CN116737975A

CN116737975A - Public health data query method and system applied to image analysis

Info

Publication number: CN116737975A
Application number: CN202310544103.3A
Authority: CN
Inventors: 陈辉; 胡绵蓉; 孙海波; 蔡苏; 宗成忠
Original assignee: Chongqing Haihuixing Technology Co ltd
Current assignee: Chongqing Haihuixing Technology Co ltd
Priority date: 2023-05-15
Filing date: 2023-05-15
Publication date: 2023-09-12

Abstract

According to the public health data query method and system applied to image analysis, event type probability arrangement corresponding to the probability arrangement to be selected is determined, various monitoring behaviors of texts can be associated with various types of description rows of the probability arrangement to be selected, and regression analysis accuracy of the probability arrangement can be improved; the priority evaluation index introduces category priority conditions between the real-time possibility arrangement and the real-time monitoring event category, so that the correlation condition between the regression analysis monitoring event category and the regression analysis possibility arrangement can be enhanced, and the regression analysis of the category and the layering category is more accurate; the artificial participation can be effectively reduced, and the artificial intelligent recognition thread can continuously work, so that the reliability of public health is guaranteed, and the readiness and the reliability of public health image data query can be improved through the artificial intelligent recognition thread when the public health image data query is carried out.

Description

Public health data query method and system applied to image analysis

Technical Field

The application relates to the technical field of data query, in particular to a public health data query method and system applied to image analysis.

Background

Image analysis is the extraction of meaningful information from an image; mainly extracted from the digital image by digital image processing technology. The image analysis task may be as simple as reading a bar code label or as complex as identifying a person from a person's face. For analyzing large amounts of data, a computer is indispensable for tasks requiring complex computation or for extraction of quantitative information.

At present, the public health data query technology and the image analysis technology are combined, so that the problem of inaccurate health data query possibly exists, and the readiness and the reliability of the public health image data query are difficult to ensure, so that a technical scheme is needed to improve the technical problem.

Disclosure of Invention

In order to improve the technical problems in the related art, the application provides a public health data query method and a public health data query system applied to image analysis.

In a first aspect, there is provided a public health data query method applied to image analysis, the method comprising at least: obtaining an important behavior monitoring description array corresponding to a public health monitoring image, and determining the monitoring event type possibility arrangement corresponding to the monitoring event type to be selected through an artificial intelligent recognition thread and the important behavior monitoring description array; obtaining a hierarchical category variable description array corresponding to the possibility arrangement to be selected, and associating the important behavior monitoring description array with the hierarchical category variable description array through the artificial intelligent recognition thread; determining event type probability arrangement corresponding to the candidate probability arrangement according to the association result; determining a category evaluation index corresponding to the public health monitoring image according to a real-time monitoring event category catalog corresponding to the public health monitoring image, the monitoring event category possibility arrangement, a real-time possibility arrangement catalog corresponding to the public health monitoring image and the event category possibility arrangement; determining a priority evaluation index corresponding to the public health monitoring image according to the monitoring event type possibility arrangement, the event type possibility arrangement and the type priority condition between the real-time monitoring event type catalog and the real-time possibility arrangement catalog; the category priority level condition refers to that the category priority level corresponding to the real-time monitoring event category directory is greater than the category priority level corresponding to the real-time possibility arrangement directory; and optimizing the artificial intelligent recognition thread according to the category evaluation index and the priority evaluation index to obtain a target artificial intelligent recognition thread, and carrying out public health data query processing by combining the target artificial intelligent recognition thread.

Further, the determining, by the artificial intelligence recognition thread and the important behavior monitoring description array, the arrangement of the monitoring event type possibility corresponding to the type of the monitoring event to be selected includes: loading the important behavior monitoring description array to a category dividing unit in the artificial intelligent recognition thread, and outputting a possibility description array corresponding to the category of the monitoring event to be selected through the category dividing unit; and obtaining the category regression analysis possibility corresponding to the category of the monitoring event to be selected according to the possibility description array, and determining the possibility arrangement formed by the category regression analysis possibility as the category possibility arrangement of the monitoring event.

Further, the obtaining the hierarchical category variable description array corresponding to the candidate possibility arrangement includes: acquiring a related data cluster corresponding to the arrangement of the possibility to be selected; obtaining a data description array corresponding to each related data one by one in the related data cluster, and splicing the data description arrays corresponding to each related data one by one to obtain a spliced vector description array; and obtaining the data number of the related data included in the related data cluster, and determining a hierarchical category variable description array corresponding to the candidate possibility arrangement according to the splicing vector description array and the data number.

Further, the number of the candidate possibility arrangements is not less than one; the at least one candidate possibility arrangement comprises a candidate possibility arrangement Xa, wherein the hierarchical type variable description array comprises a hierarchical type variable description array Ya corresponding to the candidate possibility arrangement Xa, and a is a positive integer; the method for determining event type probability arrangement corresponding to the candidate probability arrangement according to the association result includes the steps of: loading the important behavior monitoring description array and the layering type variable description array Ya into a layering type dividing unit in the artificial intelligent recognition thread; the important behavior monitoring description array and the layering type variable description array Ya are associated through the layering type dividing unit, the association possibility between the important behavior monitoring description array and the layering type variable description array Ya is determined, and the association possibility is used as layering type regression analysis possibility corresponding to the candidate possibility arrangement Xa; and when the possibility arrangement formed by the at least one hierarchical type regression analysis possibility is determined, determining the possibility arrangement formed by the at least one hierarchical type regression analysis possibility as the event type possibility arrangement.

Further, the associating, by the hierarchical category dividing unit, the important behavior monitoring description array and the hierarchical category variable description array Ya, and determining a possibility of association between the important behavior monitoring description array and the hierarchical category variable description array Ya includes: obtaining a correlation coefficient queue in the hierarchical category dividing unit; calculating the association coefficient queue, the important behavior monitoring description array and the layering type variable description array Ya to obtain a calculation vector description array corresponding to the candidate possibility arrangement Xa; and loading the calculation possibility description array to a function statistics unit, and carrying out vector conversion processing on the calculation vector description array by an analysis mode in the function statistics unit to obtain the association possibility between the important behavior monitoring description array and the layering type variable description array Ya.

Further, performing calculation processing on the association coefficient queue, the important behavior monitoring description array and the hierarchical category variable description array Ya to obtain a calculation vector description array corresponding to the candidate possibility arrangement Xa, where the calculation vector description array includes: performing queue analysis on the association coefficient queue to obtain a first analysis projection queue and a second analysis projection queue; calculating the first analysis projection queue and the important behavior monitoring description array to obtain a first projection vector; calculating the second analysis projection queue and the hierarchical category variable description array Ya to obtain a second projection vector; and carrying out calculation processing on the first projection vector and the second projection vector to obtain a calculation vector description array corresponding to the candidate possibility arrangement Xa.

Further, the determining the category evaluation index corresponding to the public health monitoring image according to the real-time monitoring event category catalog corresponding to the public health monitoring image, the monitoring event category possibility arrangement, the real-time possibility arrangement catalog corresponding to the public health monitoring image and the event category possibility arrangement includes: determining a first type evaluation index corresponding to the public health monitoring image according to the real-time monitoring event type catalogue and the monitoring event type possibility arrangement; determining a second type of evaluation index corresponding to the public health monitoring image according to the real-time possibility arrangement catalog and the event type possibility arrangement; and determining the category evaluation index according to the first category evaluation index and the second category evaluation index.

Further, the number of the types of the monitoring events to be selected is not less than one; the determining a first category evaluation index corresponding to the public health monitoring image according to the real-time monitoring event category catalogue and the monitoring event category possibility arrangement comprises the following steps: obtaining the category regression analysis possibility corresponding to the real-time monitoring event category catalogue in the monitoring event category possibility arrangement, and the category regression analysis possibility corresponding to the rest monitoring event categories; the other monitoring event types are not less than one monitoring event type to be selected except the real-time monitoring event type catalogue; obtaining a first type of performance evaluation algorithm; configuring a category matching coefficient corresponding to the real-time monitoring event category catalogue in the first category performance evaluation algorithm as an effective parameter, and generating a first staged original performance evaluation structure according to the category regression analysis possibility of the effective parameter and the real-time monitoring event category catalogue; configuring class matching coefficients corresponding to the other monitoring event classes in the first class performance evaluation algorithm as invalid vectors, and generating a second-stage original performance evaluation structure according to the class regression analysis possibility of the invalid vectors and the other monitoring event classes; and performing calculation processing on the first-stage original performance evaluation structure and the second-stage original performance evaluation structure to obtain the first-type evaluation index.

