TSP advanced geological forecast multi-signal rapid conversion system

Abstract

The invention relates to the field of geological exploration, and discloses a TSP advanced geological forecast multi-signal rapid conversion system, which comprises geological exploration equipment carrying a data acquisition unit; the forecasting unit is used for acquiring continuous geological data from the data acquisition unit, and entering a forecasting control process after determining the captured initial data of the geological features; in the forecast control process, converting subsequent data acquired after the initial data item by item to eliminate the difference with the initial data, and performing pattern recognition one by one to detect geologic feature points; updating the mode recognition model through iteration; and converting the detected geological feature points into the data before conversion to obtain coordinates of the geological feature points, and calculating a geological change trend as a forecast signal to be output to geological exploration equipment. According to the invention, the method can achieve the quick recognition and prediction of geological features, improves the accuracy and efficiency of geological exploration, facilitates the prevention of geological disasters, and guarantees the safety of exploration operation.

Description

Technical Field

[0001] The invention relates to the field of geological exploration, in particular to a TSP advanced geological prediction multi-signal fast conversion system. Background Art

[0002] In the traditional field of geological exploration, geological prediction mainly relies on manual analysis and empirical judgment. This method is not only inefficient, but also has limited accuracy, which makes it difficult to meet the needs of modern geological exploration for efficient and accurate prediction. With the development of technology, some automated geological prediction systems have begun to appear. They collect geological data and use certain algorithms to analyze them in order to improve the accuracy and efficiency of prediction. However, when processing large amounts of continuous geological data, existing geological prediction systems often face the problems of low data conversion efficiency and low pattern recognition accuracy, which limits their application in actual exploration operations.

[0003] In the process of implementing the embodiments of the present invention, the inventors found that there are at least the following problems or defects in the prior art: the existing geological prediction system is often unable to quickly and effectively identify key geological feature points from continuous geological data during data collection and processing, resulting in the timeliness and accuracy of the forecast being affected; in addition, the existing pattern recognition algorithm may not be able to accurately capture the trend of geological changes when processing complex geological data, and more efficient algorithms and data processing mechanisms are needed to improve the accuracy and response speed of the forecast. Summary of the invention

[0004] The embodiment of the present invention aims to provide a TSP advanced geological prediction multi-signal rapid conversion system to solve the technical problems raised in the prior art.

[0005] The embodiment of the present invention solves the technical problem by adopting the following technical solution:

[0006] The present invention provides a TSP advanced geological prediction multi-signal rapid conversion system, comprising:

[0007] It includes a data acquisition unit and geological exploration equipment equipped with the data acquisition unit;

[0008] Also included is a prediction unit for geological prediction electrically connected to the data acquisition unit and the geological prospecting equipment;

[0009] The forecasting unit for geological forecasting acquires continuous geological data from the data acquisition unit and enters the forecasting control process after determining the initial data that the geological features have been captured;

[0010] In the forecast control process, the subsequent data obtained after the initial data are converted item by item to eliminate the difference with the initial data; the pattern recognition is performed one by one on the converted data to detect the geological feature points; and in the pattern recognition detection one by one, the pattern recognition model is updated through iteration; the detected geological feature points are converted back to the data before the conversion to obtain the coordinates of the geological feature points in the data before the conversion; and based on the coordinates of the geological feature points, the geological change trend is calculated and output as a forecast signal to the geological exploration equipment.

[0011] Furthermore, the forecasting unit for geological forecasting includes:

[0012] A data receiving module, used for acquiring continuous geological data from a data acquisition unit;

[0013] An initial data determination module is used to capture geological features from continuous geological data and use the data after completing the geological feature capture as initial data for forecasting;

[0014] A data conversion module, used to convert subsequent data obtained after the initial data to eliminate the difference with the initial data to obtain converted data;

[0015] A geological feature extraction module is used to extract features from the initial data and the converted data to obtain a feature matrix;

[0016] The pattern recognition module is used to perform pattern recognition on the feature matrices extracted from the transformed data one by one to detect geological feature points; and in the pattern recognition detection one by one, the pattern recognition model is updated through iteration;

[0017] A coordinate conversion module is used to convert the detected geological feature points back to the data before the conversion, and obtain the coordinates of the geological feature points in the data before the conversion;

[0018] The forecast signal calculation module is used to calculate the geological change trend according to the coordinates of the geological feature points in the data before conversion one by one, and output it to the geological exploration equipment as a forecast signal.

[0019] Furthermore, in the data conversion module, the calculation process of eliminating data differences through conversion is:

[0020] 1) According to the position data of geological exploration equipment, establish the transformation matrix from the points in the actual geological coordinate system to the points in the data coordinate system:

[0021]

[0022] Where, θ is the data conversion angle; M is the transformation matrix from the data coordinate system to the actual geological coordinate system;

[0023] 2) Obtain the inverse matrix of the transformation matrix to obtain the transformation matrix from the data coordinate system to the actual geological coordinate system;

[0024]

[0025] 3) Get the data conversion angle:

[0026] θ=arctan2(y diff ,x diff )

[0027] Among them, x diff and diff are the X-direction and Y-direction components of the data difference, respectively.

