A method for identifying geological features of geological detection images and related equipment

Abstract

The present application belongs to the field of geological survey technology, and discloses a geological feature recognition method for geological detection images and related equipment. The geological detection images are first divided into a first type of images with relatively low noise content and a second type of images with relatively high noise content. Then, the geological feature contours of the first type of images are recognized by using a horizontal and vertical dimension mutual operation algorithm of the image to improve the recognition efficiency of the first type of images, and the geological feature contours of the second type of images are recognized by using a SIFT algorithm to ensure the recognition accuracy of the second type of images. Therefore, the overall recognition efficiency of the geological feature contours of multiple geological detection images can be effectively improved while ensuring the overall accuracy of the geological feature contour recognition.

Description

Technical Field

[0001] The present application relates to the field of geological survey technology, and in particular to a method for identifying geological features of geological detection images and related equipment. Background Art

[0002] At present, in the process of geological survey, survey equipment is used to scan in one direction to generate geo-detection images (or geological images) of multiple cross sections of the strata in the survey area. Then, the contours of the geological features of each geo-detection image (such as shale gas storage space, minerals, and other geological structures in the geo-detection image that are significantly different from the surrounding strata are the geological features of the geo-detection image) are extracted to obtain the characteristic values ​​of each geological feature (such as circumference, area, etc.). Then, the comprehensive characteristic values ​​of each geo-detection image are statistically analyzed to provide a reference for judging whether the survey area has development value.

[0003] In fact, common geological structures and components are relatively fixed, so the geological detection images of the same survey area are mainly divided into fixed layer images (whose number of images is relatively large) and cross layer images (whose number of images is relatively small), and the fixed layer images are relatively simple in content and have less noise. However, in the prior art, all geological detection images are generally denoised, and then the same edge detection algorithm or contour detection algorithm, such as SIFT (Scale Invariant Feature Transform) algorithm, is used to uniformly extract geological feature contours from all denoised geological detection images. Although the recognition result is more accurate, the processing efficiency is low, and due to the huge number of geological detection images, the overall processing efficiency is low.

[0004] Therefore, the prior art needs to be improved and enhanced. Summary of the invention

[0005] The purpose of this application is to provide a geological feature recognition method for geological detection images and related equipment, which can effectively improve the overall recognition efficiency of geological feature contour recognition for multiple geological detection images while ensuring the overall accuracy of geological feature contour recognition.

[0006] In a first aspect, the present application provides a method for identifying geological features of a geological detection image, comprising the steps of:

[0007] A1. Acquire multiple geoscientific detection images of the target survey area;

[0008] A2. Identify the noise content of each of the geoscientific detection images based on the pre-trained noise content recognition model, recorded as the first noise content;

[0009] A3. Dividing the geospatial detection images into a first type of image and a second type of image according to the first noise content; the noise content of the first type of image is lower than the noise content of the second type of image;

[0010] A4. Using the image horizontal and vertical mutual operation algorithm to identify the geological feature contours of the geological detection images belonging to the first category of images, and using the SIFT algorithm to identify the geological feature contours of the geological detection images belonging to the second category of images;

[0011] A5. Obtain comprehensive characteristic statistical information of the target survey area according to the geological feature contours of each of the geological detection images.

[0012] The method first divides the geological detection images into a first type of images with relatively low noise content and a second type of images with relatively high noise content, and then uses the image horizontal and vertical dimension mutual operation algorithm to identify the geological feature contours of the first type of images to improve the recognition efficiency of the first type of images, and uses the SIFT algorithm to identify the geological feature contours of the second type of images to ensure the recognition accuracy of the second type of images; thereby, the overall recognition efficiency of the geological feature contour recognition of multiple geological detection images can be effectively improved while ensuring the overall accuracy of the geological feature contour recognition.

[0013] Preferably, step A3 comprises:

[0014] A301. Obtaining the recognition accuracy threshold and the average processing time threshold of a single geoscientific detection image;

[0015] A302. Determine the target noise content cutoff value according to the recognition accuracy threshold and the average processing time threshold;

[0016] A303. Divide the geoscientific detection image into a first category of images and a second category of images according to the target noise content cutoff value and the first noise content.

