CN117933577A - Evaluation method and system for landslide disaster in high level - Google Patents

Abstract

The application relates to the technical field of data evaluation, in particular to an evaluation method and an evaluation system for landslide disasters of a high-level mountain, wherein a target secondary example high-level ramp hidden danger information set obtained after target example high-level ramp hidden danger information set analysis of a target high-level ramp hidden danger information set is used for building example important description contents through a target necessary example high-level ramp hidden danger information set, so that an analysis thread of the high-level ramp hidden danger information can pay attention to important description contents covered in the high-level ramp hidden danger information, the important description contents are ensured not to be lost while the high-level ramp hidden danger information is analyzed, and further the target high-level ramp hidden danger information of the target high-level ramp hidden danger information set obtained by analyzing the high-level ramp hidden danger information to be processed is consistent with the important description contents covered in the high-level ramp hidden danger information to be processed, and the high-level ramp hidden danger information processing effect is improved.

Description

Evaluation method and system for landslide disaster in high level

Technical Field

The application relates to the technical field of data evaluation, in particular to a method and a system for evaluating high-level landslide disasters.

Background

The high-level landslide is characterized in that the shearing outlet is higher than the ground of the slope foot, the gravity center of the landslide body and the shearing outlet are high, the temporary condition is good, and the landslide has extremely high potential energy. The high-level landslide can be divided into a sliding source area, a sliding flow area and a stacking area, and the movement speeds of the areas show different rules; the control of high-level landslide should consider the factors such as fully utilizing the topography advantage, strengthening crowd consciousness, paying attention to investigation and monitoring, especially the topography condition has important influence on the site selection of the building and disaster control structure; deep researches on a high-level landslide quantification system are urgently carried out.

At present, according to the prior art, the monitoring of the high-level landslide is very difficult, due to the complexity of the topography, the related technicians are difficult to go to the region, and the high-level landslide data acquired under such premise may be abnormal or inaccurate, so that the high-level landslide disaster is difficult to evaluate, and therefore, a technical scheme is needed to improve the technical problems.

Disclosure of Invention

In order to improve the technical problems in the related art, the application provides a method and a system for evaluating high mountain landslide disasters.

In a first aspect, a method for evaluating a landslide hazard at a high level is provided, including:

obtaining hidden danger information of a high-level ramp to be processed of a detection object;

Loading the hidden danger information of the high-level ramp to be processed into an analysis thread of hidden danger information of the high-level ramp, wherein the analysis thread of hidden danger information of the high-level ramp is configured according to quality indexes of important descriptive contents, the quality indexes of the important descriptive contents are built according to example important descriptive contents in a second example hidden danger information set of the high-level ramp obtained after analysis of a first example hidden danger information set of the detected object and example important descriptive contents in a target secondary example hidden danger information set obtained after analysis of a target example hidden danger information set of the high-level ramp;

the method comprises the steps that the high-level ramp hidden danger information analysis thread is obtained, the high-level ramp hidden danger information to be processed is analyzed to obtain target high-level ramp hidden danger information of the target high-level ramp hidden danger information set, and important description contents covered in the target high-level ramp hidden danger information are consistent with important description contents covered in the high-level ramp hidden danger information to be processed.

In the application, before loading the to-be-processed high-level ramp hidden danger information to the high-level ramp hidden danger information analysis thread, the method further comprises the following steps:

Obtaining the first example high-level ramp hidden danger information set, the target example high-level ramp hidden danger information set and a thread to be configured, wherein the thread to be configured comprises a first artificial intelligence analysis thread and a second artificial intelligence analysis thread, and the example high-level ramp hidden danger information in the first example high-level ramp hidden danger information set and the target example high-level ramp hidden danger information set both have an example catalog for representing the example important description content;

Loading the first example high-level ramp hidden danger information set to the first artificial intelligent analysis thread, analyzing to obtain a second example high-level ramp hidden danger information set of the target high-level ramp hidden danger information set with the example important description content, loading the target example high-level ramp hidden danger information set to the second artificial intelligent analysis thread, and analyzing to obtain a target secondary example high-level ramp hidden danger information set of the detection object with the example important description content;

establishing the important descriptive content quality index by combining the example important descriptive content in the target secondary example high-level ramp hidden trouble information set and the example important descriptive content in the second example high-level ramp hidden trouble information set;

and updating the thread coefficient of the thread to be configured by combining the important descriptive content quality index to obtain the high-level ramp hidden danger information analysis thread.

It can be understood that, before the information is input to the high-level ramp hidden danger information analysis thread, if the information is wrong, the data analyzed by the high-level ramp hidden danger information analysis thread is wrong, and the step is to ensure that the data is accurate when the data is input to the high-level ramp hidden danger information analysis thread.

In the present application, the obtaining the first exemplary high-level ramp hidden danger information set, the target exemplary high-level ramp hidden danger information set, and the thread to be configured includes:

Collecting the example high-level ramp hidden danger information of the detection object, and collecting the example high-level ramp hidden danger information of the target high-level ramp hidden danger information set;

carrying out annotation of the example important description content on the acquired example high-level ramp hidden danger information of the detection object to obtain the first example high-level ramp hidden danger information set;

And annotating the example important description content on the local example high-level ramp hidden danger information of the collected target high-level ramp hidden danger information set to obtain the target example high-level ramp hidden danger information set.

It can be appreciated that the accuracy and reliability of obtaining the first example high-level ramp hidden danger information set, the target example high-level ramp hidden danger information set, and the threads to be configured are ensured by accurately obtaining the example high-level ramp hidden danger information.

In the present application, the building the quality index of the important descriptive content by combining the example important descriptive content in the target secondary example high-level ramp hidden trouble information set and the example important descriptive content in the second example high-level ramp hidden trouble information set includes:

Combining the distinction of the example important descriptive content in the second example high-level ramp hidden trouble information set and the example important descriptive content in the first example high-level ramp hidden trouble information set, and constructing a first important descriptive content quality index;

combining the sample important description content in the target secondary sample high-level ramp hidden danger information set with the sample important description content in the target sample high-level ramp hidden danger information set to establish a second important description content quality index;

And generating the important descriptive content quality index by combining the first important descriptive content quality index and the second important descriptive content quality index.

It can be appreciated that when the example important descriptive contents in the target secondary example high-level ramp hidden-danger information set and the example important descriptive contents in the second example high-level ramp hidden-danger information set are combined, the problem of inaccurate distinction of the example important descriptive contents is improved, so that the important descriptive content quality index can be accurately established.

In the application, before the thread coefficient of the thread to be configured is updated by combining the important descriptive content quality index, the method further comprises:

Loading the second example high-level ramp hidden danger information set into the second artificial intelligent analysis thread, and analyzing to obtain a newly built example high-level ramp hidden danger information set of the detection object;

loading the target secondary example high-level ramp hidden danger information set into the first artificial intelligent analysis thread, and analyzing to obtain a target rebuilt example high-level ramp hidden danger information set of the target high-level ramp hidden danger information set;

Setting up a re-setting quality index by combining the re-setting example high-level ramp hidden danger information set and the target re-setting example high-level ramp hidden danger information set;

It can be appreciated that the reliability and accuracy of the construction of the quality index of the re-construction are guaranteed by the re-construction of the exemplary high-level ramp hidden danger information set.

The updating the thread coefficient of the thread to be configured by combining the important descriptive content quality index comprises the following steps: and generating a thread quality index corresponding to the thread to be configured by combining the re-establishment quality index and the important description content quality index, and updating the thread coefficient of the thread to be configured by combining the thread quality index.

It can be appreciated that the accuracy of updating the thread coefficients of the threads to be configured is improved by the thread quality index.

In the application, the construction of the quality index for re-construction by combining the re-construction example high-level ramp hidden danger information set and the target re-construction example high-level ramp hidden danger information set comprises the following steps:

Building a first rebuilt quality index by combining the rebuilt example high-level ramp hidden danger information set and the commonality score of the first example high-level ramp hidden danger information set;

Combining the target re-establishment example high-level ramp hidden danger information set and the commonality score of the target example high-level ramp hidden danger information set to establish a second re-establishment quality index;

and generating the re-construction quality index by combining the first re-construction quality index and the second re-construction quality index.

