Geological environment identification method and system based on Beidou satellite

Through the multi-threaded geological environment identification method based on Beidou satellite, combined with confidence projection and result fusion, the problem of incomplete geological environment identification in complex terrain areas is solved, and the accurate identification of geological environment is achieved.

CN118690258BActive Publication Date: 2024-11-26SICHUAN GEOLOGICAL ENVIRONMENT SURVEY & RES CENT
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202411179719.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-27
Publication Date
2024-11-26
Estimated Expiration
2044-08-27

AI Technical Summary

Technical Problem

In complex terrain areas, artificial survey equipment is difficult to access, resulting in incomplete identification of the geological environment and the inaccurate identification of the geological environment.

Method used

The geological environment identification method based on Beidou satellite is adopted, and geological environment information is processed through multiple geological environment identification threads, combined with confidence projection to the target database, and the results are fused to obtain accurate geological environment identification results.

Benefits of technology

It realizes accurate identification of the geological environment in complex terrain areas, avoids identification errors caused by differences in confidence distribution, and ensures the accuracy of geological environment information.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118690258B_ABST
    Figure CN118690258B_ABST
Patent Text Reader

Abstract

The Beidou satellite-based geological environment identification method and system provided in the present application projects the confidence levels in the geological environment identification results corresponding to different geological environment identification threads into the same confidence database to avoid the situation in which the current geological environment information label represented by the fused geological environment identification results has grammatical errors and inaccurate identification of the current geological environment information due to the different confidence distributions in several geological environment identification results. It can also take into account the ability of several geological environment identification threads to obtain the description of the current geological environment information, ensure the accuracy of the obtained current geological environment information label, and accurately identify the current geological environment information, thereby ensuring the accuracy of the current geological environment information description.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of data identification technology, and more specifically, to a method and system for identifying a geological environment based on Beidou satellites. Background Art

[0002] Geological environment: A type of natural environment, referring to the environmental system composed of the lithosphere, hydrosphere and atmosphere. In the long-term geological evolution, material migration and energy conversion take place between the lithosphere and hydrosphere, between the lithosphere and atmosphere, and between the atmosphere and hydrosphere, forming a relatively balanced open system. Humans and other organisms rely on the geological environment for survival and development, and at the same time, humans and other organisms are constantly changing the geological environment.

[0003] At present, geological environment surveys are generally carried out by manual labor and equipment. However, in some areas with complex terrain, it is difficult for humans to carry relevant equipment, which may lead to inadequate surveys and make it difficult to identify the geological environment. Therefore, a technical solution is urgently needed to improve the above technical problems. Summary of the invention

[0004] In order to improve the technical problems existing in the relevant technologies, the present application provides a geological environment identification method and system based on Beidou satellite.

[0005] In a first aspect, a method for identifying a geological environment based on a Beidou satellite is provided, comprising:

[0006] The current geological environment information is processed respectively through a plurality of geological environment identification threads to obtain a first geological environment identification result corresponding to each geological environment identification thread, wherein the first geological environment identification result includes a confidence level of each geological environment attribute in the plurality of geological environment attributes in the positioning of each geological environment attribute described in the pending current geological environment information;

[0007] For each first geological environment identification result, combining the confidence in the first geological environment identification result, projecting the confidence in the first geological environment identification result into the target confidence database, to obtain a second geological environment identification result;

[0008] Fusing the plurality of second geological environment identification results to obtain a fused geological environment identification result, wherein the fused geological environment identification result includes the confidence of each of the plurality of geological environment attributes in the positioning of each geological environment attribute described by the target current geological environment information;

[0009] Combined with the fused geological environment identification result, a target geological environment identification result is obtained.

[0010] In the present application, for each first geological environment identification result, the confidence in the first geological environment identification result is combined, and the confidence in the first geological environment identification result is projected into the target confidence database to obtain the second geological environment identification result, including:

[0011] Determine a first confidence level and a first loss value in combination with the first geological environment identification result, wherein the first confidence level is the minimum confidence level in the first geological environment identification result, and the first loss value is the maximum confidence level in the first geological environment identification result and the loss value of the first confidence level;

[0012] Determine a second confidence and a second loss value in combination with the target confidence database, wherein the second confidence is the minimum confidence in the target confidence database, and the second loss value is the loss value of the maximum confidence in the target confidence database and the second confidence;

[0013] In combination with several confidence levels in the first geological environment identification result, several third loss values ​​are determined; for each third loss value, a comparison result of the third loss value and the first loss value is determined, a product of the comparison result and the second loss value is determined, and a sum of the product and the second confidence level is determined as a third confidence level, where the third loss value is a loss value of any confidence level in the first geological environment identification result and the first confidence level;

[0014] Each confidence level in the first geological environment identification result is optimized to a corresponding third confidence level to obtain the second geological environment identification result.

[0015] In the present application, the step of combining the fused geological environment identification result to obtain the target geological environment identification result includes:

[0016] In combination with the fused geological environment identification result and the plurality of geological environment attributes, a plurality of sample geological environment information labels and the confidence corresponding to each sample geological environment information label are obtained, wherein the confidence corresponding to the sample geological environment information label indicates the confidence of the geological environment information example description obtained in combination with the fused geological environment identification result, and the geological environment attributes located at the same geological environment attribute location in different sample geological environment information labels are not completely the same;

[0017] Combined with the confidences of the several geological environment information example descriptions, the sample geological environment information label with the greatest confidence is determined as the target geological environment identification result.

[0018] In the present application, the step of combining the fused geological environment identification result to obtain the target geological environment identification result includes:

[0019] The fused geological environment identification result and the plurality of geological environment attributes are combined to determine the geological environment attribute located at each geological environment attribute location, and the determined geological environment attributes are combined into the target geological environment identification result.

[0020] In the present application, before fusing the plurality of second geological environment identification results to obtain the fused geological environment identification result, the method further includes:

[0021] Determine a second number from the first numbers corresponding to the plurality of second geological environment identification results, the second number being the largest first number among the first numbers corresponding to the plurality of second geological environment identification results, the first number corresponding to the second geological environment identification result indicating the number of geological environment attribute locations corresponding to the second geological environment identification result;

[0022] Under the premise that the first number corresponding to the third geological environment identification result is less than the second number, determining a third number, the third number being a loss value of the second number and the first number corresponding to the third geological environment identification result, the third geological environment identification result being any one of the plurality of second geological environment identification results;

[0023] In the third geological environment identification result, the confidence of each of the several geological environment attributes on the target geological environment attribute positioning of the third value is added, and the number of geological environment attribute positionings corresponding to the third geological environment identification result after adding reaches the second number, and the confidence of each geological environment attribute on the target geological environment attribute positioning is the first value, wherein, in the current geological environment information label corresponding to the third geological environment identification result after adding, the target geological environment attribute positioning is located after the remaining geological environment attribute positioning.

[0024] In the present application, the fusion of the plurality of second geological environment identification results to obtain the fused geological environment identification result is performed by a thread, and the method further includes:

[0025] Through the plurality of geological environment identification threads, the geological environment information examples are processed respectively to obtain a fourth geological environment identification result corresponding to each of the geological environment identification threads, wherein the fourth geological environment identification result includes the confidence of each of the plurality of geological environment attributes in the positioning of each geological environment attribute described in the first current geological environment information, and the first current geological environment information label is the label of the geological environment information example;

[0026] For each fourth geological environment identification result, combining the confidence in the fourth geological environment identification result, projecting the confidence in the fourth geological environment identification result into the target confidence database to obtain a fifth geological environment identification result;

[0027] Through the thread, a plurality of fifth geological environment identification results are merged to obtain a sixth geological environment identification result, wherein the sixth geological environment identification result includes the confidence of each of the plurality of geological environment attributes in the positioning of each geological environment attribute described in the second current geological environment information, and the second current geological environment information label is the label of the geological environment information example;

[0028] The thread is trained in combination with the sample label of the geological environment information example and the sixth geological environment identification result.

