CN115100587B

CN115100587B - Regional random mining monitoring method and device based on multivariate data

Info

Publication number: CN115100587B
Application number: CN202210575166.0A
Authority: CN
Inventors: 高龙华; 王少波; 袁皖华; 刘寒青; 张舒
Original assignee: Hydrology Bureau Of Zhujiang Water Resources Commission Ministry Of Water Resources
Current assignee: Hydrology Bureau Of Zhujiang Water Resources Commission Ministry Of Water Resources
Priority date: 2022-05-25
Filing date: 2022-05-25
Publication date: 2023-06-23
Anticipated expiration: 2042-05-25
Also published as: CN115100587A

Abstract

The invention discloses a regional random mining monitoring method and device based on multivariate data, wherein the method comprises the following steps: acquiring a plurality of different types of monitoring data corresponding to a target monitoring area; based on a data verification model, verifying and screening the plurality of different types of monitoring data to obtain at least two real monitoring data; determining the random mining prediction probability corresponding to the target monitoring area according to the at least two real monitoring data and the trained random mining judgment neural network model; the random acquisition judgment neural network model is obtained by training a training data set comprising a plurality of training monitoring data and corresponding probability labels; judging whether the random mining prediction probability is larger than a preset probability threshold value, if so, determining that the target monitoring area has random mining. Therefore, the invention can improve the monitoring accuracy and monitoring efficiency of the regional random mining phenomenon and provide help for environmental management.

Description

Regional random mining monitoring method and device based on multivariate data

Technical Field

The invention relates to the technical field of data processing, in particular to a regional random mining monitoring method and device based on multivariate data.

Background

Along with the gradual strengthening of the construction of ecological civilization and the supervision of water conservancy industry in China, the treatment of the phenomenon of random mining of ecological environment is gradually paid attention to government units and enterprises. However, the existing method for preventing and controlling the disordered mining is generally limited to on-site investigation or judgment by means of single monitoring sensing information, for example, whether the disordered mining phenomenon exists or not is primarily judged by means of a single remote sensing image, and the method is low in efficiency and accuracy, and is easy to make mistakes, so that the environment is possibly not managed timely, and the ecological environment is possibly damaged. It can be seen that the prior art has defects and needs to be solved.

Disclosure of Invention

The technical problem to be solved by the invention is to provide the regional random mining monitoring method and the regional random mining monitoring device based on the multivariate data, which can improve the monitoring accuracy and the monitoring efficiency of the regional random mining phenomenon and provide assistance for environmental management.

In order to solve the technical problems, the first aspect of the invention discloses a regional random mining monitoring method based on multivariate data, which comprises the following steps:

acquiring a plurality of different types of monitoring data corresponding to a target monitoring area; the type of the monitoring data comprises at least two of remote sensing image data, unmanned aerial vehicle image data, regional equipment positioning data and regional sand depth sensing data;

Based on a data verification model, verifying and screening the plurality of different types of monitoring data to obtain at least two real monitoring data;

determining the random mining prediction probability corresponding to the target monitoring area according to the at least two real monitoring data and the trained random mining judgment neural network model; the random acquisition judgment neural network model is obtained by training a training data set comprising a plurality of training monitoring data and corresponding probability labels;

judging whether the random mining prediction probability is larger than a preset probability threshold value, if so, determining that the target monitoring area has the random mining phenomenon.

As an optional implementation manner, in the first aspect of the present invention, the verifying and screening, based on the data verification model, the plurality of different types of monitoring data to obtain at least two real monitoring data includes:

based on a data verification model, performing verification calculation on any two monitoring data in the plurality of different types of monitoring data to obtain data real parameters corresponding to the two monitoring data respectively;

calculating a plurality of data real parameters corresponding to each monitoring data based on the steps;

And verifying and screening the plurality of different types of monitoring data according to the data real parameters to obtain at least two real monitoring data.

In an optional implementation manner, in a first aspect of the present invention, the verifying calculation is performed on any two monitoring data in the plurality of different types of monitoring data based on the data verification model, to obtain data real parameters corresponding to the two monitoring data respectively, where the verifying calculation includes:

for any two image data in the plurality of different types of monitoring data, determining suspected disordered acquisition areas in the two image data according to an image recognition algorithm; the image data is the remote sensing image data or the unmanned aerial vehicle image data; the suspected disordered acquisition area is an image area in which pixel color parameters in the image data are in a yellow parameter interval, and the difference value of the pixel color parameters in the area and the pixel color parameters outside the area is larger than a preset difference value threshold;

calculating the shape similarity and the first color similarity of suspected disordered acquisition areas in the two image data;

calculating a weighted sum value of the shape similarity and the first color similarity to obtain data real parameters corresponding to the two image data respectively; wherein the sum of weights corresponding to the shape similarity and the first color similarity is 1; the shape similarity is weighted more than the first color similarity.