It can be appreciated that when the re-construction example high-level ramp hidden danger information set and the target re-construction example high-level ramp hidden danger information set are combined, the problem of inaccurate commonality grading is solved, and therefore the re-construction quality index can be accurately constructed.

The thread to be configured further comprises a first analysis thread for judging the hidden danger information key of the high-level ramp on the target hidden danger information set of the high-level ramp and a second analysis thread for judging the hidden danger information key of the high-level ramp on the detection object; before the thread coefficient of the thread to be configured is updated by combining the thread quality index, the method further comprises:

loading the second exemplary high-level ramp hidden danger information set into the first analysis thread to obtain a first analysis result;

Loading the hidden danger information of the target secondary example high-level ramp into the second analysis thread to obtain a second analysis result;

constructing a quality evaluation by combining the first analysis result and the second analysis result to generate a quality index;

The step of generating the thread quality index corresponding to the thread to be configured by combining the rebuilt quality index and the important descriptive content quality index comprises the following steps: and combining the rebuilt quality index, the important descriptive content quality index and the quality index generated by quality evaluation to generate a quality index, and generating a thread quality index corresponding to the thread to be configured according to the quality index.

It will be appreciated that the accuracy of the thread quality indicator is improved by the quality indicator.

In the application, the step of constructing the combination of the first analysis result and the second analysis result to perform quality evaluation to generate a quality index includes:

loading the target example high-level ramp hidden danger information set into the first analysis thread to obtain a third analysis result;

loading the first example high-level ramp hidden danger information set into the second analysis thread to obtain a fourth analysis result;

building a first quality evaluation by combining the first analysis result and the third analysis result to generate a quality index;

constructing a second quality evaluation by combining the second analysis result and the fourth analysis result to generate a quality index;

And generating the quality index by combining the first quality index and the second quality index.

It can be understood that when the quality evaluation is performed by combining the first analysis result and the second analysis result, the problem of inaccurate analysis results is solved, so that the accuracy of generating quality indexes is improved.

The method for generating the thread quality index corresponding to the thread to be configured according to the quality index by combining the rebuilt quality index, the important descriptive content quality index and the quality index generated by quality evaluation comprises the following steps:

Obtaining the confidence coefficient of the rebuilt quality index and the confidence coefficient of the important descriptive content quality index;

Combining the confidence coefficient of the rebuilt quality index and the confidence coefficient of the important descriptive content quality index, and carrying out weighting treatment on the rebuilt quality index and the important descriptive content quality index to obtain an information quality index;

And combining the information quality index and the quality index generated by quality evaluation to generate a quality index, and generating a thread quality index corresponding to the thread to be configured according to the quality index.

It can be understood that the quality index is generated by combining the rebuilt quality index, the important description content quality index and the quality index generated by quality evaluation, and the problem of inaccurate information quality index is solved when the quality index is used, so that the thread quality index corresponding to the thread to be configured can be accurately generated.

In the present application, the updating the thread coefficient of the thread to be configured by combining the thread quality index includes:

Fixing thread coefficients of the first analysis thread and the second analysis thread, and updating the thread coefficients of the first artificial intelligent analysis thread and the second artificial intelligent analysis thread by combining the thread quality index;

And fixing the thread coefficients of the first artificial intelligence analysis thread and the second artificial intelligence analysis thread, and updating the thread coefficients of the first analysis thread and the second analysis thread by combining the quality index generated by quality evaluation.

It can be understood that the reliability of updating the thread coefficient of the thread to be configured is ensured by forming the quality index.

According to the application, the method for obtaining the target high-level ramp hidden danger information of the target high-level ramp hidden danger information set obtained by analyzing the high-level ramp hidden danger information to be processed by the high-level ramp hidden danger information analysis thread comprises the following steps:

And obtaining target high-level ramp hidden danger information of the target high-level ramp hidden danger information set obtained by analysis of the first artificial intelligent analysis thread of the high-level ramp hidden danger information analysis thread.

It can be understood that when the analysis thread of the hidden danger information of the high-level ramp is obtained to analyze the hidden danger information of the high-level ramp to be processed, the problem of inaccurate analysis is solved, so that the accuracy of the hidden danger information of the target high-level ramp of the obtained target hidden danger information set is improved.

According to the application, important description contents covered in the target high-level ramp hidden danger information are information annotated for target identification indication; after the obtaining the target high-level ramp hidden danger information of the target high-level ramp hidden danger information set obtained by analyzing the high-level ramp hidden danger information to be processed by the high-level ramp hidden danger information analysis thread, the method further comprises:

Acquiring an identification thread of a target bit ramp hidden danger information set, wherein the identification thread is used for executing the target identification instruction based on the important description content;

Obtaining a configuration set, wherein the configuration set comprises target-bit ramp hidden danger information and an important description content catalog for representing the important description content;

And configuring the identification thread in combination with the configuration set.

It can be understood that the reliability of the configuration of the identification thread is ensured by the target identification indication.

In a second aspect, there is provided an evaluation system for high mountain landslide hazard comprising a processor and a memory in communication with each other, the processor being adapted to read a computer program from the memory and execute it to implement the method described above.

According to the evaluation method and the evaluation system for the landslide hazard of the high-level mountain, the to-be-processed high-level ramp hidden danger information of the detection object is loaded to the high-level ramp hidden danger information analysis thread, the high-level ramp hidden danger information analysis thread is configured according to the important description content quality index, the important description content quality index is the example important description content in the second example high-level ramp hidden danger information set obtained after analysis according to the first example high-level ramp hidden danger information set of the detection object, and the example important description content in the target secondary example high-level ramp hidden danger information set obtained after analysis according to the target example high-level ramp hidden danger information set, so that the high-level ramp hidden danger information analysis thread can pay attention to the important description content covered in the high-level ramp hidden danger information, the important description content can not be lost while analyzing the high-level ramp hidden danger information, the target high-level ramp hidden danger information of the obtained high-level ramp hidden danger information set is analyzed according to the first example high-level ramp hidden danger information set of the detection object, and the important description content hidden danger information contained in the to be processed is consistent with the important description content hidden danger information of the to be processed, and the high-level hidden danger information to be processed is improved.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

Fig. 1 is a flowchart of an evaluation method for a landslide disaster in a high position according to an embodiment of the present application.

Detailed Description

In order to better understand the above technical solutions, the following detailed description of the technical solutions of the present application is made by using the accompanying drawings and specific embodiments, and it should be understood that the specific features of the embodiments and the embodiments of the present application are detailed descriptions of the technical solutions of the present application, and not limiting the technical solutions of the present application, and the technical features of the embodiments and the embodiments of the present application may be combined with each other without conflict.

Referring to fig. 1, a method for evaluating landslide hazard in high level is shown, which may include the following steps S210-S230.

S210, obtaining hidden danger information of the high-level ramp to be processed of the detection object.

In the embodiment of the application, the set of the hidden danger information of the high-level ramp refers to the area to which the hidden danger information of the high-level ramp belongs. The region may be partitioned according to a set of high ramp hidden trouble information.

By way of example, the detection object may be understood as a mountain.

The detection object refers to a set of high-level ramp hidden danger information to which the high-level ramp hidden danger information to be processed belongs, the high-level ramp hidden danger information to be processed can be any high-level ramp hidden danger information, the specified high-level ramp hidden danger information can be high-level ramp hidden danger information which is useful for subsequent indication, and the high-level ramp hidden danger information to be processed can be one or more and is not limited.