Further, the determining the category evaluation index according to the first category evaluation index and the second category evaluation index includes: obtaining a first calculation vector corresponding to the first type of evaluation index and a second calculation vector corresponding to the second type of evaluation index; calculating the first calculation vector and the first type of evaluation index to obtain a first calculation performance evaluation result; calculating the second calculation vector and the second type of evaluation index to obtain a second calculation performance evaluation result; and clustering the first calculation performance evaluation result and the second calculation performance evaluation result to obtain the category evaluation index.

Further, the determining the priority rating index corresponding to the public health monitoring image according to the monitoring event type likelihood arrangement, the event type likelihood arrangement, and the type priority condition between the real-time monitoring event type catalog and the real-time likelihood arrangement catalog includes: acquiring category regression analysis possibility corresponding to the real-time monitoring event category catalogue in the monitoring event category possibility arrangement, and acquiring hierarchical category regression analysis possibility corresponding to the real-time event category possibility arrangement catalogue in the event category possibility arrangement; determining the possibility of hierarchical type regression analysis corresponding to the real-time possibility arrangement catalog and the possibility comparison result between the possibility of type regression analysis corresponding to the real-time monitoring event type catalog; and determining the priority rating index according to the possibility comparison result and the category priority condition.

Further, the category priority condition is used for restricting the hierarchical category regression analysis possibility corresponding to the real-time possibility arrangement catalog and the magnitude relation between the category regression analysis possibility corresponding to the real-time monitoring event category catalog; the determining the priority rating index according to the likelihood comparison result and the category priority condition comprises the following steps: obtaining a performance expression algorithm; if the probability comparison result is positive, determining hierarchical type regression analysis probability corresponding to the real-time probability arrangement catalog, wherein the size relationship between the hierarchical type regression analysis probability corresponding to the real-time monitoring event type catalog and the type regression analysis probability corresponding to the real-time monitoring event type catalog does not accord with the type priority condition, and determining the priority evaluation index according to a target calculation vector in the performance expression algorithm and the probability comparison result; and if the probability comparison result is a negative number, determining the hierarchical type regression analysis probability corresponding to the real-time probability arrangement catalog, wherein the size relationship between the hierarchical type regression analysis probability corresponding to the real-time monitoring event type catalog and the type regression analysis probability corresponding to the real-time monitoring event type catalog accords with the type priority condition, and determining the priority evaluation index according to an invalid vector, the target calculation vector and the probability comparison result in the performance expression algorithm.

Further, optimizing the artificial intelligent recognition thread according to the category evaluation index and the priority evaluation index to obtain a target artificial intelligent recognition thread, including: obtaining a clustering calculation vector corresponding to the category evaluation index and a calculation vector section corresponding to the priority evaluation index; calculating the category evaluation index and the clustering calculation vector to obtain a target category evaluation index; calculating the priority evaluation index and the calculation vector interval to obtain a target priority evaluation index; and carrying out clustering calculation processing on the target type evaluation index and the target priority evaluation index to obtain a target performance evaluation result, and optimizing the artificial intelligent recognition thread according to the target performance evaluation result to obtain the target artificial intelligent recognition thread.

In a second aspect, there is provided a public health data query system for image analysis 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 as described above.

According to the public health data query method and the public health data query system applied to image analysis, after the important behavior monitoring description array corresponding to the public health monitoring image is obtained, the important behavior monitoring description array can be loaded into the artificial intelligent recognition thread, and the monitoring event type possibility arrangement corresponding to the type of the monitoring event to be selected can be determined through the artificial intelligent recognition thread and the important behavior monitoring description array (the type of the monitoring event can be arranged in the type of the monitoring event to be selected through the monitoring event type possibility arrangement, so that the type of the regression analysis monitoring event corresponding to the public health monitoring image is determined); through the artificial intelligent recognition thread, the important behavior monitoring description array corresponding to the public health monitoring image can be associated with the hierarchical type variable description array of the to-be-selected possibility arrangement, so that the event type possibility arrangement corresponding to the to-be-selected possibility arrangement can be determined (the regression analysis possibility arrangement corresponding to the public health monitoring image can be determined in the to-be-selected possibility arrangement through the event type possibility arrangement). The category evaluation index corresponding to the public health monitoring image can be determined through the real-time monitoring event category catalogue corresponding to the public health monitoring image, the monitoring event category possibility arrangement, the real-time possibility arrangement catalogue corresponding to the public health monitoring image and the event category possibility arrangement; the priority evaluation index corresponding to the public health monitoring image can be determined by monitoring the event type possibility arrangement, the event type possibility arrangement and the type priority condition between the real-time monitoring event type catalog and the real-time possibility arrangement catalog. And the artificial intelligent recognition thread can be optimized according to the category evaluation index and the priority evaluation index to obtain the target artificial intelligent recognition thread for text clustering. It should be understood that when determining the event type likelihood arrangement corresponding to the likelihood arrangement to be selected, the method adopts a mode of associating the important behavior monitoring description array of the text with the hierarchical type variable description array corresponding to the likelihood arrangement to be selected, so that various monitoring behaviors of the text can be associated with various types of description rows of the likelihood arrangement to be selected, in other words, when carrying out regression analysis on the likelihood arrangement of the text, the regression analysis accuracy of the likelihood arrangement can be improved; meanwhile, when the artificial intelligent recognition thread is configured, a priority level evaluation index is newly added, and the priority level evaluation index introduces category priority level conditions between real-time possibility arrangement and real-time monitoring event category, so that the correlation conditions between regression analysis monitoring event category and regression analysis possibility arrangement can be enhanced, and the regression analysis of category and layering category is more accurate; in addition, the application can effectively reduce the participation of manpower, and the artificial intelligent recognition thread can work continuously, thereby guaranteeing the reliability of public health, and improving the readiness and the reliability of the public health image data query through the artificial intelligent recognition thread when the public health image data query is carried out.

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 a public health data query method applied to image analysis 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 public health data query method applied to image analysis is shown, and the method may include the following technical solutions described in step S101 to step S105.

Step S101, an important behavior monitoring description array corresponding to the public health monitoring image is obtained, and the monitoring event type possibility arrangement corresponding to the to-be-selected monitoring event type is determined through the artificial intelligent recognition thread and the important behavior monitoring description array.

In the invention, the public health monitoring image can be used for configuring the artificial intelligent recognition thread, and the artificial intelligent recognition thread can be optimized through the configuration of the public health monitoring image, so that the regression analysis result output by the artificial intelligent recognition thread is more and more accurate.

Further, the important behavior monitoring description array corresponding to the public health monitoring image can be loaded into the artificial intelligent recognition thread, and the monitoring event type possibility arrangement corresponding to the to-be-selected monitoring event type can be determined through the artificial intelligent recognition thread and the important behavior monitoring description array of the public health monitoring image. The arrangement of the monitoring event types may be understood as a type regression analysis possibility corresponding to each of the monitoring event types to be selected, in other words, a type regression analysis possibility corresponding to each of the monitoring event types to be selected one by one, which may be referred to as a arrangement of the monitoring event types. For the monitoring description array of the thread and the important behavior through artificial intelligence recognition, the specific implementation manner of determining the monitoring event type possibility arrangement corresponding to the monitoring event type to be selected can be as follows: the important behavior monitoring description array can be loaded to a category dividing unit in the artificial intelligent recognition thread, and the possibility description array corresponding to the category of the monitoring event to be selected can be output through the category dividing unit; and then, obtaining the category regression analysis possibility corresponding to the category of the monitoring event to be selected according to the possibility description array, and determining the possibility arrangement formed by the category regression analysis possibility as the monitoring event category possibility arrangement.

