which is inserted into the image as a feature image of the cultural relic indexed by the shooting angle.

[0017] As a further solution of the present invention: the step of converting each description text based on the value of each word, obtaining a layer, inserting an image, and using the image as a feature image of the cultural relic indexed by the shooting angle includes:

[0018] Read the word segmentation results of the description text after noise removal, and query the value of each word in turn according to the order of the description text;

[0019] Create a layer using the central interpolation method. When the layer is created, insert the layer into the image to obtain a feature image.

[0020] The shooting angle of the image corresponding to the description text is read as the index of the feature image.

[0021] As a further solution of the present invention: the central interpolation method is:

[0022] With the preset origin as the center, the pixels are numbered in a clockwise order around the origin;

[0023] When interpolation is needed, query the word number of the current word in the text after word segmentation to determine the word order;

[0024] Match the corresponding number according to the word order and insert the value into the corresponding number.

[0025] As a further solution of the present invention, the step of downsampling each feature image based on the shooting angle and the level of cultural relics to obtain an image group as a feature set of cultural relics indexed by the shooting angle includes:

[0026] For each cultural relic, check the cultural relic level in the preset cultural relic registration table;

[0027] Read the characteristic image of the cultural relic and determine the downsampling step size according to the shooting angle and the level of the cultural relic;

[0028] Downsampling the feature image based on the downsampling step size and the preset top-level image size to obtain an image group as a feature set of the cultural relics indexed by the shooting angle;

[0029] Among them, the process of determining the downsampling step size is:

[0030] Where D is the downsampling step size, α and β are the preset correction coefficients, To find the minimum value of the modulus of the vector difference between the shooting angle and the normal vector of the standard six views; Indicates the shooting angle. Represents the normal vector of the standard six views, there are six in total; l is the cultural relic level.

[0031] As a further solution of the present invention: the steps of receiving the backup request sent by the display end, obtaining the display end information, selecting an image from the feature set according to the display end information, and sending the image to the display end include:

[0032] Receive the backup request sent by the display end, send an information query request to the display end, and obtain the number of visitors fed back by the display end;

[0033] Determining a minimum image clarity based on said number of visitors;

[0034] Matching a reference image in each feature set based on the minimum image definition, randomly selecting an image from images whose definition is greater than that of the reference image, and sending the image to the display end;

[0035] The minimum image clarity is proportional to the number of tourists.

[0036] As a further solution of the present invention: the step of receiving demand information fed back by the display end and updating the image selection process according to the demand information includes:

[0037] Send a request for obtaining evaluation information to the display end, and receive the ratings of different images fed back by the display end;

[0038] When the score is greater than a preset first threshold, querying a feature set corresponding to the image, and improving the minimum image clarity when selecting an image from the feature set;

[0039] When the score is less than a preset second threshold, a feature set corresponding to the image is queried to increase the number of images selected when selecting images from the feature set.

[0040] As a further solution of the present invention: the method further comprises:

[0041] When sending images to any display end, both the sending end and the receiving end obtain transmission parameters and transmission progress in real time;

[0042] When the transmission progress reaches a preset first progress threshold, pausing the transmission process, determining an encryption code according to the transmission parameter, encrypting the image based on the encryption code, and resuming the transmission process;

[0043] When the user terminal receives the encrypted target file, a decryption code is determined based on the transmission parameter and the transmission progress, and the encrypted target file is decrypted based on the decryption code.

[0044] As a further solution of the present invention: the process in which both the transmitting end and the receiving end obtain the transmission parameters and the transmission progress in real time includes:

[0045] Get the network bandwidth of the connection channel;

[0046] Calculate the predicted transmission speed according to the size of the selected image and the network bandwidth;

[0047] Determine a sampling node according to the predicted transmission speed and the preset number of sampling point bits; the sampling node has a label;

[0048] Acquire transmission parameters in real time based on the sampling nodes, sort the transmission parameters according to the labels, and generate a parameter table; the parameter table contains label items and transmission parameter items;

[0049] Generate a transmission progress item according to the numbered item in the parameter table.

[0050] As a further solution of the present invention: the step of determining the encryption code according to the transmission parameter includes:

[0051] Reading a parameter table, and converting the parameter table into an encrypted parameter table in sequence according to a preset encryption matrix library;

[0052] Generate a fitting curve group according to the encrypted parameter table;

[0053] Converting the fitting curve group into a curve image, performing feature extraction on the curve image according to a preset extraction index to obtain a feature region; the extraction index is a function of time;

[0054] The feature region is input into a trained encryption code generation model to obtain an encryption code.