determining a disarranged geographic position corresponding to the suspected disarranged area in the image data for any one of the image data and any one of the area data in the plurality of different types of monitoring data, and determining an area parameter corresponding to the disarranged geographic position from the area data; the regional data are the regional equipment positioning data or the regional sand depth sensing data; the regional parameters are the moving distance of the regional equipment position or the change amount of the regional sand depth;

determining a disarranged mining degree parameter corresponding to the regional parameter corresponding to the disarranged mining geographic position according to a preset parameter-degree corresponding relation;

according to a preset degree-ground color corresponding relation, determining a ground color parameter corresponding to the disordered mining degree parameter;

calculating second color similarity between the regional color of the suspected disordered acquisition region in the image data and the ground color parameter to obtain data real parameters corresponding to the image data and the regional data respectively;

And/or the number of the groups of groups,

and when the data types of the two area data are different, determining the random acquisition degree parameters corresponding to the area parameters in the two area data according to the parameter-degree corresponding relation respectively, and calculating the parameter similarity of the random acquisition degree parameters corresponding to the two area data to obtain the data real parameters corresponding to the two area data respectively.

In a first aspect of the present invention, the verifying and screening the plurality of different types of monitoring data according to the data real parameters to obtain at least two real monitoring data includes:

determining a data true statistical parameter corresponding to each monitoring data according to a plurality of data true parameters corresponding to each monitoring data;

sequencing all the monitoring data from large to small according to the real statistical parameters of the data to obtain a monitoring data sequence;

Determining a preset number of monitoring data of the monitoring data sequence to obtain at least two real monitoring data; the preset number is greater than 1.

In an optional implementation manner, in a first aspect of the present invention, the determining, according to a plurality of data real parameters corresponding to each of the monitored data, a data real statistical parameter corresponding to each of the monitored data includes:

for each piece of monitoring data, calculating a weighted summation average value of a plurality of data real parameters corresponding to the monitoring data to obtain a data real statistical parameter corresponding to the monitoring data; wherein the sum of weights corresponding to all the data real parameters is 1; the weight corresponding to each data real parameter is in direct proportion to the data association degree of two monitoring data corresponding to the data real parameter in the verification calculation; the data association degree is sequentially from big to small: a first degree of data association between the image data and the image data, a second degree of data association between the image data and the region data, a third degree of data association between the region data and the region data.

As an optional implementation manner, in the first aspect of the present invention, the at least two real monitoring data include the image data; the determining the random mining prediction probability corresponding to the target monitoring area according to the at least two real monitoring data and the trained random mining judgment neural network model comprises the following steps:

Judging whether a ship image exists in the image data according to a preset ship image template based on an image template matching algorithm;

if the ship image exists, determining a bearing area image and a stern water flow image corresponding to the ship image based on an image cutting algorithm;

based on a color recognition algorithm, determining whether the color parameters of the bearing area image are in a yellow parameter interval or whether the color parameters of the stern water flow image are in a soil color parameter interval, and if so, determining that the image data are sand conveying images;

and inputting the at least two real monitoring data comprising the sand transportation image into a trained disarranged mining judgment neural network model to predict and obtain disarranged mining prediction probability corresponding to the target monitoring area.

In an optional implementation manner, in a first aspect of the present invention, before inputting the at least two real monitoring data including the sand transportation image into a trained rough mining judgment neural network model to predict and obtain a rough mining prediction probability corresponding to the target monitoring area, the method further includes:

determining data type information corresponding to the at least two real monitoring data comprising the sand transporting image;

Determining a random acquisition judgment neural network model from a plurality of candidate neural network models according to the data type information; and the data type of the training monitoring data in the training data set of the random acquisition judgment neural network model is the same as the data type information.

The second aspect of the invention discloses a regional random mining monitoring device based on multivariate data, which comprises:

the acquisition module is used for acquiring a plurality of different types of monitoring data corresponding to the target monitoring area; the type of the monitoring data comprises at least two of remote sensing image data, unmanned aerial vehicle image data, regional equipment positioning data and regional sand depth sensing data;

the verification module is used for verifying and screening the plurality of different types of monitoring data based on a data verification model to obtain at least two real monitoring data;

the prediction module is used for determining the random mining prediction probability corresponding to the target monitoring area according to the at least two real monitoring data and the trained random mining judgment neural network model; the random acquisition judgment neural network model is obtained by training a training data set comprising a plurality of training monitoring data and corresponding probability labels;

And the judging module is used for judging whether the random mining prediction probability is larger than a preset probability threshold value, and if so, determining that the random mining phenomenon exists in the target monitoring area.

