A method and device for three-dimensional data acquisition of cultural relics

Abstract

The present invention discloses a method, device, equipment and storage medium for collecting three-dimensional data of cultural relics, relating to the technical field of data processing. The main purpose is to solve the problem of damage to cultural relics caused by the removal of covering objects due to artificially determined coverage areas. The method includes: obtaining three-dimensional imaging data of a target cultural relic by means of three-dimensional scanning technology, and marking the three-dimensional area of the covering object in the three-dimensional imaging data; determining at least one covering feature object matching the three-dimensional area based on the input soil quality parameters, cultural relic material parameters, and excavation time; retrieving the morphological characteristics of the cultural relic matching the excavation time, and predicting the hidden object information in the three-dimensional area according to the morphological characteristics of the cultural relic and the covering feature object; constructing three-dimensional data of the cultural relic matching the three-dimensional imaging data and the hidden object information, and outputting the same.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to a method and device for collecting three-dimensional data of cultural relics. Background Art

[0002] In order to trace historical traces, removing foreign objects from unearthed cultural relics has become an important step in cultural relic identification. In particular, newly unearthed cultural relics are often covered with some corrosives or derivatives other than soil. In order to accurately collect the three-dimensional data of cultural relics, it is usually necessary to remove these corrosives or derivatives.

[0003] Currently, it is usually based on manual removal of these covers. And because the specific shape of the covered cultural relics is unknown, it is usually necessary to manually and gently remove layer by layer for a long time, and then perform a complete collection of the three-dimensional data of the cultural relics for restoration and other work. However, due to the large coverage area of the covers, it is impossible to accurately determine the manual removal force, and it is impossible to accurately determine whether the covered parts that belong to the cultural relics will be removed, resulting in premature damage to the cultural relics. Therefore, there is an urgent need for a method for collecting three-dimensional data of cultural relics to solve the above problems. Summary of the Invention

[0004] In view of this, the present invention provides a method and device for collecting three-dimensional data of cultural relics, mainly aiming at the problem that currently, due to the large coverage area of the covers, it is impossible to accurately determine the manual removal force, and it is impossible to accurately determine whether the covered parts that belong to the cultural relics will be removed, resulting in premature damage to the cultural relics.

[0005] According to one aspect of the present invention, a method for collecting three-dimensional data of cultural relics is provided, including:

[0006] Obtaining three-dimensional imaging data obtained by scanning a target cultural relic using three-dimensional scanning technology, and marking the three-dimensional area of the cover in the three-dimensional imaging data, where the target cultural relic is a cultural relic without the cover removed;

[0007] Determining at least one covering feature object that matches the three-dimensional area based on the input soil quality parameters, cultural relic material parameters, and excavation time, where the covering feature object is used to represent the possible object covering the cover;

[0008] Retrieving the morphological features of the cultural relics that match the excavation time, and predicting the hidden object information in the three-dimensional area according to the morphological features of the cultural relics and the covering feature object, where the hidden object information at least includes one of the hidden object morphological possibility information, hidden object material possibility information, and hidden object connection structure possibility information;

[0009] Constructing the three-dimensional data of the cultural relics that matches the three-dimensional imaging data and the hidden object information, and outputting the same.

[0010] Further, the three-dimensional region of the covering in the three-dimensional imaging data is marked as follows:

[0011] Analyze the point cloud data in the three-dimensional imaging data, classify the point cloud data according to a three-dimensional surface, and render and output the classified point cloud data;

[0012] Correct the point cloud data based on the position correction coordinates of the point cloud data, and mark the corrected point cloud data as the three-dimensional region of the covering.

[0013] Further, before determining at least one covering feature object that matches the three-dimensional region based on the input soil parameters, cultural relic material parameters, and excavation time, the method further includes:

[0014] Based on the cultural relic information in the historical cultural relic resource database, establish a mapping relationship between different soil parameters, cultural relic material parameters, excavation time and different covering feature objects, and bind the authentication probability matching the covering feature object with the mapping relationship;

[0015] The determining at least one covering feature object that matches the three-dimensional region based on the input soil parameters, cultural relic material parameters, and excavation time includes:

[0016] Extract the covering feature objects that match the input soil parameters, cultural relic material parameters, and excavation time according to the mapping relationship, and determine the covering feature objects of the three-dimensional region by taking the covering features corresponding to the authentication probability greater than the preset authentication threshold.

[0017] Further, before retrieving the cultural relic morphological features that match the excavation time and predicting the hidden object information in the three-dimensional region according to the cultural relic morphological features and the covering feature objects, the method further includes:

[0018] Construct a multi-input single-output convolutional neural network, and construct a cultural relic covering morphology training sample set based on the historical cultural relic resource database. The cultural relic covering morphology training sample set contains the hidden object marks and hidden object coordinates corresponding to different cultural relic morphological features and covering feature objects;

[0019] Train the convolutional neural network based on the cultural relic covering morphology training sample set to obtain a trained convolutional neural network model. The network weight values in the convolutional neural network model are determined based on the authentication probability, and the convolutional neural network model is a three-layer network model.