S220, loading hidden danger information of the high-level ramp to be processed into a hidden danger information analysis thread of the high-level ramp, wherein the hidden danger information analysis thread of the high-level ramp is configured according to important descriptive content quality indexes, and the important descriptive content quality indexes are built according to example important descriptive contents in a second example hidden danger information set obtained after analysis of a first example hidden danger information set of the detected object and example important descriptive contents in a target secondary example hidden danger information set obtained after analysis of a target example hidden danger information set of the target high-level ramp.

By way of example, the high-level ramp hidden danger information analysis thread can be understood as an artificial intelligence thread.

In the embodiment of the application, the high-level ramp hidden danger information analysis thread is a thread which is already configured and is used for analyzing the high-level ramp hidden danger information, namely analyzing the high-level ramp hidden danger information of one high-level ramp hidden danger information set into another high-level ramp hidden danger information set, for example, analyzing the high-level ramp hidden danger information style under one camera into the high-level ramp hidden danger information style of another camera, so that the high-level ramp hidden danger information of another high-level ramp hidden danger information set can be obtained after the to-be-processed high-level ramp hidden danger information is loaded into the high-level ramp hidden danger information analysis thread.

The high-level ramp hidden danger information analysis thread can be understood as an analysis model of landslide occurrence probability.

It is noted that the high-level ramp hidden danger information analysis thread is configured according to an important description content quality index, wherein the important description content quality index is constructed according to an example important description content in a second example high-level ramp hidden danger information set obtained after analysis of a first example high-level ramp hidden danger information set of a detection object and an example important description content in a target secondary example high-level ramp hidden danger information set obtained after analysis of a target example high-level ramp hidden danger information set of a target high-level ramp hidden danger information set.

Because the analysis is performed, the example important description content in the second example high-level ramp hidden danger information set and the example important description content in the target secondary example high-level ramp hidden danger information set may be the same or different, and then the important description content quality index is built through the example important description content in the second example high-level ramp hidden danger information set and the example important description content in the target secondary example high-level ramp hidden danger information set, so that the information missing condition of the high-level ramp hidden danger information in the analysis process is represented, and further the high-level ramp hidden danger information analysis thread can improve the information missing of the high-level ramp hidden danger information in the analysis process through the important description content quality index configuration, so that the consistency of the important description content in the high-level ramp hidden danger information is maintained while the high-level ramp hidden danger information is analyzed among different domains.

In one possible implementation embodiment, the server includes a plurality of analysis threads for analyzing the hidden danger information of the high-level ramp between different hidden danger information sets of the high-level ramp, after the server obtains the hidden danger information of the high-level ramp to be processed of the detection object, the server also needs to obtain the hidden danger information set of the high-level ramp to be analyzed, for example, to obtain the hidden danger information a of the high-level ramp to be processed of the hidden danger information set of the high-level ramp and the hidden danger information of the high-level ramp to be processed need to analyze the hidden danger information set of the high-level ramp, and then searches the corresponding analysis threads for analyzing the hidden danger information of the high-level ramp according to the hidden danger information set of the high-level ramp and the hidden danger information set of the high-level ramp B to be processed, so as to load the hidden danger information of the high-level ramp to be processed into the analysis threads for analyzing the hidden danger information of the high-level ramp.

S230, obtaining target high-level ramp hidden danger information of a target high-level ramp hidden danger information set obtained by analyzing the high-level ramp hidden danger information to be processed by the high-level ramp hidden danger information analysis thread, wherein important description contents contained in the target high-level ramp hidden danger information are consistent with important description contents contained in the high-level ramp hidden danger information to be processed.

By way of example, the objective high-level ramp hidden danger information may be understood as information such as a possibility of occurrence of a landslide and a risk level of the landslide obtained by analyzing a hidden danger position of a high-level landslide, where the risk level may include a low risk, a medium risk, a high risk, and the like. Therefore, the risk places can be timely assessed, and the life safety and property safety of people are guaranteed.

In the embodiment of the application, after the hidden danger information of the high-level ramp to be processed is loaded to the hidden danger information analysis thread of the high-level ramp, the hidden danger information of the target high-level ramp of the hidden danger information set of the target high-level ramp, which is obtained by analyzing the hidden danger information of the high-level ramp to be processed by the hidden danger information analysis thread of the high-level ramp, is directly obtained, wherein, because the hidden danger information analysis thread of the high-level ramp is built according to the quality index of the important descriptive content, the important descriptive content covered in the hidden danger information of the target high-level ramp is not lost, and is identical with the important descriptive content covered in the hidden danger information of the high-level ramp to be processed, that is, the hidden danger information of the target high-level ramp and the hidden danger information of the high-level ramp to be processed are identical.

In the embodiment of the application, the high-level ramp hidden danger information analysis thread is configured according to the important description content quality index, the important description content quality index is the example important description content in the second example high-level ramp hidden danger information set obtained after analysis according to the first example high-level ramp hidden danger information set of the detection object, and the example important description content in the target secondary example high-level ramp hidden danger information set obtained after analysis according to the target example high-level ramp hidden danger information set, so that the high-level ramp hidden danger information analysis thread can pay attention to the important description content covered in the high-level ramp hidden danger information, so that the important description content is not lost while the high-level ramp hidden danger information is analyzed, and the target high-level ramp hidden danger information of the target high-level ramp hidden danger information set obtained by analyzing the high-level ramp hidden danger information to be processed is consistent with the important description content covered in the hidden danger information of the high-level ramp to be processed, and the high-level ramp hidden danger information processing effect is improved.

In one embodiment of the present application, another method for evaluating landslide hazard in high level is provided, and steps S310 to S340 are added before S220.

S310, a first example high-level ramp hidden danger information set, a target example high-level ramp hidden danger information set and a thread to be configured are obtained, wherein the thread to be configured comprises a first artificial intelligent analysis thread and a second artificial intelligent analysis thread, and the first example high-level ramp hidden danger information set and the example high-level ramp hidden danger information in the target example high-level ramp hidden danger information set are provided with an example catalog for representing important description contents of examples.

In the embodiment of the application, the first exemplary high-level ramp hidden danger information set and the target exemplary high-level ramp hidden danger information set are obtained by receiving the first exemplary high-level ramp hidden danger information transmitted by one high-level ramp hidden danger information acquisition device and receiving the target exemplary high-level ramp hidden danger information set transmitted by another high-level ramp hidden danger information acquisition device. The first example high-level ramp hidden trouble information set comprises one or more first example high-level ramp hidden trouble information, the target example high-level ramp hidden trouble information set comprises one or more target example high-level ramp hidden trouble information, and the first example high-level ramp hidden trouble information and the target example high-level ramp hidden trouble information are provided with an example catalog which is used for representing example important descriptive contents.

It should be noted that, the thread to be configured includes a first artificial intelligence analysis thread and a second artificial intelligence analysis thread, the first artificial intelligence analysis thread is used for analyzing the hidden danger information of the high-level ramp of the detected object into the hidden danger information of the high-level ramp of the target hidden danger information set, and the second artificial intelligence analysis thread is used for analyzing the hidden danger information of the high-level ramp of the target hidden danger information set into the hidden danger information of the high-level ramp of the detected object.

The obtaining of the thread to be configured may be obtaining the first artificial intelligence analysis thread and the second artificial intelligence analysis thread which are not configured from the threads, or may also obtain the configured first artificial intelligence analysis thread and second artificial intelligence analysis thread, which are not limited herein.

S320, loading the first example high-level ramp hidden danger information set to a first artificial intelligent analysis thread, analyzing to obtain a second example high-level ramp hidden danger information set with example important description content, loading the target example high-level ramp hidden danger information set to the second artificial intelligent analysis thread, and analyzing to obtain a target secondary example high-level ramp hidden danger information set with example important description content.

Loading a first example high-level ramp hidden danger information set to a first artificial intelligent analysis thread, and analyzing a second example high-level ramp hidden danger information set obtained through analysis of the first artificial intelligent analysis thread, wherein the first example high-level ramp hidden danger information set is provided with an example catalog, so that analysis is carried out to obtain an example important description content corresponding to the example catalog in the source pseudo high-level ramp hidden danger information set; and similarly, loading the target example high-level ramp hidden danger information set to a second artificial intelligence analysis thread, and analyzing to obtain the target secondary example high-level ramp hidden danger information set with the example important description content.

