WO2022062135A1

WO2022062135A1 - Method for establishing three-dimensional model database of firearms and spare parts, and database system

Info

Publication number: WO2022062135A1
Application number: PCT/CN2020/129573
Authority: WO
Inventors: 赵娟; 万由滕; 蔡昂; 魏琦; 刘晶; 宋展
Original assignee: 深圳先进技术研究院
Priority date: 2020-09-23
Filing date: 2020-11-17
Publication date: 2022-03-31
Also published as: CN112116706B; CN112116706A

Abstract

A method for establishing a three-dimensional model database of firearms and spare parts, and a database system. The method comprises: collecting attribute information, plane image information and three-dimensional structure models of firearms and spare parts (101); storing the attribute information, the plane image information and the three-dimensional structure models in a database so as to form a firearms and spare parts database (102); and when a retrieval request that includes retrieval data is received, performing matching in the firearms and spare parts database on the basis of the retrieval data, and feeding back a result obtained by means of matching (103). By means of the method, the three-dimensional reconstruction of firearms and spare parts thereof is realized, and a database of the firearms and the spare parts thereof is established, wherein various kinds of information such as attribute information, plane image information and three-dimensional structure models are all contained in the database. With regard to a database retrieval function, retrieval methods are enriched and three methods are available for selection to perform retrieval, and the accuracy and reliability of the retrieval function are improved.

Description

A method and database system for establishing a three-dimensional model database of firearms and spare parts

technical field

The invention relates to the field of firearm data management, in particular to a method and database system for establishing a three-dimensional model database of firearms and spare parts.

Background technique

Guns are powerful. Every year, my country's customs seize a large number of mass-destruction guns and bullets that are illegally smuggled into China. In order to avoid law enforcement and supervision, criminals often use various methods to dismantle, break up, or modify the guns and smuggle them into China. And the parts are not easy to identify, for example, to identify whether a part is a part of a gun? If so, which part of which model of firearm does it belong to? my country still does not have a complete database of firearms and their parts, which will require a lot of manpower and material resources for screening and verification, which will bring great difficulties to the detection of smuggling cases. In today's fast-developing digital age, it has become a popular method to quickly and accurately convert real-world objects into computers to realize information-based recognition and processing. Therefore, the establishment method and database system of a three-dimensional model database of firearms and their spare parts are of great significance to the management and quick and accurate search of illegal firearms and spare parts.

The invention patent application number 201810047270.6 discloses a gun safety control system. It includes the tracking and positioning module set on the gun, the human biometric sensor set on the gun storage device or the gun storage position, the displacement confirmation module set at the gun storage position, the background supervision command center, the authorized supervision platform and the monitoring equipment. The invention patent application number CN201911111508.8 proposes a gun control method based on the ZigBee protocol. The gun control method based on the ZigBee protocol can realize the query and statistics of the use information of guns, improve the efficiency of gun management in the army, and reduce the mistakes in manual management.

The above patents are all proposed for the safety control of firearms, and are in a blank state in the firearms database. Aiming at the blank of the existing firearm database, the invention provides a method and database system for establishing a three-dimensional database of firearms and their parts, and establishes a variety of three-dimensional model database retrieval functions to realize rapid and accurate inspection and identification of firearms and their parts.

SUMMARY OF THE INVENTION

Aiming at the defects in the prior art, the present invention proposes a method and database system for establishing a three-dimensional model database of firearms and spare parts.

Specifically, the present invention proposes the following specific embodiments:

The embodiment of the present invention also provides a method for establishing a three-dimensional model database of firearms and loose parts, the method comprising:

Collect attribute information, plane image information and three-dimensional structural models of guns and spare parts;

Store the attribute information, the plane image information and the three-dimensional structural model in a database to form a firearms and spare parts database;

When a retrieval request including retrieval data is received, matching is performed in the firearms and spare parts database based on the retrieval data, and the matching result is fed back.