[0020] Further, the retrieving the cultural relic morphological features that match the excavation time includes:

[0021] Obtain the cultural relic information that matches the unearthed time from the historical cultural relic resource database, and compare the cultural relic information according to the morphological classification categories to determine the morphological characteristics of the cultural relics;

[0022] The prediction of the hidden object information in the three-dimensional area according to the morphological characteristics of the cultural relics and the covered feature object includes:

[0023] Perform prediction processing on the morphological characteristics of the cultural relics and the covered feature object based on the trained convolutional neural network model to obtain the hidden object information in the three-dimensional area.

[0024] Furthermore, the construction of the three-dimensional cultural relic data that matches the three-dimensional imaging data and the hidden object information and the output thereof include:

[0025] Render the hidden areas of different partitions in the three-dimensional imaging data based on the hidden object information, and configure the hidden areas in the three-dimensional imaging data in an embedded manner to generate three-dimensional cultural relic data.

[0026] According to another aspect of the present invention, a three-dimensional cultural relic data acquisition device is provided, including:

[0027] An acquisition module, configured to acquire three-dimensional imaging data obtained by scanning a target cultural relic by three-dimensional scanning;

[0028] A marking module, configured to mark the three-dimensional area of the covering object in the three-dimensional imaging data, where the target cultural relic is a cultural relic without the covering object removed;

[0029] A determination module, configured to determine at least one covered feature object that matches the three-dimensional area based on the input soil quality parameters, cultural relic material parameters, and unearthed time, where the covered feature object is used to represent the possible object of the covering object with covering features;

[0030] A prediction module, configured to retrieve the morphological characteristics of the cultural relics that match the unearthed time, and predict the hidden object information in the three-dimensional area according to the morphological characteristics of the cultural relics and the covered feature object, where the hidden object information includes hidden object morphological possibility information, hidden object material possibility information, and hidden object connection structure possibility information;

[0031] An output module, configured to construct three-dimensional cultural relic data that matches the three-dimensional imaging data and the hidden object information and perform output.

[0032] Furthermore, the marking module includes:

[0033] A classification unit, configured to analyze the point cloud data in the three-dimensional imaging data, classify the point cloud data according to three-dimensional surfaces, and render and output the classified point cloud data;

[0034] A marking unit, configured to correct the point cloud data based on the position correction coordinates of the point cloud data, and mark the corrected point cloud data as a three-dimensional area of the covering object.

[0035] Further, the apparatus further includes: a building module,

[0036] The building module is configured to establish a mapping relationship between different soil property parameters, cultural relic material parameters, excavation time and different covering feature objects based on the cultural relic information in the historical cultural relic resource database, and bind the authentication probability matching the covering feature object with the mapping relationship;

[0037] The determining module is specifically configured to extract the covering feature object matching the input soil property parameters, cultural relic material parameters, and excavation time according to the mapping relationship, and determine the covering feature object corresponding to the three-dimensional area whose authentication probability is greater than a preset authentication threshold.

[0038] Further, the apparatus further includes:

[0039] A construction module, configured to construct a multi-input single-output convolutional neural network, and construct a cultural relic covering form training sample set based on the historical cultural relic resource database, where the cultural relic covering form training sample set includes hidden object marks and hidden object coordinates corresponding to different cultural relic form features and covering feature objects;

[0040] A training module, configured to perform model training on the convolutional neural network based on the cultural relic covering form training sample set to obtain a trained convolutional neural network model, where the network weight value in the convolutional neural network model is determined based on the authentication probability, and the convolutional neural network model is a three-layer network model.

[0041] Further,

[0042] The prediction module is specifically configured to obtain the cultural relic information matching the excavation time from the historical cultural relic resource database, and compare the cultural relic information according to the form classification category to determine the cultural relic form features;

[0043] The prediction module is further specifically configured to perform prediction processing on the cultural relic form features and the covering feature object based on the trained convolutional neural network model to obtain the hidden object information in the three-dimensional area.

[0044] Further, the output module is specifically configured to render the hidden areas of different partitions in the three-dimensional imaging data based on the hidden object information, and configure the hidden areas in an embedded manner in the three-dimensional imaging data to generate three-dimensional cultural relic data.

[0045] According to another aspect of the present invention, there is provided a storage medium storing at least one executable instruction that causes a processor to perform operations corresponding to the above-mentioned three-dimensional data acquisition method for cultural relics.

[0046] According to still another aspect of the present invention, there is provided a terminal, comprising: a processor, a memory, a communication interface, and a communication bus, where the processor, the memory, and the communication interface complete communication with each other through the communication bus;

[0047] The memory is used to store at least one executable instruction that causes the processor to perform operations corresponding to the above-mentioned three-dimensional data acquisition method for cultural relics.

[0048] By means of the above technical solutions, the technical solutions provided by the embodiments of the present invention have at least the following advantages:

[0049] The present invention provides a method and device for three-dimensional data acquisition of cultural relics. First, three-dimensional imaging data obtained by scanning a target cultural relic using three-dimensional scanning technology is acquired, and the three-dimensional area of the covering on the three-dimensional imaging data is marked. The target cultural relic is a cultural relic without the covering removed. Then, at least one covering feature object matching the three-dimensional area is determined based on the input soil parameters, cultural relic material parameters, and excavation time. The covering feature object is used to represent a possible object covering the covering. Then, the morphological features of the cultural relic matching the excavation time are retrieved, and the hidden object information in the three-dimensional area is predicted according to the cultural relic morphological features and the covering feature object. The hidden object information includes at least one of the hidden object morphological possibility information, the hidden object material possibility information, and the hidden object connection structure possibility information. Finally, the three-dimensional data of the cultural relics matching the three-dimensional imaging data and the hidden object information is constructed and output. Compared with the prior art, the embodiments of the present invention acquire the three-dimensional imaging data of the cultural relics by scanning and mark the covering area, and then perform multi-level screening and prediction on the covering feature object based on the cultural relic-related parameters and the cultural relic morphological features corresponding to the excavation time of the cultural relics, and output the three-dimensional imaging data containing hidden information obtained by prediction, which can provide visual information of the covering area of the target cultural relic for cultural relic workers, avoid the problem of damage to cultural relics caused by removing the covering when cultural relic workers do not know the morphology of the covering area, and improve the feasibility of removing the covering of cultural relics and the integrity of cultural relics.