[0055] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention obtains images of cultural relics at different shooting angles, and downsamples the images to obtain images of multiple different accuracies, and sends images of different accuracies according to the number of tourists at the display end, which can effectively reduce resource consumption, so that more images can be displayed under the premise of fixed costs, thereby improving the completeness of the display. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention.

[0057] Figure 1 This is a flowchart of an efficient collection and storage method for cultural relics data. DETAILED DESCRIPTION

[0058] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0059] Figure 1 The flowchart of the method for efficiently collecting and storing cultural relics data is as follows. In an embodiment of the present invention, a method for efficiently collecting and storing cultural relics data includes:

[0060] Step S100: obtaining an image of a cultural relic containing a shooting angle and its description text, converting the description text into a layer, inserting the image as a feature image of the cultural relic with the shooting angle as an index; the shooting angle is represented by a vector;

[0061] A cultural relic refers to a certain cultural relic. The same operation needs to be performed for each cultural relic that needs to be displayed; an image of the cultural relic and a description text are obtained. When obtaining the image, the shooting angle needs to be recorded. The shooting angle, as the name implies, indicates the angle from which the cultural relic is photographed. The shooting angle is represented by a vector.

[0062] Regarding the shooting angle, you need to first determine a Cartesian three-dimensional coordinate system. At this time, each face can be represented by a normal vector (three-dimensional vector). When taking an image, the center line of the lens can also be represented by a three-dimensional vector. This vector is the shooting angle.

[0063] Furthermore, each cultural relic generally has the most basic introduction information. If not, a default text will be used, such as: "This cultural relic is not described". Under the framework of the technical solution of the present invention, the descriptive text is converted into a layer and inserted into the image. Each image taken from each shooting angle corresponds to a descriptive text, and the image taken with the layer is called a feature image.

[0064] Step S200: downsampling each feature image based on the shooting angle and the level of the cultural relic to obtain an image group as a feature set of the cultural relic indexed by the shooting angle; different images in the same feature set have different clarity;

[0065] From the above content, we can know that the shooting angle indicates the direction from which the cultural relics are photographed. There are actually many shooting angles, and the importance of different shooting angles is actually different. For example, the importance of the image corresponding to the standard six views (rarely upward views) will be higher. Combined with the level of the cultural relics, the importance of each feature image can be obtained. The closer the shooting angle is to the standard six views, the higher the importance, and the higher the level of the cultural relic, the higher the importance.

[0066] Based on the shooting angle and the level of cultural relics, each feature image is downsampled to obtain an image group. The downsampling process is a process of simplifying the image. The larger the downsampling step, the fewer images in the image group and the smaller the selection space. When sending to the display end, the images under the shooting angle can only be selected within a smaller range.

[0067] The downsampled image group is called a feature set, and the feature set is indexed by the shooting angle.

[0068] Step S300: receiving a backup request sent by the display end, obtaining display end information, selecting an image from a feature set according to the display end information, and sending the image to the display end; wherein at least one image is selected from each feature set;

[0069] The display end is the port where cultural relics need to be displayed. It can be understood as a display site permitted by the cultural relics management party. The display end needs to obtain images from the cultural relics management party and then display them online. For the display end, there is actually no need to display overly clear images, especially in the online display link, where a large number of cultural relics can be displayed. If the images of these cultural relics are simplified, it can greatly alleviate resource pressure.

[0070] Specifically, when a backup request is received from the display end, the display end information is obtained. The display end information mainly includes the number of tourists the display end faces. The more tourists there are, the higher the clarity requirement is. Images are selected based on the features of all shooting angles corresponding to the cultural relics in the display end information and sent to the display end.

[0071] Since the feature set has been downsampled to different degrees according to the importance of the image, the more important the image, the larger the selected space.

[0072] Step S400: receiving demand information fed back by the display end, and updating the image selection process according to the demand information; wherein the demand information is used to characterize the demand of visitors at the display end for cultural relics, and the demand includes shooting angle and clarity;

[0073] Furthermore, the demand information fed back by the display end is received, and the image selection process is updated according to the demand information. The updated content includes the number of selected images and image clarity. The image clarity is determined by the degree of downsampling. The more downsampling times, the lower the image clarity. Among them, the demand information fed back by the display end is actually the demand information uploaded by the customers of the display end.