In a specific embodiment,

The attribute information of the gun includes: serial number, quality, size, caliber, name, structure, country of manufacture, manufacturer, model, production year, surface identification, and material information;

The attribute information of the spare parts includes: serial number, quality, size, name, structure, country of manufacture, manufacturer, model, production age, surface identification, and material information.

In a specific embodiment, the loose parts include standard parts and non-standard parts;

The three-dimensional structural model of the standard part is obtained by selecting other standard parts of the same type in the preset three-dimensional model library for parameter modification; or by forward modeling of features; or by combining mechanical measurement and forward modeling method to model;

The three-dimensional structure model of the non-standard part is obtained by first obtaining point cloud data through a surface structured light three-dimensional scanner, and then performing graphic processing and reconstruction on the point cloud data through a graphics workstation to obtain three-dimensional structure data, and then importing the three-dimensional structure data into the three-dimensional structure data. The modeling software is obtained through modeling processing; or point cloud data is obtained by scanning with a hand-held scanner, and then Geomagic Design X software is used to reverse the modeling and feature processing of the point cloud data to obtain shape data, which is then obtained through 3D modeling software. It is obtained by performing modeling processing based on the shape data.

In a specific embodiment, the plane image information is obtained by projecting the three-dimensional structural model to reduce the dimension.

In a specific embodiment, the plane image information is based on projecting the three-dimensional structural model to obtain a two-dimensional projection map with no less than 6 viewing angles; and the two-dimensional projection map is first based on a convolutional neural network The network preprocesses the image, and then uses the scale-invariant feature transformation algorithm and the ORB algorithm to extract features.

In a specific embodiment, it also includes:

After acquiring the newly added attribute information, the plane image information and the three-dimensional structural model, search and match in the firearms and spare parts database;

If the retrieval result is that there is no matching data in the firearms and spare parts database, the acquired attribute information, the plane image information and the three-dimensional structure model are stored in the firearms and spare parts database.

In a specific embodiment, it also includes:

Based on the acquired attribute information, the plane image information, and the three-dimensional structure model, data update is performed on the firearms and spare parts database; the data update includes data modification, data supplementation, and data deletion.

In a specific embodiment, it also includes:

Based on the firearms and spare parts database, query, count, supervise and analyze the information of firearms and spare parts.

In a specific embodiment, the retrieval includes: attribute information retrieval, plane image information matching screening, and three-dimensional model feature matching screening.

The embodiment of the present invention also proposes a three-dimensional model database system for firearms and spare parts, including:

The acquisition module is used to collect the attribute information, plane image information and 3D structural model of guns and spare parts;

a database module for storing the attribute information, the plane image information and the three-dimensional structure model in a database to form a firearms and spare parts database;

The retrieval module is used for matching in the firearms and spare parts database based on the retrieval data when a retrieval request including retrieval data is received, and feeding back the result obtained by the matching.

In this way, compared with the prior art, the solution of the present invention has the following advantages:

(1) Realized 3D reconstruction of guns and their parts; (2) Established a database of guns and their parts, including attribute information, plane image information, 3D structural models and other information; (3) In terms of database retrieval function, the retrieval methods are enriched, and there are three methods for retrieval selection, which improves the accuracy and reliability of retrieval functions.

Description of drawings

In order to illustrate the technical solutions of the embodiments of the present invention more clearly, the following briefly introduces the accompanying drawings used in the embodiments. It should be understood that the following drawings only show some embodiments of the present invention, and therefore do not It should be regarded as a limitation of the scope, and for those of ordinary skill in the art, other related drawings can also be obtained according to these drawings without any creative effort.