[0050] The above description is only an overview of the technical solutions of the present invention. In order to be able to understand the technical means of the present invention more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features, and advantages of the present invention more obvious and understandable, the following specifically describes the embodiments of the present invention. Brief Description of the Drawings

[0051] By reading the following detailed description of the preferred embodiments, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the preferred embodiments and are not considered to be a limitation of the present invention. Moreover, throughout the drawings, the same reference numerals are used to represent the same components. In the drawings:

[0052] Figure 1 Shows a flowchart of a method for collecting three-dimensional data of cultural relics provided by an embodiment of the present invention;

[0053] Figure 2 Shows a flowchart of another method for collecting three-dimensional data of cultural relics provided by an embodiment of the present invention;

[0054] Figure 3 Shows a block diagram of a device for collecting three-dimensional data of cultural relics provided by an embodiment of the present invention;

[0055] Figure 4 Shows a block diagram of another device for collecting three-dimensional data of cultural relics provided by an embodiment of the present invention;

[0056] Figure 5 Shows a schematic structural diagram of a terminal provided by an embodiment of the present invention. Specific embodiments

[0057] Hereinafter, exemplary embodiments of the present disclosure will be described in more detail with reference to the drawings. Although the exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be fully conveyed to those skilled in the art.

[0058] An embodiment of the present invention provides a method for collecting three-dimensional data of cultural relics, as Figure 1 shown, the method includes:

[0059] 101. Obtain three-dimensional imaging data obtained by scanning a target cultural relic using three-dimensional scanning technology, and mark the three-dimensional area of the covering object in the three-dimensional imaging data.

[0060] In an embodiment of the present invention, the execution entity may be an intelligent management system with a scanning function, such as an intelligent recognition system, a digital scanning system, etc. Exemplarily, when the execution entity is an intelligent recognition system, the target cultural relic is a cultural relic without the removal of the covering. The system can scan the target cultural relic based on three-dimensional technology. Specifically, when a laser is projected onto the target cultural relic and the three-dimensional light plane formed by the projected linear laser intersects with the target cultural relic, a bright scanning line will be formed on the surface of the target cultural relic, and all the intersection points of the target cultural relic and the projected three-dimensional light plane are included in this scanning line. Therefore, the three-dimensional imaging data of the surface points of the target cultural relic can be obtained according to the scanning line coordinates.

[0061] It should be noted that the three-dimensional imaging data in the embodiment of the present invention is recorded after the intelligent recognition system, which is the current execution entity, generates the three-dimensional imaging data of the target cultural relic based on three-dimensional imaging scanning by computer software. Then, based on data features such as the coplanarity in space and the three-dimensional space coordinate system of the three-dimensional imaging data, the deformed area of the target cultural relic can be identified. When there is a covering on the surface of the target cultural relic, the area with the covering can be regarded as a deformed area relative to the cultural relic itself. Therefore, the intelligent recognition system can identify and mark the three-dimensional area with the covering in the target cultural relic based on the three-dimensional imaging data of the target cultural relic.

[0062] 102. Determine at least one covering feature object that matches the three-dimensional area based on the input soil quality parameters, cultural relic material parameters, and excavation time.

[0063] In an embodiment of the present application, it is necessary to predict the attribute characteristics such as the shape, structure, and material of the covered area of the target cultural relic. Since these attribute characteristics of unearthed cultural relics are affected by various factors such as the dynasty of the unearthed cultural relic, the location of the unearthed cultural relic, and the plasticity of the covering, a preliminary judgment can be made based on the cultural relic itself and the cultural relic covering. Specifically, the basis for the preliminary judgment can be to determine the soil type according to the soil quality parameters. For example, whether the soil is sandy soil, clay soil, loam, etc. can be determined according to the water content of the soil, the void ratio of the soil, the saturation of the soil, etc. Thus, the possible excavation location of the target cultural relic and the possible size of the entity part of the covered area (for example, when the coverings are sandy soil and clay soil respectively, since the plasticity of sandy soil is worse than that of clay soil, for the same-sized covered area, the entity size under the sandy soil covered area will be larger than that under the clay soil covered area) can be screened out according to the soil quality parameters. In addition, the cultural relic material can be determined according to the cultural relic material parameters, such as stoneware, jadeware, porcelain, woodenware, etc. The characteristics of the unearthed cultural relic can be further determined based on the cultural relic material and the excavation time. It should be noted that there is no limit to the order of screening the possible objects of the covered area of the target cultural relic according to the soil quality parameters, cultural relic material parameters, and excavation time.