[0017] The user can set the recognition accuracy threshold and the average processing time threshold according to actual needs. Image segmentation in the above manner is conducive to ensuring that the final recognition accuracy is not lower than the recognition accuracy threshold and the average processing time does not exceed the average processing time threshold.

[0018] Preferably, step A302 includes:

[0019] Calling a query table, wherein the query table records a plurality of noise content cutoff values ​​and corresponding recognition accuracy rates and corresponding average processing times;

[0020] Filter the noise content cutoff value whose recognition accuracy is not lower than the recognition accuracy threshold from the query table as a candidate noise content cutoff value;

[0021] From the candidate noise content cutoff values ​​whose corresponding average processing time is not higher than the average processing time threshold, the candidate noise content cutoff value with the shortest average processing time is selected as the target noise content cutoff value.

[0022] Preferably, step A303 includes:

[0023] Classify the geoscientific detection image whose first noise content is not higher than the target noise content cutoff value into the first type of image;

[0024] The geoscientific detection image whose first noise content is higher than the target noise content cutoff value is classified into the second category of images.

[0025] Preferably, in step A4, the geological feature contour recognition is performed on the geoscientific detection image belonging to the first type of image using an image horizontal and vertical dimension mutual operation algorithm, including:

[0026] C1. generating a pixel data value matrix of the geoscientific detection image according to the pixel data values ​​of each pixel of the geoscientific detection image;

[0027] C2. performing a subtraction operation on two adjacent column vectors in the pixel data value matrix to obtain a plurality of difference column vectors to form a first difference matrix; the matrix size of the first difference matrix is ​​the same as the matrix size of the pixel data value matrix;

[0028] C3. Obtain pixel coordinates of non-zero elements in the first difference matrix to obtain a first pixel coordinate set;

[0029] C4. performing a subtraction operation on two adjacent row vectors in the pixel data value matrix to obtain a plurality of difference row vectors to form a second difference matrix; the matrix size of the second difference matrix is ​​the same as the matrix size of the pixel data value matrix;

[0030] C5. Obtain pixel coordinates of non-zero elements in the second difference matrix to obtain a second pixel coordinate set;

[0031] C6. Obtaining a union of the first pixel coordinate set and the second pixel coordinate set, and setting the elements in the pixel data value matrix whose pixel coordinates are not in the union to zero, to obtain a feature matrix;

[0032] C7. Generate a geological feature contour image according to the feature matrix; the geological feature contour image includes the geological feature contours of each geological feature.

[0033] To identify geological feature contours in the above manner, only simple vector subtraction operations are required, which reduces the amount of calculation and improves processing efficiency.

[0034] Preferably, step A5 comprises:

[0035] A501. According to the geological feature contours of each of the geoscience detection images, identifying the feature type of the corresponding geological feature;

[0036] A502. According to the geological feature contours of each of the geological detection images, obtain the characteristic values ​​of the corresponding geological features;

[0037] A503. Calculate the comprehensive feature value of each of the geological detection images according to the feature type and the feature value of each of the geological features as the comprehensive feature statistical information of the target survey area.

[0038] Preferably, in step A503, the comprehensive characteristic value of each of the geoscientific detection images is calculated according to the following formula:

[0039] ;

[0040] in, is the comprehensive feature value of the i-th geoscientific detection image, are the feature values ​​of the 1st to Nth geological features in the i-th geological detection image, is the attention coefficient of geological features from Category 1 to Category N.

[0041] In a second aspect, the present application provides a geological feature recognition device for a geological detection image, comprising:

[0042] An acquisition module, used for acquiring a plurality of geoscientific detection images of a target survey area;

[0043] A noise identification module, used for identifying the noise content of each of the geoscientific detection images based on a pre-trained noise content identification model, recorded as a first noise content;

[0044] A division module, used for dividing the geoscientific detection image into a first type of image and a second type of image according to the first noise content; the noise content of the first type of image is lower than the noise content of the second type of image;

[0045] A contour recognition module, used to perform geological feature contour recognition on the geological detection images belonging to the first category of images by using an image horizontal and vertical dimension mutual operation algorithm, and to perform geological feature contour recognition on the geological detection images belonging to the second category of images by using a SIFT algorithm;

[0046] The statistical module is used to obtain the comprehensive characteristic statistical information of the target survey area according to the geological characteristic contours of each of the geological detection images.

