A Method for Classifying and Identifying Formation in Logging-While-Drilling Images under Complex Drilling Scenarios

Abstract

This application relates to the technical field of logging image formation classification and recognition, and specifically relates to a method for classifying and recognizing formation of logging-while-drilling images in complex drilling scenarios. The method includes: collecting electrical imaging logging images of different formations, and extracting the foreground images of each electrical imaging logging image, denoted as logging images; performing denoising processing on the logging images according to the distribution characteristics of pixel points in the logging images to obtain denoised logging images; using an image enhancement algorithm to perform enhancement processing on all denoised logging images to obtain a deep learning model for classifying and recognizing the formation of logging-while-drilling images. This application analyzes the distribution characteristics of pixel values of pixel points in electrical imaging logging images, performs more accurate denoising processing on the electrical imaging logging images, and improves the accuracy of the deep learning model obtained by neural network training for classifying and recognizing the formation to which the logging images belong.

Description

Technical Field

[0001] This application relates to the technical field of logging image formation classification and identification, and specifically relates to a method for classifying and identifying formation of logging images while drilling in complex drilling scenarios. Background Art

[0002] Imaging logging while drilling is the main means in the process of geophysical exploration. It can send the logging images collected during the drilling process to the ground for processing in real time. Compared with conventional logging curves, it has the advantages of higher resolution and visualization. Among them, the electrical imaging logging image is one of the most widely used in all imaging logging while drilling methods, which can achieve more accurate and efficient detection of geological information and improve the exploration and development level of complex oil and gas reservoirs. In the prior art, convolutional neural networks are usually used to classify and identify the formation of electrical imaging logging images while drilling. However, in the actual acquisition process of electrical imaging logging images, the electrical imaging logging instrument will be affected by collisions with the wellbore wall, instrument electronic components and circuits, etc., resulting in different degrees of noise in the electrical imaging logging images. Therefore, in order to reduce the influence of the noise in the electrical imaging logging images on the subsequent formation classification and identification of logging images, the collected electrical imaging logging images are usually preprocessed.

[0003] During the preprocessing process, when using the electrical imaging logging images of carbonate rock formations with complex non-reservoirs to construct their formation classification and identification models, due to the similarity of the image distribution characteristics between the dense layers in the carbonate rock formations and the dot noise in the electrical imaging logging images being ignored by the existing preprocessing methods of electrical imaging logging images, and at the same time, the image halo generated by the current focusing effect during the acquisition process of electrical imaging logging images, which is an imaginary information that does not exist in the actual core or wellbore wall, is not considered. This will easily cause the preprocessed electrical imaging logging images to lose a lot of geological information of the dense layers and retain a lot of imaginary information without geological significance, reducing the accuracy of the deep learning model trained by the neural network for classifying and identifying the formation to which the logging images belong. Summary of the Invention

[0004] In order to solve the above technical problems, this application provides a method for classifying and identifying formation of logging images while drilling in complex drilling scenarios to solve the existing problems.

[0005] A method for classifying and identifying formation of logging images while drilling in complex drilling scenarios of this application adopts the following technical solutions:

[0006] An embodiment of this application provides a method for classifying and identifying formation of logging images while drilling in complex drilling scenarios, and this method includes the following steps:

[0007] S1: Collect electrical imaging logging images of different formations, and extract the foreground images of each electrical imaging logging image, denoted as logging images;

[0008] S2: Denoise the logging images according to the distribution characteristics of the pixel points in the logging images. Specifically:

[0009] S201: Use the connected component extraction algorithm and the convex hull algorithm to extract all convex hulls in each logging image, count the number of all pixel points in each convex hull, and obtain all characteristic convex hulls in each logging image;

[0010] S202: Extract the central pixel point of each characteristic convex hull, and determine the density concentration degree of each characteristic convex hull by analyzing the concentrated distribution of the central pixel points of all characteristic convex hulls;

[0011] S203: Construct a window centered on the central pixel point of each characteristic convex hull, determine the characteristic distribution value of each characteristic convex hull by analyzing the distribution of all pixel points in the window, combine the density concentration degree, determine the density characteristic value of each characteristic convex hull, and screen out the dense layer convex hulls from all characteristic convex hulls of each logging image, and denoise each logging image to obtain the denoised logging image;

[0012] S3: Use the image enhancement algorithm to enhance all denoised logging images to obtain a deep learning model for classifying and identifying the formation of the logging-while-drilling images.