[0028] Furthermore, the data conversion module interpolates the converted data to obtain a feature map including the attribute value of each feature point in the data.

[0029] Furthermore, the geological feature extraction module includes a gradient calculation module, a gradient direction histogram construction module, a normalization module and a feature matrix output module; wherein,

[0030] The gradient calculation module is used to calculate the X and Y direction gradients of the data, including the gradient amplitude and angle;

[0031] The gradient direction histogram construction module is used to divide cells and construct the gradient direction histogram of each cell;

[0032] Normalization module, used to normalize the gradient histogram;

[0033] The feature matrix output module is used to connect the normalized gradient histogram vectors together to obtain the Hog features and form a feature matrix Z for output.

[0034] Furthermore, the pattern recognition module performs geological feature detection by calculating the recognition response matrix; searching for peak values ​​in the recognition response matrix to obtain the coordinates of the geological feature points; wherein,

[0035] R=F -1 (F(G)·F(H)·F(K)+λF(K)·F(K) * )

[0036] F -1 is the inverse Fourier transform; F(G) is the Fourier transform of the pattern recognition model; F(H) is the Fourier transform of the Gaussian kernel function; λ is the regularization parameter; F(K) is the Fourier transform of the cross-correlation function; K is the cross-correlation function; G is the width parameter of the Gaussian kernel function; H is the template data of the pattern recognition model; F(H) * is the conjugate of the Fourier transform of the template data; F(Z) is the Fourier transform of the feature matrix.

[0037] Furthermore, before the pattern recognition model is updated, the initial value of the template data of the pattern recognition model is H0, and the image of a set size is intercepted and normalized in the feature map of the initial data with the coordinates of the geological feature points of the initial data as the center, and the feature matrix obtained by extracting the Hog feature is obtained;

[0038] The update formula of the pattern recognition model template data is:

[0039]

[0040] Where α is the update rate; H t-1 It is the template data of the last pattern recognition; The Hog characteristic matrix is ​​calculated by intercepting the search area with the coordinates of the geological characteristic points of the converted data of the latest frame data that has been tracked as the center.

[0041] Furthermore, before the pattern recognition model is updated, the pattern recognition model; F(G) is the Fourier transform of the Gaussian kernel function; λ is the regularization parameter; F(K) is the autocorrelation function of the initial value of the template data;

[0042] The update formula of the pattern recognition model is:

[0043]

[0044] Among them, α is the update rate; G t-1 is the relevant filtering model of the previous pattern recognition; To obtain a pattern recognition model using the updated template data.

[0045] Furthermore, in the coordinate conversion module, the coordinates of the geological feature points in the converted data are converted to the data before the conversion, and the formula for obtaining the coordinates of the geological feature points in the data before the conversion is:

[0046] (x′,y′)=(x·cosθ-y·sinθ,x·sinθ+y·cosθ)

[0047] Where (x, y) is the coordinate of the point in the data before conversion; (x′, y′) is the pixel coordinate of the data after conversion; x ,O y is the data conversion center coordinate, P x ,P y is any point of the data before conversion; θ is the data conversion angle.

[0048] Furthermore, in the forecast signal calculation module, according to the formula

[0049]

[0050] Calculate geological change trends as forecast output signals;

[0051] Among them, V x and V y is the geological change rate in the X and Y directions, (x c ,y c ) are the coordinates of the center point in the X and Y directions of the data, and is the equivalent value of geological change rate in the X and Y directions, and the calculation formula is:

[0052]

[0053] Where k is a proportional constant used to convert the coordinate difference into an equivalent value of the geological change rate.

[0054] The above-mentioned embodiments of the present invention have at least the following beneficial effects: the TSP advanced geological prediction multi-signal rapid conversion system of the present invention can improve the processing speed of geological exploration data and the accuracy of prediction. The system continuously collects geological data and quickly enters the prediction control process after capturing the initial data of geological features. In this process, the system can quickly compare and convert subsequent data with initial data, eliminate the difference between the two, and thus efficiently identify geological feature points. In addition, the system can further improve the accuracy of geological feature detection by iteratively updating the pattern recognition model. Through these technical means, the real-time and reliability of geological prediction can be ensured, providing more accurate guidance for geological exploration.

[0055] Furthermore, after detecting the geological feature points, the system of the present invention can convert their coordinates back to the data before conversion, so as to obtain the accurate position of the geological feature points in the original data. This step is crucial for understanding and predicting geological change trends. Finally, the system calculates the geological change trend based on the coordinates of these geological feature points and outputs it as a forecast signal to provide decision support for geological exploration equipment. Such a design can not only reduce the risks in the exploration process, but also improve the efficiency and safety of exploration operations, which is particularly important for exploration activities under complex geological conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] One or more embodiments are exemplarily described by the figures in the corresponding drawings, and these exemplifications do not constitute limitations on the embodiments. Elements with the same reference numerals in the drawings represent similar elements, and unless otherwise stated, the figures in the drawings do not constitute proportional limitations.