[0074] As a preferred embodiment of the technical solution of the present invention, the steps of obtaining an image of a cultural relic containing a shooting angle and its description text, converting the description text into a layer, and inserting the image as a feature image of the cultural relic indexed by the shooting angle include:

[0075] Obtain images of cultural relics with shooting angles and their description texts;

[0076] Remove noise from the description text, and perform word segmentation on the description text after noise removal;

[0077] Count the word segmentation results of all description texts and sort the words according to the preset arrangement rules;

[0078] Determine the assignment step length according to the total number of words counted, and determine the value of each word after sorting based on the assignment step length; wherein the value of each word is determined by three parameters, and the value range of the three parameters is 0 to 255;

[0079] Each description text is transformed based on the value of each word to obtain a layer, which is inserted into the image as a feature image of the cultural relic indexed by the shooting angle.

[0080] In an example of the technical solution of the present invention, the generation process of the feature image is defined as follows:

[0081] Obtain images of cultural relics with shooting angles and their description texts. For this application, these are existing data, and the acquisition process is not complicated. The description texts are subjected to noise removal, and the description texts after noise removal are segmented. The noise removal process is to delete useless symbols or characters in the text, such as punctuation marks and numbers. The description texts are segmented using existing part-of-speech recognition schemes, as long as the text can be converted into a set of words.

[0082] The word segmentation results of all description texts are counted, and the words are sorted according to the preset arrangement rules. It should be noted that all description texts are description texts of images of all cultural relics taken from all shooting angles. The purpose of this process is to count which words may appear in the description texts. In order to make these words more orderly, they can be sorted by the strokes of the first character, that is, the above arrangement rules.

[0083] The assignment step is determined according to the total number of words counted, and the value of each word after sorting is determined based on the assignment step. The purpose of this step is to convert words into color values ​​of pixels. Color values ​​have three parameters, such as RGB values. The value range of each parameter is 0 to 255, which means there are 256*256*256 cases, which is much larger than the total number of words and is sufficient to represent different words. The assignment step is obtained by dividing 256*256*256 by the total number of words, that is, the difference between the color values ​​corresponding to adjacent words. The difference between adjacent words is only the difference of one parameter, and the other two parameters are most likely the same. This process is not complicated in principle. It is to evenly distribute 256*256*256 cases to all words so that the color values ​​corresponding to any two words have the largest possible difference.

[0084] Finally, when each word corresponds to a color value, each description text is converted based on the value of each word to obtain a layer. After the layer is inserted into the image, it serves as a feature image of the cultural relic indexed by the shooting angle.

[0085] As a preferred embodiment of the technical solution of the present invention, the step of converting each description text based on the value of each word, obtaining a layer, inserting an image as a feature image of the cultural relic indexed by the shooting angle includes:

[0086] Read the word segmentation results of the description text after noise removal, and query the value of each word in turn according to the order of the description text;

[0087] Create a layer using the central interpolation method. When the layer is created, insert the layer into the image to obtain a feature image.

[0088] The shooting angle of the image corresponding to the description text is read as the index of the feature image.

[0089] In an example of the technical solution of the present invention, a specific description is given of the layer generation process. The word segmentation result of the description text after noise removal is read, and the value of each word is queried in turn according to the writing order of the description text. The writing order is the order in which the words in the text are arranged. The layer is created according to the central interpolation method. When the layer is created, the layer is inserted into the image to obtain a feature image; the shooting angle is inserted into the feature image as an index.

[0090] Specifically, the central interpolation method is:

[0091] With the preset origin as the center, the pixels are numbered in a clockwise order around the origin;

[0092] When interpolation is needed, query the word number of the current word in the text after word segmentation to determine the word order;

[0093] Match the corresponding number according to the word order and insert the value into the corresponding number.

[0094] As a preferred embodiment of the technical solution of the present invention, the step of downsampling each feature image based on the shooting angle and the cultural relic level to obtain an image group as a feature set of cultural relics indexed by the shooting angle includes:

[0095] For each cultural relic, check the cultural relic level in the preset cultural relic registration table;

[0096] Read the characteristic image of the cultural relic and determine the downsampling step size according to the shooting angle and the level of the cultural relic;

[0097] Downsampling the feature image based on the downsampling step size and the preset top-level image size to obtain an image group as a feature set of the cultural relics indexed by the shooting angle;

[0098] Among them, the process of determining the downsampling step size is:

[0099] Where D is the downsampling step size, α and β are the preset correction coefficients, To find the minimum value of the modulus of the vector difference between the shooting angle and the normal vector of the standard six views; Indicates the shooting angle. Represents the normal vector of the standard six views, there are six in total; l is the cultural relic level.