[0013] Preferably, the method for obtaining all characteristic convex hulls in each logging image is:

[0014] Take the number of all pixel points in each convex hull as the area of each convex hull;

[0015] Take the areas of all convex hulls in each logging image as the input of the threshold segmentation algorithm, output the segmentation threshold, denoted as the area threshold, and take all convex hulls with an area smaller than the area threshold as the characteristic convex hulls of each logging image.

[0016] Preferably, the extraction of the central pixel point of each characteristic convex hull includes:

[0017] Obtain the minimum bounding rectangle of each characteristic convex hull, and take the pixel point corresponding to the center of the minimum bounding rectangle as the central pixel point of each characteristic convex hull.

[0018] Preferably, the method for determining the density concentration degree of each characteristic convex hull is:

[0019] Take the central pixel points of all characteristic convex hulls in each logging image as the input of the density clustering algorithm, and output the local density of each central pixel point in all characteristic convex hulls as the density concentration degree of each characteristic convex hull.

[0020] Preferably, the method for determining the characteristic distribution value of each characteristic convex hull is:

[0021] Take the pixel values of all pixel points within the window of each feature convex hull as the input of the nearest distance algorithm, and output the empirical distribution function of each feature convex hull; construct the uniform distribution function of each feature convex hull according to the pixel values of all pixel points within the window of each feature convex hull;

[0022] Use the one-sample test algorithm to calculate the statistic between the empirical distribution function and the uniform distribution function of each feature convex hull as the feature distribution value of each feature convex hull.

[0023] Preferably, the density eigenvalue of each feature convex hull is the ratio between the density concentration degree of each feature convex hull and the feature distribution value.

[0024] Preferably, the process of screening out the tight layer convex hulls from all feature convex hulls of each logging image is as follows:

[0025] Take the density eigenvalues of all feature convex hulls in each logging image as the input of the threshold segmentation algorithm, output the segmentation threshold, denoted as the eigenvalue threshold, and take the feature convex hulls with density eigenvalues greater than the eigenvalue threshold as the tight layer convex hulls of each logging image.

[0026] Preferably, the process of denoising each logging image to obtain the denoised logging image is as follows:

[0027] In each logging image, take the pixel points in all convex hulls except the tight layer convex hulls as the input of the filtering algorithm, and output the denoised logging image;

[0028] Obtain the background image of each resistivity image logging, and merge the background image with the corresponding denoised logging image to obtain the denoised logging image.

[0029] Preferably, the image enhancement algorithm is the fast Retinex image enhancement algorithm.

[0030] Preferably, obtaining the deep learning model for classifying and identifying the formation of the logging-while-drilling image includes:

[0031] Use the open source power to label and classify all enhanced denoised logging images according to the geological categories of the carbonate formations they belong to, take the labeled denoised logging images as the input of the neural network, and output the deep learning model.

[0032] This application has at least the following beneficial effects:

[0033] In this application, by first screening out the feature convex hulls with smaller areas, the geological features in the electrical imaging logging images can be effectively distinguished from the noise. Further, by analyzing the distribution density of the feature convex hulls in the logging images, the density concentration degree is constructed, which helps to evaluate the significance of the geological features, thereby identifying the feature convex hulls representing the tight layers. Further, taking the central pixel points of each feature convex hull as the centers to construct windows, by analyzing the distribution of all pixel points within the windows and combining the density concentration degree, the density feature values are constructed, which can more accurately screen out the convex hulls of the tight layers, so as to retain important geological information during the denoising process. By comprehensively evaluating the density concentration degree and the feature distribution values of the feature convex hulls, this application accurately performs denoising and image enhancement processing on the electrical imaging logging images, reduces the misidentification and missed identification of geological features, enhances the adaptability and robustness of the deep learning model to complex drilling scenarios, and thus improves the accuracy of the deep learning model trained by the neural network for classifying and identifying the formation to which the logging images belong. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the accompanying drawings required for use in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0035] Figure 1 It is a flowchart of the steps of a method for classifying and identifying the formation of the logging-while-drilling images in a complex drilling scenario provided by an embodiment of the present application;