[0064] In the specific implementation process, cultural relics workers can pre-enter the soil parameters, cultural relic material parameters, and the unearthed time of cultural relics into the intelligent recognition system in the embodiments of the present application. For the target cultural relics whose coverings need to be removed, it is not difficult to judge their unearthed time, soil parameters, and cultural relic material parameters according to the existing technology. The relevant parameters of the target cultural relics can be manually input into the intelligent recognition system by the staff through the interaction device, so that the system can screen out the possible objects of the coverings of the target cultural relics based on the target cultural relic parameters and the preset parameters in the database.

[0065] 103. Retrieve the morphological characteristics of the cultural relics matching the unearthed time, and predict the information of the hidden objects in the three-dimensional area according to the morphological characteristics of the cultural relics and the covering feature objects.

[0066] Among them, the information of the hidden objects includes at least one of the possible information of the hidden object morphology, the possible information of the hidden object material, and the possible information of the hidden object connection structure. Further, the possible information of the hidden object morphology can be a hollow state, a flat state, an ear-vase state, a symmetric state, etc. The possible information of the hidden object material can be stoneware, jadeware, porcelain, woodenware, etc. The hidden object connection structure can be an inlay structure, a ring-shaped connection, etc. In addition, each kind of possible information can contain one or more possibilities. Exemplarily, according to the morphological characteristics of the cultural relics matching the unearthed time of the target cultural relics and the preliminarily determined covering feature objects, the possible information of the hidden objects under a certain covering area is predicted as: an ear-vase that is axially symmetric with an uncovered area, and the ear-vase has a hollow pattern, and the ear-vase is connected to the cultural relic body by bonding and is made of porcelain.

[0067] In the embodiments of the present application, cultural relics workers pre-statistics the characteristics of cultural relics under different unearthed times in advance, and input the data that can identify the mapping relationship between the morphological characteristics of cultural relics and the unearthed time of cultural relics into the intelligent recognition system, so that the intelligent recognition system as the execution subject can retrieve the morphological characteristics of the cultural relics matching the target cultural relics after obtaining the unearthed time of the target cultural relics, further narrowing the range of possible objects of the covering area and finally determining the information of the hidden objects under the covered three-dimensional area.

[0068] 104. Construct the three-dimensional data of the cultural relics that match the three-dimensional imaging data and the information of the hidden objects, and output them.

[0069] According to the above content, after the intelligent recognition system scans the target cultural relic using 3D scanning technology to obtain its 3D imaging data, it can determine one or more 3D regions of the target cultural relic with coverings. Then, based on the soil quality parameters, unearthed time, and cultural relic material parameters of each 3D region, it filters out possible objects in the covered regions. Combining with the unearthed time of the target cultural relic, for the morphological features of the cultural relic, such as the shape, material, connection structure with the main body of the cultural relic, and size of the finally predicted hidden object, based on the predicted hidden object information and the position of the covered 3D region, a 3D model and 3D imaging data of the complete target cultural relic can be generated, so that cultural relic restorers can remove the coverings on the surface of the cultural relic according to the prediction model.

[0070] The embodiment of the present invention provides another method for collecting 3D data of cultural relics. As Figure 2 shown, this method includes:

[0071] 201. Obtain the 3D imaging data obtained by scanning the target cultural relic using 3D scanning technology, and mark the 3D regions of the coverings in the 3D imaging data.

[0072] In the embodiment of the present invention, the execution entity can be an intelligent management system with a scanning function, such as an intelligent recognition system, a digital scanning system, etc. Exemplarily, when the execution entity is an intelligent recognition system, when the target cultural relic without covering removal is placed at the preset scanning position in the system, the intelligent recognition system is automatically or manually triggered to start scanning through, for example, a radar device, so that the computer software for 3D imaging scanning in the intelligent recognition system can perform cloud point collection on the target cultural relic to obtain the 3D laser point cloud data of the target cultural relic.

[0073] In one embodiment, marking the 3D regions of the coverings in the 3D imaging data can specifically include: analyzing the point cloud data in the 3D imaging data, classifying the point cloud data according to 3D surfaces, and rendering and outputting the classified point cloud data; correcting the point cloud data based on the position correction coordinates of the point cloud data, and marking the corrected point cloud data as the 3D regions of the coverings.

[0074] Generally, when using a lidar device to scan a target cultural relic, the 3D imaging data of the target cultural relic collected is the 3D point cloud data in a 3D coordinate system with the lidar device as the origin, which includes the normal vectors and curvatures corresponding to each cloud point. According to the normal vectors and curvatures of each cloud point and based on the surface segmentation technology, the 3D surface of the target cultural relic can be segmented to obtain multiple sub-surface regions, and each sub-surface region corresponds to a set of point cloud data sets. Further, according to each set of point cloud data sets for rendering and stitching, a 3D model corresponding to the target cultural relic can be obtained.

[0075] It is not difficult to understand that when the target cultural relic has a three-dimensional area with the coating not removed, there will definitely be a curved surface separation boundary between the covered area and the cultural relic itself. Therefore, when performing curved surface segmentation based on the three-dimensional point cloud data, the covered area and the target cultural relic itself can be accurately identified. It should be noted that when performing rendering and modeling based on the classified point cloud data set, the curved surfaces of the target cultural relic are spliced. Since the covered area usually presents an irregular shape, that is, the geometric features of the cloud points change irregularly, there may be errors in the splicing modeling between the covered area and the target cultural relic itself. Therefore, after the intelligent recognition system performs rendering output based on the classified point cloud data, the position of the covered area can be modified manually so that the intelligent recognition system corrects the point cloud data based on the position correction coordinates of the point cloud data and marks the covered three-dimensional area.