After the important behavior monitoring description array corresponding to the public health monitoring image is loaded to the category dividing unit, the category dividing unit can generate a possibility feature vector (which can be called a possibility description array) from the important behavior monitoring description array, wherein the possibility description array comprises one or more category regression analysis possibilities, and each category regression analysis possibility corresponds to one category of the monitoring event to be selected. Wherein, the likelihood description array contains the likelihood, namely the likelihood arrangement of the monitoring event type. The monitoring event type (which may be referred to as a regression analysis monitoring event type) to which the public health monitoring image belongs can be determined through the monitoring event type possibility arrangement, for example, the monitoring event type corresponding to the maximum type regression analysis possibility can be determined as the regression analysis monitoring event type.

Alternatively, it may be understood that the pre-configured translation unit may be deployed in an artificial intelligence recognition thread, so that the public health monitoring image may be directly loaded into the artificial intelligence recognition thread, and the important behavior monitoring description array corresponding to the public health monitoring image may be extracted by the translation unit deployed in the artificial intelligence recognition thread, and then the possibility description array (i.e. the output monitoring event type possibility arrangement) may be output according to the important behavior monitoring description array.

Step S102, a hierarchical category variable description array corresponding to the candidate possibility arrangement is obtained, the important behavior monitoring description array and the hierarchical category variable description array are associated through an artificial intelligent recognition thread, and event category possibility arrangement corresponding to the candidate possibility arrangement is determined according to an association result.

In the invention, one obtaining way of the hierarchical category variable description array corresponding to the candidate possibility arrangement can be as follows: the text information corresponding to the candidate possibility arrangement is loaded to the translation unit, and the text information is obtained directly through the translation unit. Optionally, in an embodiment of the present invention, the obtaining manner of the hierarchical category variable description array corresponding to the candidate likelihood arrangement may further be: and commonly constructing a hierarchical category variable description array for representing the corresponding candidate possibility arrangement through the data of the related data corresponding to the candidate possibility arrangement. The specific modes of the method can be as follows: related data clusters corresponding to the candidate possibility arrangement can be obtained; then, a data description array corresponding to each relevant data one by one in the relevant data cluster can be obtained, and the data description arrays corresponding to each relevant data one by one can be spliced to obtain a spliced vector description array; and then, obtaining the data number of the related data included in the related data cluster, and determining the hierarchical category variable description array corresponding to the candidate possibility arrangement according to the spliced vector description array and the data number.

Further, after the hierarchical type variable description array corresponding to the candidate possibility arrangement is obtained, the important behavior monitoring description array of the public health monitoring image and the hierarchical type variable description array can be subjected to vector association, so that the association possibility between the public health monitoring image and each candidate possibility arrangement can be obtained, the association possibility can be used as the hierarchical type regression analysis possibility corresponding to the candidate possibility arrangement, and the hierarchical type regression analysis possibility corresponding to each candidate possibility arrangement can form the event type possibility arrangement. Wherein, vector association is carried out on the important behavior monitoring description array and the layering type variable description array of the public health monitoring image.

Step S103, determining a category evaluation index corresponding to the public health monitoring image according to the real-time monitoring event category catalogue, the monitoring event category possibility arrangement, the real-time possibility arrangement catalogue and the event category possibility arrangement corresponding to the public health monitoring image.

In the invention, each public health monitoring image can be provided with two catalogue data, one catalogue data can be a real-time monitoring event catalogue corresponding to the public health monitoring image, and the other catalogue data can be a real-time main pushing layering catalogue corresponding to the public health monitoring image. Through the real-time monitoring event category catalogue, the monitoring event category possibility arrangement, the real-time possibility arrangement catalogue and the event category possibility arrangement, the category evaluation index corresponding to the public health monitoring image can be determined. The specific method comprises the following steps: the first type evaluation index corresponding to the public health monitoring image can be determined according to the real-time monitoring event type catalogue and the monitoring event type possibility arrangement; then, according to the real-time possibility arrangement catalog and the event type possibility arrangement, a second type of evaluation index corresponding to the public health monitoring image can be determined; based on the first category evaluation index and the second category evaluation index, category evaluation indexes can be determined.

The first type of evaluation index (wherein the evaluation index can be understood as a loss value) and the second type of evaluation index in the present invention can refer to negative logarithmic cross entropy loss function values in a cross entropy loss function, and the first type of evaluation index can be determined by real-time monitoring of event type catalogue and monitoring of event type probability arrangement; the second category evaluation index may be determined by a real-time likelihood arrangement catalog and an event category likelihood arrangement. The specific implementation manner for determining the first kind of evaluation index may be: the category regression analysis possibility corresponding to the real-time monitoring event category catalogue can be obtained in the monitoring event category possibility arrangement, and the category regression analysis possibility corresponding to the rest monitoring event categories can be obtained; the other monitoring event types are not less than one monitoring event type to be selected except the real-time monitoring event type catalogue; subsequently, a first type of performance evaluation algorithm may be obtained; the category matching coefficient corresponding to the real-time monitoring event category catalogue in the first category performance evaluation algorithm can be configured to be an effective parameter, and a first-stage original performance evaluation structure can be generated according to the category regression analysis possibility of the effective parameter and the real-time monitoring event category catalogue; the class matching coefficients corresponding to the other monitoring event classes in the first class performance evaluation algorithm can be configured as invalid vectors, and a second-stage original performance evaluation structure can be generated according to the regression analysis possibility of the classes corresponding to the invalid vectors and the other monitoring event classes; then, the first-stage original performance evaluation structure and the second-stage original performance evaluation structure can be subjected to calculation processing, and then the first-type evaluation index can be obtained.

It should be understood that, taking the number of possible arrangements to be selected as not less than one as an example, the specific implementation manner for determining the second kind of evaluation index may be: the hierarchical type regression analysis possibility corresponding to the real-time possibility arrangement catalog can be obtained in the event type possibility arrangement, and the hierarchical type regression analysis possibility corresponding to the rest possibility arrangement is obtained; the rest possibility arrangement is not less than one to-be-selected possibility arrangement except the real-time possibility arrangement catalog; subsequently, a second type of performance evaluation algorithm may be obtained; the hierarchical type matching coefficient corresponding to the real-time possibility arrangement catalog in the second type performance evaluation algorithm can be configured as an effective parameter, and a third-stage original performance evaluation structure can be generated according to the hierarchical type regression analysis possibility corresponding to the effective parameter and the real-time possibility arrangement catalog; the hierarchical type matching coefficients corresponding to the rest of the possibility arrangement in the second type performance evaluation algorithm can be configured as invalid vectors, and a fourth-stage original performance evaluation structure can be generated according to the hierarchical type regression analysis possibility corresponding to the invalid vectors and the rest of the possibility arrangement; and then, calculating the third-stage original performance evaluation structure and the fourth-stage original performance evaluation structure to obtain a second type of evaluation index.

It should be appreciated that after the first category rating index and the second category rating index are determined, a final category rating index (which may be referred to as a total category rating index) may be determined. The specific method comprises the following steps: a first calculation vector corresponding to the first type of evaluation index and a second calculation vector corresponding to the second type of evaluation index can be obtained; then, the first calculation vector and the first type of evaluation index can be subjected to calculation processing to obtain a first calculation performance evaluation result; the second calculation vector and the second type of evaluation index can be subjected to calculation processing to obtain a second calculation performance evaluation result; the first calculation performance evaluation result and the second calculation performance evaluation result can be clustered, and then the category evaluation index can be obtained.

Step S104, determining a priority evaluation index corresponding to the public health monitoring image according to the monitoring event type possibility arrangement, the event type possibility arrangement and the type priority condition between the real-time monitoring event type catalog and the real-time possibility arrangement catalog; the category priority condition refers to that the category priority corresponding to the category catalog of the real-time monitoring event is greater than the category priority corresponding to the real-time possibility arrangement catalog.