[0036] Figure 2 It is a schematic diagram of the denoising process of the logging image provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0037] In order to further elaborate on the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following, in combination with the accompanying drawings and preferred embodiments, details the specific implementation manners, structures, features and effects of a method for classifying and identifying the formation of the logging-while-drilling images in a complex drilling scenario proposed according to the present application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0038] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which this application belongs.

[0039] The following specifically describes the specific solution of a method for classifying and identifying formation images while drilling in a complex drilling scenario provided by the present application in conjunction with the accompanying drawings.

[0040] A method for classifying and identifying formation images while drilling in a complex drilling scenario provided by an embodiment of the present application. Specifically, a method for classifying and identifying formation images while drilling in a complex drilling scenario is provided as follows. Please refer to Figure 1 , the method includes the following steps:

[0041] S1: Collect the electrical image logging images of different formations, and extract the foreground images of each electrical image logging image, denoted as logging images.

[0042] Collect electrical image logging images with different geological characteristics of carbonate formations. A total of N electrical image logging images are collected, and all electrical image logging images are mapped into grayscale images. The geological characteristics of the carbonate formations mainly include seven types of characteristics: bedding, dense layer, shale streak, suture, induced fracture, pyrite, and chert nodule.

[0043] It should be noted that the value of the number N of electrical image logging images collected is set artificially. In this embodiment, the value of N is 3000. Implementers can also set it according to specific situations by themselves. This embodiment does not make special restrictions.

[0044] Since the geological characteristics with low resistivity are manifested as low-brightness characteristics in the electrical image logging images, and the dense layer in the carbonate formation has a high resistivity value in the conventional logging data, the dense layer has a high grayscale value and shows a high-brightness characteristic in the electrical image logging images.

[0045] Therefore, each electrical image logging image is used as the input of the threshold segmentation algorithm, and the foreground image of each electrical image logging image is output, which is used to characterize the image area with a high resistivity value in the electrical image logging image.

[0046] It should be noted that there are many commonly used threshold segmentation algorithms. In this embodiment, the Otsu threshold segmentation algorithm is used to extract the foreground image of the electrical image logging image. In the actual application process, as other implementation manners, implementers can also use other methods. Regarding the selection of the threshold segmentation method, this embodiment does not make special restrictions.

[0047] In addition, it should be supplemented that in this embodiment, whenever the threshold segmentation algorithm is involved, the Otsu threshold segmentation algorithm is used.

[0048] Among them, regarding the selection of the Otsu threshold segmentation algorithm, its specific principle will not be elaborated here.

[0049] S2: Denoise the logging images according to the distribution characteristics of the pixel points in the logging images.

[0050] In the prior art, median filtering is used to suppress the dot noise in the electrical imaging logging image. However, the dense layer in the carbonate formation (the dense lithology formation in the formation) usually appears as a densely packed and evenly arranged dot segment in the electrical imaging logging image. Therefore, directly using the median filtering algorithm to denoise the electrical imaging logging image easily causes the dot geological details belonging to the dense layer in the electrical imaging logging image to be misrecognized as dot noise and filtered out.

[0051] Therefore, to avoid the situation that the dot geological details belonging to the dense layer in the electrical imaging logging image are misrecognized as dot noise and filtered out during the image denoising process, the following processing is carried out, and the specific process is as follows:

[0052] S201: Use the connected component extraction algorithm and the convex hull algorithm to extract all convex hulls in each logging image, count the number of all pixel points in each convex hull, and obtain all characteristic convex hulls in each logging image.