[0057] Figure 1 It is a structural schematic diagram of the TSP advanced geological prediction multi-signal rapid conversion system provided by the present invention. DETAILED DESCRIPTION

[0058] In order to facilitate the understanding of the present invention, the present invention is described in more detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that when an element is described as "connecting" another element, it can be directly on another element, or there can be one or more centered elements therebetween. The orientation or positional relationship indicated by the terms "upper", "lower", "left", "right", "upper end", "lower end", "top" and "bottom" used in this specification is based on the orientation or positional relationship shown in the accompanying drawings, only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present invention. In addition, the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0059] Unless otherwise defined, all technical and scientific terms used in this specification have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0060] Combine the following Figure 1 , the TSP advanced geological prediction multi-signal rapid conversion system 100 provided in the embodiment of the present application is described in detail through a specific embodiment.

[0061] Figure 1 The structure diagram of the TSP advanced geological prediction multi-signal rapid conversion system provided by the present invention is as follows. The TSP advanced geological prediction multi-signal rapid conversion system provided by one embodiment of the present invention comprises:

[0062] Data acquisition unit 101, geological exploration equipment equipped with the data acquisition unit;

[0063] It also includes a prediction unit 102 for geological prediction electrically connected to the data acquisition unit and the geological exploration equipment;

[0064] The forecasting unit for geological forecasting acquires continuous geological data from the data acquisition unit and enters the forecasting control process after determining the initial data that the geological features have been captured;

[0065] In the forecast control process, the subsequent data obtained after the initial data are converted item by item to eliminate the difference with the initial data; the pattern recognition is performed one by one on the converted data to detect the geological feature points; and in the pattern recognition detection one by one, the pattern recognition model is updated through iteration; the detected geological feature points are converted back to the data before the conversion to obtain the coordinates of the geological feature points in the data before the conversion; and based on the coordinates of the geological feature points, the geological change trend is calculated and output as a forecast signal to the geological exploration equipment.

[0066] Specifically, the data acquisition unit is responsible for collecting various data during geological exploration, such as seismic waveforms, geological structure images, etc. These data are transmitted to the forecast unit, which includes a data receiving module, an initial data determination module, a data conversion module, a geological feature extraction module, a pattern recognition module, a coordinate conversion module, and a forecast signal calculation module. These modules work together to ensure continuous data acquisition, feature capture, conversion, recognition, and output of forecast signals. For example, the data conversion module can eliminate differences between data through specific algorithms, while the pattern recognition module uses advanced machine learning algorithms to identify geological feature points.

[0067] Preferably, in order to improve the prediction accuracy and response speed of the system, the data conversion module and the pattern recognition module can be further optimized. For example, the data conversion module can use a more efficient algorithm to process data and reduce the calculation time. The pattern recognition module can use deep learning technology to improve the accuracy of identifying geological features by training a large amount of geological data. In addition, the coordinate conversion module can introduce a more accurate mathematical model to ensure the accurate positioning of geological feature points, thereby improving the reliability of the prediction signal. These optimization measures help to further improve the prediction performance of the system and make it more adaptable to complex geological exploration environments.

[0068] In some embodiments, the prediction unit for geological prediction includes:

[0069] A data receiving module, used for acquiring continuous geological data from a data acquisition unit;

[0070] An initial data determination module is used to capture geological features from continuous geological data and use the data after completing the geological feature capture as initial data for forecasting;

[0071] A data conversion module, used to convert subsequent data obtained after the initial data to eliminate the difference with the initial data to obtain converted data;

[0072] A geological feature extraction module is used to extract features from the initial data and the converted data to obtain a feature matrix;

[0073] The pattern recognition module is used to perform pattern recognition on the feature matrices extracted from the transformed data one by one to detect geological feature points; and in the pattern recognition detection one by one, the pattern recognition model is updated through iteration;

[0074] A coordinate conversion module is used to convert the detected geological feature points back to the data before the conversion, and obtain the coordinates of the geological feature points in the data before the conversion;

[0075] The forecast signal calculation module is used to calculate the geological change trend according to the coordinates of the geological feature points in the data before conversion one by one, and output it to the geological exploration equipment as a forecast signal.

[0076] It should be noted that this embodiment describes in detail the internal structure and workflow of the forecast unit for geological forecasting. The forecast unit is the core part of the system, which realizes the whole process from data collection to geological forecasting through the collaborative work of multiple modules. These modules include a data receiving module, an initial data determination module, a data conversion module, a geological feature extraction module, a pattern recognition module, a coordinate conversion module and a forecast signal calculation module, and each module has its specific function and role.

[0077] Specifically, the data receiving module is responsible for receiving continuous geological data from the data acquisition unit, which may include seismic waveforms, geological structure images, etc. The initial data determination module identifies data with geological characteristics from the received continuous data and uses it as the initial data for prediction. The data conversion module processes the subsequent data after the initial data to eliminate the difference with the initial data, so as to facilitate the subsequent feature extraction and pattern recognition. The geological feature extraction module extracts the characteristics of geological data by calculating the gradient of the data, constructing the gradient direction histogram, etc. The pattern recognition module analyzes the extracted features and identifies the geological feature points. The coordinate conversion module converts the coordinates of the identified geological feature points back to the original data to facilitate the calculation of the prediction signal. The prediction signal calculation module calculates the geological change trend according to the coordinates of the geological feature points and outputs the prediction signal.