[0076] Specifically, the correction coordinates can be achieved by manually translating, flipping, scaling, rotating, etc. any covered three-dimensional area in the three-dimensional model, so that the coordinates of the covered area in the three-dimensional imaging data of the target cultural relic are corrected and marked, and then the three-dimensional point cloud data of the target cultural relic with the covered three-dimensional area marked is obtained.

[0077] 202. Based on the cultural relic information in the historical cultural relic resource database, establish a mapping relationship between different soil quality parameters, cultural relic material parameters, excavation time and different covered feature objects, and bind the authentication probability matching the covered feature object with the mapping relationship.

[0078] In the embodiment of the present application, the historical cultural relic resource database records at least the soil quality parameters, cultural relic material parameters, excavation time, and cultural relic attribute parameters covered when the unearthed cultural relics are unearthed, such as cultural relic size, structural characteristics, etc. Therefore, based on the historical cultural relic resource database, the characteristic information at different positions of the cultural relics can be statistically analyzed according to the relevant information of the unearthed cultural relics. In the specific implementation process, a neural network model for generating the mapping relationship can be pre-trained, and the mapping relationship in this step can be generated by using the trained cultural relic feature recognition model and cultural relic resource data, but it is not limited to this. In order to improve the accuracy of the covered feature object recognition, after creating the mapping relationship between the cultural relic information and the covered feature object, the cultural relic workers will conduct authentication and input the authentication probability for binding. Among them, the authentication probability can identify the matching degree between the covered feature and the mapping object, and the authentication probability assignment can be set according to different application scenarios.

[0079] 203. Extract the covered feature objects matching the input soil quality parameters, cultural relic material parameters, and excavation time according to the mapping relationship, and determine the covered feature objects of the three-dimensional area as the covered feature objects corresponding to the authentication probability greater than the preset authentication threshold.

[0080] Among them, the authentication threshold can be set according to different application scenarios. In the embodiments of the present application, the mapping relationship established in the above steps can be configured in the intelligent recognition system, so that the intelligent recognition system can, based on input data, such as the unearthed parameters of the target cultural relic, the cultural relic material parameters, the unearthed time, and the marked covered area location, and the mapping relationship, obtain one or more possible covered feature objects at the covered location, and then further screen the covered feature objects by the pre-bound authentication probability and the preset authentication threshold, and input the final prediction result.

[0081] In the embodiments of the present application, when making a preliminary prediction of the three-dimensional area covered on the target cultural relic, based on the pre-created mapping between the cultural relic information and the possible feature objects, and according to the soil quality parameters, cultural relic material parameters, unearthed time, covered location, and authentication probability of the covered area of the target cultural relic, multi-level screening is performed on the possible objects in the covered area, ensuring the accuracy of the three-dimensional data of the collected target cultural relic.

[0082] 204. Construct a convolutional neural network with multiple inputs and a single output, and construct a training sample set of the covered shape of cultural relics based on the historical cultural relic resource database.

[0083] Among them, the training sample set of the covered shape of cultural relics contains different cultural relic shape features, the hidden object marks corresponding to the covered feature objects, and the hidden object coordinates.

[0084] 205. Train the convolutional neural network based on the training sample set of the covered shape of cultural relics to obtain a trained convolutional neural network model.

[0085] According to the above content, after making a preliminary prediction of the covered three-dimensional area, one or more covered feature objects in each covered area can already be determined. In the embodiments of the present application, by further obtaining the cultural relic shape features matching the unearthed time of the target cultural relic, the hidden information of the covered feature objects is confirmed.

[0086] It should be noted that cultural relics produced in different dynasties have different shape features, and different shape features also show different distributions in the covered areas when unearthed. For example, the bronze jue of the Xia and Shang dynasties has the shape feature that the bottom support is three legs, and there are many protruding concave openings on the upper part. Therefore, the covered areas of such cultural relics are mostly at the concave openings and the three legs when unearthed. Therefore, in the embodiments of the present application, based on the historical cultural relic resource database, the cultural relic shape features, the hidden object identifiers and hidden object coordinates included in each covered area are extracted, a training set is created, and the convolutional neural network is trained to obtain a prediction model that can predict the hidden information under the covered object according to the cultural relic shape features.

[0087] Further, when training a neural network model using a sample set, the network weight values in the model are determined based on the authentication probability, and the convolutional neural network model can be a three-layer network model, but is not limited thereto. By pre-training a hidden information prediction model and configuring it in the intelligent recognition system, the hidden information can be predicted according to the relevant information of the target cultural relic, and at least one of the hidden object form possibility information, the hidden object material possibility information, and the hidden object connection structure possibility information can be obtained. Among them, the hidden object form possibility information identifies what form the hidden place may be, the hidden object material possibility information identifies what material the hidden place is, and the hidden object connection structure possibility information identifies what connection method the hidden place may be.

[0088] 206. Obtain the cultural relic information matching the unearthed time from the historical cultural relic resource database, and compare the cultural relic information according to the morphological classification category to determine the morphological characteristics of the cultural relic.