[0053] Use the connected component extraction algorithm to extract all connected components in each logging image, and use the convex hull algorithm to obtain the convex hull of each connected component in the foreground image respectively, to obtain all convex hulls in the foreground image of each electrical imaging logging image, so as to ensure all image regions with relatively high resistivity values in the electrical imaging logging image.

[0054] It should be understood that there are many common connected component extraction algorithms. In this embodiment, the region growing algorithm is used to obtain the connected components in the foreground image. In the actual application process, as other implementation manners, the implementer can also use other methods such as morphological operations. Regarding the selection of the connected component extraction algorithm, no special limitation is made in this embodiment.

[0055] It should be noted that there are many common convex hull algorithms. In this embodiment, the Andrew algorithm is used to obtain the convex hull in the foreground image. In the actual application process, as other implementation manners, the implementer can also use other inclusion algorithms such as the Jarvis algorithm and the Graham algorithm according to the specific situation. Regarding the selection of the convex hull algorithm, no special limitation is made in this embodiment.

[0056] Among them, both the region growing algorithm and the Andrew algorithm are well-known technologies, and the process of using the region growing algorithm to extract connected components and the specific process of using the Andrew algorithm to obtain the convex hull of the connected components will not be elaborated here.

[0057] Since the image area corresponding to the dot-like geological features of the dense layer in the carbonate formation in the electrical imaging logging image and the dot-like noise in the electrical imaging logging image have a smaller area compared to the remaining geological feature areas in the electrical imaging logging image. Therefore, the number of all pixel points within each convex hull in each logging image is counted respectively as the area of each convex hull, and the areas of all convex hulls are used as the input of the threshold segmentation algorithm, and the output segmentation threshold is denoted as the area threshold. All convex hulls with an area smaller than the area threshold are used as the characteristic convex hulls of each logging image.

[0058] S202: Extract the central pixel points of each characteristic convex hull, and determine the density concentration degree of each characteristic convex hull by analyzing the concentrated distribution of the central pixel points of all characteristic convex hulls.

[0059] Since the dense layer in the carbonate formation appears as a densely packed and evenly arranged dot-like layer segment in the electrical imaging logging image, while the dot-like noise in the electrical imaging logging image usually does not have this characteristic and is generally randomly distributed, the convex hulls of the geological feature areas belonging to the dense layer in the electrical imaging logging image have a dense and uniform convex hull distribution in the logging image where the convex hull is located.

[0060] Therefore, the density concentration degree of each characteristic convex hull can be determined by using the density degree of the convex hull distribution within the range of each convex hull in the logging image and whether the distribution of the convex hulls within the range of each convex hull is uniform, so as to evaluate whether each convex hull is the convex hull of the geological feature area belonging to the dense layer in the electrical imaging logging image. Specifically:

[0061] Obtain the minimum bounding rectangle of each characteristic convex hull, and use the pixel point corresponding to the center of the minimum bounding rectangle as the central pixel point of each characteristic convex hull.

[0062] Among them, the method for obtaining the minimum bounding rectangle is a well-known technology, and its specific obtaining process will not be elaborated.

[0063] Take the central pixel points of all characteristic convex hulls in each logging image as the input of the density clustering algorithm, and output the local density of each central pixel point within all characteristic convex hulls as the density concentration degree of each characteristic convex hull. Among them, the distance between the central pixel points is used as the metric distance of the density clustering algorithm.

[0064] It should be noted that there are many commonly used clustering algorithms. In this embodiment, the DPC density peak clustering algorithm is used to obtain the local density of each data point. In the actual application process, the implementer can also use other clustering methods such as the DBSCAN clustering algorithm. Regarding the selection of the clustering algorithm, this embodiment does not make special restrictions.

[0065] Among them, the DPC density peak clustering algorithm is a well-known technology, and its specific principle will not be elaborated.

[0066] So far, by analyzing the distribution of the characteristic convex hulls, the density concentration degree of the characteristic convex hulls has been obtained.