[0078] Preferably, in order to improve the prediction accuracy and efficiency of the system, more advanced data processing algorithms, such as wavelet transform or neural network, can be used in the data conversion module to more effectively eliminate the differences between data. In the geological feature extraction module, more feature extraction techniques, such as local binary patterns or deep learning features, can be introduced to improve the expressiveness of features. In the pattern recognition module, more complex machine learning models, such as convolutional neural networks or recurrent neural networks, can be used to improve the recognition accuracy of geological feature points. In addition, a more accurate coordinate conversion algorithm can be introduced in the coordinate conversion module to ensure the accurate positioning of geological feature points. Through these optimization measures, the prediction performance of the system can be further improved, making it more adaptable to complex geological exploration environments.

[0079] In some embodiments, in the data conversion module, the calculation process of eliminating data differences through conversion is:

[0080] 1) According to the position data of geological exploration equipment, establish the transformation matrix from the points in the actual geological coordinate system to the points in the data coordinate system:

[0081]

[0082] Where, θ is the data conversion angle; M is the transformation matrix from the data coordinate system to the actual geological coordinate system;

[0083] 2) Obtain the inverse matrix of the transformation matrix to obtain the transformation matrix from the data coordinate system to the actual geological coordinate system;

[0084]

[0085] 3) Get the data conversion angle:

[0086] θ=arctan2(y diff ,x diff )

[0087] Among them, x diff and diff are the X-direction and Y-direction components of the data difference, respectively.

[0088] It should be noted that the present embodiment further elaborates on the calculation process of eliminating data differences in the data conversion module. This process is a key step for the forecast unit to accurately identify geological features, and it involves operations such as establishing a transformation matrix, finding an inverse matrix, and calculating a data conversion angle. Through these operations, the system can effectively compare the subsequently collected data with the initial data, thereby providing support for the accurate identification of geological features.

[0089] Specifically, the data conversion module first establishes a transformation matrix based on the location data of the geological exploration equipment, which can transform points in the actual geological coordinate system to points in the data coordinate system. This transformation matrix is ​​a 3x3 matrix that contains information about the data conversion angle and coordinate system transformation. Then, the module inverts the transformation matrix to obtain the inverse matrix that converts from the data coordinate system back to the actual geological coordinate system. The data conversion angle is obtained by calculating the X-direction and Y-direction components of the data difference, and this angle is crucial for subsequent data conversion.

[0090] Preferably, in order to improve the accuracy and efficiency of data conversion, more advanced mathematical methods can be used to optimize the establishment of the transformation matrix and the solution process of the inverse matrix. For example, numerical optimization techniques can be introduced to improve the stability and accuracy of matrix calculations. In addition, the calculation of the data conversion angle can also be further optimized by introducing advanced sensor data to ensure the accuracy of the conversion angle. In practical applications, the variability of the geological exploration environment can also be considered, and the transformation matrix and inverse matrix can be adjusted in real time to adapt to different exploration conditions, thereby improving the adaptability of the system and the accuracy of the forecast. These optimization measures help to further improve the performance of the system under complex geological conditions.

[0091] In some embodiments, the data conversion module interpolates the converted data to obtain a feature map including attribute values ​​of each feature point in the data.

[0092] It should be noted that this embodiment describes a method for interpolating the converted data in the data conversion module. Interpolation is a mathematical method used to estimate the value of an unknown data point between known data points. In geological exploration, interpolation can help construct a feature map containing the attribute values ​​of each feature point in the data, which is crucial for subsequent geological feature extraction and pattern recognition.

[0093] Specifically, after the data conversion module completes the elimination of data differences, it interpolates the converted data. This process involves selecting a suitable interpolation algorithm, such as linear interpolation, cubic spline interpolation, or radial basis function interpolation. The choice of interpolation algorithm depends on the characteristics of the data and the required accuracy. During the interpolation process, the algorithm calculates and fills new data points between known data points to generate a continuous feature map containing the attribute values ​​of all feature points. This feature map can provide richer data information for the geological feature extraction module, which helps to improve the recognition accuracy of geological features.

[0094] Preferably, in order to improve the effect of interpolation processing, an adaptive interpolation algorithm can be used, which can dynamically adjust the interpolation strategy according to the local characteristics of the data. For example, linear interpolation can be used in areas where the data changes relatively slowly, and cubic spline interpolation can be used in areas where the data changes drastically. In addition, machine learning methods, such as interpolation models based on deep learning, can be introduced to further improve the accuracy and efficiency of interpolation. These optimization measures help to generate more accurate feature maps, thereby providing more reliable data support for geological prediction.

[0095] In some embodiments, the geological feature extraction module includes a gradient calculation module, a gradient direction histogram construction module, a normalization module and a feature matrix output module; wherein,

[0096] The gradient calculation module is used to calculate the X and Y direction gradients of the data, including the gradient amplitude and angle;

[0097] The gradient direction histogram construction module is used to divide cells and construct the gradient direction histogram of each cell;

[0098] Normalization module, used to normalize the gradient histogram;

[0099] The feature matrix output module is used to connect the normalized gradient histogram vectors together to obtain the Hog features and form a feature matrix Z for output.

