CN114841988A - Method and device for determining actual size of rock core - Google Patents
Method and device for determining actual size of rock core Download PDFInfo
- Publication number
- CN114841988A CN114841988A CN202210578421.7A CN202210578421A CN114841988A CN 114841988 A CN114841988 A CN 114841988A CN 202210578421 A CN202210578421 A CN 202210578421A CN 114841988 A CN114841988 A CN 114841988A
- Authority
- CN
- China
- Prior art keywords
- image
- core
- size
- feature
- feature extraction
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20021—Dividing image into blocks, subimages or windows
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30132—Masonry; Concrete
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a method and a device for determining the actual size of a core, wherein the method comprises the following steps: constructing a target segmentation model, segmenting an image to be segmented based on the target segmentation model, and determining an object prediction category and an object prediction image pixel size in the image to be segmented, wherein the object prediction category comprises a rock core and a rock core box, and the object prediction image pixel size comprises the rock core prediction image pixel size and the rock core box prediction image pixel size; constructing a proportional corresponding relation between the size of the predicted image pixel of the core box and the actual size of the core box; and determining the actual size of the rock core according to the pixel size of the rock core predicted image and the proportional corresponding relation. The invention can avoid measuring the actual size of the core manually, improve the efficiency and accuracy of the determined actual size of the core and reduce the waste of human resources.
Description
Technical Field
The invention relates to the technical field of rock core size determination, in particular to a method and a device for determining the actual size of a rock core.
Background
The core is a cylindrical rock sample taken out of a hole by using a drill bit (or other coring tools) in the geological drilling coring process, can visually reflect stratum information such as stratum deposition rules, lithology, geological structure, fracture development condition, weathering information and the like, is an important basis for scientific research, production and management personnel to know the stratum information in the geological survey process, and plays an irreplaceable important role in geological survey, geological evaluation and site suitability research in the early stage of engineering construction. Geological logging of a core is one of the fundamental tasks in geological exploration.
Due to the fact that the sizes of rock cores are various, the existing rock core recording method in geological engineering mainly depends on manual measurement of shot rock cores placed in a rock core disc. It has the disadvantages that: (1) the lengths of all the rock cores need to be measured manually, the measurement workload is large, and the waste of labor resources is serious; (2) the manual measurement inevitably has measurement errors, and the accuracy of the core record data is influenced.
Therefore, a method and a device for determining the actual size of the core are urgently needed to be provided, and the technical problems that in the prior art, the size of the core is determined by manual measurement, errors exist in the measured size of the core, and human resources are wasted are solved.
Disclosure of Invention
In view of this, it is necessary to provide a method and a device for determining the actual size of a core, so as to solve the technical problems in the prior art that the actual size of the core is determined by manual measurement, which causes errors in the measured actual size of the core and waste of human resources.
In one aspect, the invention provides a method for determining the actual size of a core, which comprises the following steps:
constructing a target segmentation model, segmenting an image to be segmented based on the target segmentation model, and determining an object prediction category and an object prediction image pixel size in the image to be segmented, wherein the object prediction category comprises a rock core and a rock core box, and the object prediction image pixel size comprises the rock core prediction image pixel size and the rock core box prediction image pixel size;
acquiring the actual size of a core box, and constructing the proportional corresponding relation between the predicted image pixel size of the core box and the actual size of the core box;
and determining the actual core size of the core according to the core predicted image pixel size and the proportional corresponding relation.
In some possible implementations, the target segmentation model includes a feature extraction module, a feature fusion module, a core box segmentation module, and a core segmentation module;
the feature extraction module is used for extracting image features in the image to be segmented to obtain a plurality of feature extraction images;
the feature fusion layer is used for extracting detail features in the feature extraction images to obtain a plurality of detail feature images, and fusing the detail feature images and the extraction images to obtain a plurality of fusion feature images;
the core box segmentation module is used for determining at least two target fusion feature images from the plurality of fusion feature images and determining the core box and the core box predicted image pixel size based on the at least two target fusion feature determination images;
the core segmentation module is to determine the core and the core predicted image pixel sizes based on the plurality of fused feature images.