[0029] In the present application, the thread is trained by combining the sample label of the geological environment information example and the sixth geological environment identification result, including:

[0030] In combination with the sample label, a sample geological environment identification result is determined, wherein the sample geological environment identification result includes the confidence of each of the several geological environment attributes in the location of each geological environment attribute described by the sample, wherein the confidence of the geological environment attribute in the sample label in the corresponding geological environment attribute location is a second value, and the confidence of the remaining geological environment attribute locations is a third value;

[0031] The thread is trained in combination with the sample geological environment identification result and the sixth geological environment identification result.

[0032] In the present application, for each fourth geological environment identification result, in combination with the confidence in the fourth geological environment identification result, the confidence in the fourth geological environment identification result is projected into the target confidence database, and the fifth geological environment identification result is obtained by optimizing the thread execution, and the sample label of the geological environment information example and the sixth geological environment identification result are combined to train the thread, including:

[0033] In combination with the sample label of the geological environment information example and the sixth geological environment identification result, the thread and the optimized thread are trained.

[0034] In the present application, the method further comprises:

[0035] Compressing the geological environment information example through the geological environment identification thread to obtain the characteristics of the geological environment information example;

[0036] By means of the geological environment identification thread, the features of the geological environment information example are decoded to obtain a seventh geological environment identification result, wherein the seventh geological environment identification result includes the confidence of each of the plurality of geological environment attributes in the positioning of each geological environment attribute described in the third current geological environment information, and the third current geological environment information label is the label of the geological environment information example;

[0037] The geological environment identification thread is trained in combination with the sample label of the geological environment information example and the seventh geological environment identification result.

[0038] In a second aspect, a geological environment identification system based on Beidou satellite is provided, comprising a processor and a memory communicating with each other, wherein the processor is used to read a computer program from the memory and execute it to implement the above method.

[0039] In the Beidou satellite-based geological environment identification method and system provided in the embodiment of the present application, different geological environment identification threads have different capabilities of obtaining the description of the current geological environment information. Therefore, when obtaining the current geological environment information label of any current geological environment information, the same current geological environment information is processed by a plurality of geological environment identification threads to obtain the geological environment identification results corresponding to each geological environment identification thread, and the confidences in the geological environment identification results corresponding to the plurality of geological environment identification threads are projected into the same confidence database, and then the geological environment identification results corresponding to the plurality of geological environment identification threads are fused, so as to obtain the current geological environment identification result using the fused geological environment identification result. The final current geological environment information label of the geological environment information is projected into the same confidence database by projecting the confidence in the geological environment identification results corresponding to different geological environment identification threads, so as to avoid the different confidence distributions in several geological environment identification results, resulting in grammatical errors in the current geological environment information label represented by the fused geological environment identification results and inaccurate identification of the current geological environment information. It can also take into account the ability of several geological environment identification threads to obtain the description of the current geological environment information, ensure the accuracy of the current geological environment information label, and accurately identify the current geological environment information, thereby ensuring the accuracy of the current geological environment information description. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0041] Figure 1 A flowchart of a method for identifying a geological environment based on Beidou satellite provided in an embodiment of the present application. DETAILED DESCRIPTION

[0042] In order to better understand the above technical scheme, the technical scheme of the present application is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present application and the specific features in the embodiments are detailed descriptions of the technical scheme of the present application, rather than limitations on the technical scheme of the present application. In the absence of conflict, the embodiments of the present application and the technical features in the embodiments can be combined with each other.

[0043] See also Figure 1 , shows a method for geological environment identification based on Beidou satellite, which may include the technical solutions described in the following steps 201-204.

[0044] 201. Process the current geological environment information through several geological environment identification threads respectively to obtain a first geological environment identification result corresponding to each geological environment identification thread, wherein the first geological environment identification result includes the confidence of each geological environment attribute among several geological environment attributes in the positioning of each geological environment attribute described by the pending current geological environment information.

[0045] Exemplarily, the current geological environment information may be geological environment information obtained through Beidou satellites, specifically including: soil state information and soil composition information, etc.

[0046] The reason for setting several geological environment recognition threads is that each geological environment recognition thread only needs to recognize one type of geological environment information, which can improve the accuracy of recognition and reduce interference. For example, one of the geological environment recognition threads is used to recognize soil state information.

[0047] The confidence in the positioning of each geological environment attribute can be understood as the accuracy of the geological environment in the corresponding regional location.

[0048] In an embodiment of the present application, the geological environment identification thread is used to obtain the geological environment identification result of the input current geological environment information to represent the current geological environment information label of the current geological environment information. Different geological environment identification threads have different capabilities to obtain the description of the current geological environment information. For the same current geological environment information, the geological environment identification results obtained by different geological environment identification threads may be different. For example, for the same current geological environment information, the data scale of the geological environment identification results obtained by different geological environment identification threads may be different. For another example, for the same current geological environment information, the semantic coherence of the description of the current geological environment information obtained by one geological environment identification thread is good, and the contextual logic of the description of the current geological environment information obtained by another geological environment identification thread is more rigorous. Therefore, when obtaining the current geological environment information label of any current geological environment information, the same current geological environment information is processed by several geological environment identification threads to obtain the geological environment identification results corresponding to each geological environment identification thread, and then the geological environment identification results corresponding to several geological environment identification threads are fused, so as to use the fused geological environment identification results to obtain the final current geological environment information label of the current geological environment information. In this way, the ability of several geological environment identification threads to obtain the description of the current geological environment information can be taken into account to ensure that the obtained current geological environment information label can accurately identify the current geological environment information to ensure the accuracy of the description of the current geological environment information.

[0049] Among them, the geological environment identification thread is any network thread, for example, the geological environment identification thread is Chat GPT (Chat Generative Pre-trained Transformer, a geological environment identification thread), LLaMA (LargeLanguage Model Meta AI, a geological environment identification thread). The current geological environment information is any type of current geological environment information. The first geological environment identification result can be represented in any form, for example, the first geological environment identification result is represented in the form of a matrix, the rows of the matrix correspond to several geological environment attributes, and the columns of the matrix correspond to the location of each geological environment attribute. Several geological environment attributes are equivalent to the geological environment attributes in the data matrix of the geological environment identification thread. The data matrix of the geological environment identification thread contains all geological environment attributes as much as possible, and the geological environment attributes contained in any current geological environment information are all included in the data matrix of the geological environment identification thread.

[0050] In the embodiment of the present application, the first geological environment identification result corresponding to the geological environment identification thread is used to represent the pending current geological environment information label corresponding to the geological environment identification thread, which can reflect the confidence of each geological environment attribute in the positioning of each geological environment attribute in the pending current geological environment information description. The positioning of each geological environment attribute in the pending current geological environment information description is used to carry the geological environment attributes contained in the pending current geological environment information label. The number of geological environment attribute positionings corresponding to the pending current geological environment information label is equivalent to the number of geological environment attributes contained in the pending current geological environment information label, that is, the corresponding geological environment attributes are filled in the positioning of each geological environment attribute in the pending current geological environment information description to obtain the pending current geological environment information label. In addition, since the capabilities of several geological environment identification threads to obtain the current geological environment information description are different, among several first geological environment identification results, the number of geological environment attribute positionings corresponding to different first geological environment identification results may be different.