Fig. 1 is the schematic flow chart of the establishment method of a kind of firearms and spare parts database proposed by the embodiment of the present invention;

2 is a schematic diagram of a framework for collecting attribute information, plane image information, and three-dimensional structural models of firearms and their spare parts in a method for establishing a three-dimensional model database of firearms and their spare parts according to an embodiment of the present invention;

3 is a schematic diagram of a basic composition framework for collecting attribute information, plane image information, and three-dimensional structural models of firearms and their spare parts in a method for establishing a three-dimensional model database of firearms and their spare parts proposed by an embodiment of the present invention;

4 is a schematic diagram of a retrieval frame in a method for establishing a three-dimensional model database of firearms and parts provided by an embodiment of the present invention;

FIG. 5 is a schematic structural diagram of a three-dimensional model database of firearms and loose parts according to an embodiment of the present invention.

detailed description

Hereinafter, various embodiments of the present disclosure will be described more fully. The present disclosure is capable of various embodiments, and adaptations and changes may be made therein. It should be understood, however, that there is no intention to limit the various embodiments of the present disclosure to the specific embodiments disclosed herein, but the present disclosure should be construed to cover various embodiments falling within the spirit and scope of the present disclosure. All adjustments, equivalents and/or alternatives.

The terminology used in the various embodiments of the present disclosure is for the purpose of describing particular embodiments only and is not intended to limit the various embodiments of the present disclosure. As used herein, the singular is intended to include the plural as well, unless the context clearly dictates otherwise. Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which various embodiments of this disclosure belong. The terms (such as those defined in commonly used dictionaries) will be interpreted as having the same meaning as the contextual meaning in the relevant technical field and will not be interpreted as having an idealized or overly formal meaning, unless explicitly defined in various embodiments of the present disclosure.

Example 1

Embodiment 1 of the present invention discloses a method for establishing a three-dimensional model database of firearms and loose parts. As shown in FIG. 1 , the method includes the following steps:

Step 101: Collect attribute information, plane image information and three-dimensional structural model of guns and spare parts;

Step 102: Store the attribute information, the plane image information and the three-dimensional structure model in a database to form a firearms and spare parts database;

Step 103: When a retrieval request including retrieval data is received, matching is performed in the firearms and spare parts database based on the retrieval data, and the result obtained by the matching is fed back.

In a specific embodiment, the attribute information of the firearm includes: serial number, quality, size, caliber, name, structure, country of manufacture, manufacturer, model, age of manufacture, surface identification, and material information; The attribute information includes: serial number, quality, size, name, structure, country of manufacture, manufacturer, model, production year, surface identification, and material information.

Specifically, the database establishment method is mainly to collect various data, as shown in Figure 2 and Figure 3, including attribute information collection, 3D structure scanning and reconstruction, 3D structure point cloud feature information collection, and 2D image feature information collection . Among them, attribute information collection is responsible for collecting the basic attributes of guns and spare parts, and drawing them into text information. The basic attribute information of gun information data is as follows: For a complete gun, the text information data includes serial number, quality, size, caliber, -, material information and other information, as well as name, structure, country of production, manufacturer, model, production year, surface identification and other information. For loose parts, the text information data includes information such as serial number, quality, size, and material information, as well as information such as name, structure, country of manufacture, manufacturer, model, production age, and surface identification. In addition, for some firearms to be inspected, some are known and some are unknown, so the collection of attribute information is optional (if you can confirm it, fill it in, if you can't confirm it, do not fill in). However, the following attribute information must be collected: the quality of the parts and the quality, length, and caliber of the gun.

In a specific embodiment, the loose parts include standard parts (such as screws, springs, barrels, etc.) and non-standard parts (such as gun handles, barrel sleeves, etc.); thus, when scanning and reconstructing three-dimensional structures At the beginning, it is necessary to classify the loose parts of the firearm, standard parts and non-standard parts.

The three-dimensional structural model of the standard part is obtained by selecting other standard parts of the same type in the preset three-dimensional model library for parameter modification; or by forward modeling of features; or by combining mechanical measurement and forward modeling It can be obtained by modeling; for standard parts, some small parts such as screws, nuts, bolts and other small parts used in daily life, these parts account for a large total in the dismantling of firearms. You can select the same type of standard parts in the Solidwork library to modify the parameters to generate the required parts. There are also some standard parts that are not in the Solidwork library, which need to be obtained by forward modeling of simple features, such as pins, etc. For these parts that can see the modeling features at a glance, mechanical measurement and normalization can be used. Modeling is carried out using a combined approach to modeling.