In a second aspect of the present invention, the verification module verifies and screens the plurality of different types of monitoring data based on a data verification model, to obtain at least two specific modes of real monitoring data, including:

In a second aspect of the present invention, the verification module performs verification calculation on any two monitoring data of the plurality of different types of monitoring data based on a data verification model, to obtain specific modes of data real parameters corresponding to the two monitoring data respectively, where the specific modes include:

and/or the number of the groups of groups,

In a second aspect of the present invention, the verification module verifies and screens the plurality of different types of monitoring data according to the data real parameters to obtain at least two specific ways of real monitoring data, including:

In a second aspect of the present invention, the verification module determines, according to a plurality of data real parameters corresponding to each of the monitored data, a specific manner of determining the data real statistical parameter corresponding to each of the monitored data, including:

As an alternative embodiment, in the second aspect of the present invention, the at least two real monitoring data include the image data; the prediction module determines a specific mode of the random mining prediction probability corresponding to the target monitoring area according to the at least two real monitoring data and the trained random mining judgment neural network model, and the specific mode comprises the following steps:

As an alternative embodiment, in the second aspect of the present invention, the apparatus further comprises a model selection module for performing the steps of:

The third aspect of the invention discloses another regional random mining monitoring device based on multivariate data, which comprises:

a memory storing executable program code;

a processor coupled to the memory;

the processor calls the executable program codes stored in the memory to execute part or all of the steps in the regional mining disorder monitoring method based on the multivariate data disclosed in the first aspect of the embodiment of the invention.

Compared with the prior art, the embodiment of the invention has the following beneficial effects:

in the embodiment of the invention, a plurality of different types of monitoring data corresponding to a target monitoring area are acquired; the type of the monitoring data comprises at least two of remote sensing image data, unmanned aerial vehicle image data, regional equipment positioning data and regional sand depth sensing data; based on a data verification model, verifying and screening the plurality of different types of monitoring data to obtain at least two real monitoring data; determining the random mining prediction probability corresponding to the target monitoring area according to the at least two real monitoring data and the trained random mining judgment neural network model; the random acquisition judgment neural network model is obtained by training a training data set comprising a plurality of training monitoring data and corresponding probability labels; judging whether the random mining prediction probability is larger than a preset probability threshold value, if so, determining that the target monitoring area has random mining. Therefore, the method and the device can improve the monitoring accuracy and the monitoring efficiency of the regional random mining phenomenon by collecting and verifying the multivariate data and judging the regional random mining according to the verified real data and the neural network model, thereby providing help for environmental management.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a schematic flow chart of a regional disarranged mining monitoring method based on multivariate data according to an embodiment of the present invention;

fig. 2 is a schematic structural diagram of a regional random mining monitoring device based on multivariate data according to an embodiment of the present invention;

fig. 3 is a schematic structural diagram of another regional random mining monitoring device based on multivariate data according to an embodiment of the present invention.

Detailed Description

In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

The terms first, second and the like in the description and in the claims and in the above-described figures are used for distinguishing between different objects and not necessarily for describing a sequential or chronological order. Furthermore, the terms "comprise" and "have," as well as any variations thereof, are intended to cover a non-exclusive inclusion. For example, a process, method, apparatus, article, or article that comprises a list of steps or elements is not limited to only those listed but may optionally include other steps or elements not listed or inherent to such process, method, article, or article.

Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the invention. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments.

The invention discloses a regional random mining monitoring method and device based on multivariate data. The following will describe in detail.

Example 1

Referring to fig. 1, fig. 1 is a schematic flow chart of a regional random mining monitoring method based on multivariate data according to an embodiment of the present invention. The method described in fig. 1 may be applied to a corresponding data processing terminal, data processing device, or data processing server, and the server may be a local server or a cloud server. As shown in fig. 1, the regional random mining monitoring method based on the multivariate data can include the following operations:

101. and acquiring a plurality of different types of monitoring data corresponding to the target monitoring area.

In the embodiment of the invention, the target monitoring area can be a river and lake area, such as a natural river and lake area needing ecological environment protection and monitoring, or other areas with ecological protection requirements, and the invention is not limited. Optionally, the types of the monitoring data may include at least two of remote sensing image data, unmanned aerial vehicle image data, regional equipment positioning data and regional sand depth sensing data. The remote sensing image data can be obtained by remote sensing photography of the target monitoring area through a remote sensing satellite, and the unmanned aerial vehicle image data can be obtained by controlling the unmanned aerial vehicle to go to the target monitoring area for photography. The regional equipment positioning data can be used for representing position information of one or more sensing equipment arranged in the target monitoring region, can be used for reflecting water and soil loss or soil movement conditions of the target monitoring region, and can be used for reflecting whether the target monitoring region is subjected to disordered mining or not. The regional sand depth sensing data can be obtained by detection through sand depth sensing equipment arranged in a target monitoring region, and can be used for reflecting the sand depth of the target monitoring region, and when the regional random mining condition exists, the sand depth can be changed, so that the regional random mining can be monitored.