[0051] 202. For each first geological environment identification result, in combination with the confidence in the first geological environment identification result, the confidence in the first geological environment identification result is projected into a target confidence database to obtain a second geological environment identification result.

[0052] Exemplarily, the first geological environment identification result includes the following contents: geological pollutant identification results, geological environment structure identification results, and underground fracture conditions, etc. The second geological environment identification result is obtained after inputting the first geological environment identification result into the database for debugging, so that the second geological environment identification result is more accurate.

[0053] In the embodiment of the present application, the confidence distributions in different first geological environment identification results are different. Under the premise that the confidence distributions in several first geological environment identification results are different, if several first geological environment identification results are directly fused, the fusion effect will be poor due to the different confidence distributions in several first geological environment identification results, which will lead to grammatical errors in the final current geological environment information label and inaccurate identification of the current geological environment information. Therefore, under the premise of obtaining several first geological environment identification results, the confidence in each first geological environment identification result is projected into the target confidence database to obtain the second geological environment identification result corresponding to each geological environment identification thread, so that the confidences in the obtained several second geological environment identification results are all in the target confidence database, that is, the confidences in the several second geological environment identification results are distributed in the same confidence database.

[0054] The target confidence database is an arbitrary numerical range, for example, the target confidence database is a preset confidence database. For another example, the target confidence database is a confidence database corresponding to a number of first geological environment identification results.

[0055] 203. Fusing several second geological environment identification results to obtain a fused geological environment identification result, wherein the fused geological environment identification result includes the confidence of each geological environment attribute among several geological environment attributes in the positioning of each geological environment attribute described by the target current geological environment information.

[0056] For example, the fusion is to combine various individual results into a complete report to comprehensively reflect the geological environment. In this way, complete geological environment information can be obtained for the area, and a comprehensive analysis can be performed to determine the geological environment.

[0057] In an embodiment of the present application, since several second geological environment identification results are used to represent the current geological environment information label of the same current geological environment information, and different second geological environment identification results are obtained by different geological environment identification threads, by fusing several second geological environment identification results, the ability of different geological environment identification threads to obtain the description of the current geological environment information is taken into account, so that the fused geological environment identification result can represent the current geological environment information label of the current geological environment information, and the target geological environment identification result corresponding to the fused geological environment identification result can accurately identify the current geological environment information.

[0058] Among them, the fusion geological environment identification result can be expressed in any form, for example, the fusion geological environment identification result is expressed in the form of a matrix, the rows of the matrix correspond to several geological environment attributes, and the columns of the matrix correspond to the locations of each geological environment attribute. The number of geological environment attribute locations corresponding to the target geological environment identification result is the same as the number of geological environment attribute locations corresponding to any pending current geological environment information tag, that is, in the pending current geological environment information tags corresponding to several geological environment identification threads, the data scale of the target current geological environment information description is the same as the data scale of any pending current geological environment information description.

[0059] 204. Combine and integrate the geological environment identification results to obtain the target geological environment identification results.

[0060] Exemplarily, the fused geological environment identification results need to be analyzed and processed before the target geological environment identification results are obtained. The specific analysis steps are as follows: extract key attribute features in the fused geological environment identification results, analyze the key features, determine the degree of influence of the key attributes on the geological environment, and determine the target geological environment identification results based on the degree of influence.

[0061] In an embodiment of the present application, the fused geological environment identification result can reflect the confidence of each of the several geological environment attributes in the positioning of each geological environment attribute described in the current geological environment information of the target. Combined with the confidence in the fused geological environment identification result, it is possible to locate and fill in the geological environment attributes for each geological environment attribute described in the current geological environment information of the target from among the several geological environment attributes to obtain the target geological environment identification result.

[0062] In the solution provided by the embodiment of the present application, different geological environment identification threads have different capabilities of obtaining the description of the current geological environment information. Therefore, when obtaining the current geological environment information label of any current geological environment information, the same current geological environment information is processed by several geological environment identification threads to obtain the geological environment identification results corresponding to each geological environment identification thread, and the confidences in the geological environment identification results corresponding to the several geological environment identification threads are projected into the same confidence database, and then the geological environment identification results corresponding to the several geological environment identification threads are fused, so as to use the fused geological environment identification results to obtain the final current geological environment information label of the current geological environment information. In this way, by projecting the confidences in the geological environment identification results corresponding to different geological environment identification threads into the same confidence database, it is avoided that the current geological environment information label represented by the fused geological environment identification result has grammatical errors and inaccurate identification of the current geological environment information due to different confidence distributions in the several geological environment identification results, and the capabilities of several geological environment identification threads to obtain the description of the current geological environment information are taken into account, so as to ensure the accuracy of the obtained current geological environment information label, and to accurately identify the current geological environment information, thereby ensuring the accuracy of the current geological environment information description.

[0063] Before fusing several second geological environment identification results, the embodiment of the present application will also optimize the number of geological environment attribute locations corresponding to the several second geological environment identification results and the confidence levels contained therein, so that the number of geological environment attribute locations corresponding to the several second geological environment identification results are the same. The specific process is detailed in the following embodiment.

[0064] Another method for identifying geological environment based on Beidou satellite provided in an embodiment of the present application is executed by:

[0065] 301. Process the current geological environment information through several geological environment identification threads respectively to obtain a first geological environment identification result corresponding to each geological environment identification thread, wherein the first geological environment identification result includes the confidence of each geological environment attribute among several geological environment attributes in the positioning of each geological environment attribute described by the pending current geological environment information.

[0066] In a possible implementation, the current geological environment information is a subject, the geological environment identification thread is a thread trained for the subject, and each geological environment identification thread is used to obtain the current geological environment information label of the subject.

[0067] For example, these two geological environment identification threads are obtained by training different general geological environment identification threads in combination with the theme. The general geological environment identification thread is used to obtain the current geological environment information label of any type of current geological environment information. The general geological environment identification thread is trained in combination with the theme so that the trained geological environment identification thread can accurately obtain the current geological environment information label of the theme.

[0068] 302. For each first geological environment identification result, in combination with the confidence in the first geological environment identification result, the confidence in the first geological environment identification result is projected into a target confidence database to obtain a second geological environment identification result.

[0069] In a possible implementation, the process of projecting the confidence in any first geological environment identification result includes: determining a first confidence and a first loss value in combination with the first geological environment identification result, the first confidence being the minimum confidence in the first geological environment identification result, and the first loss value being the loss value of the maximum confidence in the first geological environment identification result and the first confidence; determining a second confidence and a second loss value in combination with a target confidence database, the second confidence being the minimum confidence in the target confidence database, and the second loss value being the loss value of the maximum confidence in the target confidence database and the second confidence; determining a plurality of third loss values ​​in combination with a plurality of confidences in the first geological environment identification result; determining a comparison result between the third loss value and the first loss value for each third loss value, determining the product of the comparison result and the second loss value, determining the sum of the product and the second confidence as the third confidence, the third loss value being the loss value of any confidence in the first geological environment identification result and the first confidence; optimizing each confidence in the first geological environment identification result to a corresponding third confidence, and obtaining a second geological environment identification result.