[0047] In a third aspect, the present application provides an electronic device comprising a processor and a memory, wherein the memory stores a computer program executable by the processor, and when the processor executes the computer program, it runs the steps in the method for identifying geological features of geological detection images as described above.

[0048] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, runs the steps of the method for identifying geological features of geological detection images as described above.

[0049] Beneficial effect: The geological feature recognition method of geological detection images provided in the present application and the related equipment thereof first divide the geological detection images into a first type of images with relatively low noise content and a second type of images with relatively high noise content, and then use the image horizontal and vertical dimension mutual operation algorithm to recognize the geological feature contours of the first type of images to improve the recognition efficiency of the first type of images, and use the SIFT algorithm to recognize the geological feature contours of the second type of images to ensure the recognition accuracy of the second type of images; thereby, the overall recognition efficiency of the geological feature contour recognition of multiple geological detection images can be effectively improved while ensuring the overall accuracy of the geological feature contour recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 A flowchart of a method for identifying geological features of a geological detection image provided in an embodiment of the present application.

[0051] Figure 2 A schematic diagram of the structure of a geological feature recognition device for geological detection images provided in an embodiment of the present application.

[0052] Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0053] Explanation of reference numerals: 1. Acquisition module; 2. Noise recognition module; 3. Division module; 4. Contour recognition module; 5. Statistics module; 301. Processor; 302. Memory; 303. Communication bus. DETAILED DESCRIPTION

[0054] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0055] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0056] Please refer to Figure 1 , Figure 1 A method for identifying geological features of a geological detection image in some embodiments of the present application comprises the steps of:

[0057] A1. Acquire multiple geoscientific detection images of the target survey area;

[0058] A2. Identify the noise content of each geodesic detection image based on the pre-trained noise content recognition model, recorded as the first noise content;

[0059] A3. Dividing the geoscientific detection images into a first type of images and a second type of images according to the first noise content; the noise content of the first type of images is lower than the noise content of the second type of images;

[0060] A4. Use the image horizontal and vertical mutual operation algorithm to identify the geological feature contours of the geoscience detection images belonging to the first category of images, and use the SIFT algorithm to identify the geological feature contours of the geoscience detection images belonging to the second category of images;

[0061] A5. Obtain comprehensive characteristic statistical information of the target survey area based on the geological feature contours of each geological detection image.

[0062] The method first divides the geological detection images into a first type of images with relatively low noise content and a second type of images with relatively high noise content, and then uses the image horizontal and vertical dimension mutual operation algorithm to identify the geological feature contours of the first type of images to improve the recognition efficiency of the first type of images, and uses the SIFT algorithm to identify the geological feature contours of the second type of images to ensure the recognition accuracy of the second type of images; thereby, the overall recognition efficiency of the geological feature contour recognition of multiple geological detection images can be effectively improved while ensuring the overall accuracy of the geological feature contour recognition.

[0063] The geoscientific detection images are survey images of multiple cross sections of the strata in the target survey area, and the acquisition method thereof is a prior art and will not be described in detail herein. Generally, the geoscientific detection images are arranged in sequence according to the scanning direction during the survey.

[0064] Among them, the noise content recognition model can adopt the existing deep learning model. During training, multiple training images (geological detection images) can be collected and the noise content of the training images can be marked to generate a training data set; then, the noise content of the training image is used as a reference value, and each training image is input into the noise content recognition model in turn, and the loss function is calculated according to the output of the noise content recognition model and the reference value of the corresponding training image, and the model parameters of the noise content recognition model are updated according to the loss function until the loss function converges to obtain a pre-trained noise content recognition model.

[0065] In fact, the noise types in geological detection images are relatively simple. Using the noise content recognition model to identify the noise content of geological detection images can determine a relatively accurate recognition result.

[0066] Specifically, step A3 includes:

[0067] A301. Obtain the recognition accuracy threshold and the average processing time threshold of a single geoscientific detection image;

[0068] A302. Determine the target noise content cutoff value according to the recognition accuracy threshold and the average processing time threshold;

[0069] A303. According to the target noise content cutoff value and the first noise content, the geoscientific detection image is divided into a first type of image and a second type of image.