102. Based on the data verification model, verifying and screening the monitoring data of different types to obtain at least two real monitoring data.

Optionally, the data verification model includes a plurality of rules for how to perform mutual verification between different data, so that the data verification model can be used for calculating the data authenticity among a plurality of different types of monitoring data, so that the monitoring data can be screened according to the data authenticity, and the data with higher authenticity can be screened out for subsequent judgment of the random mining phenomenon.

103. And determining the random mining prediction probability corresponding to the target monitoring area according to at least two real monitoring data and the trained random mining judgment neural network model.

Specifically, the random acquisition judgment neural network model is trained by a training data set comprising a plurality of training monitoring data and corresponding probability labels. Alternatively, the random acquisition judgment neural network model may employ different feature extraction networks to perform feature extraction for different monitored data types, for example, when performing feature extraction for some region data such as position information or depth information, a simple independent thermal coding and attention mechanism may be employed, and when performing feature extraction for image data, a convolutional neural network is required.

104. Judging whether the random mining prediction probability is larger than a preset probability threshold value, and if so, determining that the random mining phenomenon exists in the target monitoring area.

Optionally, after determining that the target monitoring area has a disarranged mining phenomenon, the relevant departments or organizations may be notified to perform further governance measures on the target monitoring area, such as excluding law enforcement personnel for on-site law enforcement or further on-site verification.

Therefore, the method described by implementing the embodiment of the invention can improve the monitoring accuracy and the monitoring efficiency of the regional random mining phenomenon by collecting and verifying the multivariate data and judging the regional random mining according to the verified real data and the neural network model, thereby providing help for environmental management.

As an optional implementation manner, in the step 102, based on the data verification model, verification and screening are performed on a plurality of different types of monitoring data to obtain at least two real monitoring data, where the method includes:

based on a data verification model, performing verification calculation on any two monitoring data in a plurality of different types of monitoring data to obtain data real parameters corresponding to the two monitoring data respectively;

And verifying and screening the plurality of different types of monitoring data according to the real data parameters to obtain at least two real monitoring data.

Therefore, by implementing the optional implementation mode, a plurality of data real parameters corresponding to each monitoring data can be obtained through calculation, and according to the data real parameters, verification and screening are carried out on a plurality of different types of monitoring data, so that the prediction of the random mining phenomenon can be carried out by adopting more real and accurate monitoring data in the follow-up process, and further the monitoring efficiency and accuracy can be improved.

As an optional implementation manner, in the step, based on the data verification model, any two monitoring data in the plurality of different types of monitoring data are verified and calculated to obtain data real parameters corresponding to the two monitoring data respectively, and the method includes:

for any two image data in the plurality of different types of monitoring data, determining suspected disordered acquisition areas in the two image data according to an image recognition algorithm;

and calculating a weighted sum value of the shape similarity and the first color similarity to obtain data real parameters corresponding to the two image data respectively.

Specifically, the image data is remote sensing image data or unmanned aerial vehicle image data.

Specifically, the suspected disordered acquisition area is an image area in which pixel color parameters in the image data are in a yellow parameter interval, and a difference value between the pixel color parameters in the area and the pixel color parameters outside the area is greater than a preset difference value threshold, wherein the pixel color parameters can be RGB values, CMYK values or YIQ values of pixels in the image data, other color systems should also be included in the protection scope of the present invention, specifically, the yellow parameter interval is related to a color system to which the pixel color parameters belong, for example, when the pixel color parameters belong to the RGB color system, the yellow parameter interval is a preset RGB value interval representing yellow, which can be set by an operator according to experience or experimental results.

Optionally, the sum of the weights corresponding to the shape similarity and the first color similarity is 1, and the weight of the shape similarity is greater than the weight of the first color similarity, because there is a certain deviation of the color similarity itself due to an error of the color system or the sampling system in different image data, which makes the colors of different image data themselves larger, so that the shape similarity should be considered more when verifying the correspondence between the image data.

Therefore, by implementing the optional implementation mode, the data real parameters corresponding to any two image data can be obtained through calculation, and according to the data real parameters, verification and screening are carried out on a plurality of different types of monitoring data, so that the prediction of the random mining phenomenon can be carried out by adopting more real and accurate image data in the follow-up process, and the monitoring efficiency and accuracy can be improved.

determining the indiscriminate mining geographic position corresponding to a suspected indiscriminate mining area in the image data according to any one image data and any one area data in a plurality of different types of monitoring data, and determining the area parameter corresponding to the indiscriminate mining geographic position from the area data;

And calculating the second color similarity between the regional color of the suspected disordered acquisition region in the image data and the ground color parameter to obtain the data real parameters respectively corresponding to the image data and the regional data.