[0070] Among them, a plurality of third loss values ​​correspond to a plurality of confidences in the first geological environment identification result one by one, and the loss value of each confidence in the first geological environment identification result and the first confidence is determined as the third loss value, so that a plurality of third loss values ​​can be obtained. The third confidence is the confidence obtained by projecting the confidence in the first geological environment identification result into the target confidence database. For any confidence in the first geological environment identification result, the third confidence obtained by combining the third loss value corresponding to the confidence is the third confidence corresponding to the confidence.

[0071] In an embodiment of the present application, for each confidence level in the first geological environment identification result, each confidence level can be projected into the target confidence level database in the above manner to obtain a third confidence level corresponding to each confidence level. Each confidence level in the first geological environment identification result can be optimized to the corresponding third confidence level to obtain a second geological environment identification result to indicate the confidence level of each of a number of geological environment attributes in the positioning of each geological environment attribute described in the pending current geological environment information.

[0072] In an embodiment of the present application, the minimum confidence and maximum confidence in the first geological environment identification result and the minimum confidence and maximum confidence in the target confidence database are used to project the confidence in the first geological environment identification result into the target confidence database to ensure that the confidence obtained by the projection can both retain the size relationship between several confidences in the first geological environment identification result and ensure that the confidence in the second geological environment identification result is within the target confidence database, thereby ensuring the accuracy of the second geological environment identification result.

[0073] It should be noted that the embodiment of the present application utilizes the minimum confidence in the first geological environment identification result and the minimum confidence in the target confidence database to project the confidence in the first geological environment identification result into the target confidence database. In another embodiment, the maximum confidence in the first geological environment identification result and the maximum confidence in the target confidence database can also be utilized to project the confidence in the first geological environment identification result into the target confidence database. For example, the maximum confidence in the first geological environment identification result is called the fourth confidence, and the maximum confidence in the target confidence database is called the fifth confidence. In combination with several confidences in the first geological environment identification result, several fourth loss values ​​are determined. For each fourth loss value, the comparison result of the fourth loss value and the first loss value is determined, and the product of the comparison result and the second loss value is determined. The loss value of the fifth confidence and the product is determined as the third confidence. The fourth loss value is the loss value of the fourth confidence and any confidence in the first geological environment identification result; each confidence in the first geological environment identification result is optimized to the corresponding third confidence to obtain the second geological environment identification result.

[0074] In one possible implementation, step 302 includes: for each first geological environment identification result, optimizing it into a thread, combining the confidence in the first geological environment identification result, projecting the confidence in the first geological environment identification result into a target confidence database, and obtaining a second geological environment identification result.

[0075] The optimized thread can be any type of thread, for example, the optimized thread includes a FC (Full Connection) layer, or the optimized thread is a CNN (Convolutional Neural Network) or a Transformer.

[0076] 303. Determine a second number from the first numbers corresponding to the plurality of second geological environment identification results, the second number being the largest first number among the first numbers corresponding to the plurality of second geological environment identification results, the first number corresponding to the second geological environment identification result indicating the number of geological environment attribute locations corresponding to the second geological environment identification result.

[0077] In an embodiment of the present application, the number of geological environment attribute locations corresponding to the second geological environment identification result is equivalent to the number of geological environment attributes contained in the pending current geological environment information tag corresponding to the second geological environment identification result, and the first number is equivalent to the number of geological environment attributes contained in the pending current geological environment information tag corresponding to the geological environment identification thread. For the same current geological environment information, the data scale of the current geological environment information description obtained by different geological environment identification threads may be different, that is, the number of geological environment attributes contained in the current geological environment information tags obtained by different geological environment identification threads may be different, and the second number is equivalent to the maximum data scale of the pending current geological environment information description corresponding to several geological environment identification threads. Therefore, from the first numbers corresponding to several second geological environment identification results, the largest first number is determined, so as to optimize the number of confidences in the second geological environment identification result in combination with the largest first number, so that the data scale of the pending current geological environment information description corresponding to the optimized several second geological environment identification results is the same.

[0078] 304. On the premise that the first number corresponding to the third geological environment identification result is less than the second number, determine a third number, the third number being a loss value of the second number and the first number corresponding to the third geological environment identification result, and the third geological environment identification result being any one of the plurality of second geological environment identification results.

[0079] In an embodiment of the present application, among several second geological environment identification results, the first number corresponding to any second geological environment identification result is smaller than the second number, indicating that the data scale of the pending current geological environment information description corresponding to the second geological environment identification result is smaller than the maximum data scale, and the third number is equivalent to the loss value of the data scale of the pending current geological environment information description corresponding to the third geological environment identification result and the maximum data scale.

[0080] 305. In the third geological environment identification result, the confidence of each of the several geological environment attributes in the target geological environment attribute positioning of the third value is added, and the number of geological environment attribute positionings corresponding to the third geological environment identification result after adding reaches the second number, and the confidence of each geological environment attribute in the target geological environment attribute positioning is the first value, wherein, in the current geological environment information label corresponding to the third geological environment identification result after adding, the target geological environment attribute positioning is located after the remaining geological environment attribute positioning.

[0081] In an embodiment of the present application, the locations of the various geological environment attributes described in the pending current geological environment information are arranged in order in the pending current geological environment information label. On the premise that the first number corresponding to the third geological environment identification result is less than the second number, the confidence of each of the several geological environment attributes in the target positioning of the third numerical value is added to the third geological environment identification result, so that the data scale of the current geological environment information description corresponding to the added third geological environment identification result reaches the maximum data scale. In the current geological environment information label corresponding to the added third geological environment identification result, the target geological environment attribute positioning is located after the remaining geological environment attribute positioning, so that when the current geological environment information label is determined subsequently by the added third geological environment identification result, the original information output by the geological environment identification thread can be retained as much as possible.

[0082] The first value is an arbitrary value, for example, the first value is 0.

[0083] In an embodiment of the present application, the first number corresponding to the second geological environment identification result is equivalent to the number of geological environment attributes contained in the pending current geological environment information tag corresponding to the geological environment identification thread; for the same current geological environment information, the data scale of the current geological environment information description obtained by different geological environment identification threads may be different, that is, the number of geological environment attributes contained in the current geological environment information tags obtained by different geological environment identification threads may be different, and the second number is equivalent to the maximum data scale of the pending current geological environment information description corresponding to several geological environment identification threads. Therefore, the largest first number is determined from the first numbers corresponding to several second geological environment identification results, so that under the premise that the first number corresponding to any second confidence information is less than the largest first number, the confidence number in the second geological environment identification result is optimized in combination with the largest first number, so that the data scale of the pending current geological environment information description corresponding to the optimized several second geological environment identification results is the same, and in the current geological environment information label corresponding to the second geological environment identification result after adding the confidence, the target geological environment attribute positioning is located after the remaining geological environment attribute positioning, so as to retain the original information output by several geological environment identification threads as much as possible, so that when several second geological environment identification results are subsequently merged, the pending current geological environment information labels corresponding to several geological environment identification threads can be taken into account as much as possible to avoid damage to the original information output by the geological environment identification thread, so as to take into account the ability of each geological environment identification thread to obtain the current geological environment information description, so as to ensure that the current geological environment information label obtained subsequently can accurately identify the current geological environment information, and can reflect the meaning expressed by the original current geological environment information, thereby ensuring the accuracy of the current geological environment information description.

[0084] 306. Fusing the plurality of second geological environment identification results to obtain a fused geological environment identification result, wherein the fused geological environment identification result includes the confidence of each of the plurality of geological environment attributes in the positioning of each geological environment attribute described by the target current geological environment information.