[0070] The user can set the recognition accuracy threshold and the average processing time threshold according to actual needs. Image segmentation in the above manner is conducive to ensuring that the final recognition accuracy is not lower than the recognition accuracy threshold and the average processing time does not exceed the average processing time threshold.

[0071] In step A301, the recognition accuracy threshold and the average processing time threshold can be set by the user according to actual needs, or the system default values ​​can be used.

[0072] Wherein, step A302 includes:

[0073] Calling a query table, the query table records multiple noise content cutoff values ​​and corresponding recognition accuracy rates and corresponding average processing times;

[0074] Selecting a noise content cutoff value whose recognition accuracy is not less than a recognition accuracy threshold from the query table as a candidate noise content cutoff value;

[0075] From the candidate noise content cutoff values ​​whose corresponding average processing times are not higher than the average processing time threshold, the candidate noise content cutoff value with the shortest average processing time is selected as the target noise content cutoff value.

[0076] The target noise content cutoff value can be quickly determined in the above manner to ensure that the final recognition accuracy is not lower than the recognition accuracy threshold and the average processing time does not exceed the average processing time threshold.

[0077] Furthermore, the query table can be generated in the following manner (that is, before step A2, it also includes the steps):

[0078] B1. Obtain multiple historical geoscientific detection images with known noise content (i.e., geoscientific detection images collected in history);

[0079] B2. taking the noise content of each historical geological detection image as the second noise content, and dividing the historical geological detection image into the first type of image and the second type of image based on a plurality of different noise content cutoff values ​​according to the second noise content, and obtaining a division result corresponding to each noise content cutoff value;

[0080] B3. For each division result, the image horizontal and vertical mutual operation algorithm is used to identify the geological feature contours of the historical geological detection images belonging to the first category of images, and the SIFT algorithm is used to identify the geological feature contours of the historical geological detection images belonging to the second category of images (refer to step A4 for the specific process);

[0081] B4. For each division result, the recognition accuracy of geological feature contour recognition and the average processing time of a single historical geological detection image are statistically analyzed to form a query table containing the mapping relationship between the noise content cutoff value and the recognition accuracy and average processing time.

[0082] Among them, multiple different noise content cutoff values ​​can be set according to actual needs, such as 50%, 25%, 12.5%, ..., 50% / 2 v , where v is a preset positive integer, but is not limited thereto.

[0083] In step B2, for each noise content threshold value H, the historical geological detection images whose second noise content is not higher than H are classified into the first category of images, and the historical geological detection images whose second noise content is higher than H are classified into the second category of images. Thus, each noise content threshold value H corresponds to a classification result (each classification result includes a group of first category images and a group of second category images).

[0084] In step B4, the similarity between the contour recognition result of each geological feature of the historical geological detection image and the actual contour can be calculated, and then the average value of the similarity of all geological features is calculated as the recognition accuracy of geological feature contour recognition; but it is not limited to this.

[0085] In step B4, the processing time for identifying geological feature contours of each historical geological detection image may be obtained, and then the average processing time of each historical geological detection image may be calculated as the average processing time of a single historical geological detection image.

[0086] Wherein, step A303 includes:

[0087] Classify the geoscientific detection images whose first noise content is not higher than the target noise content cutoff value into the first category of images;

[0088] The geoscientific detection images whose first noise content is higher than the target noise content cutoff value are classified into the second type of images.

[0089] Specifically, in step A4, the image horizontal and vertical dimension mutual operation algorithm is used to identify the geological feature contours of the geoscience detection image belonging to the first type of image, including:

[0090] C1. Generate a pixel data value matrix of the geoscientific detection image according to the pixel data value of each pixel of the geoscientific detection image;

[0091] C2. performing a subtraction operation on two adjacent column vectors in the pixel data value matrix to obtain a plurality of difference column vectors to form a first difference matrix; the matrix size of the first difference matrix is ​​the same as the matrix size of the pixel data value matrix;

[0092] C3. Obtain pixel coordinates of non-zero elements in the first difference matrix to obtain a first pixel coordinate set;

[0093] C4. performing a subtraction operation on two adjacent row vectors in the pixel data value matrix to obtain a plurality of difference row vectors to form a second difference matrix; the matrix size of the second difference matrix is ​​the same as the matrix size of the pixel data value matrix;

[0094] C5. Obtain pixel coordinates of non-zero elements in the second difference matrix to obtain a second pixel coordinate set;

[0095] C6. Obtain the union of the first pixel coordinate set and the second pixel coordinate set, and set the elements in the pixel data value matrix whose pixel coordinates are not in the union to zero, to obtain a feature matrix;

[0096] C7. Generate a geological feature contour image according to the feature matrix; the geological feature contour image includes the geological feature contours of each geological feature.

