An image processing method and system based on artificial intelligence

Abstract

The present application discloses an image processing method and system based on artificial intelligence, which relates to the field of artificial intelligence technology. The image processing method based on artificial intelligence includes the following steps: S1: acquiring data to be processed according to an initial acquisition frequency; S2: preprocessing the data to be processed to obtain preprocessed data; S3: acquiring a target mining data set according to an acquisition code, analyzing the preprocessed data through the target mining data set to obtain a current difference value, and analyzing the current difference value. If the current difference value falls within a normal range, S1 is executed; if the current difference value falls within a critical range, the initial acquisition frequency is adjusted to obtain a new acquisition frequency, data to be processed is acquired according to the new acquisition frequency, and S2 is executed; if the current difference value falls within an abnormal range, an alarm message is sent. The present application can correct the data to be processed, thereby improving data quality, and can provide timely warnings of disaster hazards.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to an image processing method and system based on artificial intelligence. Background Art

[0002] Oil and gas exploration involves three stages: field seismic data acquisition, seismic data processing, and seismic data interpretation. Field data acquisition primarily utilizes artificial seismic methods (e.g., blasting or using a seismic vehicle to generate artificial seismic waves) to collect seismic waves returning from the ground, generating field observation data. The mission of seismic field acquisition is to obtain raw data, the quality of which directly impacts the quality of digital processing and the accuracy of interpretation. Seismic field acquisition encompasses various stages, including site investigation, construction design, testing, and production. It requires close coordination among various disciplines, including measurement, drilling, excitation, reception, and interpretation.

[0003] During the oil and gas production process, multi-temporal remote sensing images are used to monitor surface changes in the oil and gas exploration area, such as ground subsidence and crack expansion, to promptly detect geological hazards and environmental problems during the oil and gas production process. However, existing image processing methods for detecting geological hazards during oil and gas production have the following problems:

[0004] (1) Due to the different geological environments and acquisition methods in oil and gas mining areas, the collected data may be affected by a variety of factors such as bad weather (such as heavy rain, dense fog, etc.), poor lighting conditions (such as deep sea or underground mining environment), and imaging methods, resulting in blurred images, low contrast, and key geological disaster features (such as stratum cracks and signs of landslides) that are difficult to clearly identify. There are also data quality issues such as geometric deformation caused by terrain undulations and changes in camera posture.

[0005] (2) Oil and gas production is a dynamic process, and geological disasters may occur at any time. Rapid image analysis is required to provide timely warnings. However, the large amount of image data combined with complex processing algorithms will result in processing speeds that cannot keep up with actual needs, making it difficult to detect disasters in a timely manner. Summary of the Invention

[0006] The purpose of this application is to provide an artificial intelligence-based image processing method and system, which can correct the data to be processed, thereby improving the data quality, and can provide timely warnings of disaster risks.

[0007] To achieve the above-mentioned purpose, the present application provides an artificial intelligence-based image processing method, comprising the following steps: S1: acquiring data to be processed according to an initial acquisition frequency, wherein the data to be processed includes: acquisition code, acquisition time and acquisition data, wherein the acquisition data includes at least: optical remote sensing data and synthetic aperture radar data; S2: preprocessing the data to be processed to obtain preprocessed data, wherein the preprocessed data includes at least: acquisition code, acquisition time, first correction data and second correction data; S3: acquiring a target mining data set according to the acquisition code, analyzing the preprocessed data through the target mining data set to obtain a current difference value, and analyzing the current difference value. If the current difference value belongs to a normal range, execute S1; if the current difference value belongs to a critical range, adjust the initial acquisition frequency to obtain a new acquisition frequency, acquire the data to be processed according to the new acquisition frequency, and execute S2; if the current difference value belongs to an abnormal range, send an alarm message.

[0008] As above, the data to be processed are preprocessed, and the sub-steps for obtaining preprocessed data are as follows: S21: query the standard database according to the acquisition code, and use the pre-mining data in the standard data packet whose query code is consistent with the acquisition code as the target data; S22: use the target data to correct the optical remote sensing data in the acquired data to obtain the first corrected data; S23: filter the synthetic aperture radar data in the acquired data to obtain the second corrected data; S24: use the acquisition code, acquisition time, first corrected data and second corrected data as preprocessed data.

[0009] As above, the sub-steps of using the target data to correct the optical remote sensing data in the collected data and obtain the first correction data are as follows: S221: selecting multiple evenly distributed ground control points on the original optical remote sensing data in the target data; S222: using the ground control points to perform imaging correction processing on the optical remote sensing data in the collected data to obtain imaging correction data; S223: using the ground control points to perform deformation correction processing on the imaging correction data to obtain deformation correction data, and use the deformation correction data as the first correction data.

[0010] As above, wherein the optical remote sensing data in the collected data is imaged and corrected using ground control points, and the sub-steps for obtaining the imaging correction data are as follows: S2221: Analyze the optical remote sensing data in the collected data using a first classification model based on artificial intelligence to determine the imaging category to which the optical remote sensing data in the collected data belongs, wherein the imaging category includes at least: central projection imaging, multi-center projection imaging, and array imaging; S2222: Select a corresponding imaging correction model according to the imaging category, and use the corresponding imaging correction model and ground control points to correct the optical remote sensing data in the collected data, and output initial imaging correction data, wherein the imaging correction model includes at least: a polynomial correction model, a physical model, and a rational function model; S2223: Perform an accuracy assessment on the initial imaging correction data to obtain a first accuracy assessment result. If the first accuracy assessment result is qualified, the initial imaging correction data is used as the imaging correction data; if the first accuracy assessment result is unqualified, re-perform imaging correction on the optical remote sensing data in the collected data.