[0085] In an embodiment of the present application, the confidences in the current several second geological environment identification results are in the same confidence database, and the data scale of the pending current geological environment information description corresponding to the current several second geological environment identification results is the same, and the several second geological environment identification results are fused to ensure the accuracy of the fused geological environment identification results.

[0086] In a possible implementation, step 306 includes: fusing a plurality of second geological environment identification results to obtain fused information, and fusing the fused information to obtain a fused geological environment identification result.

[0087] In the embodiment of the present application, several second geological environment identification results are first fused, so as to fuse the fusion information, so that different second geological environment identification results can be fully integrated, thereby ensuring the accuracy of the fused geological environment identification result.

[0088] In a possible implementation, step 306 includes: fusing a plurality of second geological environment identification results through a thread to obtain a fused geological environment identification result.

[0089] Here, the thread is any type of thread, for example, the thread is composed of 2 FC (Fully Connected Layer) and ReLU (Rectified Linear Unit, an activation function) layers; or, the thread is CNN or Transformer.

[0090] 307. Combine and integrate the geological environment identification results to obtain the target geological environment identification results.

[0091] In a possible implementation, step 307 includes the following two methods.

[0092] The first method: Combine the fusion geological environment identification results and several geological environment attributes to obtain several sample geological environment information labels and the confidence corresponding to each sample geological environment information label. The confidence corresponding to the sample geological environment information label indicates the confidence of the geological environment information example description obtained in combination with the fusion geological environment identification results. The geological environment attributes located at the same geological environment attribute location in different sample geological environment information labels are not exactly the same; combined with the confidence of several geological environment information example descriptions, the sample geological environment information label with the largest confidence is determined as the target geological environment identification result.

[0093] In an embodiment of the present application, the fused geological environment identification result includes the confidence of each geological environment attribute in the positioning of each geological environment attribute described in the current geological environment information of the target. Then, according to the fused geological environment identification result, in the process of locating and determining the geological environment attributes for each geological environment attribute described in the current geological environment information of the target, several sample geological environment information labels can be determined, and the geological environment attributes located at the same geological environment attribute location in different sample geological environment information labels are not exactly the same, and combined with the confidence of each geological environment attribute in each sample geological environment information label in the positioning of each geological environment attribute, each sample geological environment information label is determined. The confidence of the example description of geological environment information indicates the confidence of the example description of geological environment information obtained by combining the fusion geological environment identification result. It can also reflect the accuracy of the sample geological environment information label in identifying the current geological environment information, and can also reflect the accuracy of the meaning reflected by the sample geological environment information label, so as to determine the sample geological environment information label with the highest confidence as the target geological environment identification result, to ensure that the target geological environment identification result can accurately identify the current geological environment information, and can reflect the meaning expressed by the original current geological environment information, thereby ensuring the accuracy of the current geological environment information description.

[0094] In one possible implementation, the process of determining an example description of geological environment information includes: determining a geological environment attribute for each geological environment attribute from a number of geological environment attributes, forming a sample geological environment information label with the determined geological environment attributes, and determining the confidence of the example description of geological environment information based on the confidence of the geological environment attributes contained in the sample geological environment information label at the location of each geological environment attribute.

[0095] In an embodiment of the present application, the Viterbi algorithm is adopted, combined with several geological environment attributes and fused geological environment identification results, to determine several sample geological environment information labels and determine the confidence of each geological environment information example description to ensure the accuracy of the obtained confidence.

[0096] Optionally, the process of determining the confidence of the example description of geological environment information includes: determining the confidence of each geological environment attribute in the sample geological environment information label at the corresponding geological environment attribute location from the fused geological environment identification result, and determining the target confidence corresponding to each geological environment attribute in the sample geological environment information label; and determining the product of the confidence of each geological environment attribute in the sample geological environment information label at the corresponding geological environment attribute location and the target confidence corresponding to each geological environment attribute as the confidence of the example description of geological environment information. Wherein, taking the geological environment attribute located at the first geological environment attribute location in the current geological environment information label as an example, the target confidence of the confidence corresponding to the geological environment attribute indicates the confidence of determining the geological environment attribute as the geological environment attribute located at the first geological environment attribute location on the premise that the geological environment attribute at the second geological environment attribute location has been determined, and the second geological environment attribute location is the geological environment attribute location before the first geological environment attribute location.

[0097] In an embodiment of the present application, for the sample geological environment information label, not only the confidence of each geological environment attribute in the sample geological environment information label at the corresponding geological environment attribute location is determined, but also the target confidence corresponding to each geological environment attribute is determined, so as to determine the confidence of the geological environment attribute as the geological environment attribute located at the next geological environment attribute location on the premise that the geological environment attribute of the previous geological environment attribute location has been determined, and then the determined confidence product is used as the confidence of the geological environment information example description, so that the confidence of the geological environment information example description can reflect the semantic coherence and contextual relevance of the geological environment information example description, and then reflect the degree of matching between the sample geological environment information label and the current geological environment information, so that the confidence can reflect the quality of the geological environment information example description and ensure the accuracy of the confidence.

[0098] Optionally, the process of determining the target confidence corresponding to the geological environment attribute includes: determining the first word type to which the geological environment attribute at the first geological environment attribute location in the current geological environment information tag belongs and the second word type to which the geological environment attribute at the second geological environment attribute location belongs; combining the first word type and the second word type, querying the word type information, and determining the queried confidence as the target confidence corresponding to the geological environment attribute at the first geological environment attribute location.

[0099] The second method: Combine and integrate the geological environment identification results and several geological environment attributes, determine the geological environment attributes located at each geological environment attribute location, and compose the determined geological environment attributes into the target geological environment identification results.

[0100] In an embodiment of the present application, the geological environment identification result and several geological environment attributes are combined and integrated, and the geological environment attributes at each geological environment attribute location are determined separately according to the order in which each geological environment attribute is located in the target geological environment identification result, so as to ensure that the confidence corresponding to the determined geological environment attribute is large enough, and then the determined geological environment attributes are composed of the target geological environment identification result, so as to ensure that the target geological environment identification result can accurately identify the current geological environment information, can reflect the meaning expressed by the original current geological environment information, and thus ensure the accuracy of the description of the current geological environment information.

[0101] Optionally, for any geological environment attribute location, the confidence of several geological environment attributes on the geological environment attribute location is combined, and the geological environment attribute corresponding to the maximum confidence is determined as the geological environment attribute located on the geological environment attribute location, and the determined geological environment attributes constitute the target geological environment identification result.

[0102] In an embodiment of the present application, the confidence of several geological environment attributes on the same geological environment attribute location, the greater the confidence, the greater the confidence that the geological environment attribute is located at the geological environment attribute location. Therefore, combined with the confidence of several geological environment attributes on the geological environment attribute location, the geological environment attribute corresponding to the maximum confidence is determined as the geological environment attribute located at the geological environment attribute location, and the determined geological environment attributes constitute the target geological environment identification result to ensure the accuracy of the description of the determined target current geological environment information.