In some possible implementations, the feature extraction module includes a first feature extraction layer, a second feature extraction layer, a third feature extraction layer, a fourth feature extraction layer, and a fifth feature extraction layer;
the first feature extraction layer is used for extracting image features in the image to be segmented to obtain a first feature extraction image;
the second feature extraction layer is used for carrying out downsampling on the first feature extraction image to obtain a second feature extraction image;
the third feature extraction layer is used for performing downsampling on the second feature extraction image to obtain a third feature extraction image;
the fourth feature extraction layer is used for performing downsampling on the third feature extraction image to obtain a fourth feature extraction image;
the fifth feature extraction layer is configured to perform downsampling on the fourth feature extraction image to obtain a fifth feature extraction image.
In some possible implementations, the feature fusion module includes a first feature fusion layer, a second feature fusion layer, a third feature fusion layer, and a fourth feature fusion layer;
the first feature fusion layer is used for performing convolution on the fifth feature extraction image to obtain a first fusion feature image;
the second feature fusion layer is used for performing convolution on the fourth feature extraction image to obtain a first convolution image, performing up-sampling on the first fusion feature image to obtain a first up-sampling image, and fusing the first convolution image and the first up-sampling image to obtain a second fusion feature image;
the third feature fusion layer is used for convolving the third feature extraction image to obtain a second convolution image, upsampling the second fusion feature image to obtain a second upsampled image, and fusing the second convolution image and the second upsampled image to obtain a third fusion feature image;
the fourth feature fusion layer is used for convolving the second feature extraction image to obtain a third convolution image, upsampling the third fusion feature image to obtain a third upsampled image, and fusing the third convolution image and the third upsampled image to obtain a fourth fusion feature image.
In some possible implementations, the target fused feature image is the first fused feature image and the second fused feature image.
In some possible implementations, the target segmentation model further includes a partition segmentation module, the object prediction class further includes a partition, the object predicted image pixel size further includes a partition predicted image pixel size; the method for determining the actual size of the core further comprises the following steps:
determining a partition in the image to be partitioned and a partition predicted image pixel size based on the partition partitioning module;
acquiring the actual size of the partition, and constructing the proportional association relationship between the pixel size of the partition predicted image and the actual size of the partition;
determining the core verification size according to the core predicted image pixel size and the proportional correlation;
and verifying the reliability of the actual size of the rock core according to the rock core verification size.
In some possible implementations, the verifying the reliability of the actual size of the core according to the core verification size includes:
judging whether the difference value between the rock core verification size and the actual size of the rock core is smaller than a threshold difference value or not;
if the difference value between the core verification size and the actual core size is smaller than the threshold difference value, the actual core size is reliable;
and if the difference value between the core verification size and the actual core size is larger than or equal to the threshold difference value, the actual core size is unreliable.
In some possible implementations, the constructing the object segmentation model includes:
acquiring a real core box image, and performing enhancement processing on the real core box image to obtain an enhanced core box image;
marking the real core box image and the enhanced core box image to obtain a core box image sample set;
and constructing an initial segmentation model, and training the initial segmentation model based on the core box image sample set to obtain the target segmentation model.
In some possible implementations, the segmenting the image to be segmented based on the target segmentation model, and determining the pixel size of the image predicted by the object in the image to be segmented includes:
inputting the image to be segmented into the target segmentation model to obtain a plurality of core prediction frames;
and screening the plurality of core prediction frames based on a non-maximum value inhibition method to obtain a plurality of target core prediction frames, and determining the pixel size of the core prediction image according to the plurality of target core prediction frames.