[0100] It should be noted that the present embodiment describes in detail the composition and working mechanism of the geological feature extraction module. The geological feature extraction module is a key part of the forecast unit, which is responsible for extracting features that are helpful for geological forecasting from the converted data. The module includes a gradient calculation module, a gradient direction histogram construction module, a normalization module, and a feature matrix output module. Each module has a specific task and completes the extraction of geological features together.

[0101] Specifically, the gradient calculation module first calculates the gradient of the data in the X and Y directions, which includes the magnitude and angle of the gradient. Gradient is an important concept used for edge detection in image processing, which reflects the intensity and direction of the change in image brightness. After calculating the gradient, the gradient direction histogram construction module divides the data into multiple small areas called cells and constructs a gradient direction histogram in each cell, which helps capture the local structural information of the image. The normalization module normalizes the histogram to eliminate the brightness and contrast differences between different images, making the features more comparable. Finally, the feature matrix output module connects the normalized gradient histogram vectors together to form Hog features, which are organized into a feature matrix Z for output.

[0102] Preferably, in order to improve the accuracy and robustness of geological feature extraction, a more advanced gradient calculation method, such as Gaussian difference method or Sobel operator, can be used for the gradient calculation module. In the gradient direction histogram construction module, the size and number of cells can be adjusted to adapt to geological data of different resolutions and complexities. The normalization module can adopt different normalization strategies, such as L2 normalization or MinMax normalization, to optimize the distribution of features. In addition, the feature matrix output module can introduce more feature descriptors, such as color histograms or texture features, to enrich the information content of the feature matrix Z. These optimization measures help to improve the performance of geological feature extraction, thereby improving the forecast accuracy of the entire forecast system.

[0103] In some embodiments, the pattern recognition module performs geological feature detection by calculating an identification response matrix; searching for peak values ​​in the identification response matrix to obtain the coordinates of geological feature points; wherein,

[0104] R=F -1 (F(G)·F(H)·F(K)+λF(K)·F(K) * )

[0105] F -1 is the inverse Fourier transform; F(G) is the Fourier transform of the pattern recognition model; F(H) is the Fourier transform of the Gaussian kernel function; λ is the regularization parameter; F(K) is the Fourier transform of the cross-correlation function; K is the cross-correlation function; G is the width parameter of the Gaussian kernel function; H is the template data of the pattern recognition model; F(H) * is the conjugate of the Fourier transform of the template data; F(Z) is the Fourier transform of the feature matrix.

[0106] It should be noted that the pattern recognition module is the key part of the forecast unit for identifying geological feature points. It determines the coordinates of geological feature points by calculating the recognition response matrix and searching for peaks in it. This process involves complex mathematical operations such as Fourier transform, Gaussian kernel function, and cross-correlation function to ensure accurate identification of geological features.

[0107] Specifically, the pattern recognition module first calculates the recognition response matrix, which is achieved by multiplying the Fourier transform of the pattern recognition model with the Fourier transform of the Gaussian kernel function and the Fourier transform of the cross-correlation function. In this process, the regularization parameter is used to control the smoothness of the model to avoid overfitting. The calculation formula of the recognition response matrix includes the Fourier transform of the pattern recognition model, the Fourier transform of the Gaussian kernel function, the regularization parameter, the Fourier transform of the cross-correlation function, and the conjugate of the Fourier transform of the template data. The selection of these parameters and functions is crucial to the accuracy of identifying geological feature points. After obtaining the recognition response matrix, the coordinates of the geological feature points are determined by searching for the peaks therein.

[0108] Preferably, in order to improve the accuracy and efficiency of pattern recognition, more advanced pattern recognition algorithms, such as deep convolutional neural networks, can be used to improve the recognition ability of complex geological features. When calculating the recognition response matrix, an adaptive Gaussian kernel function can be introduced, and its width parameter can be dynamically adjusted according to the characteristics of the geological data to better adapt to different geological conditions. In addition, the calculation efficiency can be improved by optimizing the calculation method of Fourier transform, such as using the fast Fourier transform FFT algorithm. When searching for peaks, more sophisticated search strategies, such as multi-scale search or non-local maximum suppression, can be used to improve the accuracy of peak detection. These optimization measures help to improve the performance of the pattern recognition module, thereby improving the forecast accuracy and reliability of the entire forecast system.

[0109] In some embodiments, before the pattern recognition model is updated, the initial value of the template data of the pattern recognition model is H0, and the image of a set size is intercepted and normalized in the feature map of the initial data with the coordinates of the geological feature points of the initial data as the center, and the feature matrix obtained by extracting the Hog feature is obtained;

[0110] The update formula of the pattern recognition model template data is:

[0111]

[0112] Where α is the update rate; H t-1 It is the template data of the last pattern recognition; The Hog characteristic matrix is ​​calculated by intercepting the search area with the coordinates of the geological characteristic points of the converted data of the latest frame data that has been tracked as the center.