[0103] In the solution provided by the embodiment of the present application, different geological environment identification threads have different capabilities of obtaining the description of the current geological environment information. Therefore, when obtaining the current geological environment information label of any current geological environment information, the same current geological environment information is processed by several geological environment identification threads to obtain the geological environment identification results corresponding to each geological environment identification thread, and the confidences in the geological environment identification results corresponding to the several geological environment identification threads are projected into the same confidence database, and then the geological environment identification results corresponding to the several geological environment identification threads are fused, so as to use the fused geological environment identification results to obtain the final current geological environment information label of the current geological environment information. In this way, by projecting the confidences in the geological environment identification results corresponding to different geological environment identification threads into the same confidence database, it is avoided that the current geological environment information label represented by the fused geological environment identification result has grammatical errors and inaccurate identification of the current geological environment information due to different confidence distributions in the several geological environment identification results, and the capabilities of several geological environment identification threads to obtain the description of the current geological environment information are taken into account, so as to ensure the accuracy of the obtained current geological environment information label, and to accurately identify the current geological environment information, thereby ensuring the accuracy of the current geological environment information description.

[0104] Another method for identifying a geological environment based on a Beidou satellite is provided in an embodiment of the present application. The method is executed by:

[0105] 601. Through several geological environment identification threads, the geological environment information example is processed respectively to obtain a fourth geological environment identification result corresponding to each geological environment identification thread, the fourth geological environment identification result includes the confidence of each geological environment attribute of several geological environment attributes in the positioning of each geological environment attribute described in the first current geological environment information, and the first current geological environment information label is the label of the geological environment information example.

[0106] The geological environment information example is any type of current geological environment information, for example, the geological environment information example is a current geological environment information of a subject or a residual type. The fourth geological environment identification result can be represented in any form, for example, the fourth geological environment identification result is represented in the form of a matrix, the rows of the matrix correspond to a number of geological environment attributes, and the columns of the matrix correspond to the location of each geological environment attribute.

[0107] In an embodiment of the present application, the fourth geological environment identification result corresponding to the geological environment identification thread is used to represent the first current geological environment information label corresponding to the geological environment identification thread, which can reflect the confidence of each of the several geological environment attributes at the respective geological environment attribute locations described in the first current geological environment information. The respective geological environment attribute locations described in the first current geological environment information are used to carry the geological environment attributes contained in the first current geological environment information label. The number of geological environment attribute locations corresponding to the first current geological environment information label is equivalent to the number of geological environment attributes contained in the first current geological environment information label, that is, by filling in the corresponding geological environment attributes at the respective geological environment attribute locations described in the first current geological environment information, the first current geological environment information label can be obtained.

[0108] It should be noted that the process of obtaining the fourth geological environment identification result corresponding to the geological environment identification thread is the same as the above step 301 and will not be repeated here.

[0109] 602. For each fourth geological environment identification result, combined with the confidence in the fourth geological environment identification result, the confidence in the fourth geological environment identification result is projected into the target confidence database to obtain a fifth geological environment identification result.

[0110] 603. Through the thread, several fifth geological environment identification results are integrated to obtain a sixth geological environment identification result, the sixth geological environment identification result includes the confidence of each of the several geological environment attributes in the positioning of each geological environment attribute described in the second current geological environment information, and the second current geological environment information label is the label of the geological environment information example.

[0111] 604. Train the thread based on the sample label of the geological environment information example and the sixth geological environment identification result.

[0112] In the embodiment of the present application, the sample label of the geological environment information example is the real label of the geological environment information example, and the sample label can accurately identify the geological environment information example. The sixth geological environment identification result is obtained through the thread, and the sixth geological environment identification result is used to represent the second current geological environment information label. Then, in combination with the sample label and the sixth geological environment identification result, the difference between the sample label and the second current geological environment information label can be determined to reflect the accuracy of the thread. Therefore, in combination with the sample label and the sixth geological environment identification result, the thread is trained so that when the sixth geological environment identification result is obtained through the trained thread, the second current geological environment information label represented by the sixth geological environment identification result is similar enough to the sample label to improve the accuracy of the thread.

[0113] In one possible implementation, step 604 includes: determining a sample geological environment identification result in combination with the sample label, the sample geological environment identification result including the confidence of each of several geological environment attributes in the positioning of each geological environment attribute described by the sample, wherein the confidence of the geological environment attribute in the sample label in the corresponding geological environment attribute positioning is a second value, and the confidence in the positioning of the remaining geological environment attributes is a third value; training the thread in combination with the sample geological environment identification result and the sixth geological environment identification result.

[0114] In an embodiment of the present application, a sample geological environment identification result is determined in combination with a sample label, so that the sample geological environment identification result can reflect the confidence of each of the several geological environment attributes in the positioning of each geological environment attribute described by the sample, so as to compare the sample geological environment identification result with the confidence in the sixth geological environment identification result, and it is easier to determine the difference between the two, which can reflect the accuracy of the thread. Therefore, the thread is trained in combination with the sample geological environment identification result and the sixth geological environment identification result to improve the accuracy of the thread.

[0115] Optionally, the number of geological environment attribute locations corresponding to the sample label is different from the number of geological environment attribute locations corresponding to the second current geological environment information label, that is, for the sample geological environment identification result and the sixth geological environment identification result, according to the above steps 303-305, the number of geological environment attribute locations corresponding to the sample geological environment identification result and the confidence level included are optimized, or the number of geological environment attribute locations corresponding to the sixth geological environment identification result and the confidence level included are optimized so that the number of geological environment attribute locations corresponding to the sample geological environment identification result and the sixth geological environment identification result are the same. Afterwards, the thread is trained in combination with the optimized sample geological environment identification result and the sixth geological environment identification result. Alternatively, the thread is trained in combination with the sample geological environment identification result and the optimized sixth geological environment identification result.

[0116] In one possible implementation, taking the optimization of the sample geological environment identification results as an example, the thread training process includes: combining the optimized sample geological environment identification results and the sixth geological environment identification results, determining the first loss value, and training the thread in combination with the first loss value.

[0117] In an embodiment of the present application, the number of geological environment attribute locations corresponding to the optimized sample geological environment identification result and the sixth geological environment identification result is the same, and a loss value is determined to represent the difference between the optimized sample geological environment identification result and the sixth geological environment identification result. The thread is trained in combination with the loss value to ensure the accuracy of the thread training.

[0118] It should be noted that the above embodiment is to train a thread in combination with a geological environment information example as an example, and in another embodiment, in accordance with the above method, the thread is iteratively trained in combination with several geological environment information examples and sample labels of the geological environment information examples. When the number of iterations reaches the number threshold, or when the current loss value is less than the loss value threshold, the thread training is stopped. The number threshold is an arbitrary value, and the loss value threshold is an arbitrary value.

[0119] In one possible implementation, the process of obtaining a sample description includes: obtaining a geological environment information example and thread indication information, wherein the thread indication information instructs the geological environment identification thread to process the input current geological environment information to obtain the current geological environment information label; by generating a thread, combining the thread indication information, processing the geological environment information example, and obtaining a sample label of the geological environment information example.

[0120] In the embodiment of the present application, the labels are optimized to ensure that the sample labels can accurately identify the geological environment information examples to ensure the accuracy of the sample description.

[0121] Optionally, on the premise of obtaining several geological environment information examples, multiple types of geological environment information examples are evenly sampled to balance the number of geological environment information examples of each type. According to the above method, the label of each geological environment information example obtained by sampling is obtained by generating a thread, and the obtained label is manually optimized, and the optimized label is used as the sample label of the geological environment information example.

[0122] The process of projecting the confidence to the target confidence database is performed by the optimized thread. In the process of training the thread, the optimized thread is also trained at the same time. That is, the process of training the thread includes: combining the sample label of the geological environment information example and the sixth geological environment identification result, training the thread and the optimized thread.