In another aspect, the present invention further provides a device for determining the actual size of a core, comprising:
the image segmentation unit is used for constructing a target segmentation model, segmenting an image to be segmented based on the target segmentation model, and determining an object prediction category and an object prediction image pixel size in the image to be segmented, wherein the object prediction category comprises a rock core and a rock core box, and the object prediction image pixel size comprises the rock core prediction image pixel size and the rock core box prediction image pixel size;
the size corresponding relation construction unit is used for acquiring the actual size of the core box and constructing the proportional corresponding relation between the predicted image pixel size of the core box and the actual size of the core box;
and the actual core size determining unit is used for determining the actual core size according to the predicted image pixel size of the core and the corresponding proportion relation.
The beneficial effects of adopting the above embodiment are: according to the method for determining the actual size of the core, the core box and the predicted image pixel size of the core box in the image to be segmented and the corresponding relation of the ratio of the predicted image pixel size of the core box to the actual size of the core box are determined based on the constructed target segmentation model, so that the actual size of the core can be determined according to the predicted image pixel size of the core and the corresponding relation of the ratio. The method can avoid measuring the actual size of the core manually, improve the efficiency and accuracy of the actual size of the core, and reduce the waste of human resources.
Furthermore, the method determines the actual size of the rock core with larger size change by constructing the proportional corresponding relation between the actual size of the rock core box with fixed size and the predicted image pixel size of the rock core box, and can further improve the accuracy of the determined actual size of the rock core.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a schematic flow chart of an embodiment of a method for determining the actual size of a core provided by the present invention;
FIG. 2 is a schematic structural diagram of an embodiment of a target segmentation model provided in the present invention;
FIG. 3 is a schematic flow chart illustrating verification of reliability of actual dimensions of a core provided by the present invention;
FIG. 4 is a flowchart illustrating an embodiment of S304 of FIG. 3 according to the present invention;
FIG. 5 is a flowchart illustrating an embodiment of determining the pixel size of the core prediction image in S101 of FIG. 1 according to the present invention;
FIG. 6 is a flowchart illustrating an embodiment of constructing a target segmentation model in S101 of FIG. 1 according to the present invention;
FIG. 7 is a schematic structural view of an embodiment of the device for determining the actual size of the core provided by the present invention;
fig. 8 is a schematic structural diagram of an embodiment of an electronic device provided in the present invention.
Detailed Description
The technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. It is to be understood that the described embodiments are merely exemplary of the invention, and not restrictive of the full scope of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be understood that the schematic drawings are not necessarily to scale. The flowcharts used in this disclosure illustrate operations implemented according to some embodiments of the present invention. It should be understood that the operations of the flow diagrams may be performed out of order, and that steps without logical context may be performed in reverse order or concurrently. One skilled in the art, under the direction of this summary, may add one or more other operations to, or remove one or more operations from, the flowchart.
Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in the form of software, or in one or more hardware modules or integrated circuits, or in different networks and/or processor systems and/or microcontroller systems.
Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the invention. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is explicitly and implicitly understood by one skilled in the art that the embodiments described herein can be combined with other embodiments.
The embodiment of the invention provides a method and a device for determining the actual size of a rock core, which are respectively explained below.
Fig. 1 is a schematic flow chart of an embodiment of a method for determining an actual size of a core provided by the present invention, and as shown in fig. 1, the method for determining an actual size of a core includes:
s101, constructing a target segmentation model, segmenting an image to be segmented based on the target segmentation model, and determining an object prediction type and an object prediction image pixel size in the image to be segmented, wherein the object prediction type comprises a rock core and a rock core box, and the object prediction image pixel size comprises the rock core prediction image pixel size and the rock core box prediction image pixel size;
s102, acquiring the actual size of the core box, and constructing the proportional corresponding relation between the predicted image pixel size of the core box and the actual size of the core box;
s103, determining the actual size of the rock core according to the pixel size of the rock core predicted image and the proportion corresponding relation.
Compared with the prior art, the method for determining the actual size of the rock core provided by the embodiment of the invention determines the predicted image pixel sizes of the rock core, the rock core box and the rock core and the predicted image pixel size of the rock core box in the image to be segmented based on the constructed target segmentation model, and constructs the proportional corresponding relation between the predicted image pixel size of the rock core box and the actual size of the rock core box, so that the actual size of the rock core can be determined according to the predicted image pixel size of the rock core and the proportional corresponding relation. The method can avoid measuring the actual size of the rock core manually, improve the efficiency and accuracy of the determined actual size of the rock core, and reduce the waste of human resources.