[0113] It should be noted that in the geological prediction system, the accuracy of the pattern recognition model is crucial for the accurate identification of geological feature points. Therefore, the system needs to continuously update the pattern recognition model to adapt to changes in geological data. This process involves the initialization, update and fusion of template data with the latest frame data to ensure that the model can reflect the latest changes in geological features in a timely manner.

[0114] Specifically, before the pattern recognition model is updated, the initial value of the template data is determined based on the coordinates of the geological feature points of the initial data. In the feature map of the initial data, an image area of ​​a certain size is intercepted with the geological feature points as the center, and normalized, and then the Hog features are extracted to form a feature matrix. This process involves the size of the image interception, the normalization method, and the extraction parameters of the Hog features. In the update formula, the update rate is a key parameter that determines the degree of influence of the new data on the template data. The choice of the update rate needs to balance the stability of the model and its adaptability to new data.

[0115] Preferably, in order to improve the efficiency and accuracy of pattern recognition model updating, more advanced image processing and feature extraction techniques can be used. For example, more complex image capture strategies, such as content-based image region selection, can be used to ensure that the captured image region contains as much geological feature information as possible. In the normalization process, more advanced normalization methods, such as contrast-limited adaptive histogram equalization CLAHE, can be used to improve the contrast of the image and highlight the geological features. In terms of feature extraction, in addition to the Hog feature, other types of features, such as deep learning features, can also be considered to provide richer information. In addition, the update rate in the update formula can be dynamically adjusted according to the performance of the model and the changes in geological data to achieve a better model update effect. These optimization measures help to improve the update quality of the pattern recognition model, thereby improving the forecast accuracy and response speed of the entire forecast system.

[0116] In some embodiments, before the pattern recognition model is updated, the pattern recognition model; F(G) is the Fourier transform of the Gaussian kernel function; λ is the regularization parameter; F(K) is the autocorrelation function of the initial value of the template data;

[0117] The update formula of the pattern recognition model is:

[0118]

[0119] Among them, α is the update rate; G t-1 is the relevant filtering model of the previous pattern recognition; To obtain a pattern recognition model using the updated template data.

[0120] It should be noted that the pattern recognition model can adapt to the changes in geological data and improve the accuracy of forecasts by continuously updating. This process involves multiple concepts such as Gaussian kernel function, regularization parameter, and autocorrelation function of the initial value of template data, which work together to update the pattern recognition model.

[0121] Specifically, the update of the pattern recognition model starts with the initial value of the template data, which is determined based on the coordinates of the geological feature points. The autocorrelation function is used to describe the internal correlation of the initial value of the template data, which helps to capture the inherent laws of geological features. The Fourier transform of the Gaussian kernel function is used for smoothing to reduce the noise interference of the model. The regularization parameter is used to control the complexity of the model to prevent overfitting. In the update formula, the update rate is a key parameter, which determines the degree of influence of the new data on the template data and needs to be adjusted according to the actual situation.

[0122] Preferably, in order to improve the efficiency and accuracy of pattern recognition model updating, more advanced data processing and machine learning technologies can be used. For example, a model update strategy based on deep learning can be introduced to automatically learn the characteristics and change laws of geological data using neural networks. In the selection of Gaussian kernel function, an adaptive Gaussian kernel can be used, and its width parameter can be dynamically adjusted according to the characteristics of geological data to better adapt to different geological conditions. In addition, the selection of regularization parameters can be based on the cross-validation method to ensure the generalization ability of the model on different geological data. When updating the template data, it is possible to consider introducing a multi-scale update strategy, that is, updating the template data at different scales at the same time to capture geological features at different scales. These optimization measures help to improve the update quality of the pattern recognition model, thereby improving the forecast accuracy and response speed of the entire forecast system.

[0123] In some embodiments, in the coordinate conversion module, the coordinates of the geological feature points in the converted data are converted to the data before the conversion, and the formula for obtaining the coordinates of the geological feature points in the data before the conversion is:

[0124] (x′,y′)=(x·cosθ-y·sinθ,x·sinθ+y·cosθ)

[0125] Where (x, y) is the coordinate of the point in the data before conversion; (x′, y′) is the pixel coordinate of the data after conversion; x ,O y is the data conversion center coordinate, P x ,P y is any point of the data before conversion; θ is the data conversion angle.

[0126] It should be noted that this embodiment describes in detail the function of the coordinate conversion module, which is responsible for converting the coordinates of the geological feature points in the converted data back to the corresponding coordinates in the data before conversion. This step is crucial to ensure the accuracy of geological prediction because it allows the system to accurately locate the geological feature points in the context of the original data.

[0127] Specifically, the coordinate conversion module uses a specific formula to calculate the position of the geological feature points in the converted data in the data before conversion. This process involves parameters such as the data conversion center coordinates, the coordinates of any point in the data before conversion, and the data conversion angle. These parameters jointly determine the accuracy of the coordinate conversion. The data conversion angle is calculated by analyzing the data difference, which reflects the direction change of the data during the conversion process. The conversion center coordinates define the reference point of the coordinate conversion.