S330, constructing important descriptive content quality indexes according to the example important descriptive content in the target secondary example high-level ramp hidden danger information set and the example important descriptive content in the second example high-level ramp hidden danger information set.

And S340, updating the thread coefficient of the thread to be configured according to the important description content quality index to obtain the high-order ramp hidden danger information analysis thread.

And updating the thread coefficients of the threads to be configured according to the quality indexes of the important descriptive contents, such as updating the thread coefficients of the first artificial intelligence analysis thread and the second artificial intelligence analysis thread respectively until the configured high-level ramp hidden danger information analysis threads are obtained when the first artificial intelligence analysis thread and the second artificial intelligence analysis thread are converged.

In the embodiment of the application, the analysis of the high-level ramp hidden danger information set is conveniently realized through the artificial intelligence analysis thread, and the important descriptive content quality index is built through the example important descriptive content in the target secondary example high-level ramp hidden danger information set and the example important descriptive content in the second example high-level ramp hidden danger information set, so that the thread to be configured pays attention to the important descriptive content while analyzing, thereby realizing the non-deletion of the important descriptive content; the method further comprises the steps of analyzing the hidden danger information of the high-level ramp of the detection object to the target hidden danger information set through two artificial intelligent analysis threads, and converting the hidden danger information of the high-level ramp of the detection object back to the target hidden danger information set, so that the situation that the hidden danger information of the high-level ramp of all the detection objects is analyzed to be the same hidden danger information of the high-level ramp in the target hidden danger information set by the threads to be configured is avoided.

The embodiment of the application provides another evaluation method S310 of the landslide hazard in high level, which is expanded into S410-S440.

S410, collecting the example high-level ramp hidden danger information of the detection object, and collecting the example high-level ramp hidden danger information of the target high-level ramp hidden danger information set.

In the embodiment of the application, the hidden danger information content of the high-level ramp of the example hidden danger information of the detected object is the same as the hidden danger information content of the high-level ramp of the example hidden danger information of the target hidden danger information set of the high-level ramp, and the hidden danger information styles of the high-level ramp are different.

In an example, the same scene may be acquired by different high-level ramp hidden danger information acquisition devices, so as to obtain the example high-level ramp hidden danger information and the target high-level ramp hidden danger information set of the detection object.

In an example, the example high-level ramp hidden-danger information of the detection object and the example high-level ramp hidden-danger information of the target high-level ramp hidden-danger information set may be captured by a data capturing technique.

S420, annotating the acquired example high-level ramp hidden danger information of the detection object with example important description content to obtain a first example high-level ramp hidden danger information set.

In the embodiment of the application, all the acquired exemplary high-level ramp hidden danger information of the detection object is annotated, specifically, the important description content of the examples in all the exemplary high-level ramp hidden danger information is annotated to generate the example catalog, and then all the exemplary high-level ramp hidden danger information in the first exemplary high-level ramp hidden danger information set is provided with the example catalog.

In an example, local exemplary high-level ramp hidden danger information of the detected object may be annotated, so as to obtain a first exemplary high-level ramp hidden danger information set.

The annotating of the example important description content may be annotating the example important description content through a rectangular frame, or annotating the example important description content.

S430, annotating the local example high-level ramp hidden danger information of the collected target high-level ramp hidden danger information set with example important description content to obtain the target example high-level ramp hidden danger information set.

In the embodiment of the application, local example high-level ramp hidden danger information of the collected target high-level ramp hidden danger information set is annotated to annotate example important description content to generate an example catalog, and then the example high-level ramp hidden danger information in the target example high-level ramp hidden danger information set is provided with the example catalog.

In an example, the local example high-level ramp hidden-danger information may be high-level ramp hidden-danger information of a preset proportion of all the example high-level ramp hidden-danger information of the target high-level ramp hidden-danger information set, for example, the local example high-level ramp hidden-danger information is 1/20 of all the example high-level ramp hidden-danger information of the target high-level ramp hidden-danger information set, and at this time, the local example high-level ramp hidden-danger information is selected from all the example high-level ramp hidden-danger information of the target high-level ramp hidden-danger information set for annotation.

In another example, the example high-level ramp hidden-danger information of the collected target high-level ramp hidden-danger information set is annotated randomly with example important descriptive content, wherein the random number is less than the number of all the example high-level ramp hidden-danger information of the target high-level ramp hidden-danger information set.

It should be understood that the annotation mode of the exemplary high-level ramp hidden danger information of the detection object may be the same as or different from the annotation mode of the exemplary high-level ramp hidden danger information of the target high-level ramp hidden danger information set.

S440, obtaining the thread to be configured.

According to the embodiment of the application, the example high-level ramp hidden danger information of different high-level ramp hidden danger information sets is collected, all the example high-level ramp hidden danger information of the detection object is annotated, only the local example high-level ramp hidden danger information of the target high-level ramp hidden danger information set is annotated, so that a large amount of data annotation is avoided, and the configuration efficiency of threads is increased.

The embodiment of the application also provides another evaluation method of the high-level landslide hazard, and S330 is expanded into S510-S530.

S510, according to the distinction between the example important descriptive content in the second example high-level ramp hidden trouble information set and the example important descriptive content in the first example high-level ramp hidden trouble information set, constructing a first important descriptive content quality index.

In the embodiment of the application, the first important descriptive content quality index is specific to the first generator, and the first important descriptive content quality index is used for representing the missing condition of the information of the hidden danger information of the high-level ramp in the analysis process of the first generator.

The distinguishing of the example important description content in the second example high grade slope risk information set from the example important description content in the first example high grade slope risk information set includes a number distinction.

In one possible implementation embodiment, the distinguishing of the example important descriptive content in the second example high-level ramp hidden-danger information set from the example important descriptive content in the first example high-level ramp hidden-danger information set includes a similarity distinguishing in addition to the number distinguishing, so that the similarity between the example important descriptive content in the second example high-level ramp hidden-danger information set and the example important descriptive content in the first example high-level ramp hidden-danger information set can be calculated, and the first important descriptive content quality index is built in combination with the number distinguishing and the similarity.

S520, according to the distinction between the example important description content in the target secondary example high-level ramp hidden danger information set and the example important description content in the target example high-level ramp hidden danger information set, a second important description content quality index is built.

In the embodiment of the present application, the second important descriptive content quality index is set for the second generator, and consistent with setting up the first important descriptive content quality index, the setting up may be differentiated according to the number of the example important descriptive content in the target secondary example high-level ramp hidden danger information set and the example important descriptive content in the target example high-level ramp hidden danger information set, or the number of the example important descriptive content in the target secondary example high-level ramp hidden danger information set and the example important descriptive content in the target example high-level ramp hidden danger information set, and the second important descriptive content quality index may be set up according to the similarity.

S530, generating an important descriptive content quality index according to the first important descriptive content quality index and the second important descriptive content quality index.

In an example, a sum of the first important descriptive content quality index and the second important descriptive content quality index may be taken as the important descriptive content quality index.

In one example, a confidence level may be configured for the first important descriptive content quality indicator and the second important descriptive content quality indicator, and a weighted sum of the first important descriptive content quality indicator and the second important descriptive content is taken as the important descriptive content quality indicator.

In the embodiment of the application, the quality index of the built important description content is distinguished by the example important description content in the analyzed example high-level ramp hidden danger information set and the example important description content in the example high-level ramp hidden danger information set before analysis, so that the loss of the high-level ramp hidden danger information in the analysis process is represented, and the thread can pay attention to the important description content in the subsequent process configuration.

In one embodiment of the present application, another method for evaluating a landslide hazard in high-level mountain is also provided, and S340 extends to S640.

S610, loading the second example high-level ramp hidden danger information set into a second artificial intelligent analysis thread, and analyzing to obtain a newly-built example high-level ramp hidden danger information set of the detection object.