[0089] The historical cultural relic resource database contains various attribute information of cultural relics, such as the unearthed time, material, production process, development history, etc. of the cultural relics. After receiving the unearthed time of the target cultural relic, the intelligent recognition system in the embodiment of the present application extracts the cultural relic information of the cultural relic at this unearthed time from the historical cultural relic resources data based on technologies such as web crawling according to the unearthed time of the target cultural relic, and then combines the morphological classification category, such as the cultural relic being symmetric, having a smooth arc surface, being a cube, having a hollow surface, etc., and compares it with the extracted cultural relic information to determine the main morphological characteristics of the cultural relic unearthed at the current unearthed time.

[0090] 207. Perform prediction processing on the cultural relic morphological characteristics and the covered feature object based on the trained convolutional neural network model to obtain the hidden object information in the three-dimensional region.

[0091] Specifically, according to the cultural relic morphological information corresponding to the unearthed time of the target cultural relic output by the pre-trained cultural relic form prediction model, the initially screened covered feature objects can be filtered and screened. Exemplarily, if it is determined in step 203 above that the covered feature objects at area A of the target cultural relic covering area in the Xia, Shang, and Zhou dynasties are a bronze round zun and a bronze square zun, in this step, according to the unearthed time of the Xia, Shang, and Zhou dynasties and using the pre-trained cultural relic morphological feature prediction model for prediction, the morphological characteristics of the cultural relic at this unearthed time are obtained as circular. Therefore, according to this output result, the initially selected covered feature objects are screened, and the final hidden object information is a bronze round zun.

[0092] 208. Render the hidden areas of different partitions in the three-dimensional imaging data based on the hidden object information, and configure the hidden areas in the three-dimensional imaging data in an embedded manner to generate three-dimensional cultural relic data.

[0093] In the embodiments of the present application, each covered area may include one or more hidden objects. The hidden object information predicted based on the above steps may further include the position coordinates of the hidden objects and the hidden object identifiers corresponding to each hidden object, so that when the three-dimensional point cloud data corresponding to the hidden objects is merged with the three-dimensional point cloud data of the target cultural relic body, the hidden objects can be corresponded to the hidden areas based on the hidden object information, and at the same time, the hidden areas can be corresponded to the positions in the target cultural relics, and then each hidden area can be embedded into the three-dimensional imaging data of the target cultural relics.

[0094] Further, after generating the complete three-dimensional imaging data of the target cultural relics according to the hidden information of each hidden area of the target cultural relics, the execution main body intelligent management system in the embodiments of the present application can generate and output a cultural relic model corresponding to the prediction result according to the comprehensive three-dimensional point cloud data, so as to provide visual cultural relic information for cultural relic workers.

[0095] Further, as an implementation of the above Figure 1 shown method, an embodiment of the present invention provides a three-dimensional data acquisition device for cultural relics, as Figure 3 shown, the device includes: an acquisition module 31, a marking module 32, a marking module 32, a determination module 33, a prediction module 34, and an output module 35.

[0096] The acquisition module 31 is used to acquire the three-dimensional imaging data obtained by scanning the target cultural relics using three-dimensional scanning technology;

[0097] The marking module 32 is used to mark the three-dimensional area of the covering object in the three-dimensional imaging data, and the target cultural relics are cultural relics without removing the covering object;

[0098] The determination module 33 is used to determine at least one covering feature object matching the three-dimensional area based on the input soil quality parameters, cultural relic material parameters, and excavation time, and the covering feature object is used to represent the possible object covering the covering object;

[0099] The prediction module 34 is used to retrieve the morphological characteristics of the cultural relics matching the excavation time, and predict the hidden object information in the three-dimensional area according to the morphological characteristics of the cultural relics and the covering feature object, and the hidden object information includes at least one of the hidden object morphological possibility information, the hidden object material possibility information, and the hidden object connection structure possibility information;

[0100] The output module 35 is used to construct the three-dimensional data of the cultural relics matching the three-dimensional imaging data and the hidden object information, and output it.

[0101] In a specific application scenario, Figure 4 shown, the marking module 32 includes:

[0102] A classification unit 321 for parsing the point cloud data in the 3D imaging data, classifying the point cloud data according to 3D surfaces, and rendering and outputting the classified point cloud data;

[0103] A marking unit 322 for correcting the point cloud data based on the position correction coordinates of the point cloud data and marking the corrected point cloud data as the 3D area of the covering object.

[0104] In a specific application scenario, such as Figure 4 As shown, the device further includes: a establishing module 36,

[0105] The establishing module 36 is used to establish a mapping relationship between different soil property parameters, cultural relic material parameters, excavation time and different covering feature objects based on the cultural relic information in the historical cultural relic resource database, and bind the authentication probability matching the covering feature object with the mapping relationship;

[0106] The determining module 33 is specifically used to extract the covering feature object matching the input soil property parameters, cultural relic material parameters, and excavation time according to the mapping relationship, and determine the covering feature object of the 3D area by taking the covering feature corresponding to the authentication probability greater than the preset authentication threshold.