[0128] Preferably, in order to improve the accuracy and efficiency of coordinate transformation, more advanced mathematical models and algorithms can be used. For example, a three-dimensional coordinate transformation model can be used to process more complex geological data, or an algorithm based on machine learning can be introduced to automatically optimize the coordinate transformation parameters. In addition, an iterative method can be used to gradually refine the coordinate transformation results to improve the accuracy of the transformation. In practical applications, the noise and errors of geological data can also be considered, and the influence of these factors on the coordinate transformation results can be reduced by introducing an error correction mechanism. These optimization measures help to improve the performance of the coordinate transformation module, thereby improving the accuracy and reliability of the entire forecast system in geological forecasting.

[0129] In some embodiments, in the forecast signal calculation module, according to the formula

[0130]

[0131] Calculate geological change trends as forecast output signals;

[0132] Among them, V x and V y is the geological change rate in the X and Y directions, (x c ,y c ) are the coordinates of the center point in the X and Y directions of the data, and is the equivalent value of geological change rate in the X and Y directions, and the calculation formula is:

[0133]

[0134] Where k is a proportional constant used to convert the coordinate difference into an equivalent value of the geological change rate.

[0135] Specifically, the forecast signal calculation module uses a specific formula to calculate the geological change trend. This formula involves parameters such as the coordinates of the center point in the X and Y directions of the data, the geological change rate, and the proportional constant. The geological change rate is estimated by analyzing the changes in the coordinates of the geological feature points, which reflects the changes in the geological structure in time and space. The proportional constant is used to convert the coordinate difference into an equivalent value of the geological change rate. The setting of this constant needs to be determined according to the actual geological conditions and exploration needs.

[0136] Preferably, in order to improve the accuracy and reliability of the forecast signal calculation, more advanced mathematical models and calculation methods can be used. For example, a physics-based model can be introduced to more accurately describe the geological change process, or a machine learning algorithm can be used to learn the pattern of geological change from historical data. In addition, a multi-parameter fusion method can be used to combine the geological change rate with other geological parameters, such as seismic wave velocity, rock density, etc., to provide a more comprehensive forecast signal. In practical applications, the complexity of the geological environment can also be considered, and the proportional constant and other parameters can be adjusted in real time to adapt to different geological conditions. These optimization measures help to improve the performance of the forecast signal calculation module, thereby improving the accuracy and practicality of the entire forecast system in geological forecasting.

[0137] The above-mentioned embodiments of the present invention have the following beneficial effects: The TSP advanced geological prediction multi-signal rapid conversion system of the present invention can improve the efficiency and accuracy of geological exploration. Through the integrated data acquisition unit and prediction unit, the system can acquire and process geological data in real time, quickly identify geological features and make predictions. The application of such rapid conversion and pattern recognition technology makes geological predictions more timely and accurate, helps prevent potential geological disasters, and ensures the safety of exploration operations.

[0138] In addition, the system can continuously optimize the accuracy of forecasts by iteratively updating the pattern recognition model and the application of the coordinate conversion module. The forecast signal calculation module further converts the detected geological feature points into geological change trends, providing direct forecast signal output for geological exploration equipment. The comprehensive application of these technologies can not only improve the response speed of forecasts, but also enhance the practicality of forecast results, providing strong technical support for geological exploration.

[0139] Furthermore, the storage medium of the embodiment of the present application stores program instructions that can implement all the above methods, wherein the program instructions can be stored in the above storage medium in the form of a software product, including several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or terminal devices such as a computer, a server, a mobile phone, and a tablet.

[0140] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them. Under the concept of the present invention, the technical features in the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other changes in different aspects of the present invention as above, which are not provided in detail for the sake of simplicity. Although the present invention is described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A TSP advanced geological prediction multi-signal rapid conversion system, comprising a data acquisition unit and a geological exploration device equipped with the data acquisition unit, characterized in that: Also included is a prediction unit for geological prediction electrically connected to the data acquisition unit and the geological prospecting equipment; The forecasting unit for geological forecasting acquires continuous geological data from the data acquisition unit and enters the forecasting control process after determining the initial data that the geological features have been captured; In the forecast control process, the subsequent data obtained after the initial data are converted item by item to eliminate the difference with the initial data; the pattern recognition is performed one by one on the converted data to detect the geological feature points; and in the pattern recognition detection one by one, the pattern recognition model is updated through iteration; the detected geological feature points are converted back to the data before the conversion to obtain the coordinates of the geological feature points in the data before the conversion; and based on the coordinates of the geological feature points, the geological change trend is calculated and output as a forecast signal to the geological exploration equipment.