[0011] As above, the sub-steps of performing accuracy assessment on the initial imaging correction data and obtaining the first accuracy assessment result are as follows: S22231: Calculate the residual between each ground control point and the corresponding corrected coordinates in the initial imaging correction data; S22232: Perform statistical analysis on all residuals to obtain the imaging comprehensive error value; S22233: Use the imaging accuracy threshold to analyze the imaging comprehensive error value to generate the first accuracy assessment result; if the imaging comprehensive error value is less than the imaging accuracy threshold, the generated first accuracy assessment result is qualified; if the imaging comprehensive error value is greater than or equal to the imaging accuracy threshold, the generated first accuracy assessment result is unqualified.

[0012] As above, the sub-steps of using ground control points to perform deformation correction processing on the imaging correction data to obtain deformation correction data, and using the deformation correction data as the first correction data are as follows: S2231: Using a second classification model based on artificial intelligence to analyze the imaging correction data, determine the geometric deformation category to which the imaging correction data belongs; wherein the geometric deformation category includes at least: terrain geometric deformation, systematic geometric deformation and non-systematic geometric deformation; S2232: Selecting a corresponding deformation correction model according to the geometric deformation category, and using the corresponding deformation correction model and ground control points to correct the imaging correction data, and outputting initial deformation correction data, wherein the deformation correction model includes at least: an orthorectification model, a sensor-based physical correction model, a polynomial correction model, and a correction model with compensation terms added to a strict physical model; S2233: Performing an accuracy assessment on the initial deformation correction data to obtain a second accuracy assessment result. If the second accuracy assessment result is qualified, the initial deformation correction data is used as the deformation correction data, and the deformation correction data is used as the first correction data; if the second accuracy assessment result is unqualified, the imaging correction data is deformed again.

[0013] As above, the sub-steps of performing accuracy assessment on the initial deformation correction data and obtaining the second accuracy assessment result are as follows: S22331: Analyze the corrected coordinates corresponding to each ground control point in the initial deformation correction data to obtain a first deformation error value; S22332: Select multiple evenly distributed ground inspection points on the original optical remote sensing data in the target data, and analyze the corrected coordinates corresponding to each ground inspection point in the initial deformation correction data to obtain a second deformation error value, wherein the ground inspection point is different from the ground control point; S22333: Obtain a deformation comprehensive error value based on the first deformation error value and the second deformation error value, and analyze the deformation comprehensive error value using the deformation accuracy threshold to generate a second accuracy assessment result; if the deformation comprehensive error value is less than the deformation accuracy threshold, the generated second accuracy assessment result is qualified; if the deformation comprehensive error value is greater than or equal to the deformation accuracy threshold, the generated second accuracy assessment result is unqualified.

[0014] As above, wherein the target mining data set is obtained according to the acquisition code, the pre-processed data is analyzed through the target mining data set, and the sub-steps for obtaining the current difference value are as follows: S31: The standard database is traversed according to the acquisition code, and the mining data set in the standard data packet whose query code is consistent with the acquisition code is used as the comparison data set, wherein the comparison data set includes: multiple mining period data; each mining period data includes: post-mining acquisition time, post-mining optical remote sensing data and post-mining synthetic aperture radar data; S32: The mining period data and the pre-processed data are sorted in order from the first to the last of the multiple post-mining acquisition times and acquisition times, and a corresponding analysis serial number is generated to obtain the data to be analyzed; wherein each data to be analyzed includes: analysis serial number, analysis acquisition time, analysis optical remote sensing data and analysis synthetic aperture radar data; the analysis serial number increases in sequence from the first to the last of the multiple post-mining acquisition times and acquisition times; S33: The data to be analyzed is analyzed to obtain the current difference value.

[0015] As above, the expression of the current difference value is: Where Dcy is the current difference value; λ1 is the pixel difference weight; ΔCxs k is the average value of the difference between all corresponding pixels in the k+1th analyzed optical remote sensing data and the kth analyzed optical remote sensing data; ΔCxs K is the average value of the difference between the analyzed optical remote sensing data in the K+1th data to be analyzed and the analyzed optical remote sensing data in the Kth data to be analyzed; K+1 is the total number of data to be analyzed, K is a natural number; λ2 is the disaster similarity weight; Gzh K+1 is the similarity value between the analyzed synthetic aperture radar data in the K+1th data to be analyzed and the pre-built geological hazard feature set.

[0016] The present application also provides an artificial intelligence-based image processing system, comprising: at least one data acquisition node and an artificial intelligence image processing center; wherein the data acquisition node: acquires data to be processed according to an initial acquisition frequency or a new acquisition frequency, and sends the data to be processed to the artificial intelligence image processing center; receives alarm information; the artificial intelligence image processing center: is used to execute the above-mentioned artificial intelligence-based image processing method.

[0017] The beneficial effects achieved by this application are as follows:

[0018] (1) The artificial intelligence-based image processing method and system of the present application can correct the data to be processed, thereby improving the data quality.