[0100] In an example of the technical solution of the present invention, the downsampling process is limited. For each cultural relic, the cultural relic level is queried in a preset cultural relic registration table, assuming that the more important the cultural relic is, the higher the cultural relic level is; the characteristic image of the cultural relic is read, each document corresponds to many characteristic images (there is a characteristic image at each shooting angle), and the downsampling step is determined according to the shooting angle and the cultural relic level. The closer the shooting angle is to the standard six-view, the more important the corresponding image is, the smaller the downsampling step is, and the larger the image selection space is when it is sent to the display end; the higher the cultural relic level is, the more important the corresponding image is, the smaller the downsampling step is, and the larger the image selection space is when it is sent to the display end.

[0101] The feature image is downsampled based on the downsampling step size and the preset top-level image size to obtain an image group, which serves as a feature set of the cultural relics indexed by the shooting angle; finally, each feature image will correspond to multiple downsampled images, which are retained in the form of an image group.

[0102] Among them, the top-level image size is a preset size to prevent the image from being too small. Because the downsampling process continuously deletes rows and columns of the image, the image will become smaller and smaller. An image that is too small, such as one with only a single-digit number of pixels, has no display meaning. Therefore, there needs to be a downsampling boundary, that is, the above-mentioned top-level image size.

[0103] As a preferred embodiment of the technical solution of the present invention, the steps of receiving a backup request sent by a display terminal, obtaining display terminal information, selecting an image from a feature set according to the display terminal information, and sending the image to the display terminal include:

[0104] Receive the backup request sent by the display end, send an information query request to the display end, and obtain the number of visitors fed back by the display end;

[0105] Determining a minimum image clarity based on said number of visitors;

[0106] Matching a reference image in each feature set based on the minimum image definition, randomly selecting an image from images whose definition is greater than that of the reference image, and sending the image to the display end;

[0107] The minimum image clarity is proportional to the number of tourists.

[0108] Receive a backup request sent by the display end, send an information query request to the display end, and obtain the number of tourists fed back by the display end. In this process, the display end information is limited to the number of tourists, and the minimum image clarity is determined according to the number of tourists. Then, based on the minimum image clarity, match the benchmark image in each feature set, randomly select images from images with a clarity greater than that of the benchmark image, and send them to the display end. The meaning of this process is that after the minimum clarity is determined, randomly select images within a range not lower than the minimum clarity, and the number of selected images is at least one and sent to the display end.

[0109] It should be noted that the more visitors there are, the higher the exposure of the corresponding display terminal. At this time, the minimum image clarity should be larger, making the randomly selected image clearer.

[0110] As a preferred embodiment of the technical solution of the present invention, the step of receiving demand information fed back by the display end and updating the image selection process according to the demand information includes:

[0111] Send a request for obtaining evaluation information to the display end, and receive the ratings of different images fed back by the display end;

[0112] When the score is greater than a preset first threshold, querying a feature set corresponding to the image, and improving the minimum image clarity when selecting an image from the feature set;

[0113] When the score is less than a preset second threshold, a feature set corresponding to the image is queried to increase the number of images selected when selecting images from the feature set.

[0114] Regarding the update phase of the image selection process, a request for obtaining evaluation information is sent to the display end, and the ratings of different images fed back by the display end are received. The ratings are actually the ratings of each image given by the visitors to the display end. When the ratings are greater than a preset first threshold, it means that the image is more popular with visitors. At this time, the feature set corresponding to the image is queried to improve the minimum image clarity when selecting an image from the feature set, so that the image selected next time is clearer.

[0115] When the score is less than the preset second threshold, it means that the tourist does not like the image, perhaps because he does not like the shooting angle or is not satisfied with the clarity. At this time, it is necessary to select several more images at this shooting angle to try to satisfy the tourist. The specific operation process is to query the feature set corresponding to the image and increase the number of images selected when selecting images in the feature set.

[0116] As a preferred embodiment of the technical solution of the present invention, the method further includes:

[0117] When sending images to any display end, both the sending end and the receiving end obtain transmission parameters and transmission progress in real time;

[0118] When the transmission progress reaches a preset first progress threshold, pausing the transmission process, determining an encryption code according to the transmission parameter, encrypting the image based on the encryption code, and resuming the transmission process;

[0119] When the user terminal receives the encrypted target file, a decryption code is determined based on the transmission parameter and the transmission progress, and the encrypted target file is decrypted based on the decryption code.