In the invention, when determining the performance evaluation result for configuring and optimizing the artificial intelligent recognition thread, the invention also newly introduces a priority evaluation index in addition to the above-mentioned kind of evaluation index. The priority rating index may be determined by monitoring in real time the category priority condition between the event category and the real time likelihood arrangement catalog. The specific modes of the method can be as follows: the category regression analysis possibility corresponding to the real-time monitoring event category catalogue can be obtained in the monitoring event category possibility arrangement, and the layering category regression analysis possibility corresponding to the real-time event category possibility arrangement catalogue is obtained in the event category possibility arrangement; then, determining the possibility of hierarchical type regression analysis corresponding to the real-time possibility arrangement catalog and the possibility comparison result between the type regression analysis corresponding to the real-time monitoring event type catalog; and determining a priority evaluation index according to the probability comparison result and the category priority condition.

The category priority level condition can be used for restraining the hierarchical category regression analysis possibility corresponding to the real-time possibility arrangement catalog and the magnitude relation between the category regression analysis possibility corresponding to the real-time monitoring event category catalog; the specific method for determining the priority rating index according to the likelihood comparison result and the category priority condition may be: a performance expression algorithm can be obtained; if the probability comparison result is positive, determining the hierarchical type regression analysis probability corresponding to the real-time probability arrangement catalog, wherein the size relationship between the hierarchical type regression analysis probability corresponding to the real-time monitoring event type catalog does not accord with the type priority condition, and determining a priority evaluation index according to the target calculation vector in the performance expression algorithm and the probability comparison result; if the probability comparison result is negative, the hierarchical type regression analysis probability corresponding to the real-time probability arrangement catalog can be determined, the size relationship between the hierarchical type regression analysis probability corresponding to the real-time monitoring event type catalog accords with the type priority condition, and the priority evaluation index can be determined according to the invalid vector, the target calculation vector and the probability comparison result in the performance expression algorithm.

It should be appreciated that for the identification of the upper class (i.e., the class with the greater range and higher class priority) it is easy to identify the lower class (i.e., the hierarchical class with the finer granularity, smaller range and lower class priority), in other words, the lower class is more deeply hierarchical, finer granularity, and the difficulty of fine granularity should be greater, then the regression analysis probability should also be smaller than that of the upper class. The priority evaluation index is determined according to the category priority conditions of the upper and lower categories, and the artificial intelligent recognition thread is optimized according to the priority evaluation index, so that the consistency between the upper and lower categories (such as regression analysis monitoring event category and regression analysis possibility arrangement) output by the configured artificial intelligent recognition thread can be enhanced.

And step 105, optimizing the artificial intelligent recognition thread according to the category evaluation index and the priority evaluation index to obtain a target artificial intelligent recognition thread, and inquiring public health data based on the target artificial intelligent recognition thread.

According to the invention, the total target performance evaluation result can be determined according to the category evaluation index and the priority evaluation index, and then the artificial intelligent recognition thread is optimized according to the target performance evaluation result, so that the target artificial intelligent recognition thread is obtained. The specific method comprises the following steps: a clustering calculation vector corresponding to the category evaluation index and a calculation vector interval corresponding to the priority evaluation index can be obtained; subsequently, the category evaluation index and the clustering calculation vector can be calculated to obtain a target category evaluation index; the priority evaluation index and the calculation vector interval can be calculated to obtain a target priority evaluation index; and clustering calculation processing can be carried out on the target type evaluation index and the target priority evaluation index to obtain a target performance evaluation result, and the artificial intelligent recognition thread is optimized according to the target performance evaluation result to obtain the target artificial intelligent recognition thread.

The specific method for obtaining the target artificial intelligent recognition thread by optimally configuring the artificial intelligent recognition thread according to the target performance evaluation result comprises the following steps: whether the target performance evaluation result meets the model convergence condition can be determined, if the target performance evaluation result does not meet the model convergence condition, the artificial intelligent recognition thread can be optimized (for example, the model coefficient in the artificial intelligent recognition thread is optimized) according to the target performance evaluation result, a new round of configuration is performed after the optimization, a new target performance evaluation result is obtained, if the new target performance evaluation result meets the model convergence condition, the regression analysis result of the artificial intelligent recognition thread can be considered to be accurate enough at the moment, the artificial intelligent recognition thread can not be optimized according to the new target performance evaluation result, and the optimized artificial intelligent recognition thread can be determined as the target artificial intelligent recognition thread. It will be appreciated that public health data query processing may be performed based on the target artificial intelligence recognition thread. For example, loading text data to be clustered into a target artificial intelligent recognition thread, outputting monitoring event type probability arrangement corresponding to the text data to be clustered through the target artificial intelligent recognition thread, and determining regression analysis monitoring event types corresponding to the text data to be clustered through the monitoring event type probability arrangement; the target artificial intelligent recognition thread can output event type probability arrangement corresponding to the text data to be clustered, and regression analysis probability arrangement corresponding to the text data to be clustered can be determined through the event type probability arrangement.

In the embodiment of the invention, after the important behavior monitoring description array corresponding to the public health monitoring image is obtained, the important behavior monitoring description array can be loaded into an artificial intelligent recognition thread, and the monitoring event type probability arrangement corresponding to the type of the monitoring event to be selected can be determined through the artificial intelligent recognition thread and the important behavior monitoring description array (the type of the monitoring event to be selected can be in the type of the monitoring event to be selected through the monitoring event type probability arrangement, so that the type of the regression analysis monitoring event corresponding to the public health monitoring image is determined); through the artificial intelligent recognition thread, the important behavior monitoring description array corresponding to the public health monitoring image can be associated with the hierarchical type variable description array of the to-be-selected possibility arrangement, so that the event type possibility arrangement corresponding to the to-be-selected possibility arrangement can be determined (the regression analysis possibility arrangement corresponding to the public health monitoring image can be determined in the to-be-selected possibility arrangement through the event type possibility arrangement). The category evaluation index corresponding to the public health monitoring image can be determined through the real-time monitoring event category catalogue corresponding to the public health monitoring image, the monitoring event category possibility arrangement, the real-time possibility arrangement catalogue corresponding to the public health monitoring image and the event category possibility arrangement; the priority evaluation index corresponding to the public health monitoring image can be determined by monitoring the event type possibility arrangement, the event type possibility arrangement and the type priority condition between the real-time monitoring event type catalog and the real-time possibility arrangement catalog. And the artificial intelligent recognition thread can be optimized according to the category evaluation index and the priority evaluation index to obtain the target artificial intelligent recognition thread for text clustering. It should be understood that when determining the event type likelihood arrangement corresponding to the likelihood arrangement to be selected, the method adopts a mode of associating the important behavior monitoring description array of the text with the hierarchical type variable description array corresponding to the likelihood arrangement to be selected, so that various monitoring behaviors of the text can be associated with various types of description rows of the likelihood arrangement to be selected, in other words, when carrying out regression analysis on the likelihood arrangement of the text, the consideration of the hierarchy is deeper, and the regression analysis accuracy of the likelihood arrangement can be improved; meanwhile, when the artificial intelligent recognition thread is configured, a priority level evaluation index is newly added, and the priority level evaluation index introduces category priority level conditions between real-time possibility arrangement and real-time monitoring event category, so that the correlation conditions between regression analysis monitoring event category and regression analysis possibility arrangement can be enhanced, and the regression analysis of category and layering category is more accurate; in addition, the invention can effectively reduce the participation of manpower, and the artificial intelligent recognition thread can work continuously, thereby guaranteeing the reliability of public health, and improving the readiness and the reliability of the public health image data query through the artificial intelligent recognition thread when the public health image data query is carried out.

The embodiment of the invention provides a public health data query method applied to image analysis, which carries out vector association on an important behavior monitoring description array and a layering type variable description array of a public health monitoring image, and determines the flow of event type possibility arrangement according to association results. The process is illustrated by taking the example that the number of the candidate possibility arrangements is not less than one, the at least one candidate possibility arrangement comprises a candidate possibility arrangement Xa (a is a positive integer), and the hierarchical type variable description array comprises a hierarchical type variable description array Ya corresponding to the candidate possibility arrangement Xa. The flow may include at least the following step S501 to step S503.

In step S501, the important behavior monitoring description array and the hierarchical category variable description array Ya are loaded to the hierarchical category dividing unit in the artificial intelligence recognition thread.