Optionally, the parameter-degree correspondence is used to represent a relationship between the magnitude of the value of the regional parameter and the corresponding severity of the disagreement, which may be set by an operator according to experience or experimental results. Optionally, the degree-ground color correspondence is used to represent the correspondence between different severity of the disarranged mining and the color of the ground (sand or soil surface), which can also be set by the operator according to experience or experimental results.

Specifically, in this embodiment, the area data is area equipment positioning data or area sand depth sensing data, and the corresponding area parameter is an area equipment position moving distance or an area sand depth variation.

Therefore, by implementing the optional implementation mode, the data real parameters corresponding to any one image data and any one region data can be obtained through calculation, and according to the data real parameters, verification and screening are carried out on a plurality of different types of monitoring data, so that the prediction of the random mining phenomenon can be carried out by adopting more real and accurate image or region data in the follow-up process, and further the monitoring efficiency and accuracy can be improved.

for any two region data in the plurality of different types of monitoring data;

and when the data types of the two region data are the same, calculating the data similarity of the two region data to obtain the data real parameters corresponding to the two region data respectively.

When the data types of the two area data are different, determining the disordered mining degree parameters corresponding to the area parameters in the two area data according to the parameter-degree correspondence, and calculating the parameter similarity of the disordered mining degree parameters corresponding to the two area data to obtain the data real parameters corresponding to the two area data.

Alternatively, the data similarity or the parameter similarity may be a similarity between values, such as a difference between values or a statistical parameter such as variance.

Therefore, by implementing the optional implementation mode, the data real parameters corresponding to any two area data can be obtained through calculation, and according to the data real parameters, verification and screening are carried out on a plurality of different types of monitoring data, so that the prediction of the random mining phenomenon can be carried out by adopting more real and accurate area data in the follow-up process, and the monitoring efficiency and accuracy can be improved.

As an optional implementation manner, in the step, according to the real parameters of the data, verifying and screening the plurality of different types of monitoring data to obtain at least two real monitoring data, including:

determining a data real statistical parameter corresponding to each monitoring data according to a plurality of data real parameters corresponding to each monitoring data;

determining a preset number of monitoring data of a monitoring data sequence to obtain at least two real monitoring data; the preset number is greater than 1.

Therefore, by implementing the optional implementation mode, the real data statistical parameter corresponding to each monitoring data can be calculated, and the plurality of different types of monitoring data are verified and screened according to the real data statistical parameter, so that the more real and accurate monitoring data can be adopted for predicting the random mining phenomenon in the follow-up process, and the monitoring efficiency and accuracy can be improved.

As an optional implementation manner, in the step, determining the data true statistical parameter corresponding to each monitoring data according to the plurality of data true parameters corresponding to each monitoring data includes:

And for each monitoring data, calculating a weighted sum average value of a plurality of data real parameters corresponding to the monitoring data to obtain the data real statistical parameters corresponding to the monitoring data.

Specifically, the sum of the weights corresponding to all the data real parameters is 1, and the weight corresponding to each data real parameter is in direct proportion to the data association degree of two monitoring data corresponding to the data real parameter in verification calculation, wherein the data association degree is sequentially from large to small: a first degree of data association between the image data and the image data, a second degree of data association between the image data and the region data, a third degree of data association between the region data and the region data. The reason for this is that mutual authentication between image data is more intuitive and accurate, and association between image data and region data is less, but the degree of data association between region data is generally higher, and since the degree of similarity between region data is generally higher, the weight thereof should be the lowest because both are harder to tamper with, otherwise the characterization ability of the data true statistical parameters may be deteriorated.

Therefore, by implementing the optional implementation manner, for each monitoring data, a weighted sum average value of a plurality of data real parameters corresponding to the monitoring data can be calculated to obtain the data real statistical parameters corresponding to the monitoring data, so that a plurality of different types of monitoring data can be verified and screened according to the data real statistical parameters, and further the monitoring efficiency and accuracy can be improved.

As an optional implementation manner, the at least two pieces of real monitoring data obtained by the screening include image data, and correspondingly, in the step, determining a disarranged mining prediction probability corresponding to the target monitoring area according to the at least two pieces of real monitoring data and the trained disarranged mining judgment neural network model includes:

and inputting at least two real monitoring data comprising the sand transportation image into a trained disarranged mining judgment neural network model to predict and obtain disarranged mining prediction probability corresponding to the target monitoring area.

Alternatively, the ship image template may be obtained by performing template learning on ship images of a plurality of target monitoring areas in advance, for example, may be obtained by calculating a parameter average value or a profile characterization parameter of a plurality of ship images.