2. The TSP advanced geological prediction multi-signal rapid conversion system according to claim 1 is characterized in that: The forecasting unit for geological forecasting comprises: A data receiving module, used for acquiring continuous geological data from a data acquisition unit; An initial data determination module is used to capture geological features from continuous geological data and use the data after completing the geological feature capture as initial data for forecasting; A data conversion module, used to convert subsequent data obtained after the initial data to eliminate the difference with the initial data to obtain converted data; A geological feature extraction module is used to extract features from the initial data and the converted data to obtain a feature matrix; The pattern recognition module is used to perform pattern recognition on the feature matrices extracted from the transformed data one by one to detect geological feature points; and in the pattern recognition detection one by one, the pattern recognition model is updated through iteration; A coordinate conversion module is used to convert the detected geological feature points back to the data before the conversion, and obtain the coordinates of the geological feature points in the data before the conversion; The forecast signal calculation module is used to calculate the geological change trend according to the coordinates of the geological feature points in the data before conversion one by one, and output it to the geological exploration equipment as a forecast signal.

3. The TSP advanced geological prediction multi-signal rapid conversion system according to claim 2 is characterized in that: In the data conversion module, the calculation process of eliminating data differences through conversion is: 1) According to the position data of geological exploration equipment, establish the transformation matrix from the points in the actual geological coordinate system to the points in the data coordinate system: Where, θ is the data conversion angle; M is the transformation matrix from the data coordinate system to the actual geological coordinate system; 2) Obtain the inverse matrix of the transformation matrix to obtain the transformation matrix from the data coordinate system to the actual geological coordinate system; 3) Get the data conversion angle: θ=arctan 2(and diff ,x diff ) Among them, x diff and diff are the X-direction and Y-direction components of the data difference, respectively.

4. The TSP advanced geological prediction multi-signal rapid conversion system according to claim 2 is characterized in that: In the data conversion module, the converted data is interpolated to obtain a feature map including the attribute values ​​of each feature point in the data.

5. The TSP advanced geological prediction multi-signal rapid conversion system according to claim 4 is characterized in that: The geological feature extraction module includes a gradient calculation module, a gradient direction histogram construction module, a normalization module and a feature matrix output module; among them, The gradient calculation module is used to calculate the X and Y direction gradients of the data, including the gradient amplitude and angle; The gradient direction histogram construction module is used to divide cells and construct the gradient direction histogram of each cell; Normalization module, used to normalize the gradient histogram; The feature matrix output module is used to connect the normalized gradient histogram vectors together to obtain the Hog features and form a feature matrix Z for output.

6. The TSP advanced geological prediction multi-signal rapid conversion system according to claim 5 is characterized in that: The pattern recognition module performs geological feature detection by calculating the recognition response matrix; searching for peak values ​​in the recognition response matrix to obtain the coordinates of the geological feature points; wherein, R=F -1 (F(G)·F(H)·F(K)+λF(K)·F(K) * ) F -1 is the inverse Fourier transform; F(G) is the Fourier transform of the pattern recognition model; F(H) is the Fourier transform of the Gaussian kernel function; λ is the regularization parameter; F(k) is the Fourier transform of the cross-correlation function; k is the cross-correlation function; G is the width parameter of the Gaussian kernel function; H is the template data of the pattern recognition model; F(H) * is the conjugate of the Fourier transform of the template data; F(Z) is the Fourier transform of the feature matrix.

7. The TSP advanced geological prediction multi-signal rapid conversion system according to claim 6 is characterized in that: Before the pattern recognition model is updated, the initial value of the template data of the pattern recognition model is H0. Taking the coordinates of the geological feature points of the initial data as the center, an image of a set size is intercepted from the feature map of the initial data for normalization, and then the Hog feature is extracted to obtain the feature matrix; The update formula of the pattern recognition model template data is: Where α is the update rate; H t-1 It is the template data of the last pattern recognition; The Hog characteristic matrix is ​​calculated by intercepting the search area with the coordinates of the geological characteristic points of the converted data of the latest frame data that has been tracked as the center.

8. The TSP advanced geological prediction multi-signal rapid conversion system according to claim 7 is characterized in that: Before the pattern recognition model is updated, the pattern recognition model; F(G) is the Fourier transform of the Gaussian kernel function; λ is the regularization parameter; F(K) is the autocorrelation function of the initial value of the template data; The update formula of the pattern recognition model is: Where F is the update rate; G t-1 is the relevant filtering model of the previous pattern recognition; To obtain a pattern recognition model using the updated template data.

9. The TSP advanced geological prediction multi-signal rapid conversion system according to claim 8 is characterized in that: In the coordinate conversion module, the coordinates of the geological feature points in the converted data are converted to the data before conversion, and the formula for obtaining the coordinates of the geological feature points in the data before conversion is: (x′,y′)=(x·cosθ-y·sinθ,x·sinθ+y·cosθ) Where (x, y) is the coordinate of the point in the data before conversion; (x′, y′) is the pixel coordinate of the data after conversion; x ,O y is the data conversion center coordinate, P x ,P y is any point of the data before conversion; θ is the data conversion angle.

10. The TSP advanced geological prediction multi-signal rapid conversion system according to claim 7 is characterized in that: In the forecast signal calculation module, according to the formula Calculate geological change trends as forecast output signals; Among them, V x and V y is the geological change rate in the X and Y directions, (x c ,y c ) are the coordinates of the center point in the X and Y directions of the data, and is the equivalent value of geological change rate in the X and Y directions, and the calculation formula is: Where k is a proportional constant used to convert the coordinate difference into an equivalent value of the geological change rate.

Images