[0107] In a specific application scenario, such as Figure 4 As shown, the device further includes:

[0108] A constructing module 37 for constructing a multi-input single-output convolutional neural network and constructing a cultural relic covering form training sample set based on the historical cultural relic resource database. The cultural relic covering form training sample set contains the hidden object marks and hidden object coordinates corresponding to different cultural relic morphological features and covering feature objects;

[0109] A training module 38 for training the convolutional neural network based on the cultural relic covering form training sample set to obtain a trained convolutional neural network model. The network weight values in the convolutional neural network model are determined based on the authentication probability, and the convolutional neural network model is a three-layer network model.

[0110] In a specific application scenario, such as Figure 4 As shown,

[0111] The prediction module 34 is specifically used to obtain the cultural relic information matching the excavation time from the historical cultural relic resource database, and compare the cultural relic information according to the morphological classification category to determine the cultural relic morphological features;

[0112] The prediction module 34 is specifically further configured to perform prediction processing on the cultural relic form features and the covered feature objects based on the trained convolutional neural network model to obtain the hidden object information in the three-dimensional region.

[0113] In a specific application scenario, such as Figure 4 shown

[0114] The output module 35 is specifically configured to render the hidden areas of different partitions in the three-dimensional imaging data based on the hidden object information, and configure the hidden areas in the three-dimensional imaging data in an embedded manner to generate three-dimensional cultural relic data.

[0115] The present invention provides a method and device for collecting three-dimensional cultural relic data. First, three-dimensional imaging data obtained by scanning a target cultural relic using three-dimensional scanning technology is acquired, and the three-dimensional region of the covering object in the three-dimensional imaging data is marked. The target cultural relic is a cultural relic without the covering object removed. Then, at least one covered feature object matching the three-dimensional region is determined based on the input soil parameters, cultural relic material parameters, and excavation time. The covered feature object is used to represent the possible object covering the covering object. Then, the cultural relic form features matching the excavation time are retrieved, and the hidden object information in the three-dimensional region is predicted according to the cultural relic form features and the covered feature object. The hidden object information at least includes one of the hidden object form possibility information, the hidden object material possibility information, and the hidden object connection structure possibility information. Finally, three-dimensional cultural relic data matching the three-dimensional imaging data and the hidden object information is constructed and output. Compared with the prior art, in the embodiment of the present invention, the three-dimensional imaging data of the cultural relic is obtained by scanning and the covered area is marked, and then the covered feature object is screened and predicted at multiple levels based on the cultural relic-related parameters and the cultural relic form features corresponding to the excavation time of the cultural relic, and the three-dimensional imaging data containing hidden information obtained by prediction is output, which can provide visual information of the covered area of the target cultural relic for cultural relic workers, avoid the problem of cultural relic damage caused by removing the covering object when cultural relic workers do not know the form of the covered area, and improve the feasibility of removing the cultural relic covering object and the integrity of the cultural relic.

[0116] According to an embodiment of the present invention, a storage medium is provided. The storage medium stores at least one executable instruction, and the computer executable instruction can execute the method for collecting three-dimensional cultural relic data in any of the above method embodiments.

[0117] Figure 5 The structural schematic diagram of a terminal provided according to an embodiment of the present invention is shown. The specific implementation of the terminal is not limited in the specific embodiments of the present invention.

[0118] Such as Figure 5As shown in the figure, the terminal may include: a processor 502, a communications interface 504, a memory 506, and a communication bus 508.

[0119] Among them: The processor 502, the communication interface 504, and the memory 506 communicate with each other through the communication bus 508.

[0120] The communication interface 504 is used to communicate with network elements of other devices such as clients or other servers.

[0121] The processor 502 is used to execute the program 510, and specifically can execute the relevant steps in the above-mentioned embodiments of the three-dimensional data acquisition method for cultural relics.

[0122] Specifically, the program 510 may include program code, and the program code includes computer operation instructions.

[0123] The processor 502 may be a central processing unit CPU, or a specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present invention. One or more processors included in the terminal may be of the same type of processor, such as one or more CPUs; or may be of different types of processors, such as one or more CPUs and one or more ASICs.

[0124] The memory 506 is used to store the program 510. The memory 506 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk memory.

[0125] The program 510 is specifically used to enable the processor 502 to perform the following operations:

[0126] Obtain three-dimensional imaging data obtained by scanning a target cultural relic using three-dimensional scanning technology, and mark the three-dimensional area of the covering object in the three-dimensional imaging data, where the target cultural relic is a cultural relic without the covering object removed;

[0127] Based on the input soil parameters, cultural relic material parameters, and excavation time, determine at least one covering feature object that matches the three-dimensional area, and the covering feature object is used to characterize the possible object covering the covering object;

[0128] Retrieve the morphological characteristics of the cultural relics that match the unearthed time, and predict the hidden object information in the three-dimensional area based on the morphological characteristics of the cultural relics and the covered feature objects. The hidden object information includes at least one of the hidden object form possibility information, the hidden object material possibility information, and the hidden object connection structure possibility information;

[0129] Construct the three-dimensional data of the cultural relics that match the three-dimensional imaging data and the hidden object information, and output it.

[0130] Obviously, those skilled in the art should understand that the above-mentioned modules or steps of the present invention can be implemented by a general-purpose computing device. They can be concentrated on a single computing device or distributed on a network composed of multiple computing devices. Optionally, they can be implemented by program codes executable by the computing device. Thus, they can be stored in a storage device and executed by the computing device. And in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules respectively, or multiple modules or steps among them can be made into a single integrated circuit module to implement. In this way, the present invention is not limited to any specific combination of hardware and software.