The re-building example high-level ramp hidden danger information set refers to the re-built example high-level ramp hidden danger information after twice analysis of the first example high-level ramp hidden danger information.

It should be understood that the artificial intelligence analysis thread may generate pseudo high-level ramp hidden danger information with a consistent style, and in order to ensure that the key content after the analysis of the high-level ramp hidden danger information is unchanged, for example, whether the position of the high-level ramp hidden danger information after the position analysis of the lane line in the high-level ramp hidden danger information before the analysis is the lane line, because the second artificial intelligence analysis thread is used for analyzing the high-level ramp hidden danger information of the target high-level ramp hidden danger information set into the high-level ramp hidden danger information of the detection object, the second exemplary high-level ramp hidden danger information set of the target high-level ramp hidden danger information set can be loaded into the second artificial intelligence analysis thread, the high-level ramp hidden danger information of the detection object is generated through the second artificial intelligence analysis thread, and the high-level ramp hidden danger information generated by the second artificial intelligence analysis thread is the high-level ramp hidden danger information after the first exemplary high-level ramp hidden danger information is rebuilt.

S620, loading the target secondary example high-level ramp hidden danger information set into a first artificial intelligent analysis thread, and analyzing to obtain a target rebuilt example high-level ramp hidden danger information set of the target high-level ramp hidden danger information set.

As described above, the first artificial intelligence analysis thread is configured to analyze the detected object's hidden-high-level-ramp hidden-risk information into the target hidden-high-level-ramp hidden-risk information set, so that the target secondary sample hidden-high-level-ramp hidden-risk information set of the detected object can be loaded into the first artificial intelligence analysis thread, and the hidden-high-level-ramp hidden-risk information re-built for the target sample hidden-high-level-ramp hidden-risk information set is obtained through analysis, that is, the target re-built sample hidden-high-level-ramp hidden-risk information set.

S630, building a re-building quality index according to the re-building example high-level ramp hidden danger information set and the target re-building example high-level ramp hidden danger information set.

In the embodiment of the application, because the re-construction example high-level ramp hidden danger information and the target re-construction example high-level ramp hidden danger information are both the re-construction example high-level ramp hidden danger information, the re-construction quality index can be constructed based on the key content change of the re-construction example high-level ramp hidden danger information, so that the change condition of the key content of the analyzed high-level ramp hidden danger information can be represented by the re-construction quality index.

S640, generating a thread quality index corresponding to the thread to be configured according to the rebuilt quality index and the important description content quality index, and updating the thread coefficient of the thread to be configured according to the thread quality index.

Because the newly built quality index can represent the change condition of key content of the hidden danger information of the high-level ramp after analysis, the quality index of the important descriptive content can represent the missing condition of the hidden danger information of the high-level ramp in the analysis process; therefore, the thread quality index corresponding to the thread to be configured can be generated based on the rebuilt quality index and the important descriptive content quality index, and the thread coefficient information of the thread to be configured is updated from two aspects.

In an example, the thread quality index corresponding to the thread to be configured is generated according to the re-establishment quality index and the important description content quality index, and the sum of the re-establishment quality index and the important description content quality index is taken as the thread quality index corresponding to the thread to be configured.

In another example, the thread quality index corresponding to the thread to be configured may also be generated by respectively configuring the confidence level for the quality index of the re-establishment and the quality index of the important description content, and taking the weighted summation of the quality index of the re-establishment and the quality index of the important description content as the thread quality index corresponding to the thread to be configured.

In the embodiment of the application, the updating of the thread coefficients of the thread to be configured according to the thread quality index may be to update the thread coefficients of the first artificial intelligence analysis thread and the second artificial intelligence analysis thread respectively according to the thread quality index until the first artificial intelligence analysis thread and the second artificial intelligence analysis thread converge.

In the embodiment of the application, the secondary example high-level ramp hidden danger information set is loaded into the corresponding artificial intelligent analysis thread again for analysis, so that the rebuilt high-level ramp hidden danger information of the original example high-level ramp hidden danger information set is obtained, the deletion of important description content in the high-level ramp hidden danger information is concerned, the change condition of the key content of the analyzed high-level ramp hidden danger information is also concerned, and further, the analyzed graph is more consistent with the original high-level ramp hidden danger information during the process configuration.

In one embodiment of the present application, another method for evaluating landslide hazard in high level is also provided, and S630 is extended to S710-S730.

S710, building a first rebuilt quality index according to the commonality score of the rebuilt example high-level ramp hidden danger information set and the first example high-level ramp hidden danger information set.

It can be understood that whether the key content after analysis changes is judged, and the rebuilt high-level ramp hidden danger information after analysis and the original high-level ramp hidden danger information can be compared, so that the rebuilt quality index is measured through similarity between the rebuilt high-level ramp hidden danger information and the original high-level ramp hidden danger information. Therefore, in the embodiment of the application, the first rebuilt quality index is set up according to the common score of the rebuilt example high-level ramp hidden danger information set and the first example high-level ramp hidden danger information set, and the first rebuilt quality index is characterized by the rebuilt quality index of the detection object of the first generator, namely, the change condition of the key content after the first example high-level ramp hidden danger information set of the detection object analyzes to the second example high-level ramp hidden danger information set of the target high-level ramp hidden danger information set.

Optionally, the similarity score of the rebuilt example high-level ramp hidden danger information set and the first example high-level ramp hidden danger information set may be calculated through cosine similarity, or may also be calculated through distance.

S720, building a second rebuilt quality index according to the commonality scores of the target rebuilt example high-level ramp hidden danger information set and the target example high-level ramp hidden danger information set.

In the embodiment of the application, the second rebuilt quality index represents a rebuilt quality index of the target high-level ramp hidden danger information set of the second generator, namely, the change condition of key content after the target example high-level ramp hidden danger information set of the target high-level ramp hidden danger information set is analyzed to the target secondary example high-level ramp hidden danger information set of the detection object.

The similarity between the target re-establishment example high-level ramp hidden danger information set and the target example high-level ramp hidden danger information set can be calculated through cosine similarity, and can also be calculated through distance.

And S730, generating a re-construction quality index according to the first re-construction quality index and the second re-construction quality index.

In an example, the sum of the first and second re-construction quality indicators is taken as the re-construction quality indicator.

In another example, a confidence level may be configured for the first and second re-construction quality indicators, with a weighted sum of the first and second re-construction quality indicators being taken as the re-construction quality indicators.

In the embodiment of the application, the rebuilt quality index is built through the commonality score of the rebuilt high-level ramp hidden danger information and the original high-level ramp hidden danger information, so that the change condition of the key content of the analyzed high-level ramp hidden danger information is represented, and then the key content of the analyzed high-level ramp hidden danger information can be focused by a module during configuration.

In one embodiment of the present application, another method for evaluating high mountain landslide hazard is also provided to extend S640 to S810-S840.

S810, loading the second example high-level ramp hidden danger information set into a first analysis thread to obtain a first analysis result.

In the embodiment of the application, the second exemplary high-level ramp hidden danger information set is loaded into the first analysis thread, and then the first analysis thread judges the key of the second exemplary high-level ramp hidden danger information set and outputs a first analysis result.

S820, loading the hidden danger information of the target secondary example high-order ramp into a second analysis thread to obtain a second analysis result.

And loading the target secondary example high-level ramp hidden danger information set into a second analysis thread, judging the key of the target secondary example high-level ramp hidden danger information set by the second analysis thread, and outputting a second analysis result.

S830, constructing a quality evaluation according to the first analysis result and the second analysis result to generate a quality index.

In the embodiment of the application, quality indexes are generated by performing quality evaluation through key construction of the key of the second exemplary high-level ramp hidden danger information set and the key construction of the target secondary exemplary high-level ramp hidden danger information set.

And S840, generating a thread quality index corresponding to the thread to be configured according to the rebuilt quality index, the important descriptive content quality index and the quality evaluation generation quality index, and updating the thread coefficient of the thread to be configured according to the thread quality index.