Optionally, the color parameters of the bearing area image or the stern water flow image are RGB values, CMYK values or YIQ values, and other color systems should be included in the protection scope of the present invention, specifically, the yellow parameter interval or the clay color parameter interval is related to the color system to which the color parameters belong, for example, when the color parameters belong to the RGB color system, the yellow parameter interval is a preset RGB value interval representing yellow, and the clay color parameter interval is a preset RGB value interval representing brown or turquoise, which can be set by an operator according to experience or experimental results.

Therefore, by implementing the optional implementation mode, the detection of the image of the sand carrier possibly existing in the image data can be realized, and the image is further predicted through the neural network when the detection is carried out, so that the efficiency and the accuracy of monitoring the random mining phenomenon can be improved.

As an optional implementation manner, in the step, at least two real monitoring data including the sand transportation image are input into the trained rough mining judgment neural network model, so as to predict and obtain the rough mining prediction probability corresponding to the target monitoring area, and before the method further includes:

Determining data type information corresponding to at least two real monitoring data comprising sand conveying images;

and determining a random acquisition judgment neural network model from the plurality of candidate neural network models according to the data type information, wherein the data type of training monitoring data in the training data set of the random acquisition judgment neural network model is the same as the data type information.

Alternatively, a plurality of candidate neural network models may be obtained in advance through training by combining training monitoring data of a plurality of different data types.

Therefore, by implementing the optional implementation manner, the random mining judgment neural network model with the same training data type can be determined from the multiple candidate neural network models according to the data type information, so that random mining prediction can be performed according to the network model with the same training data type in the follow-up process, and the efficiency and accuracy for monitoring the random mining phenomenon are further improved.

Example two

Referring to fig. 2, fig. 2 is a schematic structural diagram of a regional random mining monitoring device based on multivariate data according to an embodiment of the present invention. The apparatus described in fig. 2 may be applied to a corresponding data processing terminal, data processing device, or data processing server, where the server may be a local server or a cloud server, and embodiments of the present invention are not limited. As shown in fig. 2, the apparatus may include:

An acquiring module 201, configured to acquire a plurality of different types of monitoring data corresponding to a target monitoring area; the type of the monitoring data comprises at least two of remote sensing image data, unmanned aerial vehicle image data, regional equipment positioning data and regional sand depth sensing data;

the verification module 202 is configured to verify and screen a plurality of different types of monitoring data based on a data verification model, so as to obtain at least two real monitoring data;

the prediction module 203 is configured to determine a disarranged mining prediction probability corresponding to the target monitoring area according to at least two real monitoring data and the trained disarranged mining judgment neural network model; the random acquisition judgment neural network model is obtained by training a training data set comprising a plurality of training monitoring data and corresponding probability labels;

and the judging module 204 is configured to judge whether the disarranged mining prediction probability is greater than a preset probability threshold, and if yes, determine that the disarranged mining phenomenon exists in the target monitoring area.

As an alternative embodiment, the verification module 202 performs verification and screening on a plurality of different types of monitoring data based on a data verification model, to obtain a specific manner of at least two real monitoring data, including:

As an optional implementation manner, the verification module 202 performs verification calculation on any two monitoring data in the plurality of different types of monitoring data based on the data verification model, to obtain specific modes of data real parameters corresponding to the two monitoring data respectively, where the specific modes include:

For any two image data in the plurality of different types of monitoring data, determining suspected disordered acquisition areas in the two image data according to an image recognition algorithm; the image data is remote sensing image data or unmanned aerial vehicle image data; the suspected disordered acquisition area is an image area in which pixel color parameters in the image data are in a yellow parameter interval, and the difference value of the pixel color parameters in the area and the pixel color parameters outside the area is larger than a preset difference value threshold value;

determining the indiscriminate mining geographic position corresponding to a suspected indiscriminate mining area in the image data according to any one image data and any one area data in a plurality of different types of monitoring data, and determining the area parameter corresponding to the indiscriminate mining geographic position from the area data; the regional data are regional equipment positioning data or regional sand depth sensing data; the regional parameter is the regional equipment position moving distance or regional sand depth variation;

for any two area data in the plurality of different types of monitoring data, when the data types of the two area data are the same, calculating the data similarity of the two area data to obtain data real parameters corresponding to the two area data respectively, when the data types of the two area data are different, determining the disarranged mining degree parameters corresponding to the area parameters in the two area data according to the parameter-degree corresponding relation respectively, and calculating the parameter similarity of the disarranged mining degree parameters corresponding to the two area data to obtain the data real parameters corresponding to the two area data respectively.