[0019] (2) The artificial intelligence-based image processing method and system of the present application can improve the accuracy of disaster hazard analysis results and provide timely warnings of disaster hazards. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments recorded in this application. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0021] Figure 1 A schematic structural diagram of an embodiment of an artificial intelligence-based image processing system;

[0022] Figure 2 The present invention is a flowchart of an embodiment of an image processing method based on artificial intelligence. DETAILED DESCRIPTION

[0023] The following is a clear and complete description of the technical solutions in the embodiments of the present invention, in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0024] like Figure 1 As shown, the present application provides an artificial intelligence-based image processing system, comprising: at least one data acquisition node 110 and an artificial intelligence image processing center 120.

[0025] Among them, the data collection node 110: collects the data to be processed according to the initial collection frequency or the new collection frequency, and sends the data to be processed to the artificial intelligence image processing center 120; receives alarm information.

[0026] Artificial intelligence image processing center 120: used to execute the following artificial intelligence-based image processing method.

[0027] Furthermore, the artificial intelligence image processing center 120 includes: a transceiver unit, a preprocessing unit, an analysis unit and a storage unit.

[0028] Among them, the transceiver unit receives the data to be processed according to the initial collection frequency or the new collection frequency, and sends the data to be processed to the pre-processing unit.

[0029] Preprocessing unit: preprocesses the data to be processed to obtain preprocessed data.

[0030] Analysis unit: obtains the target mining data set according to the acquisition code, analyzes the pre-processed data through the target mining data set, obtains the current difference value, and analyzes the current difference value. If the current difference value belongs to the normal range, the transceiver unit obtains the data to be processed according to the initial acquisition frequency; if the current difference value belongs to the critical range, the initial acquisition frequency is adjusted to obtain a new acquisition frequency, and the transceiver unit obtains the data to be processed according to the new acquisition frequency; if the current difference value belongs to the abnormal range, an alarm message is sent.

[0031] Storage unit: used to store the standard database, imaging correction model database and deformation correction model database.

[0032] The standard database includes multiple standard data packages, each corresponding to a query code. Each standard data package includes pre-mining data and a mining data set. Pre-mining data includes the original acquisition time, original optical remote sensing data, and original synthetic aperture radar data. The mining data set includes data from multiple mining periods; each mining period includes the post-mining acquisition time, post-mining optical remote sensing data, and post-mining synthetic aperture radar data.

[0033] Specifically, the original collection time is the period during which the pre-mining data is collected.

[0034] The original optical remote sensing data is: complete optical remote sensing image data collected before mining, covering the collection area corresponding to the query code.

[0035] The original synthetic aperture radar data is: the complete synthetic aperture radar data collected before mining, covering the collection area corresponding to the query code.

[0036] The imaging correction model database includes: multiple standard imaging correction models, one standard imaging correction model corresponds to one standard imaging category, and each standard imaging correction model includes: one imaging correction module or a combination of multiple imaging correction models.

[0037] Specifically, the standard imaging categories include at least central projection imaging, multi-center projection imaging, and area array imaging. A standard imaging correction model corresponding to each standard imaging category is set according to actual conditions.

[0038] The deformation correction model database includes: multiple standard deformation correction models, one standard deformation correction model corresponds to one standard geometric deformation category, and each standard deformation correction model includes: one deformation correction module or a combination of multiple deformation correction models.

[0039] Specifically, the standard geometric deformation categories include at least: terrain geometric deformation, systematic geometric deformation, and non-systematic geometric deformation. A standard deformation correction model corresponding to each standard geometric deformation category is set according to actual conditions.

[0040] like Figure 2 As shown, the present application provides an image processing method based on artificial intelligence, comprising the following steps:

[0041] S1: Acquire data to be processed according to an initial acquisition frequency, wherein the data to be processed includes: acquisition code, acquisition time and acquisition data, wherein the acquisition data at least includes: optical remote sensing data and synthetic aperture radar data.

[0042] Specifically, the acquisition code is an identification code of the acquisition area for acquiring the acquired data. One acquisition area corresponds to one acquisition code, and different acquisition areas have different acquisition codes.

[0043] The collection time is the period during which the collected data is obtained.

[0044] The collected data are high-resolution remote sensing satellite data covering the oil and gas production area, including at least: optical remote sensing data and synthetic aperture radar data.

[0045] S2: Preprocessing the data to be processed to obtain preprocessed data, wherein the preprocessed data at least includes: acquisition code, acquisition time, first correction data and second correction data.

[0046] Furthermore, the data to be processed is preprocessed, and the sub-steps of obtaining the preprocessed data are as follows:

[0047] S21: querying the standard database according to the acquisition code, and taking the pre-mining data in the standard data package whose query code is consistent with the acquisition code as the target data.

[0048] Specifically, the target data includes: original acquisition time, original optical remote sensing data, and original synthetic aperture radar data.

[0049] S22: Correcting the optical remote sensing data in the collected data using the target data to obtain first corrected data.

[0050] Furthermore, the sub-steps of correcting the optical remote sensing data in the collected data using the target data to obtain first corrected data are as follows:

[0051] S221: Selecting a plurality of evenly distributed ground control points on the original optical remote sensing data in the target data.

[0052] Specifically, the ground control points are ground features that are easily identifiable on the image and have accurate geographic coordinates, such as road intersections, building corners, etc.