Specifically, the artificial intelligence recognition thread may further include a hierarchical category classification unit in addition to the category classification unit. What is adopted by the hierarchical category classification unit is a vector association mechanism. The important behavior monitoring description array and the hierarchical category variable description array (such as the hierarchical category variable description array Ya) can be vector-associated in the hierarchical category dividing unit.

In step S502, the hierarchical category dividing unit associates the important behavior monitoring description array and the hierarchical category variable description array Ya, determines the association possibility between the important behavior monitoring description array and the hierarchical category variable description array Ya, and uses the association possibility as the hierarchical category regression analysis possibility corresponding to the candidate possibility arrangement Xa.

Specifically, for associating the important behavior monitoring description array with the hierarchical category variable description array Ya, a specific implementation manner for determining the association possibility between the important behavior monitoring description array and the hierarchical category variable description array Ya may be: the association coefficient queue in the layering type dividing unit can be obtained; then, the association coefficient queue, the important behavior monitoring description array and the layering type variable description array Ya can be subjected to calculation processing to obtain a calculation vector description array corresponding to the candidate possibility arrangement Xa; the calculation possibility description array can be loaded to the function statistics unit, and vector conversion processing can be carried out on the calculation vector description array through an analysis mode in the function statistics unit, so that the correlation possibility between the important behavior monitoring description array and the layering type variable description array Ya is obtained.

In other words, the hierarchical type dividing unit may include a correlation coefficient (may be referred to as a correlation coefficient queue), and the correlation coefficient, the important behavior monitoring description array and the hierarchical type variable description array Ya are calculated to obtain a correlation vector (may be referred to as a calculation vector description array) corresponding to the candidate likelihood arrangement, and the correlation vector is loaded into the function statistics unit to output a likelihood coefficient corresponding to the correlation vector, where the likelihood coefficient may be used as a hierarchical type regression analysis likelihood corresponding to the candidate likelihood arrangement Xa.

Optionally, in one possible implementation embodiment, to simplify the calculation, the correlation coefficient queue may be parsed to obtain a plurality of parsed projection queues (with lower dimensions), and then the parsed projection queues are used to calculate the important behavior monitoring description array and the hierarchical category variable description array (such as the hierarchical category variable description array Ya). The specific method comprises the following steps: the correlation coefficient queue can be subjected to queue analysis to obtain a first analysis projection queue and a second analysis projection queue; then, the first analysis projection queue and the important behavior monitoring description array can be subjected to calculation processing to obtain a first projection vector; calculating the second analysis projection queue and the hierarchical category variable description array Ya to obtain a second projection vector; the first projection vector and the second projection vector can be calculated to obtain a calculation vector description array corresponding to the candidate possibility arrangement Xa.

In step S503, when it is determined that the at least one candidate likelihood arrangement corresponds to one-by-one hierarchical seed regression analysis likelihood, the likelihood arrangement formed by the at least one hierarchical seed regression analysis likelihood is determined as the event type likelihood arrangement.

Specifically, when it is determined that at least one of the candidate likelihood arrangements corresponds to one-by-one hierarchical seed regression analysis likelihood, the event type likelihood arrangement corresponding to the candidate likelihood arrangement can be determined.

In the embodiment of the invention, when the event type possibility arrangement corresponding to the possibility arrangement to be selected is determined (namely, when the hierarchical type regression analysis possibility corresponding to the possibility arrangement to be selected is determined), the important behavior monitoring description array of the public health monitoring image is related to the semantic information of the possibility arrangement to be selected by means of bilinear tensor analysis and vector association, so that the semantic information of the possibility arrangement to be selected is effectively utilized, the identification capability of the artificial intelligent identification thread to the possibility arrangement to be selected can be enhanced by means of vector association, and the type identification accuracy of the configured artificial intelligent identification thread is effectively improved.

In a possible implementation embodiment, the obtaining the hierarchical category variable description array corresponding to the candidate likelihood arrangement includes: acquiring a related data cluster corresponding to the arrangement of the possibility to be selected; obtaining a data description array corresponding to each related data one by one in the related data cluster, and splicing the data description arrays corresponding to each related data one by one to obtain a spliced vector description array; and obtaining the data number of the related data included in the related data cluster, and determining a hierarchical category variable description array corresponding to the candidate possibility arrangement according to the splicing vector description array and the data number.

It can be understood that the hierarchical category variable description array can be more accurately obtained by splicing the data description arrays corresponding to each related data one by one.

In a possible embodiment, the number of the candidate possibility arrangements is not less than one; the at least one candidate possibility arrangement comprises a candidate possibility arrangement Xa, wherein the hierarchical type variable description array comprises a hierarchical type variable description array Ya corresponding to the candidate possibility arrangement Xa, and a is a positive integer; the method for determining event type probability arrangement corresponding to the candidate probability arrangement according to the association result includes the steps of: loading the important behavior monitoring description array and the layering type variable description array Ya into a layering type dividing unit in the artificial intelligent recognition thread; the important behavior monitoring description array and the layering type variable description array Ya are associated through the layering type dividing unit, the association possibility between the important behavior monitoring description array and the layering type variable description array Ya is determined, and the association possibility is used as layering type regression analysis possibility corresponding to the candidate possibility arrangement Xa; and when the possibility arrangement formed by the at least one hierarchical type regression analysis possibility is determined, determining the possibility arrangement formed by the at least one hierarchical type regression analysis possibility as the event type possibility arrangement.

It can be understood that when the important behavior monitoring description array and the hierarchical category variable description array are associated by the artificial intelligence recognition thread, the problem of inaccurate association is solved, so that the event category probability arrangement corresponding to the candidate probability arrangement can be determined more accurately according to the association result.

In one possible implementation embodiment, the associating, by the hierarchical category dividing unit, the important behavior monitoring description array and the hierarchical category variable description array Ya, and determining the possibility of association between the important behavior monitoring description array and the hierarchical category variable description array Ya includes: obtaining a correlation coefficient queue in the hierarchical category dividing unit; calculating the association coefficient queue, the important behavior monitoring description array and the layering type variable description array Ya to obtain a calculation vector description array corresponding to the candidate possibility arrangement Xa; and loading the calculation possibility description array to a function statistics unit, and carrying out vector conversion processing on the calculation vector description array by an analysis mode in the function statistics unit to obtain the association possibility between the important behavior monitoring description array and the layering type variable description array Ya.

It can be understood that when the hierarchical category dividing unit is used for associating the important behavior monitoring description array and the hierarchical category variable description array Ya, the problem of association abnormality is improved, so that the association possibility between the important behavior monitoring description array and the hierarchical category variable description array Ya can be more accurately determined

In one possible implementation embodiment, the calculating the association coefficient queue, the important behavior monitoring description array and the hierarchical category variable description array Ya to obtain a calculation vector description array corresponding to the candidate likelihood arrangement Xa includes: performing queue analysis on the association coefficient queue to obtain a first analysis projection queue and a second analysis projection queue; calculating the first analysis projection queue and the important behavior monitoring description array to obtain a first projection vector; calculating the second analysis projection queue and the hierarchical category variable description array Ya to obtain a second projection vector; and carrying out calculation processing on the first projection vector and the second projection vector to obtain a calculation vector description array corresponding to the candidate possibility arrangement Xa.

It can be understood that when the association coefficient queue, the important behavior monitoring description array and the hierarchical category variable description array Ya are subjected to calculation processing, the problem of abnormal calculation is solved, so that the calculation vector description array corresponding to the candidate possibility arrangement Xa can be obtained more accurately.

In a possible implementation embodiment, the determining the category evaluation index corresponding to the public health monitoring image according to the real-time monitoring event category catalog corresponding to the public health monitoring image, the monitoring event category possibility arrangement, the real-time possibility arrangement catalog corresponding to the public health monitoring image and the event category possibility arrangement includes: determining a first type evaluation index corresponding to the public health monitoring image according to the real-time monitoring event type catalogue and the monitoring event type possibility arrangement; determining a second type of evaluation index corresponding to the public health monitoring image according to the real-time possibility arrangement catalog and the event type possibility arrangement; and determining the category evaluation index according to the first category evaluation index and the second category evaluation index.