[0097] To identify geological feature contours in the above manner, only simple vector subtraction operations are required, which results in low computational effort and high processing efficiency. Since the noise content of the first type of image is low, although the calculation process of the above recognition algorithm is simple, the recognition accuracy rate that meets the requirements can still be achieved for the first type of image.

[0098] For example, in step C1, the pixel data value matrix can be recorded as:

[0099] ;

[0100] in, is the pixel data value matrix, is the pixel data value of the pixel at the mth row and nth column (i.e., the pixel with pixel coordinates (m, n)), m is the number of rows in the pixel data value matrix, and n is the number of columns in the pixel data value matrix. The pixel data value may be a brightness value or a grayscale value, etc.

[0101] In step C2, each difference column vector can be calculated using the following formula to form a first difference matrix:

[0102] ;

[0103] in, is the jth difference column vector in the first difference matrix, is the j-1th difference column vector in the first difference matrix, is the j+1th column vector in the pixel data value matrix (i.e., the pixel data value matrix ), a vector consisting of the data in the j+1th column of is the jth column vector in the pixel data value matrix (i.e., the pixel data value matrix Thus, the first difference matrix It can be expressed as: , is the column vector of differences from the 1st to the nth.

[0104] In which, it is assumed that the element in the i-th row and j-th column of the first difference matrix Not zero (i.e. is a non-zero element), then The pixel coordinates are (i, j).

[0105] In step C4, each difference row vector can be calculated using the following formula to form a second difference matrix:

[0106] ;

[0107] in, is the i-th difference row vector in the second difference matrix, is the i-1th difference row vector in the second difference matrix, is the i+1th row vector in the pixel data value matrix (i.e., the pixel data value matrix ), a vector consisting of the i+1th row of data in is the i-th row vector in the pixel data value matrix (i.e., the pixel data value matrix Thus, the second difference matrix It can be expressed as: , is the row vector of differences from the 1st to the mth value.

[0108] In which, it is assumed that the element in the i-th row and j-th column of the second difference matrix Not zero (i.e. is a non-zero element), then The pixel coordinates are (i, j).

[0109] In step C7, a first image with a pixel size of m*n can be generated, and the pixel data values ​​of the pixels in the first image whose pixel coordinates are the same as the pixel coordinates of the zero elements in the feature matrix are set to the first preset value, and the pixel data values ​​of the pixels in the first image whose pixel coordinates are the same as the pixel coordinates of the non-zero elements in the feature matrix are set to the second preset value, and finally the first image is used as the geological feature contour image. The first preset value and the second preset value are different, and can be set according to actual needs.

[0110] Alternatively, in step C7, the geological detection image can be copied to obtain a second image, and the pixel data values ​​of the pixels in the second image whose pixel coordinates are the same as the pixel coordinates of the zero element in the feature matrix are set to zero, and finally the second image is used as the geological feature contour image.

[0111] Among them, the SIFT algorithm is an existing technology and will not be described in detail here.

[0112] Specifically, step A5 includes:

[0113] A501. Identify the feature type of the corresponding geological feature according to the geological feature contours of each geological detection image;

[0114] A502. According to the geological feature contours of each geological detection image, obtain the characteristic values ​​of the corresponding geological features;

[0115] A503. According to the feature type and feature value of each geological feature, the comprehensive feature value of each geological detection image is calculated as the comprehensive feature statistical information of the target survey area.