The three-dimensional structure model of the non-standard part is obtained by first obtaining point cloud data through a surface structured light three-dimensional scanner, and then performing graphic processing and reconstruction on the point cloud data through a graphics workstation to obtain three-dimensional structure data, and then importing the three-dimensional structure data into the three-dimensional structure data. The modeling software is obtained through modeling processing; or point cloud data is obtained by scanning with a hand-held scanner, and then Geomagic Design X software is used to reverse the modeling and feature processing of the point cloud data to obtain shape data, which is then obtained through 3D modeling software. It is obtained by performing modeling processing based on the shape data. For non-standard parts and local complex parts, it refers to parts with complex curves in some local areas, such as the springs of firearms. Although the modeling features of these parts can be seen, it is not easy to obtain their local curves. Therefore, the point cloud data can be obtained by using a surface structured light 3D scanner, and the 3D structural data can be reconstructed through the graphics processing of the graphics workstation and then imported into Solidworks for modeling processing. A finer three-dimensional structure is obtained. There are also overall complex parts, which refer to parts with complex curves and surfaces in many parts of the whole part, and the modeling features cannot be seen, such as large parts such as gun bodies. The hand-held scanner scans to obtain point cloud data STL files, and then uses Geomagic Design X software for reverse modeling and feature processing to obtain shape data, and SolidWorks for modeling and processing to obtain finer three-dimensional structures.

Therefore, for standard parts, such as screws, springs, barrels, etc., use mechanical modeling software to draw 3D structural models; for non-standard parts, such as gun handles, barrel sleeves, etc., use 3D scanning technology to reconstruct 3D structural models ( Including but not limited to line laser scanning and surface structured light scanning). When storing the 3D structure model in the database, the file storage type should be unified, and the basic attribute text should be introduced for each 3D structure model, and the basic information such as gun name, gun category, part name, part category, material, size, etc. should be written.

During the three-dimensional reconstruction of guns and their parts, their point cloud data will be obtained. The point cloud data is obtained by the three-dimensional data acquisition equipment. The information on the surface of the object is measured by the data acquisition instrument and converted into three-dimensional coordinate data. Get the corresponding point cloud data set. After obtaining the point cloud data set, it is necessary to perform feature extraction on the point cloud data. The main method of point cloud feature extraction is to extract the features of the scattered parts through the geometric information and topology information of the grid model on the basis of grid modeling. With the point cloud data on the surface of the object, the 3D structural model can be quickly and accurately reconstructed.

Further, the surface structured light scanning or line laser scanning of the 3D structure scanning and reconstruction of the loose parts includes at least: a surface structured light 3D scanner, a handheld 3D scanner, a graphics workstation, and professional 3D modeling software.

Surface structured light 3D scanner is used for digital 3D scanning of parts and bullets. It has the advantages of automation, speed and high precision. It can scan different types and structures of gun parts and bullets. Surface structured light 3D scanner Consists of projection equipment and cameras.

Handheld 3D scanners are line laser scanners, which are small and portable, allowing operators to scan objects from any angle and splicing them into a 360° 3D model, with excellent scanning accuracy and acquisition speed. Different scanners can be selected according to the type and size of guns and loose parts.

The graphics workstation has powerful graphics processing capabilities, and is equipped with software supporting the 3D scanner. According to the point cloud data obtained from the 3D scanner, the 3D structural data is reconstructed through graphics processing. At this time, the obtained 3D structure data is relatively rough and may be defective, and will be imported into professional 3D modeling software, and a finer 3D structure will be obtained through modeling processing. At the same time, the size and volume of the parts, the length, caliber and volume of the bullet can be directly obtained in the professional 3D modeling software. Professional 3D modeling software such as Solidworks, 3D MAX, UG, ProE, etc., and other software such as Geomagic Studio and Geomagic Design X can also be used together. The data output format of the three-dimensional structure is a general three-dimensional file format, such as stp, igs, stl, etc.