As an optional implementation manner, the verification module 202 verifies and screens the plurality of different types of monitoring data according to the real parameters of the data, so as to obtain at least two specific modes of real monitoring data, which include:

As an optional implementation manner, the verification module 202 determines, according to a plurality of data real parameters corresponding to each monitoring data, a specific manner of data real statistical parameters corresponding to each monitoring data, including:

for each monitoring data, calculating a weighted sum average value of a plurality of data real parameters corresponding to the monitoring data to obtain a data real statistical parameter corresponding to the monitoring data; wherein the sum of weights corresponding to all data real parameters is 1; the weight corresponding to each data real parameter is in direct proportion to the data association degree of two monitoring data corresponding to the data real parameter in verification calculation; the data association degree is as follows from big to small in turn: a first degree of data association between the image data and the image data, a second degree of data association between the image data and the region data, a third degree of data association between the region data and the region data.

As an alternative embodiment, the at least two real monitoring data comprise image data; the specific mode of determining the disarranged mining prediction probability corresponding to the target monitoring area by the prediction module 203 according to at least two real monitoring data and the trained disarranged mining judgment neural network model comprises the following steps:

As an alternative embodiment, the apparatus further comprises a model selection module for performing the steps of:

determining a random acquisition judgment neural network model from a plurality of candidate neural network models according to the data type information; the data type of training monitoring data in the training data set of the random acquisition judgment neural network model is the same as the data type information.

Example III

Referring to fig. 3, fig. 3 is a schematic structural diagram of another regional random mining monitoring device based on multivariate data according to an embodiment of the present invention. As shown in fig. 3, the apparatus may include:

a memory 301 storing executable program code;

a processor 302 coupled with the memory 301;

The processor 302 invokes the executable program code stored in the memory 301 to perform some or all of the steps in the multi-data-based regional mining disorder monitoring method disclosed in the embodiment of the present invention.

Example IV

The embodiment of the invention discloses a computer storage medium which stores computer instructions, wherein the computer instructions are used for executing part or all of the steps in the regional random mining monitoring method based on multi-data.

The foregoing describes certain embodiments of the present disclosure, other embodiments being within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results. Furthermore, the processes depicted in the accompanying drawings do not necessarily have to be in the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for apparatus, devices, non-transitory computer readable storage medium embodiments, the description is relatively simple, as it is substantially similar to method embodiments, with reference to portions of the description of method embodiments being relevant.

The apparatus, the device, the nonvolatile computer readable storage medium and the method provided in the embodiments of the present disclosure correspond to each other, and therefore, the apparatus, the device, and the nonvolatile computer storage medium also have similar advantageous technical effects as those of the corresponding method, and since the advantageous technical effects of the method have been described in detail above, the advantageous technical effects of the corresponding apparatus, device, and nonvolatile computer storage medium are not described herein again.

In the 90 s of the 20 th century, improvements to one technology could clearly be distinguished as improvements in hardware (e.g., improvements to circuit structures such as diodes, transistors, switches, etc.) or software (improvements to the process flow). However, with the development of technology, many improvements of the current method flows can be regarded as direct improvements of hardware circuit structures. Designers almost always obtain corresponding hardware circuit structures by programming improved method flows into hardware circuits. Therefore, an improvement of a method flow cannot be said to be realized by a hardware entity module. For example, a programmable logic device (Programmable Logic Device, PLD) (e.g., a field programmable gate array (Field Programmable gate array, FPGA)) is an integrated circuit whose logic function is determined by the user programming the device. A designer programs to "integrate" a digital system onto a PLD without requiring the chip manufacturer to design and fabricate application-specific integrated circuit chips. Moreover, nowadays, instead of manually manufacturing integrated circuit chips, such programming is mostly implemented by using "logic compiler" software, which is similar to the software compiler used in program development and writing, and the original code before the compiling is also written in a specific programming language, which is called hardware description language (Hardware Description Language, HDL), but not just one of the hdds, but a plurality of kinds, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware DescriptionLanguage), confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), lava, lola, myHDL, PALASM, RHDL (RubyHardware Description Language), etc., VHDL (Very-High-SpeedIntegrated Circuit Hardware Description Language) and Verilog are currently most commonly used. It will also be apparent to those skilled in the art that a hardware circuit implementing the logic method flow can be readily obtained by merely slightly programming the method flow into an integrated circuit using several of the hardware description languages described above.

The controller may be implemented in any suitable manner, for example, the controller may take the form of, for example, a microprocessor or processor and a computer readable medium storing computer readable program code (e.g., software or firmware) executable by the (micro) processor, logic gates, switches, application specific integrated circuits (Application Specific Integrated Circuit, ASIC), programmable logic controllers, and embedded microcontrollers, examples of which include, but are not limited to, the following microcontrollers: ARC 625D, atmel AT91SAM, microchip PIC18F26K20, and Silicone Labs C8051F320, the memory controller may also be implemented as part of the control logic of the memory. Those skilled in the art will also appreciate that, in addition to implementing the controller in a pure computer readable program code, it is well possible to implement the same functionality by logically programming the method steps such that the controller is in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers, etc. Such a controller may thus be regarded as a kind of hardware component, and means for performing various functions included therein may also be regarded as structures within the hardware component. Or even means for achieving the various functions may be regarded as either software modules implementing the methods or structures within hardware components.