[0053] Furthermore, the specific number of ground control points is set according to the resolution, range and correction accuracy requirements of the image. In this application, the preferred number of ground control points is greater than or equal to six.

[0054] S222: Perform imaging correction processing on the optical remote sensing data in the collected data using the ground control points to obtain imaging correction data.

[0055] Furthermore, the optical remote sensing data in the collected data is subjected to imaging correction processing using ground control points. The sub-steps for obtaining imaging correction data are as follows:

[0056] S2221: Analyze the optical remote sensing data in the collected data using a first classification model based on artificial intelligence to determine the imaging category to which the optical remote sensing data in the collected data belongs, wherein the imaging category includes at least: central projection imaging, multi-center projection imaging, and array imaging.

[0057] Specifically, central projection imaging (such as aerial remote sensing) is a centrally projected image. For centrally projected images, image point displacement is the main source of geometric deformation and is related to the flight altitude and the distance between the image point and the image principal point.

[0058] The geometric deformation characteristics of multi-center projection imaging (such as linear array push-broom remote sensing satellites) are different in the flight direction and scanning direction.

[0059] The geometric deformation of array imaging (such as some high-resolution satellites) is mainly caused by systematic factors (such as satellite orbit and attitude changes).

[0060] S2222: Select a corresponding imaging correction model according to the imaging category, and use the corresponding imaging correction model and ground control points to correct the optical remote sensing data in the collected data, and output initial imaging correction data, wherein the imaging correction model includes at least: a polynomial correction model, a physical model, and a rational function model.

[0061] Furthermore, as an embodiment, the imaging correction model database is traversed according to the imaging category, and a standard imaging correction model whose standard imaging category is consistent with the imaging category is determined as the corresponding imaging correction model.

[0062] Specifically, the corresponding imaging correction model is an imaging correction module or a combination of multiple imaging correction models.

[0063] For example: when the imaging category is central projection imaging, the selected imaging correction model is a polynomial correction model, but it is not limited to a polynomial correction model. This application preferably uses a polynomial correction model, which can fit complex deformation curves and correct geometric deformations caused by terrain undulations and changes in camera posture.

[0064] When the imaging category is multi-center projection imaging, the selected imaging correction model is the physical model, but it is not limited to the physical model. It can also be a physical model combined with a polynomial correction model for correction. This application preferably combines the physical model with the polynomial correction model. The physical model is used to consider prior information such as satellite orbit parameters and attitude parameters to describe the geometric relationship in the imaging process; the polynomial correction model is used to compensate for the remaining systematic errors and random errors.

[0065] When the imaging category is array imaging, the selected imaging correction model is a strict physical model or a rational function model, but is not limited to a strict physical model or a rational function model. This application preferably uses a rational function model, which can construct the imaging geometric relationship based on the precise parameters of the satellite orbit and attitude.

[0066] Furthermore, the optical remote sensing data in the collected data is corrected using the corresponding imaging correction model and ground control points, and the sub-steps of outputting the initial imaging correction data are as follows:

[0067] S22221: Calculate imaging correction parameters using the imaging correction model and ground control points.

[0068] Specifically, the object space coordinates of the ground control point (i.e., the precise three-dimensional space coordinates on the actual ground) and the pixel coordinates corresponding to the ground control point in the optical remote sensing data in the collected data are substituted into the imaging correction model, and the imaging correction parameters are calculated by the imaging correction model.

[0069] S22222: Correct the optical remote sensing data in the collected data according to the imaging correction parameters to obtain initial imaging correction data.

[0070] Specifically, the coordinates in the optical remote sensing data in the collected data are converted into corrected coordinates through the imaging correction parameters, thereby obtaining initial imaging correction data.

[0071] S2223: Perform accuracy assessment on the initial imaging correction data to obtain a first accuracy assessment result. If the first accuracy assessment result is qualified, the initial imaging correction data is used as the imaging correction data; if the first accuracy assessment result is unqualified, re-perform imaging correction processing on the optical remote sensing data in the collected data.

[0072] Furthermore, the initial imaging correction data is subjected to an accuracy evaluation, and the sub-steps for obtaining a first accuracy evaluation result are as follows:

[0073] S22231: Calculate the residual between each ground control point and the corresponding corrected coordinates in the initial imaging correction data.

[0074] Furthermore, the expression of the residual is:

[0075] ΔXc n =Xjz n -Xcs n ;

[0076] ΔYc n =Yjz n -Ycs n ;

[0077] Where ΔXc n is the residual error in the x direction between the nth ground control point and the corresponding corrected coordinate in the initial imaging correction data; ΔYc n is the residual error in the y direction between the nth ground control point and the corresponding corrected coordinate in the initial imaging correction data; (Xjz n ,Yjz n ) is the corrected coordinate corresponding to the nth ground control point in the initial imaging correction data; (Xcs n ,Ycs n ) is the object space coordinate of the nth ground control point; n∈[1,N], N is the total number of ground control points, and N is a natural number.

[0078] S22232: Perform statistical analysis on all residuals to obtain a comprehensive imaging error value.

[0079] Furthermore, the expression of the comprehensive imaging error value is:

[0080]

[0081] Where Wczc is the comprehensive imaging error value; ΔXc n is the residual error in the x direction between the nth ground control point and the corresponding corrected coordinate in the initial imaging correction data; ΔYc n is the residual in the y direction between the nth ground control point and the corresponding corrected coordinate in the initial imaging correction data; n∈[1,N], N is the total number of ground control points, and N is a natural number.