[0120] The cultural relics data needs to be encrypted in the process of being sent to the display end. When the transmission progress reaches a certain level, such as 90%, the file is encrypted according to the file transmission parameters. This encryption process is based on the transmission parameters. The transmission parameters can only be obtained by the server and the user terminal, and the security is extremely high. The encryption process and the decryption process are inverse processes of each other. Under the same technical solution, the two should be designed at the same time.

[0121] Furthermore, the process in which both the sending end and the receiving end obtain the transmission parameters and the transmission progress in real time includes:

[0122] Get the network bandwidth of the connection channel;

[0123] Calculate the predicted transmission speed according to the size of the selected image and the network bandwidth;

[0124] Determine a sampling node according to the predicted transmission speed and the preset number of sampling point bits; the sampling node has a label;

[0125] Acquire transmission parameters in real time based on the sampling nodes, sort the transmission parameters according to the labels, and generate a parameter table; the parameter table contains label items and transmission parameter items;

[0126] Generate a transmission progress item according to the numbered item in the parameter table.

[0127] The transmission process changes in real time. The smaller the sampling granularity, the greater the demand for computing resources. Therefore, it is necessary to specifically design the sampling granularity. The specific solution is: the predicted transmission duration can be determined based on the size of the image to be transmitted and the average network transmission speed (network bandwidth of the connection channel), and then the predicted transmission duration can be divided according to the number of preset sampling points to determine the sampling nodes.

[0128] Specifically, the step of determining the encryption code according to the transmission parameter includes:

[0129] Reading a parameter table, and converting the parameter table into an encrypted parameter table in sequence according to a preset encryption matrix library;

[0130] Generate a fitting curve group according to the encrypted parameter table;

[0131] Converting the fitting curve group into a curve image, performing feature extraction on the curve image according to a preset extraction index to obtain a feature region; the extraction index is a function of time;

[0132] The feature region is input into a trained encryption code generation model to obtain an encryption code.

[0133] The above content specifically limits the generation process of the encryption code. First, the parameter table containing the transmission parameters and transmission progress is read, and then the parameter table is encrypted. The encryption process is to convert the data in the parameter table through a preset encryption matrix; finally, a fitting curve group is generated according to the converted parameter table; wherein the fitting curve group contains characteristic information.

[0134] Then, the fitting curve group is converted into a curve image. With the help of image processing technology, certain areas can be extracted, and the extraction method (extraction index) is pre-set; then, an encryption code can be generated based on these feature areas. This process includes multiple steps, and the calculation amount of each step is not large. Multiple steps ensure security.

[0135] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An efficient collection and storage method for cultural relics data, characterized in that: The method comprises: Obtain an image of a cultural relic containing a shooting angle and its description text, convert the description text into a layer, insert the image as a feature image of the cultural relic with the shooting angle as an index; the shooting angle is represented by a vector; Down-sampling each feature image based on the shooting angle and the level of the cultural relic to obtain an image group as a feature set of the cultural relic indexed by the shooting angle; different images in the same feature set have different clarity; Receive a backup request sent by the display end, obtain display end information, select an image from a feature set according to the display end information, and send the image to the display end; wherein at least one image is selected from each feature set; Receive demand information fed back by the display end, and update the image selection process according to the demand information; wherein the demand information is used to characterize the demand of visitors at the display end for cultural relics, including shooting angle and clarity; The steps of obtaining an image of a cultural relic containing a shooting angle and a description text thereof, converting the description text into a layer, and inserting the image as a feature image of the cultural relic with the shooting angle as an index include: Obtain images of cultural relics with shooting angles and their description texts; Remove noise from the description text, and perform word segmentation on the description text after noise removal; Count the word segmentation results of all description texts and sort the words according to the preset arrangement rules; Determine the assignment step length according to the total number of words counted, and determine the value of each word after sorting based on the assignment step length; wherein the value of each word is determined by three parameters, and the value range of the three parameters is 0 to 255; Each description text is converted based on the value of each word to obtain a layer, and an image is inserted as a feature image of the cultural relic indexed by the shooting angle; The steps of converting each description text based on the value of each word, obtaining a layer, inserting an image as a feature image of the cultural relic indexed by the shooting angle include: Read the word segmentation results of the description text after noise removal, and query the value of each word in turn according to the order of the description text; Create a layer using the central interpolation method. When the layer is created, insert the layer into the image to obtain a feature image. The shooting angle of the image corresponding to the description text is read as the index of the feature image.

Images