The system, apparatus, module or unit set forth in the above embodiments may be implemented in particular by a computer chip or entity, or by a product having a certain function. One typical implementation is a computer. In particular, the computer may be, for example, a personal computer, a laptop computer, a cellular telephone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

For convenience of description, the above devices are described as being functionally divided into various units, respectively. Of course, the functions of each element may be implemented in one or more software and/or hardware elements when implemented in the present specification.

It will be appreciated by those skilled in the art that the present description may be provided as a method, system, or computer program product. Accordingly, the present specification embodiments may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present description embodiments may take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.

The present description is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the specification. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

In one typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include volatile memory in a computer-readable medium, random Access Memory (RAM) and/or nonvolatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of computer-readable media.

Computer readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Disks (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises the element.

The description may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.

In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for system embodiments, since they are substantially similar to method embodiments, the description is relatively simple, as relevant to see a section of the description of method embodiments.

Finally, it should be noted that: the embodiment of the invention discloses a regional random mining monitoring method and device based on multi-element data, which are disclosed by the embodiment of the invention and are only used for illustrating the technical scheme of the invention, but not limiting the technical scheme; although the invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art will understand that; the technical scheme recorded in the various embodiments can be modified or part of technical features in the technical scheme can be replaced equivalently; such modifications and substitutions do not depart from the spirit and scope of the corresponding technical solutions.

Claims

1. The regional random mining monitoring method based on the multivariate data is characterized by comprising the following steps of:

judging whether the random mining prediction probability is larger than a preset probability threshold value, if so, determining that the target monitoring area has a random mining phenomenon;

the data verification model is based on verifying and screening the plurality of different types of monitoring data to obtain at least two real monitoring data, and the data verification model comprises the following steps:

verifying and screening the plurality of different types of monitoring data according to the data real parameters to obtain at least two real monitoring data;

the data verification model is based on verifying and calculating any two monitoring data in the plurality of different types of monitoring data to obtain data real parameters corresponding to the two monitoring data respectively, and the data verification model comprises the following steps:

2. The multi-data-based regional disaggregation monitoring method of claim 1, wherein the image data is the remote sensing image data or the unmanned aerial vehicle image data; the suspected disordered acquisition area is an image area in which pixel color parameters in the image data are in a yellow parameter interval, and the difference value of the pixel color parameters in the area and the pixel color parameters outside the area is larger than a preset difference value threshold.

3. The regional random mining monitoring method based on multivariate data according to claim 2, wherein the verifying calculation is performed on any two monitoring data in the plurality of different types of monitoring data based on a data verification model to obtain data real parameters respectively corresponding to the two monitoring data, and the method comprises the following steps:

And/or the number of the groups of groups,

4. The method for regional random mining monitoring based on multivariate data according to claim 3, wherein the verifying and screening the plurality of different types of monitoring data according to the data real parameters to obtain at least two real monitoring data comprises:

5. The multi-data-based regional disagreement monitoring device according to claim 4, wherein the determining the data true statistical parameter corresponding to each monitoring data according to the plurality of data true parameters corresponding to each monitoring data comprises:

6. The multi-data-based regional disagreement monitoring device of claim 2, wherein the at least two real monitoring data comprise the image data; the determining the random mining prediction probability corresponding to the target monitoring area according to the at least two real monitoring data and the trained random mining judgment neural network model comprises the following steps:

7. The regional disagreement monitoring device based on multivariate data according to claim 6, wherein before inputting the at least two real monitoring data including the sand transportation image to a trained disagreement judgment neural network model to predict and obtain a disagreement prediction probability corresponding to the target monitoring region, the method further comprises:

8. Regional random mining monitoring device based on multivariate data, which is characterized by comprising:

The judging module is used for judging whether the random mining prediction probability is larger than a preset probability threshold value, and if so, determining that the target monitoring area has a random mining phenomenon;

the verification module verifies and screens the plurality of different types of monitoring data based on a data verification model to obtain at least two specific modes of real monitoring data, and the specific modes comprise:

the verification module performs verification calculation on any two monitoring data in the plurality of different types of monitoring data based on a data verification model to obtain a specific mode of data real parameters corresponding to the two monitoring data respectively, and the specific mode comprises the following steps:

9. Regional random mining monitoring device based on multivariate data, which is characterized by comprising:

a memory storing executable program code;

a processor coupled to the memory;

the processor invokes the executable program code stored in the memory to perform the multi-data based regional mining disorder monitoring method of any one of claims 1-7.