In an example, the sum of the rebuilt quality index, the important descriptive content quality index, and the quality evaluation generation quality index may be used as the thread quality index corresponding to the thread to be configured.

In another example, the confidence level may be configured for the quality index re-built, the quality index of the important descriptive content, and the quality evaluation generated quality index, and the weighted summation of the quality index re-built, the quality index of the important descriptive content, and the quality evaluation generated quality index is used as the thread quality index corresponding to the thread to be configured.

In one embodiment of the present application, S830 is extended to S1010-S1050.

S1010, loading the target example high-level ramp hidden danger information set into the first analysis thread to obtain a third analysis result.

In the embodiment of the application, when the first analysis thread performs key discrimination on the hidden danger information of the second example high-level ramp, the hidden danger information of the high-level ramp output by the first generator is discriminated from the real hidden danger information of the high-level ramp, so that the target example hidden danger information set (namely the real hidden danger information of the target hidden danger information set of the high-level ramp) also needs to be loaded into the first analysis thread, and the key of the target example hidden danger information set of the high-level ramp is discriminated through the first analysis thread.

S1020, loading the first example high-level ramp hidden danger information set into a second analysis thread to obtain a fourth analysis result.

Similarly, the first example high-level ramp hidden danger information set (namely the real high-level ramp hidden danger information of the detection object) is loaded into the second analysis thread, and the key of the first example high-level ramp hidden danger information set is judged through the second analysis thread.

S1030, building a first quality evaluation according to the first analysis result and the third analysis result to generate a quality index.

Through the key of the second example high-level ramp hidden danger information set and the key of the first example high-level ramp hidden danger information set, a first quality evaluation generation quality index is built, so that the first analysis thread can distinguish the second example high-level ramp hidden danger information set and the first example high-level ramp hidden danger information set as far as possible, and the high-level ramp hidden danger information output by the first artificial intelligent analysis thread is found out.

S1040, constructing a second quality evaluation according to the second analysis result and the fourth analysis result to generate a quality index.

And establishing a second process quality evaluation to generate a quality index through the key of the target secondary example high-level ramp hidden danger information set and the key of the target example high-level ramp hidden danger information set, so that the second analysis thread can distinguish the target secondary example high-level ramp hidden danger information set and the target example high-level ramp hidden danger information set as far as possible, and finding out the high-level ramp hidden danger information output by the second artificial intelligence analysis thread.

S1050, generating quality indexes according to the first quality evaluation generation quality index and the second quality evaluation generation quality index, and generating quality indexes according to the second quality evaluation generation quality index.

In the embodiment of the application, the sum of the first quality evaluation generation quality index and the second quality evaluation generation quality index can be used as the quality evaluation generation quality index.

In an example, a confidence level may be configured for the first quality-evaluation-performed quality index and the second quality-evaluation-performed quality index, and the first quality-evaluation-performed quality index and the second quality-evaluation-performed quality index may be weighted and summed according to the confidence level to obtain the quality-evaluation-performed quality index.

In the embodiment of the application, the high-level ramp hidden danger information generated by the artificial intelligent analysis thread is distinguished from the original real high-level ramp hidden danger information by the analysis thread, and further the quality index is generated by carrying out quality evaluation on the key of the high-level ramp hidden danger information generated by the artificial intelligent analysis thread and the key construction of the original real high-level ramp hidden danger information, so that more real high-level ramp hidden danger information can be generated during thread configuration.

In one embodiment of the present application, another method for evaluating high mountain landslide hazard is also provided, and S840 is extended to S1110-S1130.

S1110, obtaining the confidence coefficient of the rebuilt quality index and the confidence coefficient of the important descriptive content quality index.

In the embodiment of the application, the confidence levels configured for the quality indexes of the rebuilt and the quality indexes of the important descriptive contents are preset, so that the confidence levels of the quality indexes of the rebuilt and the important descriptive contents can be respectively obtained, wherein the confidence level of the quality indexes of the important descriptive contents is larger than that of the quality indexes of the rebuilt.

S1120, weighting the rebuilt quality index and the important descriptive content quality index according to the confidence level of the rebuilt quality index and the confidence level of the important descriptive content quality index to obtain an information quality index.

In the embodiment of the application, the rebuilt quality index and the important descriptive content quality index are both specific to the generator, and the information quality index specific to the generator is obtained by carrying out weighting processing on the rebuilt quality index and the important descriptive content quality index according to the confidence level of the rebuilt quality index and the confidence level of the important descriptive content quality index.

S1130, generating a thread quality index corresponding to the thread to be configured according to the information quality index and the quality evaluation generation quality index, and updating the thread coefficient of the thread to be configured according to the thread quality index.

In an example, the quality indicator generated by performing quality assessment is specific to the arbiter, so that the sum of the information quality indicator and the quality indicator generated by performing quality assessment can be used as the thread quality indicator corresponding to the thread to be configured.

In another example, the information quality index and the confidence level of the quality evaluation generation quality index can be obtained, and then the thread quality index corresponding to the thread to be configured is obtained by weighting the information quality index and the quality evaluation generation quality index.

In the embodiment of the application, the quality index of the thread to be configured is weighted by the quality index of the rebuilt and the quality index of the important descriptive content, so that the thread quality index corresponding to the thread to be configured is more focused on the related information of the hidden danger information of the high-order ramp.

The embodiment of the application provides another evaluation method of high-level landslide disasters, and S1130 is expanded to S1210 and S1230.

S1210, generating a thread quality index corresponding to the thread to be configured according to the information quality index and the quality index generated by quality evaluation.

S1220, fixing thread coefficients of the first analysis thread and the second analysis thread, and updating the thread coefficients of the first artificial intelligence analysis thread and the second artificial intelligence analysis thread according to the thread quality index.

S1230, fixing thread coefficients of the first artificial intelligence analysis thread and the second artificial intelligence analysis thread, and updating the thread coefficients of the first analysis thread and the second analysis thread according to quality index generated by quality evaluation.

In the embodiment of the application, the thread coefficient of the artificial intelligence analysis thread is updated through the fixed analysis thread, and then the thread coefficient of the analysis thread is updated through the fixed artificial intelligence analysis thread, so that the artificial intelligence analysis thread and the analysis thread mutually perform quality evaluation, and the capability of the artificial intelligence analysis thread is stronger.

The embodiment of the application provides another evaluation method of high mountain landslide disasters, and S230 is expanded to S1310.

S1310, obtaining target high-level ramp hidden danger information of a target high-level ramp hidden danger information set obtained by analysis of a first artificial intelligent analysis thread of the high-level ramp hidden danger information analysis thread.

As described above, the thread quality index is used to update the thread coefficients of the first artificial intelligence analysis thread and the second artificial intelligence analysis thread, the key content of the pseudo high-level ramp hidden danger information set and the key content of the true high-level ramp hidden danger information set output by the first artificial intelligence analysis thread and the second artificial intelligence analysis thread are the same, the important description content in the pseudo high-level ramp hidden danger information set is identical to the important description content of the true high-level ramp hidden danger information set, the key of the high-level ramp hidden danger information cannot be distinguished by the first analysis thread and the second discriminator, the to-be-processed high-level ramp hidden danger information of the detection object is loaded to the high-level ramp hidden danger information analysis thread, specifically the first artificial intelligence analysis thread loaded to the high-level ramp hidden danger information analysis thread, and then the target high-level ramp hidden danger information of the target high-level ramp hidden danger information set is obtained by analysis of the first artificial intelligence analysis thread, and the target high-level ramp hidden danger information of the target high-level ramp hidden danger information set is the same as the key content of the to be processed, and the important description content is identical.

In the embodiment of the application, the analyzed hidden danger information of the high-level ramp outputted by the first artificial intelligent analysis thread is the same as the key content of the original hidden danger information of the high-level ramp, and the important description content in the analyzed hidden danger information of the high-level ramp is consistent with the important description content of the original hidden danger information of the high-level ramp.