[0082] S22233: Use the imaging accuracy threshold to analyze the imaging comprehensive error value and generate a first accuracy evaluation result; if the imaging comprehensive error value is less than the imaging accuracy threshold, the generated first accuracy evaluation result is qualified; if the imaging comprehensive error value is greater than or equal to the imaging accuracy threshold, the generated first accuracy evaluation result is unqualified.

[0083] Specifically, the smaller the imaging comprehensive error value is, the higher the overall correction accuracy is.

[0084] Furthermore, the specific value of the imaging accuracy threshold is set according to actual conditions.

[0085] S223: Perform deformation correction processing on the imaging correction data using the ground control points to obtain deformation correction data, and use the deformation correction data as first correction data.

[0086] Furthermore, the sub-steps of performing deformation correction processing on the imaging correction data using the ground control points to obtain deformation correction data and using the deformation correction data as the first correction data are as follows:

[0087] S2231: Analyze the imaging correction data using a second classification model based on artificial intelligence to determine the geometric deformation category to which the imaging correction data belongs; wherein the geometric deformation category includes at least: terrain geometric deformation, systematic geometric deformation, and non-systematic geometric deformation.

[0088] Specifically, terrain geometric deformation refers to the geometric deformation of the image caused by projection differences due to the terrain in a collection area with large terrain undulations.

[0089] Systematic geometric deformation is the geometric deformation caused by sensor attitude instability (such as roll, pitch, yaw) or changes in orbital parameters.

[0090] Non-systematic geometric deformation is caused by complex factors such as atmospheric refraction and the rotation of the earth.

[0091] S2232: Select a corresponding deformation correction model according to the geometric deformation category, and use the corresponding deformation correction model and ground control points to correct the imaging correction data, and output the initial deformation correction data, wherein the deformation correction model at least includes: an orthorectification model, a sensor-based physical correction model, a polynomial correction model, and a correction model with compensation terms added to a strict physical model.

[0092] Furthermore, as an embodiment, the deformation correction model database is traversed according to the geometric deformation category, and a standard deformation correction model whose standard geometric deformation category is consistent with the geometric deformation category is determined as the corresponding deformation correction model.

[0093] For example: when the geometric deformation category is terrain geometric deformation, the selected deformation correction model is the orthorectification model, but it is not limited to the orthorectification model. This application preferably uses the orthorectification model, which is based on the digital elevation model (DEM) and eliminates the displacement caused by terrain undulation by projecting the image points onto the reference plane according to the terrain undulation.

[0094] When the geometric deformation category is systematic geometric deformation, the selected deformation correction model is a sensor-based physical correction model, but is not limited to a sensor-based physical correction model. This application preferably uses a sensor-based physical correction model, which can construct a strict imaging geometric relationship based on the internal parameters of the sensor (such as focal length, pixel size) and external parameters (satellite position, attitude) to accurately correct the image.

[0095] When the geometric deformation category is non-systematic geometric deformation, the selected deformation correction model is a polynomial correction model or a correction model with compensation terms added on the basis of a strict physical model, but it is not limited to a polynomial correction model or a correction model with compensation terms added on the basis of a strict physical model. The present application preferably uses a polynomial correction model, which can fit complex deformation curves and determine the polynomial coefficients by selecting a sufficient number of control points on the image to correct the deformation.

[0096] Specifically, a suitable correction model is selected according to the imaging mode and geometric deformation characteristics of the remote sensing image, and the optical remote sensing data in the collected data is corrected multiple times using the selected correction model to obtain first corrected data, which can improve the correction accuracy and thus improve the data quality.

[0097] Furthermore, the imaging correction data is corrected using the corresponding deformation correction model and ground control points, and the sub-steps of outputting the initial deformation correction data are as follows:

[0098] S22321: Calculate deformation correction parameters using the deformation correction model and ground control points.

[0099] Specifically, the object space coordinates of the ground control point (i.e., the precise three-dimensional space coordinates on the actual ground) and the pixel coordinates corresponding to the ground control point in the imaging correction data are substituted into the deformation correction model, and the deformation correction parameters are calculated by the deformation correction model.

[0100] S22322: Correct the imaging correction data according to the deformation correction parameters to obtain initial deformation correction data.

[0101] Specifically, the coordinates in the imaging correction data are converted into corrected coordinates using the deformation correction parameters, thereby obtaining initial deformation correction data.

[0102] S2233: Perform accuracy evaluation on the initial deformation correction data to obtain a second accuracy evaluation result. If the second accuracy evaluation result is qualified, the initial deformation correction data is used as the deformation correction data, and the deformation correction data is used as the first correction data; if the second accuracy evaluation result is unqualified, the imaging correction data is re-deformed.

[0103] Furthermore, the sub-steps of performing accuracy evaluation on the initial deformation correction data and obtaining a second accuracy evaluation result are as follows:

[0104] S22331: Analyze the corrected coordinates of each ground control point and the initial deformation correction data to obtain a first deformation error value.

[0105] Furthermore, the expression of the first deformation error value is:

[0106]

[0107] Where Wczb is the first deformation error value; (Xcs n ,Ycs n ) is the object space coordinate of the nth ground control point; n∈[1,N], N is the total number of ground control points, N is a natural number; (Xxz n ,Yxz n ) is the corrected coordinate corresponding to the nth ground control point in the initial deformation correction data.