[0116] Among them, in step A501, the minimum circumscribed rectangular frame of the geological feature contour of each geological feature can be generated as a marking frame in the geological feature contour image corresponding to each geological detection image, and then each marking frame can be copied to the corresponding geological detection image to obtain the segmentation frame of each geological feature (the position and posture of the segmentation frame in the geological detection image is the same as the position and posture of the corresponding marking frame in the corresponding geological feature contour image, and the position and posture include position and posture angle), and the image in the segmentation frame is extracted for image recognition to identify the feature type of the corresponding geological feature (for example, through a deep learning model for recognition, which is a prior art and will not be described in detail here).

[0117] The characteristic value of the geological feature may include perimeter and / or area. It is a prior art to obtain the area or perimeter of a corresponding object according to the outline of the object in an image, which will not be described in detail here.

[0118] In step A503, the comprehensive feature value of each geography detection image is calculated according to the following formula:

[0119] ;

[0120] in, is the comprehensive feature value of the i-th geoscientific detection image, is the characteristic value of the 1st to Nth geological features in the i-th geoscience detection image, is the attention coefficient of geological features from Category 1 to Category N. It can be set according to actual needs. For geological feature types that are not of concern, the corresponding attention coefficient can be set to 0, and for geological feature types that are of concern, the corresponding attention coefficient can be set to a larger value. The comprehensive feature value comprehensively reflects the amount of geological structure feature information of concern contained in the corresponding geological detection image, which can be used as a basis for judging whether the target survey area has development value.

[0121] Generally, since the geological detection images of each region are sequentially arranged according to the scanning direction during the survey, in step A503, after the comprehensive characteristic values ​​of the geological detection images of each region are calculated, the comprehensive characteristic values ​​can be arranged according to the order of the corresponding geological detection images to form a comprehensive characteristic value vector, and the comprehensive characteristic value vector is used as the comprehensive characteristic statistical information of the target survey area. Thus, the comprehensive characteristic value vector not only reflects the amount of geological structure characteristic information noted in the target survey area, but also reflects the change of the geological structure characteristic information noted along the scanning direction, so as to analyze whether the target survey area has development value according to the comprehensive characteristic value vector.

[0122] As can be seen from the above, the geological feature recognition method of the geological detection image obtains multiple geological detection images of the target survey area; identifies the noise content of each geological detection image based on a pre-trained noise content recognition model, which is recorded as the first noise content; divides the geological detection images into a first type of image and a second type of image according to the first noise content; the noise content of the first type of image is lower than the noise content of the second type of image; adopts the image horizontal and vertical mutual operation algorithm to identify the geological feature contours of the geological detection images belonging to the first type of image, and adopts the SIFT algorithm to identify the geological feature contours of the geological detection images belonging to the second type of image; obtains the comprehensive feature statistical information of the target survey area according to the geological feature contours of each geological detection image; thereby, it is possible to effectively improve the overall recognition efficiency of the geological feature contour recognition of multiple geological detection images under the premise of ensuring the overall accuracy of the geological feature contour recognition.

[0123] refer to Figure 2 The present application provides a geological feature recognition device for a geological detection image, comprising:

[0124] Acquisition module 1, used to acquire multiple geoscientific detection images of the target survey area (refer to step A1 above for details);

[0125] Noise recognition module 2, used to identify the noise content of each geodesic detection image based on a pre-trained noise content recognition model, recorded as the first noise content (refer to step A2 above for details);

[0126] A division module 3 is used to divide the geoscientific detection image into a first type of image and a second type of image according to the first noise content; the noise content of the first type of image is lower than the noise content of the second type of image (refer to the above step A3 for details);

[0127] Contour recognition module 4, used to use the image horizontal and vertical dimension mutual operation algorithm to perform geological feature contour recognition on the geoscience detection image belonging to the first category of images, and use the SIFT algorithm to perform geological feature contour recognition on the geoscience detection image belonging to the second category of images (refer to step A4 above for details);

[0128] The statistical module 5 is used to obtain comprehensive characteristic statistical information of the target survey area according to the geological feature contours of each geological detection image (refer to step A5 above for details).