[0131] The above are only the preferred embodiments of the present invention and are not used to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for collecting three-dimensional data of cultural relics, characterized in that, Including: Obtaining three-dimensional imaging data of a target cultural relic scanned by three-dimensional scanning technology, and marking a three-dimensional area of the covering in the three-dimensional imaging data, where the target cultural relic is a cultural relic without the covering removed; Determining at least one covering feature object matching the three-dimensional area based on the input soil parameters, cultural relic material parameters, and excavation time, where the covering feature object is used to represent a possible object covering the covering; Retrieving the morphological features of the cultural relic matching the excavation time, and predicting the hidden object information in the three-dimensional area according to the cultural relic morphological features and the covering feature object, where the hidden object information includes at least one of the possible information of the hidden object morphology, the possible information of the hidden object material, and the possible information of the hidden object connection structure; Constructing three-dimensional data of the cultural relic that matches the three-dimensional imaging data and the hidden object information, and outputting it.

2. The method according to claim 1, characterized in that The marking of the three-dimensional area of the covering in the three-dimensional imaging data includes: Analyzing the point cloud data in the three-dimensional imaging data, classifying the point cloud data according to a three-dimensional surface, and rendering and outputting the classified point cloud data; Correcting the point cloud data based on the position correction coordinates of the point cloud data, and marking the corrected point cloud data as the three-dimensional area of the covering.

3. The method according to claim 1, characterized in that, Before determining at least one covering feature object matching the three-dimensional area based on the input soil parameters, cultural relic material parameters, and excavation time, the method further includes: Based on the cultural relic information in the historical cultural relic resource database, establishing a mapping relationship between different soil parameters, cultural relic material parameters, excavation times and different covering feature objects, and binding the authentication probability matching the covering feature object and the mapping relationship; The determining at least one covering feature object matching the three-dimensional area based on the input soil parameters, cultural relic material parameters, and excavation time includes: Extracting the covering feature objects matching the input soil parameters, cultural relic material parameters, and excavation time according to the mapping relationship, and determining the covering feature objects of the three-dimensional area as the covering feature objects corresponding to the authentication probability greater than the preset authentication threshold.

4. The method according to claim 1, characterized in that, Before retrieving the morphological features of the cultural relic matching the excavation time, and predicting the hidden object information in the three-dimensional area according to the cultural relic morphological features and the covering feature object, the method further includes: Constructing a multi-input single-output convolutional neural network, and constructing a cultural relic covering morphology training sample set based on the historical cultural relic resource database, where the cultural relic covering morphology training sample set contains the hidden object markings and hidden object coordinates corresponding to different cultural relic morphological features and covering feature objects; Training the convolutional neural network based on the cultural relic covering morphology training sample set to obtain a trained convolutional neural network model, where the network weight values in the convolutional neural network model are determined based on the authentication probability, and the convolutional neural network model is a three-layer network model.

5. The method according to claim 4, characterized in that, The retrieving the morphological features of the cultural relic matching the excavation time includes: Obtain the cultural relic information that matches the unearthed time from the historical cultural relic resource database, and compare the cultural relic information according to the morphological classification categories to determine the morphological characteristics of the cultural relics; The predicting the hidden object information in the three-dimensional area according to the cultural relic morphological characteristics and the covered feature object includes: Performing a prediction process on the cultural relic morphological characteristics and the covered feature object based on a pre-trained convolutional neural network model to obtain the hidden object information in the three-dimensional area.

6. The method according to claim 1, wherein The constructing the three-dimensional cultural relic data that matches the three-dimensional imaging data and the hidden object information and outputting it includes: Rendering the hidden areas of different partitions in the three-dimensional imaging data based on the hidden object information, and configuring the hidden areas in an embedded manner in the three-dimensional imaging data to generate the three-dimensional cultural relic data.

7. A three-dimensional data acquisition device for cultural relics, characterized in that, Including: An acquisition module, configured to acquire three-dimensional imaging data obtained by three-dimensional scanning of a target cultural relic; A marking module, configured to mark the three-dimensional area of the covering object in the three-dimensional imaging data, where the target cultural relic is a cultural relic without the covering object removed; A determination module, configured to determine at least one covered feature object that matches the three-dimensional area based on the input soil quality parameters, cultural relic material parameters, and unearthed time, where the covered feature object is used to characterize the possible object of the covering object with covered features; A prediction module, configured to retrieve the morphological characteristics of the cultural relics that match the unearthed time, and predict the hidden object information in the three-dimensional area according to the cultural relic morphological characteristics and the covered feature object, where the hidden object information includes hidden object morphological possibility information, hidden object material possibility information, and hidden object connection structure possibility information; An output module, configured to construct the three-dimensional cultural relic data that matches the three-dimensional imaging data and the hidden object information and output it.

8. A storage medium, in which at least one executable instruction is stored, and the executable instruction causes a processor to perform the operations corresponding to the three-dimensional cultural relic data acquisition method according to any one of claims 1-6.

9. A terminal, comprising: A processor, a memory, a communication interface, and a communication bus, where the processor, the memory, and the communication interface complete communication with each other through the communication bus; The memory is used to store at least one executable instruction, and the executable instruction causes the processor to perform the operations corresponding to the three-dimensional cultural relic data acquisition method according to any one of claims 1-6.

Images