[0123] In an embodiment of the present application, the sixth geological environment identification result is obtained by optimizing into threads and threads, and the sixth geological environment identification result is used to represent the second current geological environment information label. Then, in combination with the sample label and the sixth geological environment identification result, the difference between the sample label and the second current geological environment information label can be determined to reflect the accuracy of the optimization into threads and threads. Therefore, in combination with the sample label and the sixth geological environment identification result, the threads and threads optimized into threads are trained so that when the sixth geological environment identification result is obtained again through the trained threads and threads optimized into threads, the second current geological environment information label represented by the sixth geological environment identification result is sufficiently similar to the sample label to improve the accuracy of the thread.

[0124] In one possible implementation, step 604 includes: determining a sample geological environment identification result in combination with a sample label, the sample geological environment identification result including the confidence of each of a number of geological environment attributes in the positioning of each geological environment attribute described by the sample, wherein the confidence of the geological environment attribute in the sample label in the corresponding geological environment attribute positioning is a second value, and the confidence in the positioning of the remaining geological environment attributes is a third value; training the thread and optimizing it into a thread in combination with the sample geological environment identification result and the sixth geological environment identification result.

[0125] In an embodiment of the present application, the sample geological environment identification result is determined in combination with the sample label, so as to compare the sample geological environment identification result with the sixth geological environment identification result, and then determine the difference between the two. The difference can reflect the accuracy of the thread. Therefore, in combination with the sample geological environment identification result and the sixth geological environment identification result, the thread and the thread optimized are trained to improve the accuracy of the thread and the thread optimized.

[0126] Before the training thread, the embodiment of the present application also needs to train the geological environment identification thread, and the training process of the geological environment identification thread includes: compressing the geological environment information example through the geological environment identification thread to obtain the characteristics of the geological environment information example; decoding the characteristics of the geological environment information example through the geological environment identification thread to obtain the seventh geological environment identification result, the seventh geological environment identification result includes the confidence of each geological environment attribute in the positioning of each geological environment attribute described in the third current geological environment information, and the third current geological environment information label is the label of the geological environment information example; combining the sample label of the geological environment information example and the seventh geological environment identification result, the geological environment identification thread is trained.

[0127] Among them, the feature of the geological environment information example is used to characterize the geological environment information example, and the feature can be expressed in any form, for example, the feature is expressed in the form of a vector. The fourth geological environment identification result is the same as the first geological environment identification result mentioned above, and will not be repeated here. In an embodiment of the present application, the geological environment information example is processed by the geological environment identification thread, and the difference between the sixth geological environment identification result and the sample label obtained can reflect the accuracy of the thread. Then, the geological environment identification thread is trained in combination with the sample label of the geological environment information example and the seventh geological environment identification result, so that when the seventh geological environment identification result is obtained by the trained geological environment identification thread, the current geological environment information label represented by the seventh geological environment identification result is sufficiently similar to the sample label to improve the accuracy of the geological environment identification thread.

[0128] In the solution provided by the embodiment of the present application, the current geological environment information label of the current geological environment information is obtained by using several geological environment identification threads, optimized into threads and threads, and the current geological environment information labels obtained by different geological environment identification threads for the same current geological environment information are fused to ensure that the obtained current geological environment information labels can accurately identify the current geological environment information, so as to ensure the accuracy of the description of the current geological environment information. Moreover, in the process of training the thread, the thread parameters of the thread can be optimized in a self-learning manner, so that the thread has the ability to fuse the current geological environment information description of the same current geological environment information output by several geological environment identification threads, so as to ensure the fusion effect of the thread.

[0129] In the solution provided by the embodiment of the present application, the current geological environment information geological environment identification results of the same current geological environment information are obtained through several geological environment identification threads, and the confidences in the several current geological environment information geological environment identification results are projected into the same area, so that the confidence databases corresponding to the several current geological environment information geological environment identification results are aligned, so as to avoid the differences in the confidence databases corresponding to the several current geological environment information geological environment identification results and affect the subsequent fusion effect, and the number of geological environment attribute positioning corresponding to the several current geological environment information geological environment identification results is optimized, so that the current geological environment information labels represented by the optimized several current geological environment information geological environment identification results are the same, and then the optimized several current geological environment information geological environment identification results are fused, so as to retain the original information output by the geological environment identification thread, thereby avoiding artificial damage to the natural language expression in the output results of the several geological environment identification threads, and combining the fused geological environment identification results to obtain the final current geological environment information label of the current geological environment information, so that the current geological environment information label can accurately identify the current geological environment information, so as to ensure the accuracy of the current geological environment information description.

[0130] On the basis of the above, a geological environment identification device based on Beidou satellite is provided, and the device includes:

[0131] A first result obtaining module is used to process the current geological environment information through a plurality of geological environment identification threads respectively, and obtain a first geological environment identification result corresponding to each geological environment identification thread, wherein the first geological environment identification result includes the confidence of each geological environment attribute in the plurality of geological environment attributes in the positioning of each geological environment attribute described in the pending current geological environment information;

[0132] A second result module is used for, for each first geological environment identification result, combining the confidence in the first geological environment identification result, projecting the confidence in the first geological environment identification result into a target confidence database to obtain a second geological environment identification result;

[0133] A result fusion module is used to fuse the plurality of second geological environment identification results to obtain a fused geological environment identification result, wherein the fused geological environment identification result includes the confidence of each of the plurality of geological environment attributes in the positioning of each geological environment attribute described by the target current geological environment information;

[0134] The result recognition module is used to combine the fused geological environment recognition result to obtain the target geological environment recognition result.

[0135] On the basis of the above, a geological environment identification system based on Beidou satellite is shown, including a processor and a memory that communicate with each other, and the processor is used to read and execute a computer program from the memory to implement the above method.

[0136] Based on the above, a computer-readable storage medium is also provided, on which a computer program stored implements the above method when running.

[0137] In summary, based on the above scheme, different geological environment identification threads have different abilities to obtain the description of the current geological environment information. Therefore, when obtaining the current geological environment information label of any current geological environment information, the same current geological environment information is processed by several geological environment identification threads to obtain the geological environment identification results corresponding to each geological environment identification thread, and the confidences in the geological environment identification results corresponding to the several geological environment identification threads are projected into the same confidence database, and then the geological environment identification results corresponding to the several geological environment identification threads are fused, so as to use the fused geological environment identification results to obtain the final current geological environment information label of the current geological environment information. In this way, the confidences in the geological environment identification results corresponding to different geological environment identification threads are projected into the same confidence database to avoid the situation in which the current geological environment information label represented by the fused geological environment identification results has grammatical errors and inaccurate identification of the current geological environment information due to different confidence distributions in the several geological environment identification results, and the ability of several geological environment identification threads to obtain the description of the current geological environment information is taken into account, so as to ensure the accuracy of the obtained current geological environment information label, and to accurately identify the current geological environment information, thereby ensuring the accuracy of the current geological environment information description.

[0138] It should be understood that the system and its modules shown above can be implemented in various ways. For example, in some embodiments, the system and its modules can be implemented by hardware, software, or a combination of software and hardware. Among them, the hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or a dedicated design hardware. Those skilled in the art will understand that the above methods and systems can be implemented using computer executable instructions and / or included in a processor control code, such as a carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. Such code is provided on the carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The system and its modules of the present application can not only be implemented by hardware circuits such as ultra-large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., but can also be implemented by software such as executed by various types of processors, and can also be implemented by a combination of the above hardware circuits and software (e.g., firmware).

[0139] It should be noted that different embodiments may produce different beneficial effects. In different embodiments, the beneficial effects that may be produced may be any one or a combination of the above, or any other beneficial effects that may be obtained.