[0129] Please refer to Figure 3 , Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The present application provides an electronic device, including: a processor 301 and a memory 302. The processor 301 and the memory 302 are interconnected and communicate with each other through a communication bus 303 and / or other forms of connection mechanisms (not shown). The memory 302 stores a computer program executable by the processor 301. When the electronic device is running, the processor 301 executes the computer program to execute the geological feature recognition method of the geological detection image in any optional implementation of the above embodiment to achieve the following functions: obtaining multiple geological detection images of the target survey area; identifying the noise content of each geological detection image based on a pre-trained noise content recognition model, recorded as a first noise content; dividing the geological detection images into a first category of images and a second category of images according to the first noise content; the noise content of the first category of images is lower than the noise content of the second category of images; using an image horizontal and vertical dimension mutual operation algorithm to identify the geological feature contours of the geological detection images belonging to the first category of images, and using a SIFT algorithm to identify the geological feature contours of the geological detection images belonging to the second category of images; and obtaining comprehensive feature statistical information of the target survey area according to the geological feature contours of each geological detection image.

[0130] An embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for identifying geological features of a geological detection image in any optional implementation of the above-mentioned embodiment is executed to achieve the following functions: acquiring multiple geological detection images of a target survey area; identifying the noise content of each geological detection image based on a pre-trained noise content recognition model, recorded as a first noise content; dividing the geological detection images into a first category of images and a second category of images according to the first noise content; the noise content of the first category of images is lower than the noise content of the second category of images; using an image horizontal and vertical dimension mutual operation algorithm to identify the geological feature contours of the geological detection images belonging to the first category of images, and using a SIFT algorithm to identify the geological feature contours of the geological detection images belonging to the second category of images; and obtaining comprehensive feature statistical information of the target survey area according to the geological feature contours of each geological detection image.

[0131] Among them, the computer-readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable red-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0132] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0133] In addition, the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0134] Furthermore, the functional modules in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

[0135] In this document, relational terms such as first and second, etc. are used merely to distinguish one entity or operation from another entity or operation, but do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0136] The above description is only an embodiment of the present application and is not intended to limit the protection scope of the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for identifying geological features of a geological detection image, characterized in that: Includes steps: A1. Acquire multiple geoscientific detection images of the target survey area; A2. Identify the noise content of each of the geoscientific detection images based on the pre-trained noise content recognition model, recorded as the first noise content; A3. Dividing the geospatial detection images into a first type of image and a second type of image according to the first noise content; the noise content of the first type of image is lower than the noise content of the second type of image; A4. Using the image horizontal and vertical mutual operation algorithm to identify the geological feature contours of the geological detection images belonging to the first category of images, and using the SIFT algorithm to identify the geological feature contours of the geological detection images belonging to the second category of images; A5. Obtaining comprehensive characteristic statistical information of the target survey area according to the geological feature contours of each of the geoscience detection images; In step A4, the image horizontal and vertical dimension mutual operation algorithm is used to identify the geological feature contour of the geoscience detection image belonging to the first type of image, including: C1. generating a pixel data value matrix of the geoscientific detection image according to the pixel data values ​​of each pixel of the geoscientific detection image; C2. performing a subtraction operation on two adjacent column vectors in the pixel data value matrix to obtain a plurality of difference column vectors to form a first difference matrix; the matrix size of the first difference matrix is ​​the same as the matrix size of the pixel data value matrix; C3. Obtain pixel coordinates of non-zero elements in the first difference matrix to obtain a first pixel coordinate set; C4. performing a subtraction operation on two adjacent row vectors in the pixel data value matrix to obtain a plurality of difference row vectors to form a second difference matrix; the matrix size of the second difference matrix is ​​the same as the matrix size of the pixel data value matrix; C5. Obtain pixel coordinates of non-zero elements in the second difference matrix to obtain a second pixel coordinate set; C6. Obtaining a union of the first pixel coordinate set and the second pixel coordinate set, and setting the elements in the pixel data value matrix whose pixel coordinates are not in the union to zero, to obtain a feature matrix; C7. Generate a geological feature contour image according to the feature matrix; the geological feature contour image includes the geological feature contours of each geological feature.

2. The method for identifying geological features of a geological detection image according to claim 1, characterized in that: Step A3 includes: A301. Obtaining the recognition accuracy threshold and the average processing time threshold of a single geoscientific detection image; A302. Determine the target noise content cutoff value according to the recognition accuracy threshold and the average processing time threshold; A303. Divide the geoscientific detection image into a first category of images and a second category of images according to the target noise content cutoff value and the first noise content.