It can be understood that according to the real-time monitoring event category catalogue corresponding to the public health monitoring image, the monitoring event category possibility arrangement, the real-time possibility arrangement catalogue corresponding to the public health monitoring image and the event category possibility arrangement, the problem of inaccurate possibility distribution is solved, and therefore the category evaluation index corresponding to the public health monitoring image can be determined more accurately.

In a possible embodiment, the number of the types of the monitoring events to be selected is not less than one; the determining a first category evaluation index corresponding to the public health monitoring image according to the real-time monitoring event category catalogue and the monitoring event category possibility arrangement comprises the following steps: obtaining the category regression analysis possibility corresponding to the real-time monitoring event category catalogue in the monitoring event category possibility arrangement, and the category regression analysis possibility corresponding to the rest monitoring event categories; the other monitoring event types are not less than one monitoring event type to be selected except the real-time monitoring event type catalogue; obtaining a first type of performance evaluation algorithm; configuring a category matching coefficient corresponding to the real-time monitoring event category catalogue in the first category performance evaluation algorithm as an effective parameter, and generating a first staged original performance evaluation structure according to the category regression analysis possibility of the effective parameter and the real-time monitoring event category catalogue; configuring class matching coefficients corresponding to the other monitoring event classes in the first class performance evaluation algorithm as invalid vectors, and generating a second-stage original performance evaluation structure according to the class regression analysis possibility of the invalid vectors and the other monitoring event classes; and performing calculation processing on the first-stage original performance evaluation structure and the second-stage original performance evaluation structure to obtain the first-type evaluation index.

It can be understood that when the real-time monitoring event category catalogue and the monitoring event category possibility are distributed, the problem of inaccurate possibility distribution is solved, so that the first category evaluation index corresponding to the public health monitoring image can be more accurately determined.

In a possible embodiment, the determining the category evaluation index according to the first category evaluation index and the second category evaluation index includes: obtaining a first calculation vector corresponding to the first type of evaluation index and a second calculation vector corresponding to the second type of evaluation index; calculating the first calculation vector and the first type of evaluation index to obtain a first calculation performance evaluation result; calculating the second calculation vector and the second type of evaluation index to obtain a second calculation performance evaluation result; and clustering the first calculation performance evaluation result and the second calculation performance evaluation result to obtain the category evaluation index.

In a possible implementation embodiment, the determining the priority evaluation index corresponding to the public health monitoring image according to the monitoring event type likelihood arrangement, the event type likelihood arrangement, and the type priority condition between the real-time monitoring event type catalog and the real-time likelihood arrangement catalog includes: acquiring category regression analysis possibility corresponding to the real-time monitoring event category catalogue in the monitoring event category possibility arrangement, and acquiring hierarchical category regression analysis possibility corresponding to the real-time event category possibility arrangement catalogue in the event category possibility arrangement; determining the possibility of hierarchical type regression analysis corresponding to the real-time possibility arrangement catalog and the possibility comparison result between the possibility of type regression analysis corresponding to the real-time monitoring event type catalog; and determining the priority rating index according to the possibility comparison result and the category priority condition.

It can be understood that when the category priority condition between the monitored event category directory and the real-time monitored event category directory is set according to the monitored event category likelihood arrangement, the event category likelihood arrangement and the category priority condition between the real-time monitored event category directory and the real-time likelihood arrangement directory, the problem of inaccurate category priority condition determination is solved, so that the priority evaluation index corresponding to the public health monitoring image can be more accurately determined.

In a possible implementation embodiment, the category priority condition is used for constraining a hierarchical category regression analysis possibility corresponding to the real-time likelihood arrangement catalog and a magnitude relation between the category regression analysis possibility corresponding to the real-time monitoring event category catalog; the determining the priority rating index according to the likelihood comparison result and the category priority condition comprises the following steps: obtaining a performance expression algorithm; if the probability comparison result is positive, determining hierarchical type regression analysis probability corresponding to the real-time probability arrangement catalog, wherein the size relationship between the hierarchical type regression analysis probability corresponding to the real-time monitoring event type catalog and the type regression analysis probability corresponding to the real-time monitoring event type catalog does not accord with the type priority condition, and determining the priority evaluation index according to a target calculation vector in the performance expression algorithm and the probability comparison result; and if the probability comparison result is a negative number, determining the hierarchical type regression analysis probability corresponding to the real-time probability arrangement catalog, wherein the size relationship between the hierarchical type regression analysis probability corresponding to the real-time monitoring event type catalog and the type regression analysis probability corresponding to the real-time monitoring event type catalog accords with the type priority condition, and determining the priority evaluation index according to an invalid vector, the target calculation vector and the probability comparison result in the performance expression algorithm.

In a possible implementation embodiment, the optimizing the artificial intelligence recognition thread according to the category evaluation index and the priority evaluation index to obtain a target artificial intelligence recognition thread includes: obtaining a clustering calculation vector corresponding to the category evaluation index and a calculation vector section corresponding to the priority evaluation index; calculating the category evaluation index and the clustering calculation vector to obtain a target category evaluation index; calculating the priority evaluation index and the calculation vector interval to obtain a target priority evaluation index; and carrying out clustering calculation processing on the target type evaluation index and the target priority evaluation index to obtain a target performance evaluation result, and optimizing the artificial intelligent recognition thread according to the target performance evaluation result to obtain the target artificial intelligent recognition thread.

On the basis of the above, there is provided a public health data query device applied to image analysis, the device comprising:

the arrangement determining module is used for obtaining an important behavior monitoring description array corresponding to the public health monitoring image, and determining the monitoring event type possibility arrangement corresponding to the monitoring event type to be selected through the artificial intelligent identification thread and the important behavior monitoring description array;

An array association module for obtaining a hierarchical category variable description array corresponding to the candidate possibility arrangement, associating the important behavior monitoring description array with the hierarchical category variable description array through the artificial intelligent recognition thread,

the index determining module is used for determining event type probability arrangement corresponding to the candidate probability arrangement according to the association result;

determining a category evaluation index corresponding to the public health monitoring image according to a real-time monitoring event category catalog corresponding to the public health monitoring image, the monitoring event category possibility arrangement, a real-time possibility arrangement catalog corresponding to the public health monitoring image and the event category possibility arrangement;

the index evaluation module is used for determining a priority evaluation index corresponding to the public health monitoring image according to the monitoring event type possibility arrangement, the event type possibility arrangement and the type priority condition between the real-time monitoring event type catalog and the real-time possibility arrangement catalog; the category priority level condition refers to that the category priority level corresponding to the real-time monitoring event category directory is greater than the category priority level corresponding to the real-time possibility arrangement directory;

And the data query module is used for optimizing the artificial intelligent recognition thread according to the category evaluation index and the priority evaluation index to obtain a target artificial intelligent recognition thread, and carrying out public health data query processing by combining the target artificial intelligent recognition thread.