Claims

1. A method for identifying geological environment based on Beidou satellite, characterized in that: The method comprises: The current geological environment information is processed respectively through a plurality of geological environment identification threads to obtain a first geological environment identification result corresponding to each geological environment identification thread, wherein the first geological environment identification result includes a confidence level of each geological environment attribute in the plurality of geological environment attributes in the positioning of each geological environment attribute described in the pending current geological environment information; For each first geological environment identification result, combining the confidence in the first geological environment identification result, projecting the confidence in the first geological environment identification result into the target confidence database, to obtain a second geological environment identification result; Fusing the plurality of second geological environment identification results to obtain a fused geological environment identification result, wherein the fused geological environment identification result includes the confidence of each of the plurality of geological environment attributes in the positioning of each geological environment attribute described by the target current geological environment information; Combined with the fused geological environment identification result, a target geological environment identification result is obtained.

2. The method according to claim 1, characterized in that For each first geological environment identification result, combining the confidence in the first geological environment identification result, projecting the confidence in the first geological environment identification result into the target confidence database to obtain the second geological environment identification result, including: Determine a first confidence level and a first loss value in combination with the first geological environment identification result, wherein the first confidence level is the minimum confidence level in the first geological environment identification result, and the first loss value is the maximum confidence level in the first geological environment identification result and the loss value of the first confidence level; Determine a second confidence and a second loss value in combination with the target confidence database, wherein the second confidence is the minimum confidence in the target confidence database, and the second loss value is the loss value of the maximum confidence in the target confidence database and the second confidence; In combination with several confidence levels in the first geological environment identification result, several third loss values ​​are determined; for each third loss value, a comparison result of the third loss value and the first loss value is determined, a product of the comparison result and the second loss value is determined, and a sum of the product and the second confidence level is determined as a third confidence level, where the third loss value is a loss value of any confidence level in the first geological environment identification result and the first confidence level; Each confidence level in the first geological environment identification result is optimized to a corresponding third confidence level to obtain the second geological environment identification result.

3. The method according to claim 1, characterized in that The step of combining the fused geological environment identification result to obtain the target geological environment identification result includes: In combination with the fused geological environment identification result and the plurality of geological environment attributes, a plurality of sample geological environment information labels and the confidence corresponding to each sample geological environment information label are obtained, wherein the confidence corresponding to the sample geological environment information label indicates the confidence of the geological environment information example description obtained in combination with the fused geological environment identification result, and the geological environment attributes located at the same geological environment attribute location in different sample geological environment information labels are not completely the same; Combined with the confidences of the several geological environment information example descriptions, the sample geological environment information label with the greatest confidence is determined as the target geological environment identification result.

4. The method according to claim 1, characterized in that: The step of combining the fused geological environment identification result to obtain the target geological environment identification result includes: The fused geological environment identification result and the plurality of geological environment attributes are combined to determine the geological environment attribute located at each geological environment attribute location, and the determined geological environment attributes are combined into the target geological environment identification result.

5. The method according to claim 1, characterized in that Before fusing the plurality of second geological environment identification results to obtain a fused geological environment identification result, the method further includes: Determine a second number from the first numbers corresponding to the plurality of second geological environment identification results, the second number being the largest first number among the first numbers corresponding to the plurality of second geological environment identification results, the first number corresponding to the second geological environment identification result indicating the number of geological environment attribute locations corresponding to the second geological environment identification result; Under the premise that the first number corresponding to the third geological environment identification result is less than the second number, determining a third number, the third number being a loss value of the second number and the first number corresponding to the third geological environment identification result, the third geological environment identification result being any one of the plurality of second geological environment identification results; In the third geological environment identification result, the confidence of each of the several geological environment attributes in the third number of target geological environment attribute locations is added, and the number of geological environment attribute locations corresponding to the third geological environment identification result after addition reaches the second number, and the confidence of each geological environment attribute in the target geological environment attribute location is a first value, wherein, in the current geological environment information label corresponding to the third geological environment identification result after addition, the target geological environment attribute location is located after the remaining geological environment attribute locations.

6. The method according to claim 1, characterized in that The fusing of the plurality of second geological environment identification results to obtain the fused geological environment identification result is performed by a thread, and the method further includes: Through the plurality of geological environment identification threads, the geological environment information examples are processed respectively to obtain a fourth geological environment identification result corresponding to each of the geological environment identification threads, wherein the fourth geological environment identification result includes the confidence of each of the plurality of geological environment attributes in the positioning of each geological environment attribute described in the first current geological environment information, and the first current geological environment information label is the label of the geological environment information example; For each fourth geological environment identification result, combining the confidence in the fourth geological environment identification result, projecting the confidence in the fourth geological environment identification result into the target confidence database to obtain a fifth geological environment identification result; Through the thread, a plurality of fifth geological environment identification results are merged to obtain a sixth geological environment identification result, wherein the sixth geological environment identification result includes the confidence of each of the plurality of geological environment attributes in the positioning of each geological environment attribute described in the second current geological environment information, and the second current geological environment information label is the label of the geological environment information example; The thread is trained in combination with the sample label of the geological environment information example and the sixth geological environment identification result.

7. The method according to claim 6, characterized in that The step of training the thread in combination with the sample label of the geological environment information example and the sixth geological environment identification result includes: In combination with the sample label, a sample geological environment identification result is determined, wherein the sample geological environment identification result includes the confidence of each of the several geological environment attributes in the location of each geological environment attribute described by the sample, wherein the confidence of the geological environment attribute in the sample label in the corresponding geological environment attribute location is a second value, and the confidence of the remaining geological environment attribute locations is a third value; The thread is trained in combination with the sample geological environment identification result and the sixth geological environment identification result.

8. The method according to claim 6, characterized in that The step of: for each fourth geological environment identification result, combining the confidence in the fourth geological environment identification result, projecting the confidence in the fourth geological environment identification result into the target confidence database, and obtaining the fifth geological environment identification result is performed by optimizing the thread, and combining the sample label of the geological environment information example and the sixth geological environment identification result to train the thread, including: In combination with the sample label of the geological environment information example and the sixth geological environment identification result, the thread and the optimized thread are trained.

9. The method according to claim 1, characterized in that: The method further comprises: Compressing the geological environment information example through the geological environment identification thread to obtain the characteristics of the geological environment information example; By means of the geological environment identification thread, the features of the geological environment information example are decoded to obtain a seventh geological environment identification result, wherein the seventh geological environment identification result includes the confidence of each of the plurality of geological environment attributes in the positioning of each geological environment attribute described in the third current geological environment information, and the third current geological environment information label is the label of the geological environment information example; The geological environment identification thread is trained in combination with the sample label of the geological environment information example and the seventh geological environment identification result.

10. A geological environment identification system based on Beidou satellite, characterized in that: The invention comprises a processor and a memory communicating with each other, wherein the processor is used to read a computer program from the memory and execute the computer program to implement the method according to any one of claims 1 to 9.

Citation Information

Patent Citations

  • Method and equipment for identifying and positioning unfavorable geologic body through airborne electromagnetic method in railway tunnel

    CN115346114A

  • Multi-source geological survey data fusion method and device, equipment and storage medium

    CN116304977A

  • Geotechnical engineering geological disaster prediction method and system based on deep learning

    CN117093919A

  • Method for classifying eco-geological environment types based on coal resource exploitation

    US20200234170A1

  • Displaying confidence values in wellbore inversion modeling using a visual indicator

    US20240068348A1