3. The method for identifying geological features of geological detection images according to claim 2, characterized in that: Step A302 includes: Calling a query table, wherein the query table records a plurality of noise content cutoff values ​​and corresponding recognition accuracy rates and corresponding average processing times; Filter the noise content cutoff value whose recognition accuracy is not lower than the recognition accuracy threshold from the query table as a candidate noise content cutoff value; From the candidate noise content cutoff values ​​whose corresponding average processing time is not higher than the average processing time threshold, the candidate noise content cutoff value with the shortest average processing time is selected as the target noise content cutoff value.

4. The method for identifying geological features of geological detection images according to claim 2, characterized in that: Step A303 includes: Classify the geoscientific detection image whose first noise content is not higher than the target noise content cutoff value into the first type of image; The geoscientific detection image whose first noise content is higher than the target noise content cutoff value is classified into the second category of images.

5. The method for identifying geological features of geological detection images according to claim 1, characterized in that: Step A5 includes: A501. According to the geological feature contours of each of the geoscience detection images, identifying the feature type of the corresponding geological feature; A502. According to the geological feature contours of each of the geological detection images, obtain the characteristic values ​​of the corresponding geological features; A503. Calculate the comprehensive feature value of each of the geological detection images according to the feature type and the feature value of each of the geological features as the comprehensive feature statistical information of the target survey area.

6. The method for identifying geological features of geological detection images according to claim 5, characterized in that: In step A503, the comprehensive characteristic value of each of the geoscientific detection images is calculated according to the following formula: ； in, is the comprehensive feature value of the i-th geoscientific detection image, are the feature values ​​of the 1st to Nth geological features in the i-th geological detection image, is the attention coefficient of geological features from Category 1 to Category N.

7. A geological feature recognition device for geological detection images, characterized in that: include: An acquisition module, used for acquiring a plurality of geoscientific detection images of a target survey area; A noise identification module, used for identifying the noise content of each of the geoscientific detection images based on a pre-trained noise content identification model, recorded as a first noise content; A division module, used for dividing the geoscientific detection image into a first type of image and a second type of image according to the first noise content; the noise content of the first type of image is lower than the noise content of the second type of image; A contour recognition module, used to perform geological feature contour recognition on the geological detection images belonging to the first category of images by using an image horizontal and vertical dimension mutual operation algorithm, and to perform geological feature contour recognition on the geological detection images belonging to the second category of images by using a SIFT algorithm; A statistical module, used for obtaining comprehensive characteristic statistical information of the target survey area according to the geological characteristic contours of each of the geological detection images; The image horizontal and vertical mutual operation algorithm is used to identify the geological feature contour of the geoscience detection image belonging to the first type of image, including: C1. generating a pixel data value matrix of the geoscientific detection image according to the pixel data values ​​of each pixel of the geoscientific detection image; C2. performing a subtraction operation on two adjacent column vectors in the pixel data value matrix to obtain a plurality of difference column vectors to form a first difference matrix; the matrix size of the first difference matrix is ​​the same as the matrix size of the pixel data value matrix; C3. Obtain pixel coordinates of non-zero elements in the first difference matrix to obtain a first pixel coordinate set; C4. performing a subtraction operation on two adjacent row vectors in the pixel data value matrix to obtain a plurality of difference row vectors to form a second difference matrix; the matrix size of the second difference matrix is ​​the same as the matrix size of the pixel data value matrix; C5. Obtain pixel coordinates of non-zero elements in the second difference matrix to obtain a second pixel coordinate set; C6. Obtaining a union of the first pixel coordinate set and the second pixel coordinate set, and setting the elements in the pixel data value matrix whose pixel coordinates are not in the union to zero, to obtain a feature matrix; C7. Generate a geological feature contour image according to the feature matrix; the geological feature contour image includes the geological feature contours of each geological feature.

8. An electronic device, characterized in that: It comprises a processor and a memory, wherein the memory stores a computer program executable by the processor, and when the processor executes the computer program, it runs the steps in the method for identifying geological features of geological detection images as described in any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for identifying geological features of a geoscientific detection image are performed as described in any one of claims 1 to 6.

Images