[0108] S22332: Select multiple evenly distributed ground check points on the original optical remote sensing data in the target data, and analyze the corrected coordinates of each ground check point and the corresponding initial deformation correction data to obtain a second deformation error value, wherein the ground check point is different from the ground control point.

[0109] Specifically, the ground check points are points in the target data other than the ground control points. The corresponding positions of the ground check points in the initial deformation correction data are evenly distributed, which can comprehensively evaluate the calibration accuracy of different positions in the image.

[0110] Furthermore, the expression of the second deformation error value is:

[0111]

[0112] Where Wmae is the second deformation error value; (Xjy m ,Yjy m ) is the object space coordinate of the mth ground check point; m∈[1,M], M is the total number of ground check points, M is a natural number; (Xxz m ,Yxz m ) is the corrected coordinate corresponding to the mth ground check point in the initial deformation correction data.

[0113] S22333: Obtain a comprehensive deformation error value based on the first deformation error value and the second deformation error value, and use the deformation accuracy threshold to analyze the comprehensive deformation error value to generate a second accuracy evaluation result; if the comprehensive deformation error value is less than the deformation accuracy threshold, the generated second accuracy evaluation result is qualified; if the comprehensive deformation error value is greater than or equal to the deformation accuracy threshold, the generated second accuracy evaluation result is unqualified.

[0114] Specifically, the smaller the deformation comprehensive error value, the higher the overall correction accuracy. The specific value of the deformation accuracy threshold is set according to the actual situation.

[0115] Furthermore, the expression of the comprehensive deformation error value is:

[0116]

[0117] Among them, Zhc is the comprehensive deformation error value; Wczb is the first deformation error value; Wmae is the second deformation error value; η1 is the weight of the first deformation error value; η2 is the weight of the second deformation error value.

[0118] Specifically, the specific values ​​of η1 and η2 are set according to actual conditions.

[0119] S23: Filtering the synthetic aperture radar data in the collected data to obtain second correction data.

[0120] Specifically, existing techniques are used to filter the synthetic aperture radar (SAR) data within the acquired data to remove noise and obtain second corrected data. For example, multi-look processing is used to reduce speckle noise within the SAR data. The second corrected data is a high-quality, corrected remote sensing image. The radiometric values ​​of the image in the second corrected data accurately reflect the reflective characteristics of the ground objects, accurately position the objects, and suppress image noise to a predetermined degree, facilitating subsequent accurate analysis.

[0121] S24: The acquisition code, acquisition time, first correction data, and second correction data are used as pre-processing data.

[0122] Specifically, the artificial intelligence-based image processing method and system of the present application can correct the data to be processed, thereby improving the data quality. Due to the high data quality, the accuracy of the disaster hazard analysis results can be further improved.

[0123] S3: Obtain the target mining data set according to the acquisition code, analyze the pre-processed data through the target mining data set, obtain the current difference value, and analyze the current difference value. If the current difference value belongs to the normal range, execute S1; if the current difference value belongs to the critical range, adjust the initial acquisition frequency to obtain a new acquisition frequency, obtain the data to be processed according to the new acquisition frequency, and execute S2; if the current difference value belongs to the abnormal range, send an alarm message.

[0124] Specifically, the specific value of the normal range is set according to the actual situation. If the current difference value falls within the normal range, it means that there is no disaster risk.

[0125] The specific value of the critical range is set according to the actual situation. If the current difference value belongs to the critical range, it means that the disaster risk has not yet occurred, but the probability of a disaster risk occurring is very high.

[0126] The specific value of the abnormal range is set according to the actual situation. If the current difference value falls within the abnormal range, it means that a disaster risk has occurred. If the current difference value falls within the abnormal range, an alarm message will be sent, allowing relevant departments and personnel to deal with the disaster risk in a timely manner.

[0127] Furthermore, the alarm information includes at least: the alarm time, the type of disaster hazard and the degree of the disaster hazard.

[0128] Furthermore, the new acquisition frequency is greater than the initial acquisition frequency, that is, within the same time period, the amount of to-be-processed data acquired by the new acquisition frequency is greater than the amount of to-be-processed data acquired by the initial acquisition frequency.

[0129] Specifically, the new collection frequency is set based on actual conditions. Within the same time period, a higher collection frequency results in more data being collected, and a greater amount of data to be processed is obtained. When the current difference value falls within the critical range, the initial collection frequency is adjusted to a higher frequency. This new frequency is used to obtain more data to be processed, enabling timely analysis and early warning of potential disasters.

[0130] Furthermore, the target mining data set is obtained according to the acquisition code, and the pre-processed data is analyzed by the target mining data set to obtain the current difference value. The sub-steps are as follows:

[0131] S31: Traverse the standard database according to the acquisition code, and use the mining data set in the standard data packet whose query code is consistent with the acquisition code as the comparison data set, wherein the comparison data set includes: multiple mining period data; each mining period data includes: post-mining acquisition time, post-mining optical remote sensing data and post-mining synthetic aperture radar data.

[0132] S32: Sort the mining period data and pre-processed data in order of multiple post-mining acquisition times and acquisition times from earliest to latest, and generate corresponding analysis serial numbers to obtain data to be analyzed; wherein each data to be analyzed includes: analysis serial number, analysis acquisition time, analysis optical remote sensing data and analysis synthetic aperture radar data; the analysis serial number increases in sequence according to the multiple post-mining acquisition times and acquisition times from earliest to latest.