Wherein, S1410-S1430 are described in detail as follows:

s1410, acquiring an identification thread of the target bit ramp hidden danger information set, wherein the identification thread is used for executing target identification indication based on the important description content.

In the embodiment of the application, after the target high-level ramp hidden danger information of the target high-level ramp hidden danger information set is obtained, the target high-level ramp hidden danger information of the target high-level ramp hidden danger information set can be applied, and as described above, important description contents contained in the target high-level ramp hidden danger information are information annotated for target identification instructions, so that an identification thread of the target high-level ramp hidden danger information set can be configured based on the target high-level ramp hidden danger information, the identification thread is used for executing target identification instructions based on the important description contents, such as the important description contents are vehicle body information, and the identification thread is used for judging vehicle types based on the vehicle body information.

It should be appreciated that the identified thread may be a deep neural thread, and the thread structure of the identified thread is not limited herein.

S1420, obtaining a configuration set, wherein the configuration set comprises target bit ramp hidden danger information and an important description content catalog for representing important description content.

It should be noted that, if the to-be-processed high-level ramp hidden danger information of the detection object annotates the important description content, the to-be-processed high-level ramp hidden danger information has an important description content directory for representing the important description content, and further, the obtained target high-level ramp hidden danger information is analyzed to also have the important description content directory, so that the target high-level ramp hidden danger information carrying the important description content directory can be directly applied to the configuration set.

If the to-be-processed high-level ramp hidden danger information of the detection object does not annotate the important description content, annotating the important description content contained in the target high-level ramp hidden danger information after analyzing the to-be-detected object to obtain the important description content catalog, and further applying the target high-level ramp hidden danger information and the important description content catalog to the configuration set.

The other high-level ramp hidden danger information of the target high-level ramp hidden danger information set is a small amount of high-level ramp hidden danger information with important description content catalogs, so that the target high-level ramp hidden danger information, the other high-level ramp hidden danger information of the target high-level ramp hidden danger information set and the important description content catalogs can be combined to obtain a configuration set.

S1430, configuring the identification thread according to the configuration set.

In the embodiment of the application, the analyzed target high-level ramp hidden danger information is applied to the configuration of the identification thread, so that on the basis of marking a large number of existing high-level ramp hidden danger information sets with high-level ramp hidden danger information, only a small amount of new high-level ramp hidden danger information set data is required to be marked, a better identification algorithm can be realized on the new high-level ramp hidden danger information sets, and a large number of data marks are effectively avoided.

In order to facilitate understanding, the embodiment of the present application further provides a method for evaluating a high-level landslide disaster, which is described by way of a specific example, where the method for evaluating a high-level landslide disaster includes:

S1510, obtaining a catalog of the hidden danger information of the high-level ramp in the existing hidden danger information set of the high-level ramp, and marking the catalog of the hidden danger information of the high-level ramp in the hidden danger information set of a small amount of new high-level ramp.

And after marking, taking a small amount of high-level ramp hidden danger information in the high-level ramp hidden danger information set A and all marked high-level ramp hidden danger information in the high-level ramp hidden danger information set B as configuration sets.

S1520, designing a quality evaluation generation thread.

S1530, quality evaluation is defined to generate a quality index, a rebuilt quality index and an important descriptive content quality index.

S1540, configuring a quality evaluation generation thread.

S1550, analyzing the hidden danger information of the high-level ramp in the existing hidden danger information set of the high-level ramp to a new hidden danger information set of the high-level ramp.

S1560, developing a new high-order ramp hidden danger information set identification algorithm on the data set obtained by analysis.

On the basis of the above, an evaluation device for landslide hazard in high level is provided, the device includes:

the information acquisition module is used for acquiring hidden danger information of the high-level ramp to be processed of the detection object;

The information analysis module is used for loading the hidden danger information of the high-level ramp to be processed to an analysis thread of the hidden danger information of the high-level ramp, wherein the analysis thread of the hidden danger information of the high-level ramp is configured according to the quality index of important descriptive contents, the quality index of the important descriptive contents is set up according to the example important descriptive contents in the hidden danger information set of the second example obtained after the analysis of the hidden danger information set of the first example of the detected object of the high-level ramp and the example important descriptive contents in the hidden danger information set of the target secondary example obtained after the analysis of the hidden danger information set of the target high-level ramp;

The hidden danger information determining module is used for obtaining target hidden danger information of the target hidden danger information set, which is obtained by analyzing the hidden danger information of the high-level ramp to be processed by the hidden danger information analyzing thread of the high-level ramp, wherein important description content covered in the hidden danger information of the target high-level ramp is consistent with important description content covered in the hidden danger information of the high-level ramp to be processed.

On the above basis, an evaluation system of high mountain landslide hazard is shown, comprising a processor and a memory in communication with each other, the processor being adapted to read a computer program from the memory and execute it for carrying out the method described above.

On the basis of the above, there is also provided a computer readable storage medium on which a computer program stored which, when run, implements the above method.

In summary, based on the above scheme, by loading the to-be-processed high-level ramp hidden danger information of the detection object to the high-level ramp hidden danger information analysis thread, since the high-level ramp hidden danger information analysis thread is configured according to the quality index of the important description content, the quality index of the important description content is configured according to the example important description content in the second example high-level ramp hidden danger information set obtained after the analysis of the first example high-level ramp hidden danger information set of the detection object, and the example important description content in the target secondary example high-level ramp hidden danger information set obtained after the analysis of the target example high-level ramp hidden danger information set of the target high-level ramp hidden danger information set, the high-level ramp hidden danger information analysis thread can pay attention to the important description content covered in the high-level ramp hidden danger information, so that the important description content is ensured not to be lost while analyzing the high-level ramp hidden danger information, and the target high-level ramp hidden danger information of the target high-level ramp hidden danger information set obtained by the high-level ramp hidden danger information analysis thread is analyzed to the to be processed, thereby improving the processing effect of the high-level ramp hidden danger information.

It should be appreciated that the systems and modules thereof shown above may be implemented in a variety of ways. For example, in some embodiments, the system and its modules may be implemented in hardware, software, or a combination of software and hardware. Wherein the hardware portion may be implemented using dedicated logic; the software portions may then be stored in a memory and executed by a suitable instruction execution system, such as a microprocessor or special purpose design hardware. Those skilled in the art will appreciate that the methods and systems described above may be implemented using computer executable instructions and/or embodied in processor control code, such as provided on a carrier medium such as a magnetic disk, CD or DVD-ROM, a programmable memory such as read only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The system of the present application and its modules may be implemented not only with hardware circuitry such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., but also with software executed by various types of processors, for example, and with a combination of the above hardware circuitry and software (e.g., firmware).

It should be noted that, the advantages that may be generated by different embodiments may be different, and in different embodiments, the advantages that may be generated may be any one or a combination of several of the above, or any other possible advantages that may be obtained.

Claims (10)

1. The method for evaluating the landslide hazard at high level is characterized by comprising the following steps of:

obtaining hidden danger information of a high-level ramp to be processed of a detection object;

Loading the hidden danger information of the high-level ramp to be processed into an analysis thread of hidden danger information of the high-level ramp, wherein the analysis thread of hidden danger information of the high-level ramp is configured according to quality indexes of important descriptive contents, the quality indexes of the important descriptive contents are built according to example important descriptive contents in a second example hidden danger information set of the high-level ramp obtained after analysis of a first example hidden danger information set of the detected object and example important descriptive contents in a target secondary example hidden danger information set obtained after analysis of a target example hidden danger information set of the high-level ramp;

the method comprises the steps that the high-level ramp hidden danger information analysis thread is obtained, the high-level ramp hidden danger information to be processed is analyzed to obtain target high-level ramp hidden danger information of the target high-level ramp hidden danger information set, and important description contents covered in the target high-level ramp hidden danger information are consistent with important description contents covered in the high-level ramp hidden danger information to be processed.