At the same time, point cloud data of guns and their parts are obtained when 3D scanning is performed. The point cloud data of the object can be collected by the 3D scanner, and then converted into 3D coordinate data, and the corresponding point cloud data set can be obtained. After obtaining the point cloud data set, it is necessary to perform feature extraction on the point cloud data. The main method of point cloud feature extraction is to extract the features of the scattered parts through the geometric information and topology information of the grid model on the basis of grid modeling. Points, and then you can generate feature lines and feature surfaces based on these feature points. The feature extraction of 3D point cloud data is the basis of the database for 3D model retrieval application.

In a specific embodiment, the plane image information is obtained by projecting the three-dimensional structural model to reduce the dimension. Further, the plane image information is based on projecting the three-dimensional structural model to obtain a two-dimensional projection map with no less than 6 viewing angles; and the two-dimensional projection map is first based on the convolutional neural network. Then use the scale-invariant feature transformation algorithm and the ORB algorithm to perform feature extraction.

In addition, after reconstructing the three-dimensional structural model, it is necessary to use projection technology to reduce the dimensionality to obtain two-dimensional plane graphics. Projection refers to the process of converting three-dimensional coordinates into two-dimensional coordinates. The projection used here is perspective projection, which is characterized by all the Projection lines are projected from a point in space, that is, the viewpoint or the projection center. Objects close to the viewpoint are relatively large, while objects farther away are relatively small. When using perspective projection on a 3D structural model, obtain a 2D projection at no less than 6 viewing angles. After obtaining these two-dimensional projection maps, the images are processed based on the convolutional neural network, and then the scale-invariant feature transformation (SIFT) algorithm and the ORB (Oriented FAST and Rotated BRIEF) algorithm are used for feature extraction. That is, two-dimensional image feature points of multiple angles of each model can be obtained.

Further, it also includes: after acquiring the newly added attribute information, the plane image information and the three-dimensional structural model, searching and matching in the firearms and spare parts database;

In a specific embodiment, data updating is performed on the firearms and spare parts database based on the acquired attribute information, the planar image information, and the three-dimensional structural model; the data updating includes data modification, data supplementation, and Data deletion; in addition, based on the firearms and spare parts database, query, statistics, supervise and analyze the information of firearms and spare parts.

Specifically, the information management of firearms and spare parts in the database is also very important, including the following management processes:

(1) For newly added guns and spare parts, enter their basic attribute information, plane image information, and 3D structure information obtained by 3D structure scanning and reconstruction in the gun information collection method, compare them through the information retrieval function, and check the information in the database. Whether the parts match the above collected information; if no matching data is found, create the data in the database.

(2) According to the development pace of the current situation, gradually update the firearms and spare parts data in the database, including data modification, addition, deletion, etc.

(3) In the database, the information of firearms and spare parts can be inquired, counted, monitored, analyzed, etc., such as type statistics, structure statistics, source statistics, etc.

In a specific embodiment, as shown in FIG. 4 , the retrieval includes: attribute information retrieval, plane image information matching screening, and three-dimensional model feature matching screening.