[0067] S203: Construct a window centered on the central pixel point of each characteristic convex hull. By analyzing the distribution of all pixel points within the window, determine the characteristic distribution value of each characteristic convex hull. Combine the density concentration degree to determine the density characteristic value of each characteristic convex hull, so as to screen out the dense layer convex hulls from all the characteristic convex hulls of each logging image, and perform denoising processing on each logging image to obtain a denoised logging image.

[0068] (1) In each logging image, construct a square window centered on the central pixel point of each characteristic convex hull. In this embodiment, the value of the side length M of the window is 5. Implementers can also set it by themselves according to specific situations. This embodiment does not make special restrictions.

[0069] (2) Take the pixel values of all pixel points within the window of each characteristic convex hull as the input of the nearest distance algorithm, and output the empirical distribution function of each characteristic convex hull; construct the uniform distribution function of each characteristic convex hull according to the pixel values of all pixel points within the window of each characteristic convex hull;

[0070] (3) Use the one-sample Kolmogorov-Smirnov test algorithm to calculate the statistic between the empirical distribution function and the uniform distribution function of each characteristic convex hull as the characteristic distribution value of each characteristic convex hull; used to evaluate whether the distribution of the convex hull in the local image area where the convex hull is located in the electrical imaging logging image follows a uniform distribution. The larger the statistic, the less the distribution of the convex hulls in the local image area where the convex hulls are located in the electrical imaging logging image follows a uniform distribution.

[0071] Among them, the process of obtaining the empirical distribution function using the nearest distance algorithm, the calculation method of the uniform distribution function, and the one-sample Kolmogorov-Smirnov test algorithm are all well-known technologies, and their specific principle processes will not be elaborated here.

[0072] (4) Further, take the ratio between the density concentration degree and the characteristic distribution value of each characteristic convex hull as the density characteristic value of each characteristic convex hull, which is used to evaluate whether the geological characteristic area corresponding to the convex hull in the electrical imaging logging image has the distribution characteristics of a dense layer in the carbonate formation, that is, a densely packed and evenly arranged dot-like layer section.

[0073] It can be understood from the density eigenvalue of each feature convex hull that if the density concentration degree of the feature convex hull is larger and the feature distribution value is smaller, the density eigenvalue is larger, indicating that the convex hull distribution in the local image area where the convex hull is located in the electrical imaging logging image is more uniform, and the area corresponding to the convex hull in the electrical imaging logging image is more likely to be a dense layer in the carbonate formation; on the contrary, if the density concentration degree of the feature convex hull is smaller and the feature distribution value is larger, the density eigenvalue is smaller, indicating that the convex hull distribution in the local image area where the convex hull is located in the electrical imaging logging image is more uneven, and the area corresponding to the convex hull in the electrical imaging logging image is more likely to be punctate noise.

[0074] (5) Further, take the density eigenvalues of all feature convex hulls in each logging image as the input of the threshold segmentation algorithm, and output the segmentation threshold, denoted as the eigenvalue threshold. The feature convex hulls with density eigenvalues greater than the eigenvalue threshold are used as the dense layer convex hulls of each logging image; they are used to represent the convex hulls with the punctate geological features of the dense layer in the carbonate formation in the electrical imaging logging image.

[0075] (6) In this embodiment, in each logging image, take the pixel points in all convex hulls except the dense layer convex hulls as the input of the median filtering algorithm, and output the denoised logging image;

[0076] Among them, the median filtering algorithm is a well-known technology, and the specific process of using the median filtering algorithm to denoise the image will not be elaborated here.

[0077] Further, obtain the background image of each electrical imaging logging image, and merge the background image with the corresponding denoised logging image to obtain the denoised logging image, so as to suppress the punctate noise in the electrical imaging logging image, and avoid the situation that the punctate noise, which is meaningless false information, is misrecognized as geological details in the image and enhanced during the subsequent image enhancement process; at the same time, avoid the situation that the punctate texture detail information belonging to the dense layer in the image is misrecognized as punctate noise and filtered out.

[0078] It should be noted that the process of merging the logging image, that is, the foreground image, with the background image is a well-known technology, and the specific principle process will not be elaborated here.