[0133] Specifically, the total number of mining period data and pre-processing data is K+1, and the analysis sequence number of the pre-processing data is K+1, where K is a natural number.

[0134] The acquisition time of the preprocessed data after the corresponding analysis serial number is generated is the analysis acquisition time, the first correction data is the analysis optical remote sensing data, and the second correction data is the analysis synthetic aperture radar data.

[0135] The post-mining acquisition time of the mining period data after the corresponding analysis serial number is generated is the analysis acquisition time, the post-mining optical remote sensing data is the analysis optical remote sensing data, and the post-mining synthetic aperture radar data is the analysis synthetic aperture radar data.

[0136] S33: Analyze the data to be analyzed to obtain the current difference value.

[0137] Furthermore, the expression of the current difference value is:

[0138]

[0139] Where Dcy is the current difference value; λ1 is the pixel difference weight; ΔCxs k is the average value of the difference between all corresponding pixels in the k+1th analyzed optical remote sensing data and the kth analyzed optical remote sensing data; ΔCxs K is the average value of the difference between the analyzed optical remote sensing data in the K+1th data to be analyzed and the analyzed optical remote sensing data in the Kth data to be analyzed; K+1 is the total number of data to be analyzed, K is a natural number; λ2 is the disaster similarity weight; Gzh K+1 is the similarity value between the analyzed synthetic aperture radar data in the K+1th data to be analyzed and the pre-built geological hazard feature set.

[0140] Specifically, λ1 and λ2 are set according to actual conditions. Through an artificial intelligence-based analysis model, the similarity value between the analyzed synthetic aperture radar data in the K+1th data to be analyzed and the pre-constructed geological disaster feature set can be obtained. The greater the similarity value, the greater the probability of the occurrence of such geological disasters. The geological disaster feature set is a set composed of features that can clearly express geological disasters extracted from actual collection or research data. The average value of the difference between all corresponding pixels in the analyzed optical remote sensing data of two data to be analyzed with adjacent collection times can be achieved according to existing technologies, so it will not be repeated here.

[0141] The beneficial effects achieved by this application are as follows:

[0142] (1) The artificial intelligence-based image processing method and system of the present application can correct the data to be processed, thereby improving the data quality.

[0143] (2) The artificial intelligence-based image processing method and system of the present application can improve the accuracy of disaster hazard analysis results and provide timely warnings of disaster hazards.

[0144] Although preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they become aware of the underlying inventive concepts. Therefore, the scope of protection of this application is intended to include the preferred embodiments and all changes and modifications that fall within the scope of this application. Obviously, those skilled in the art may make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if such changes and modifications of this application fall within the scope of protection of this application and its equivalents, then this application is intended to include such changes and modifications.

Claims

1. An image processing method based on artificial intelligence, characterized in that: The steps include: S1: Acquire data to be processed according to an initial acquisition frequency, wherein the data to be processed includes: acquisition code, acquisition time, and acquisition data, wherein the acquisition data includes at least: optical remote sensing data and synthetic aperture radar data; S2: Preprocessing the data to be processed to obtain preprocessed data, wherein the preprocessed data at least includes: acquisition code, acquisition time, first correction data and second correction data; S3: Obtain the target mining data set according to the acquisition code, analyze the pre-processed data through the target mining data set, obtain the current difference value, and analyze the current difference value. If the current difference value is within the normal range, execute S1; if the current difference value is within the critical range, adjust the initial acquisition frequency to obtain a new acquisition frequency, obtain the data to be processed according to the new acquisition frequency, and execute S2; if the current difference value is within the abnormal range, send an alarm message; The sub-steps for preprocessing the data to be processed and obtaining the preprocessed data are as follows: S21: querying the standard database according to the acquisition code, and taking the pre-mining data in the standard data package with the query code consistent with the acquisition code as the target data; S22: Correcting the optical remote sensing data in the collected data using the target data to obtain first corrected data; S23: Filtering the synthetic aperture radar data in the collected data to obtain second correction data; S24: Using the acquisition code, acquisition time, first correction data and second correction data as pre-processing data; The sub-steps of correcting the optical remote sensing data in the collected data using the target data to obtain the first corrected data are as follows: S221: selecting a plurality of evenly distributed ground control points on the original optical remote sensing data in the target data; S222: performing imaging correction processing on the optical remote sensing data in the collected data using the ground control points to obtain imaging correction data; S223: Performing deformation correction processing on the imaging correction data using the ground control points to obtain deformation correction data, and using the deformation correction data as first correction data; Among them, the optical remote sensing data in the collected data is imaged and corrected using ground control points. The sub-steps for obtaining the imaged and corrected data are as follows: S2221: Analyze the optical remote sensing data in the collected data using a first classification model based on artificial intelligence to determine an imaging category to which the optical remote sensing data in the collected data belongs, wherein the imaging category includes at least: central projection imaging, multi-center projection imaging, and area array imaging; S2222: Selecting a corresponding imaging correction model according to the imaging category, and correcting the optical remote sensing data in the collected data using the corresponding imaging correction model and ground control points, and outputting initial imaging correction data, wherein the imaging correction model includes at least: a polynomial correction model, a physical model, and a rational function model; S2223: Performing an accuracy assessment on the initial imaging correction data to obtain a first accuracy assessment result. If the first accuracy assessment result is qualified, the initial imaging correction data is used as the imaging correction data; if the first accuracy assessment result is unqualified, re-performing imaging correction processing on the optical remote sensing data in the collected data; The sub-steps of performing deformation correction processing on the imaging correction data using the ground control points to obtain the deformation correction data and using the deformation correction data as the first correction data are as follows: S2231: Analyze the imaging correction data using a second classification model based on artificial intelligence to determine a geometric deformation category to which the imaging correction data belongs; wherein the geometric deformation category includes at least topographic geometric deformation, systematic geometric deformation, and non-systematic geometric deformation; S2232: Selecting a corresponding deformation correction model based on the geometric deformation category, and correcting the imaging correction data using the corresponding deformation correction model and ground control points, outputting initial deformation correction data. The deformation correction model includes at least: an orthorectification model, a sensor-based physical correction model, a polynomial correction model, and a correction model that adds compensation terms to a strict physical model. S2233: Perform accuracy evaluation on the initial deformation correction data to obtain a second accuracy evaluation result. If the second accuracy evaluation result is qualified, the initial deformation correction data is used as the deformation correction data, and the deformation correction data is used as the first correction data; if the second accuracy evaluation result is unqualified, the imaging correction data is re-deformed.