2. The method of claim 1, wherein prior to loading the to-be-processed high ramp potential information into a high ramp potential information analysis thread, the method further comprises:

Obtaining the first example high-level ramp hidden danger information set, the target example high-level ramp hidden danger information set and a thread to be configured, wherein the thread to be configured comprises a first artificial intelligence analysis thread and a second artificial intelligence analysis thread, and the example high-level ramp hidden danger information in the first example high-level ramp hidden danger information set and the target example high-level ramp hidden danger information set both have an example catalog for representing the example important description content;

Loading the first example high-level ramp hidden danger information set to the first artificial intelligent analysis thread, analyzing to obtain a second example high-level ramp hidden danger information set of the target high-level ramp hidden danger information set with the example important description content, loading the target example high-level ramp hidden danger information set to the second artificial intelligent analysis thread, and analyzing to obtain a target secondary example high-level ramp hidden danger information set of the detection object with the example important description content;

establishing the important descriptive content quality index by combining the example important descriptive content in the target secondary example high-level ramp hidden trouble information set and the example important descriptive content in the second example high-level ramp hidden trouble information set;

and updating the thread coefficient of the thread to be configured by combining the important descriptive content quality index to obtain the high-level ramp hidden danger information analysis thread.

3. The method of claim 2, wherein the obtaining the first example set of high ramp potential information, the target example set of high ramp potential information, and the thread to be configured comprises:

Collecting the example high-level ramp hidden danger information of the detection object, and collecting the example high-level ramp hidden danger information of the target high-level ramp hidden danger information set;

carrying out annotation of the example important description content on the acquired example high-level ramp hidden danger information of the detection object to obtain the first example high-level ramp hidden danger information set;

And annotating the example important description content on the local example high-level ramp hidden danger information of the collected target high-level ramp hidden danger information set to obtain the target example high-level ramp hidden danger information set.

4. The method of claim 2, wherein the building the important descriptive quality indicator in combination with the example important descriptive in the target secondary set of example high ramp risk information and the example important descriptive in the second set of example high ramp risk information comprises:

Combining the distinction of the example important descriptive content in the second example high-level ramp hidden trouble information set and the example important descriptive content in the first example high-level ramp hidden trouble information set, and constructing a first important descriptive content quality index;

combining the sample important description content in the target secondary sample high-level ramp hidden danger information set with the sample important description content in the target sample high-level ramp hidden danger information set to establish a second important description content quality index;

And generating the important descriptive content quality index by combining the first important descriptive content quality index and the second important descriptive content quality index.

5. The method of claim 2, wherein prior to updating the thread coefficients of the thread to be configured in conjunction with the important descriptive content quality indicator, the method further comprises:

Loading the second example high-level ramp hidden danger information set into the second artificial intelligent analysis thread, and analyzing to obtain a newly built example high-level ramp hidden danger information set of the detection object;

loading the target secondary example high-level ramp hidden danger information set into the first artificial intelligent analysis thread, and analyzing to obtain a target rebuilt example high-level ramp hidden danger information set of the target high-level ramp hidden danger information set;

Setting up a re-setting quality index by combining the re-setting example high-level ramp hidden danger information set and the target re-setting example high-level ramp hidden danger information set;

The updating the thread coefficient of the thread to be configured by combining the important descriptive content quality index comprises the following steps: and generating a thread quality index corresponding to the thread to be configured by combining the re-establishment quality index and the important description content quality index, and updating the thread coefficient of the thread to be configured by combining the thread quality index.

6. The method of claim 5, wherein the combining the re-construction example high-level ramp hidden danger information set with the target re-construction example high-level ramp hidden danger information set to construct a re-construction quality index comprises:

Building a first rebuilt quality index by combining the rebuilt example high-level ramp hidden danger information set and the commonality score of the first example high-level ramp hidden danger information set;

Combining the target re-establishment example high-level ramp hidden danger information set and the commonality score of the target example high-level ramp hidden danger information set to establish a second re-establishment quality index;

and generating the re-construction quality index by combining the first re-construction quality index and the second re-construction quality index.

7. The method of claim 5, wherein the thread to be configured further comprises a first analysis thread for determining that the high ramp potential information is critical on the target high ramp potential information set, and a second analysis thread for determining that the high ramp potential information is critical on the detection object;

before the thread coefficient of the thread to be configured is updated by combining the thread quality index, the method further comprises:

loading the second exemplary high-level ramp hidden danger information set into the first analysis thread to obtain a first analysis result;

Loading the hidden danger information of the target secondary example high-level ramp into the second analysis thread to obtain a second analysis result;

constructing a quality evaluation by combining the first analysis result and the second analysis result to generate a quality index;

The step of generating the thread quality index corresponding to the thread to be configured by combining the rebuilt quality index and the important descriptive content quality index comprises the following steps: and combining the rebuilt quality index, the important descriptive content quality index and the quality index generated by quality evaluation to generate a quality index, and generating a thread quality index corresponding to the thread to be configured according to the quality index.

8. The method of claim 7, wherein the constructing a quality assessment to generate a quality indicator in combination with the first analysis result and the second analysis result comprises:

loading the target example high-level ramp hidden danger information set into the first analysis thread to obtain a third analysis result;

loading the first example high-level ramp hidden danger information set into the second analysis thread to obtain a fourth analysis result;

building a first quality evaluation by combining the first analysis result and the third analysis result to generate a quality index;

constructing a second quality evaluation by combining the second analysis result and the fourth analysis result to generate a quality index;

And generating the quality index by combining the first quality index and the second quality index.

9. The method of claim 7, wherein the generating the thread quality indicator corresponding to the thread to be configured according to the quality indicator by combining the re-building quality indicator, the important descriptive content quality indicator, and the performing quality assessment to generate the quality indicator fusion to generate the quality indicator comprises:

Obtaining the confidence coefficient of the rebuilt quality index and the confidence coefficient of the important descriptive content quality index;

Combining the confidence coefficient of the rebuilt quality index and the confidence coefficient of the important descriptive content quality index, and carrying out weighting treatment on the rebuilt quality index and the important descriptive content quality index to obtain an information quality index;

Combining the information quality index and the quality index fusion generated by performing quality evaluation to generate a quality index, and generating a thread quality index corresponding to the thread to be configured according to the quality index;

the updating the thread coefficient of the thread to be configured by combining the thread quality index comprises the following steps:

Fixing thread coefficients of the first analysis thread and the second analysis thread, and updating the thread coefficients of the first artificial intelligent analysis thread and the second artificial intelligent analysis thread by combining the thread quality index;

Fixing thread coefficients of the first artificial intelligence analysis thread and the second artificial intelligence analysis thread, and updating the thread coefficients of the first analysis thread and the second analysis thread by combining the quality index generated by quality evaluation;

The step of obtaining the target high-level ramp hidden danger information of the target high-level ramp hidden danger information set, which is obtained by analyzing the high-level ramp hidden danger information to be processed by the high-level ramp hidden danger information analysis thread, comprises the following steps:

Acquiring target high-level ramp hidden danger information of the target high-level ramp hidden danger information set obtained by analysis of a first artificial intelligent analysis thread of the high-level ramp hidden danger information analysis thread;

The important description content contained in the target high-level ramp hidden trouble information is information annotated for target identification indication; after the obtaining the target high-level ramp hidden danger information of the target high-level ramp hidden danger information set obtained by analyzing the high-level ramp hidden danger information to be processed by the high-level ramp hidden danger information analysis thread, the method further comprises:

Acquiring an identification thread of a target bit ramp hidden danger information set, wherein the identification thread is used for executing the target identification instruction based on the important description content;

Obtaining a configuration set, wherein the configuration set comprises target-bit ramp hidden danger information and an important description content catalog for representing the important description content;

And configuring the identification thread in combination with the configuration set.

10. An evaluation system for high mountain landslide disasters, comprising a processor and a memory in communication with each other, the processor being adapted to read a computer program from the memory and execute it to implement the method of any one of claims 1-9.