[0079] Preferably, the schematic diagram of the logging image denoising process provided in this embodiment is as Figure 2 shown.

[0080] S3: Use the image enhancement algorithm to perform enhancement processing on all denoised logging images to obtain a deep learning model for classifying and identifying the formation of the logging-while-drilling image.

[0081] When acquiring imaging logging data using an electrical imaging logging tool, due to the current focusing effect, the current always flows towards areas with high conductivity, resulting in a white halo, a geological-meaningless false image, outside the high-conductivity regions in the electrical imaging logging image. For example, when acquiring an electrical imaging logging image of a geological layer containing pyrite, since pyrite is an excellent electrical conductor, a bright white halo will form around the area where pyrite is located in the acquired electrical imaging logging image. However, this geological-meaningless halo in the electrical imaging logging image does not exist in the actual core or wellbore wall. If this halo in the image is not suppressed, it will affect the accurate identification of different geological features in the formation by the formation classification and recognition model of the logging-while-drilling image constructed subsequently.

[0082] Therefore, the image halo suppression technology is used to enhance the electrical imaging logging image to suppress the halo area in the image and enhance the geological detail information in the image. Specifically:

[0083] In this embodiment, each denoised logging image is used as the input of the fast Retinex image enhancement algorithm, and the enhanced denoised logging image is output.

[0084] Among them, the fast Retinex image enhancement algorithm is a well-known technology, and the specific process will not be elaborated.

[0085] Furthermore, the Labelme open-source tool is used to label and classify all the enhanced denoised logging images according to the geological categories of the carbonate formations to which they belong. The labeled denoised logging images are used as the input of the convolutional neural network, and a deep learning model is output. Among them, the cross-entropy loss function and the stochastic gradient descent method are used as the loss function and the optimization algorithm of the convolutional neural network respectively. The training of the deep learning model is a well-known technology, and the specific training process will not be elaborated.

[0086] So far, based on the convex hull of the tight layer obtained from the point feature values of the tight layer in this embodiment, the median filtering algorithm is used to complete the denoising process of the electrical imaging logging image, which can avoid the situation that the point noise, a geological-meaningless interference information in the electrical imaging logging image, is misinterpreted as geological details in the image and enhanced during the subsequent image enhancement process, and at the same time avoid the situation that the point geological details of the tight layer in the carbonate formation in the electrical imaging logging image are misinterpreted as point noise and filtered out; furthermore, by using the image halo suppression technology to enhance the denoised electrical imaging logging image, the accuracy of the deep learning model trained by the neural network for classifying and recognizing the formation to which the logging image belongs.

[0087] It should be noted that: The above order of the embodiments of the present application is only for description and does not represent the superiority or inferiority of the embodiments. In addition, the specific embodiments of this specification have been described. Moreover, the processes depicted in the drawings do not necessarily require the specific order or consecutive order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0088] The various embodiments in this specification are all described in a progressive manner. For the same or similar parts between the various embodiments, reference can be made to each other, and each embodiment focuses on the differences from other embodiments.

[0089] The above-described embodiments are only used to illustrate the technical solutions of the present application and are not intended to limit them; modifying the technical solutions recorded in the foregoing embodiments or equivalently replacing some of the technical features does not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the various embodiments of the present application, and should all be included within the protection scope of the present application.

Claims

1. A method for classification and recognition of formations in logging while drilling images under complex drilling scenarios, characterized in that: The method comprises the following steps: S1: Collect electrical imaging logging images of different formations, and extract the foreground image of each electrical imaging logging image, which is recorded as the logging image; S2: According to the distribution characteristics of the pixels in the well logging image, the well logging image is denoised. Specifically: S201: using a connected domain extraction algorithm and a convex hull algorithm to extract all convex hulls in each well logging image, counting the number of all pixel points in each convex hull, and obtaining all feature convex hulls in each well logging image; S202: extracting the central pixel point of each feature convex hull, and determining the density concentration of each feature convex hull by analyzing the concentrated distribution of the central pixel points of all feature convex hulls; S203: constructing a window with the central pixel point of each characteristic convex hull as the center, determining the characteristic distribution value of each characteristic convex hull by analyzing the distribution of all pixel points in the window, and determining the density characteristic value of each characteristic convex hull in combination with the density concentration, so as to screen out the dense layer convex hull from all characteristic convex hulls of each logging image, and denoising each logging image to obtain a denoised logging image; S3: All denoised logging images are enhanced using an image enhancement algorithm to obtain a deep learning model for classifying and identifying formations in LWD images.