2. The image processing method based on artificial intelligence according to claim 1, characterized in that: The sub-steps of performing accuracy assessment on the initial imaging correction data and obtaining the first accuracy assessment result are as follows: S22231: Calculate the residual between each ground control point and the corresponding corrected coordinates in the initial imaging correction data; S22232: Perform statistical analysis on all residuals to obtain the comprehensive imaging error value; S22233: Use the imaging accuracy threshold to analyze the imaging comprehensive error value and generate a first accuracy evaluation result; if the imaging comprehensive error value is less than the imaging accuracy threshold, the generated first accuracy evaluation result is qualified; if the imaging comprehensive error value is greater than or equal to the imaging accuracy threshold, the generated first accuracy evaluation result is unqualified.

3. The image processing method based on artificial intelligence according to claim 1, characterized in that: The sub-steps for performing accuracy assessment on the initial deformation correction data and obtaining the second accuracy assessment result are as follows: S22331: Analyze the corrected coordinates of each ground control point and the initial deformation correction data to obtain a first deformation error value; S22332: selecting a plurality of evenly distributed ground check points on the original optical remote sensing data in the target data, and analyzing the corrected coordinates of each ground check point and the corresponding corrected coordinates in the initial deformation correction data to obtain a second deformation error value, wherein the ground check point is different from the ground control point; S22333: Obtain a comprehensive deformation error value based on the first deformation error value and the second deformation error value, and use the deformation accuracy threshold to analyze the comprehensive deformation error value to generate a second accuracy evaluation result; if the comprehensive deformation error value is less than the deformation accuracy threshold, the generated second accuracy evaluation result is qualified; if the comprehensive deformation error value is greater than or equal to the deformation accuracy threshold, the generated second accuracy evaluation result is unqualified.

4. The image processing method based on artificial intelligence according to claim 1, characterized in that: The sub-steps for obtaining the target mining data set based on the acquisition code and analyzing the pre-processed data using the target mining data set to obtain the current difference value are as follows: S31: Traversing the standard database according to the acquisition code, taking the mining data set in the standard data packet whose query code is consistent with the acquisition code as the comparison data set, wherein the comparison data set includes: multiple mining period data; each mining period data includes: post-mining acquisition time, post-mining optical remote sensing data, and post-mining synthetic aperture radar data; S32: sorting the mining period data and the pre-processed data in descending order of the plurality of post-mining acquisition times and the acquisition times, and generating corresponding analysis serial numbers, thereby obtaining data to be analyzed; wherein each data to be analyzed includes: an analysis serial number, an analysis acquisition time, analysis optical remote sensing data, and analysis synthetic aperture radar data; the analysis serial numbers are incremented in descending order of the plurality of post-mining acquisition times and the acquisition times; S33: Analyze the data to be analyzed to obtain the current difference value.

5. The image processing method based on artificial intelligence according to claim 4, characterized in that: The expression for the current difference value is: Where Dcy is the current difference value; λ1 is the pixel difference weight; ΔCxs k is the average value of the difference between all corresponding pixels in the k+1th analyzed optical remote sensing data and the kth analyzed optical remote sensing data; ΔCxs K is the average value of the difference between the analyzed optical remote sensing data in the K+1th data to be analyzed and the analyzed optical remote sensing data in the Kth data to be analyzed; K+1 is the total number of data to be analyzed, K is a natural number; λ2 is the disaster similarity weight; Gzh K+1 is the similarity value between the analyzed synthetic aperture radar data in the K+1th data to be analyzed and the pre-built geological hazard feature set.

6. An image processing system based on artificial intelligence, characterized in that: include: At least one data acquisition node and artificial intelligence image processing center; Among them, the data collection node: collects the data to be processed according to the initial collection frequency or the new collection frequency, and sends the data to be processed to the artificial intelligence image processing center; receives alarm information; Artificial intelligence image processing center: used to execute the artificial intelligence-based image processing method described in any one of claims 1-5.

Images