On the basis of the above, a public health data query system for image analysis 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 to implement 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, after obtaining the important behavior monitoring description array corresponding to the public health monitoring image, the important behavior monitoring description array may be loaded into the artificial intelligent recognition thread, and the monitoring event type probability arrangement corresponding to the to-be-selected monitoring event type may be determined through the artificial intelligent recognition thread and the important behavior monitoring description array (the regression analysis monitoring event type corresponding to the public health monitoring image may be determined in the to-be-selected monitoring event type through the monitoring event type probability arrangement); through the artificial intelligent recognition thread, the important behavior monitoring description array corresponding to the public health monitoring image can be associated with the hierarchical type variable description array of the to-be-selected possibility arrangement, so that the event type possibility arrangement corresponding to the to-be-selected possibility arrangement can be determined (the regression analysis possibility arrangement corresponding to the public health monitoring image can be determined in the to-be-selected possibility arrangement through the event type possibility arrangement). The category evaluation index corresponding to the public health monitoring image can be determined through the real-time monitoring event category catalogue corresponding to the public health monitoring image, the monitoring event category possibility arrangement, the real-time possibility arrangement catalogue corresponding to the public health monitoring image and the event category possibility arrangement; the priority evaluation index corresponding to the public health monitoring image can be determined by monitoring the event type possibility arrangement, the event type possibility arrangement and the type priority condition between the real-time monitoring event type catalog and the real-time possibility arrangement catalog. And the artificial intelligent recognition thread can be optimized according to the category evaluation index and the priority evaluation index to obtain the target artificial intelligent recognition thread for text clustering. It should be understood that when determining the event type likelihood arrangement corresponding to the likelihood arrangement to be selected, the method adopts a mode of associating the important behavior monitoring description array of the text with the hierarchical type variable description array corresponding to the likelihood arrangement to be selected, so that various monitoring behaviors of the text can be associated with various types of description rows of the likelihood arrangement to be selected, in other words, when carrying out regression analysis on the likelihood arrangement of the text, the regression analysis accuracy of the likelihood arrangement can be improved; meanwhile, when the artificial intelligent recognition thread is configured, a priority level evaluation index is newly added, and the priority level evaluation index introduces category priority level conditions between real-time possibility arrangement and real-time monitoring event category, so that the correlation conditions between regression analysis monitoring event category and regression analysis possibility arrangement can be enhanced, and the regression analysis of category and layering category is more accurate; in addition, the invention can effectively reduce the participation of manpower, and the artificial intelligent recognition thread can work continuously, thereby guaranteeing the reliability of public health, and improving the readiness and the reliability of the public health image data query through the artificial intelligent recognition thread when the public health image data query is carried out.

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.

While the basic concepts have been described above, it will be apparent to those skilled in the art that the foregoing detailed disclosure is by way of example only and is not intended to be limiting. Although not explicitly described herein, various modifications, improvements and adaptations of the application may occur to one skilled in the art. Such modifications, improvements, and modifications are intended to be suggested within the present disclosure, and therefore, such modifications, improvements, and adaptations are intended to be within the spirit and scope of the exemplary embodiments of the present disclosure.

Meanwhile, the present application uses specific words to describe embodiments of the present application. Reference to "one embodiment," "an embodiment," and/or "some embodiments" means that a particular feature, structure, or characteristic is associated with at least one embodiment of the application. Thus, it should be emphasized and should be appreciated that two or more references to "an embodiment" or "one embodiment" or "an alternative embodiment" in various positions in this specification are not necessarily referring to the same embodiment. Furthermore, certain features, structures, or characteristics of one or more embodiments of the application may be combined as suitable.

Furthermore, those skilled in the art will appreciate that the various aspects of the application are illustrated and described in the context of a number of patentable categories or circumstances, including any novel and useful procedures, machines, products, or materials, or any novel and useful modifications thereof. Accordingly, aspects of the application may be performed entirely by hardware, entirely by software (including firmware, resident software, micro-code, etc.) or by a combination of hardware and software. The above hardware or software may be referred to as a "data block," module, "" engine, "" unit, "" component, "or" system. Furthermore, aspects of the application may take the form of a computer product, comprising computer-readable program code, embodied in one or more computer-readable media.

The computer storage medium may contain a propagated data signal with the computer program code embodied therein, for example, on a baseband or as part of a carrier wave. The propagated signal may take on a variety of forms, including electro-magnetic, optical, etc., or any suitable combination thereof. A computer storage medium may be any computer readable medium that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code located on a computer storage medium may be propagated through any suitable medium, including radio, cable, fiber optic cable, RF, or the like, or a combination of any of the foregoing.

Computer program code necessary for operation of portions of the present application may be written in any one or more programming languages, including an object oriented programming language such as Java, scala, smalltalk, eiffel, JADE, emerald, C ++, c#, vb net, python, and the like, a conventional programming language such as C language, visualBasic, fortran2003, perl, COBOL2002, PHP, ABAP, a dynamic programming language such as Python, ruby, and Groovy, or other programming languages, and the like. The program code may execute entirely on the user's computer or as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any form of network, such as a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet), or the use of services such as software as a service (SaaS) in a cloud computing environment.

Furthermore, the order in which the elements and sequences are presented, the use of numerical letters, or other designations are used in the application is not intended to limit the sequence of the processes and methods unless specifically recited in the claims. While certain presently useful inventive embodiments have been discussed in the foregoing disclosure, by way of example, it is to be understood that such details are merely illustrative and that the appended claims are not limited to the disclosed embodiments, but, on the contrary, are intended to cover all modifications and equivalent arrangements included within the spirit and scope of the embodiments of the application. For example, while the system components described above may be implemented by hardware devices, they may also be implemented solely by software solutions, such as installing the described system on an existing server or mobile device.

Similarly, it should be noted that in order to simplify the description of the present disclosure and thereby aid in understanding one or more inventive embodiments, various features are sometimes grouped together in a single embodiment, figure, or description thereof. This method of disclosure, however, is not intended to imply that more features than are required by the subject application. Indeed, less than all of the features of a single embodiment disclosed above.

In some embodiments, numbers describing the components, number of attributes are used, it being understood that such numbers being used in the description of embodiments are modified in some examples by the modifier "about," approximately, "or" substantially. Unless otherwise indicated, "about," "approximately," or "substantially" indicate that the numbers allow for adaptive variation. Accordingly, in some embodiments, numerical parameters set forth in the specification and claims are approximations that may vary depending upon the desired properties sought to be obtained by the individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and employ a method for preserving the general number of digits. Although the numerical ranges and parameters set forth herein are approximations in some embodiments for use in determining the breadth of the range, in particular embodiments, the numerical values set forth herein are as precisely as possible.

Each patent, patent application publication, and other material, such as articles, books, specifications, publications, documents, etc., cited herein is hereby incorporated by reference in its entirety. Except for the application history file that is inconsistent or conflicting with this disclosure, the file (currently or later attached to this disclosure) that limits the broadest scope of the claims of this disclosure is also excluded. It is noted that the description, definition, and/or use of the term in the appended claims controls the description, definition, and/or use of the term in this application if there is a discrepancy or conflict between the description, definition, and/or use of the term in the appended claims.

Finally, it should be understood that the embodiments described herein are merely illustrative of the principles of the embodiments of the present application. Other variations are also possible within the scope of the application. Thus, by way of example, and not limitation, alternative configurations of embodiments of the application may be considered in keeping with the teachings of the application. Accordingly, the embodiments of the present application are not limited to the embodiments explicitly described and depicted herein.

The foregoing is merely exemplary of the present application and is not intended to limit the present application. Various modifications and variations of the present application will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. which come within the spirit and principles of the application are to be included in the scope of the claims of the present application.

Claims

1. A public health data query method applied to image analysis, the method at least comprising:

obtaining an important behavior monitoring description array corresponding to a public health monitoring image, and determining the monitoring event type possibility arrangement corresponding to the monitoring event type to be selected through an artificial intelligent recognition thread and the important behavior monitoring description array;

obtaining a hierarchical category variable description array corresponding to the candidate possibility arrangement, associating the important behavior monitoring description array with the hierarchical category variable description array through the artificial intelligent recognition thread,

determining event type probability arrangement corresponding to the candidate probability arrangement according to the association result; determining a category evaluation index corresponding to the public health monitoring image according to a real-time monitoring event category catalog corresponding to the public health monitoring image, the monitoring event category possibility arrangement, a real-time possibility arrangement catalog corresponding to the public health monitoring image and the event category possibility arrangement;

determining a priority evaluation index corresponding to the public health monitoring image according to the monitoring event type possibility arrangement, the event type possibility arrangement and the type priority condition between the real-time monitoring event type catalog and the real-time possibility arrangement catalog; the category priority level condition refers to that the category priority level corresponding to the real-time monitoring event category directory is greater than the category priority level corresponding to the real-time possibility arrangement directory;

And optimizing the artificial intelligent recognition thread according to the category evaluation index and the priority evaluation index to obtain a target artificial intelligent recognition thread, and carrying out public health data query processing by combining the target artificial intelligent recognition thread.

2. The method according to claim 1, wherein the determining, by the artificial intelligence recognition thread and the important behavior monitoring description array, a monitoring event type likelihood arrangement corresponding to a candidate monitoring event type includes:

loading the important behavior monitoring description array to a category dividing unit in the artificial intelligent recognition thread, and outputting a possibility description array corresponding to the category of the monitoring event to be selected through the category dividing unit;

and obtaining the category regression analysis possibility corresponding to the category of the monitoring event to be selected according to the possibility description array, and determining the possibility arrangement formed by the category regression analysis possibility as the category possibility arrangement of the monitoring event.