2. The method for classification and identification of formations in logging while drilling images in a complex drilling scenario according to claim 1, characterized in that: The method for obtaining all feature convex hulls in each logging image is as follows: The number of all pixels in each convex hull is taken as the area of ​​each convex hull; The area of ​​all convex hulls in each logging image is used as the input of the threshold segmentation algorithm, and the segmentation threshold is output, which is recorded as the area threshold. All convex hulls with an area smaller than the area threshold are used as the feature convex hull of each logging image.

3. The method for classification and identification of formations in logging while drilling images in a complex drilling scenario according to claim 1, characterized in that: The step of extracting the central pixel point of each feature convex hull comprises: Get the minimum circumscribed rectangle of each feature convex hull, and use the pixel point corresponding to the center of the minimum circumscribed rectangle as the center pixel point of each feature convex hull.

4. The method for classification and identification of formations in LWD images in complex drilling scenarios according to claim 1, characterized in that: The method for determining the density concentration of each feature convex hull is: The central pixel points of all feature convex hulls in each logging image are used as the input of the density clustering algorithm, and the local density of the central pixel points in all feature convex hulls is output as the density concentration of all feature convex hulls.

5. The method for classification and identification of formations in logging while drilling images in a complex drilling scenario according to claim 1, characterized in that: The method for determining the characteristic distribution value of each characteristic convex hull is: The pixel values ​​of all pixels in the window of each feature convex hull are used as the input of the nearest distance algorithm, and the empirical distribution function of each feature convex hull is output; the uniform distribution function of each feature convex hull is constructed according to the pixel values ​​of all pixels in the window of each feature convex hull; The single sample test algorithm is used to calculate the statistic between the empirical distribution function and the uniform distribution function of each feature convex hull as the feature distribution value of each feature convex hull.

6. The method for classification and identification of formations in logging while drilling images in a complex drilling scenario according to claim 1, characterized in that: The density characteristic value of each feature convex hull is the ratio between the density concentration and the characteristic distribution value of each feature convex hull.

7. The method for classification and identification of formations in logging while drilling images in a complex drilling scenario according to claim 1, characterized in that: The process of selecting the dense layer convex hull from all the characteristic convex hulls of each logging image is as follows: The density eigenvalues ​​of all feature convex hulls in each logging image are used as the input of the threshold segmentation algorithm, and the segmentation threshold is output, which is recorded as the eigenvalue threshold. The feature convex hull with a density eigenvalue greater than the eigenvalue threshold is used as the dense layer convex hull of each logging image.

8. The method for classification and identification of formations in logging while drilling images in a complex drilling scenario according to claim 1, characterized in that: The process of performing denoising on each well logging image to obtain a denoised well logging image is as follows: In each well logging image, all pixel points in the convex hull except the convex hull of the dense layer are used as the input of the filtering algorithm, and the denoised well logging image is output; A background image of each electrical imaging logging image is obtained, and the background image is merged with the corresponding denoised logging image to obtain a denoised logging image.

9. The method for classification and identification of formations in logging while drilling images in a complex drilling scenario according to claim 1, characterized in that: The image enhancement algorithm is a fast Retinex image enhancement algorithm.

10. The method for classification and identification of formations in logging while drilling images in a complex drilling scenario according to claim 1, characterized in that: The method of obtaining a deep learning model for classifying and identifying formations in logging while drilling images includes: All enhanced denoised logging images are labeled and classified according to the geological categories of carbonate formations using open source power. The labeled denoised logging images are used as the input of the neural network to output the deep learning model.

Images