The following is a detailed description of the rapid retrieval and matching method of gun parts in the database:

For the gun parts to be retrieved, collect attribute information and plane image information on the gun information collection, and use the three-dimensional structure scanning and reconstruction module to obtain the three-dimensional structure information, and obtain the mass M, maximum length L, volume V, two The 3D image feature set P and the 3D structure point cloud feature information T are used to retrieve and compare the bulk data in the database. The two-dimensional image feature set P refers to a feature set extracted from multiple images with different angles and directions, P={p ₁ , p ₂ , p ₃ ..., p _k }, k is the number of images . Feature extraction algorithms include scale-invariant feature transformation (SIFT), local feature (SURF), and k-means clustering algorithms. The three-dimensional structure point cloud information is obtained by a three-dimensional data scanner. By converting the scanned parts into three-dimensional coordinate data, the corresponding point cloud data set can be obtained. The point cloud feature contains the key geometric and topological information of the surface of the loose part, which can accurately describe an object. Feature extraction refers to the process of extracting those points that can represent the geometric characteristics or texture features of objects from the target point cloud data, and then generating objects inversely based on these feature points. The main target of feature extraction is these shape feature points, which can help us quickly match the data in the database during retrieval.

Next, several methods are explained. The first is the text retrieval of basic information of keywords. When you know some basic information of a certain inspection and identification of gun parts (such as identification, type, country of production, manufacturer, model, production age, etc.), These texts are input to compare the direct similarity with the information in the database, and when the same text information is retrieved, the n pieces of information with the highest similarity are returned. If you know the numerical information such as the mass M ₁ , the maximum length L ₁ , and the volume V ₁ of the component, you can search for |M ₁ /M-1|＜γ _M ,|L ₁ /L-1|＜γ _L ,| V ₁ /V-1|<γ _V set of all parts {B}, γ is the error control coefficient, mainly considering the errors caused by various factors such as machining, measurement, and wear of the parts, so the returned matching score is the highest The n component information of . If the input is a 2D image or a 3D model, the database will perform feature extraction, and the extracted 2D image feature information or 3D structure point cloud feature information will be compared with the 2D image feature set P in the database or the 3D structure point. The cloud feature information T carries out feature matching, sorts the parts according to the matching score, and returns the information of the n parts with the highest matching score. Existing feature matching methods include a combination of machine learning, deep learning, and morphological processing. Finally, the inspectors manually identify the returned n components, including basic attributes, two-dimensional images, three-dimensional shape features, etc., to determine whether the matching is successful.

Example 2

Embodiment 2 of the present invention also discloses a three-dimensional model database system for firearms and loose parts, as shown in FIG. 5 , including:

The collection module 201 is used to collect the attribute information, plane image information and three-dimensional structure model of the gun and its parts;

The database module 202 is used for storing the attribute information, the plane image information and the three-dimensional structure model in a database to form a firearms and spare parts database;

The retrieval module 203 is configured to perform matching in the firearms and spare parts database based on the retrieval data when a retrieval request including retrieval data is received, and feedback the result obtained by the matching.

The firearms and spare parts database is mainly composed of database server and database system management software. The database server has powerful storage capacity and fast data access speed. According to application requirements and layout conditions, it can be a local physical server or a cloud server. Database management software such as Oracle, SQL Server, MySQL, etc. are installed on the server.

The three-dimensional structure model of the non-standard part is obtained by first obtaining point cloud data through a surface structured light three-dimensional scanner, and then performing graphic processing and reconstruction on the point cloud data through a graphics workstation to obtain three-dimensional structure data, and then importing the three-dimensional structure data into the three-dimensional structure data. The modeling software is obtained through modeling processing; or point cloud data is obtained by scanning with a handheld scanner, and then Geomagic Design X software is used to perform reverse modeling and feature processing on the point cloud data to obtain shape data, which is then obtained through 3D modeling software. It is obtained by performing modeling processing based on the shape data.

In a specific embodiment, it also includes:

A new module is used for searching and matching in the firearms and spare parts database after acquiring the newly added attribute information, the plane image information and the three-dimensional structure model;

In a specific embodiment, the system further includes:

The updating module is used for data updating of the firearms and spare parts database based on the acquired attribute information, the plane image information and the three-dimensional structure model; the data updating includes data modification, data supplementation and data deletion.

Those skilled in the art can understand that the accompanying drawing is only a schematic diagram of a preferred implementation scenario, and the modules or processes in the accompanying drawing are not necessarily necessary to implement the present invention.