3. The method according to claim 1, wherein the obtaining the hierarchical category variable description array corresponding to the candidate likelihood arrangement includes:

Acquiring a related data cluster corresponding to the arrangement of the possibility to be selected;

obtaining a data description array corresponding to each related data one by one in the related data cluster, and splicing the data description arrays corresponding to each related data one by one to obtain a spliced vector description array;

and obtaining the data number of the related data included in the related data cluster, and determining a hierarchical category variable description array corresponding to the candidate possibility arrangement according to the splicing vector description array and the data number.

4. The method according to claim 1, wherein the number of candidate likelihood arrangements is not less than one; the at least one candidate possibility arrangement comprises a candidate possibility arrangement Xa, wherein the hierarchical type variable description array comprises a hierarchical type variable description array Ya corresponding to the candidate possibility arrangement Xa, and a is a positive integer; the method for determining event type probability arrangement corresponding to the candidate probability arrangement according to the association result includes the steps of:

loading the important behavior monitoring description array and the layering type variable description array Ya into a layering type dividing unit in the artificial intelligent recognition thread;

The important behavior monitoring description array and the layering type variable description array Ya are associated through the layering type dividing unit, the association possibility between the important behavior monitoring description array and the layering type variable description array Ya is determined, and the association possibility is used as layering type regression analysis possibility corresponding to the candidate possibility arrangement Xa;

when the possibility arrangement formed by the possibility of the hierarchical species regression analysis is determined, determining the possibility arrangement formed by the possibility of the hierarchical species regression analysis as the event type possibility arrangement;

the correlation between the important behavior monitoring description array and the hierarchical category variable description array Ya is determined by the hierarchical category dividing unit, which comprises the following steps:

obtaining a correlation coefficient queue in the hierarchical category dividing unit; calculating the association coefficient queue, the important behavior monitoring description array and the layering type variable description array Ya to obtain a calculation vector description array corresponding to the candidate possibility arrangement Xa;

Loading the calculation possibility description array into a function statistics unit, and performing vector conversion processing on the calculation vector description array in an analysis mode in the function statistics unit to obtain the association possibility between the important behavior monitoring description array and the layering type variable description array Ya;

the calculating the association coefficient queue, the important behavior monitoring description array and the hierarchical category variable description array Ya to obtain a calculation vector description array corresponding to the candidate possibility arrangement Xa includes:

performing queue analysis on the association coefficient queue to obtain a first analysis projection queue and a second analysis projection queue;

calculating the first analysis projection queue and the important behavior monitoring description array to obtain a first projection vector;

calculating the second analysis projection queue and the hierarchical category variable description array Ya to obtain a second projection vector;

and carrying out calculation processing on the first projection vector and the second projection vector to obtain a calculation vector description array corresponding to the candidate possibility arrangement Xa.

5. The method according to claim 1, wherein the determining the category evaluation index corresponding to the public health monitoring image according to the real-time monitoring event category catalog corresponding to the public health monitoring image, the monitoring event category likelihood arrangement, the real-time likelihood arrangement catalog corresponding to the public health monitoring image, and the event category likelihood arrangement includes:

Determining a first type evaluation index corresponding to the public health monitoring image according to the real-time monitoring event type catalogue and the monitoring event type possibility arrangement;

determining a second type of evaluation index corresponding to the public health monitoring image according to the real-time possibility arrangement catalog and the event type possibility arrangement;

determining the category evaluation index according to the first category evaluation index and the second category evaluation index;

wherein the number of the types of the monitoring events to be selected is not less than one; the determining a first category evaluation index corresponding to the public health monitoring image according to the real-time monitoring event category catalogue and the monitoring event category possibility arrangement comprises the following steps:

obtaining the category regression analysis possibility corresponding to the real-time monitoring event category catalogue in the monitoring event category possibility arrangement, and the category regression analysis possibility corresponding to the rest monitoring event categories; the other monitoring event types are not less than one monitoring event type to be selected except the real-time monitoring event type catalogue; obtaining a first type of performance evaluation algorithm; configuring a category matching coefficient corresponding to the real-time monitoring event category catalogue in the first category performance evaluation algorithm as an effective parameter, and generating a first staged original performance evaluation structure according to the category regression analysis possibility of the effective parameter and the real-time monitoring event category catalogue; configuring class matching coefficients corresponding to the other monitoring event classes in the first class performance evaluation algorithm as invalid vectors, and generating a second-stage original performance evaluation structure according to the class regression analysis possibility of the invalid vectors and the other monitoring event classes; and performing calculation processing on the first-stage original performance evaluation structure and the second-stage original performance evaluation structure to obtain the first-type evaluation index.

6. The method of claim 4, wherein determining the category rating measure based on the first category rating measure and the second category rating measure comprises:

obtaining a first calculation vector corresponding to the first type of evaluation index and a second calculation vector corresponding to the second type of evaluation index;

calculating the first calculation vector and the first type of evaluation index to obtain a first calculation performance evaluation result;

calculating the second calculation vector and the second type of evaluation index to obtain a second calculation performance evaluation result;

and clustering the first calculation performance evaluation result and the second calculation performance evaluation result to obtain the category evaluation index.

7. The method according to claim 1, wherein the determining the priority evaluation index corresponding to the public health monitoring image according to the monitoring event type likelihood arrangement, the event type likelihood arrangement, and the type priority condition between the real-time monitoring event type catalog and the real-time likelihood arrangement catalog includes:

Acquiring category regression analysis possibility corresponding to the real-time monitoring event category catalogue in the monitoring event category possibility arrangement, and acquiring hierarchical category regression analysis possibility corresponding to the real-time event category possibility arrangement catalogue in the event category possibility arrangement;

determining the possibility of hierarchical type regression analysis corresponding to the real-time possibility arrangement catalog and the possibility comparison result between the possibility of type regression analysis corresponding to the real-time monitoring event type catalog;

determining the priority evaluation index according to the likelihood comparison result and the category priority condition;

the category priority condition is used for restraining the hierarchical category regression analysis possibility corresponding to the real-time possibility arrangement catalog and the magnitude relation between the category regression analysis possibility corresponding to the real-time monitoring event category catalog; the determining the priority rating index according to the likelihood comparison result and the category priority condition comprises the following steps:

obtaining a performance expression algorithm; if the probability comparison result is positive, determining hierarchical type regression analysis probability corresponding to the real-time probability arrangement catalog, wherein the size relationship between the hierarchical type regression analysis probability corresponding to the real-time monitoring event type catalog and the type regression analysis probability corresponding to the real-time monitoring event type catalog does not accord with the type priority condition, and determining the priority evaluation index according to a target calculation vector in the performance expression algorithm and the probability comparison result;

And if the probability comparison result is a negative number, determining the hierarchical type regression analysis probability corresponding to the real-time probability arrangement catalog, wherein the size relationship between the hierarchical type regression analysis probability corresponding to the real-time monitoring event type catalog and the type regression analysis probability corresponding to the real-time monitoring event type catalog accords with the type priority condition, and determining the priority evaluation index according to an invalid vector, the target calculation vector and the probability comparison result in the performance expression algorithm.

8. The method of claim 1, wherein optimizing the artificial intelligence recognition thread according to the category evaluation index and the priority evaluation index to obtain a target artificial intelligence recognition thread comprises:

obtaining a clustering calculation vector corresponding to the category evaluation index and a calculation vector section corresponding to the priority evaluation index;

calculating the category evaluation index and the clustering calculation vector to obtain a target category evaluation index;

calculating the priority evaluation index and the calculation vector interval to obtain a target priority evaluation index;

and carrying out clustering calculation processing on the target type evaluation index and the target priority evaluation index to obtain a target performance evaluation result, and optimizing the artificial intelligent recognition thread according to the target performance evaluation result to obtain the target artificial intelligent recognition thread.

9. A public health data query system for image analysis, 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 of claims 1-8.