Those skilled in the art can understand that the modules in the device in the implementation scenario may be distributed in the device in the implementation scenario according to the description of the implementation scenario, or may be located in one or more devices different from the implementation scenario with corresponding changes. The modules of the above implementation scenarios may be combined into one module, or may be further split into multiple sub-modules.

The above serial numbers of the present invention are only for description, and do not represent the pros and cons of the implementation scenarios.

The above disclosures are only a few specific implementation scenarios of the present invention, however, the present invention is not limited thereto, and any changes that can be conceived by those skilled in the art should fall within the protection scope of the present invention.

Claims

A method for establishing a three-dimensional model database of firearms and loose parts, characterized in that the method comprises:

Collect attribute information, plane image information and three-dimensional structural models of guns and spare parts;

Store the attribute information, the plane image information and the three-dimensional structural model in a database to form a firearms and spare parts database;

When a retrieval request including retrieval data is received, matching is performed in the firearms and spare parts database based on the retrieval data, and the matching result is fed back.
The method of claim 1, wherein:

The attribute information of the gun includes: serial number, quality, size, caliber, name, structure, country of manufacture, manufacturer, model, production year, surface identification, and material information;

The attribute information of the spare parts includes: serial number, quality, size, name, structure, country of manufacture, manufacturer, model, production age, surface identification, and material information.
The method of claim 1, wherein the loose parts include standard parts and non-standard parts;

The three-dimensional structural model of the standard part is obtained by selecting other standard parts of the same type in the preset three-dimensional model library for parameter modification; or by forward modeling of features; or by combining mechanical measurement and forward modeling method to model;

The three-dimensional structure model of the non-standard part is obtained by first obtaining point cloud data through a surface structured light three-dimensional scanner, and then performing graphic processing and reconstruction on the point cloud data through a graphics workstation to obtain three-dimensional structure data, and then importing the three-dimensional structure data into the three-dimensional structure data. The modeling software is obtained through modeling processing; or point cloud data is obtained by scanning with a hand-held scanner, and then Geomagic Design X software is used to reverse the modeling and feature processing of the point cloud data to obtain shape data, which is then obtained through 3D modeling software. It is obtained by performing modeling processing based on the shape data.
The method of claim 1, wherein the plane image information is obtained by projecting the three-dimensional structural model to reduce the dimension.
The method according to claim 1 or 4, wherein the plane image information is based on projecting the three-dimensional structural model to obtain a two-dimensional projection map with no less than six viewing angles; The projection map is first preprocessed based on the convolutional neural network, and then the feature is extracted by the scale-invariant feature transformation algorithm and the ORB algorithm.
The method of claim 1, further comprising:

After acquiring the newly added attribute information, the plane image information and the three-dimensional structural model, search and match in the firearms and spare parts database;

If the retrieval result is that there is no matching data in the firearms and spare parts database, the acquired attribute information, the plane image information and the three-dimensional structure model are stored in the firearms and spare parts database.
The method of claim 1, further comprising:

Based on the acquired attribute information, the plane image information, and the three-dimensional structure model, data update is performed on the firearms and spare parts database; the data update includes data modification, data supplementation, and data deletion.
The method of claim 1, further comprising:

Based on the firearms and spare parts database, query, count, supervise and analyze the information of firearms and spare parts.
The method of claim 1, wherein the retrieval comprises: attribute information retrieval, plane image information matching screening, and three-dimensional model feature matching screening.
A three-dimensional model database system for firearms and loose parts, characterized in that it includes:

The acquisition module is used to collect the attribute information, plane image information and 3D structural model of guns and spare parts;

a database module for storing the attribute information, the plane image information and the three-dimensional structure model in a database to form a firearms and spare parts database;

The retrieval module is used for matching in the firearms and spare parts database based on the retrieval data when a retrieval request including retrieval data is received, and feeding back the result obtained by the matching.