Furthermore, the method and the device determine the actual size of the rock core with larger size change by constructing the proportional corresponding relation between the actual size of the rock core box with fixed size and the predicted image pixel size of the rock core box, and can further improve the accuracy of the determined actual size of the rock core.
Because the size of the core box is fixed, and the size of the core is variable, in order to reduce the calculation amount and improve the determination speed of the actual size of the core while ensuring the segmentation accuracy, in some embodiments of the invention, as shown in fig. 2, the target segmentation model includes a feature extraction module, a feature fusion module, a core box segmentation module and a core segmentation module;
the characteristic extraction module is used for extracting image characteristics in an image to be segmented to obtain a plurality of characteristic extraction images;
the feature fusion layer is used for extracting detail features in the feature extraction images to obtain a plurality of detail feature images, and fusing the detail feature images and the extraction images to obtain a plurality of fusion feature images;
the core box segmentation module is used for determining at least two target fusion characteristic images from the multiple fusion characteristic images and determining the core box and the core box prediction image pixel size based on the at least two target fusion characteristic images;
the core segmentation module is used for determining a core and a core predicted image pixel size based on the plurality of fused feature images.
According to the embodiment of the invention, the core box segmentation model is set to determine the predicted image pixel sizes of the core box and the core box based on the at least two target fusion feature images, and the core predicted image pixel sizes are determined based on the plurality of fusion feature images, so that the number of fusion feature images for determining the predicted image pixel sizes of the core box and the core box can be reduced, the speed for determining the predicted image pixel sizes of the core box and the core box can be increased, and the determination speed for the actual size of the core can be further increased.
In a specific embodiment of the present invention, as shown in fig. 2, the feature extraction module includes a first feature extraction layer C1, a second feature extraction layer C2, a third feature extraction layer C3, a fourth feature extraction layer C4, and a fifth feature extraction layer C5;
the first feature extraction layer C1 is configured to extract image features in an image to be segmented, to obtain a first feature extraction image;
the second feature extraction layer C2 is configured to down-sample the first feature extraction image to obtain a second feature extraction image;
the third feature extraction layer C3 is configured to down-sample the second feature extraction image to obtain a third feature extraction image;
the fourth feature extraction layer C4 is configured to perform downsampling on the third feature extraction image to obtain a fourth feature extraction image;
the fifth feature extraction layer C5 is configured to down-sample the fourth feature extraction image to obtain a fifth feature extraction image.
According to the embodiment of the invention, by arranging five feature extraction layers C1-C5, multi-scale extraction can be carried out on the image to be segmented, and a basis is provided for subsequent accurate segmentation.
In some embodiments of the present invention, as shown in FIG. 2, the feature fusion module includes a first feature fusion layer P5, a second feature fusion layer P4, a third feature fusion layer P3, and a fourth feature fusion layer P2;
the first feature fusion layer P5 is configured to convolve the fifth feature extraction image to obtain a first fusion feature image;
the second feature fusion layer P4 is configured to convolve the fourth feature extraction image to obtain a first convolved image, perform upsampling on the first fusion feature image to obtain a first upsampled image, and fuse the first convolved image and the first upsampled image to obtain a second fusion feature image;
the third feature fusion layer P3 is configured to convolve the third feature extraction image to obtain a second convolved image, perform upsampling on the second fusion feature image to obtain a second upsampled image, and fuse the second convolved image and the second upsampled image to obtain a third fusion feature image;
the fourth feature fusion layer P2 is configured to convolve the second feature extraction image to obtain a third convolved image, perform upsampling on the third fusion feature image to obtain a third upsampled image, and fuse the third convolved image and the third upsampled image to obtain a fourth fusion feature image.
According to the embodiment of the invention, the four characteristic fusion layers P2-P5 are arranged, so that the number of the obtained image characteristics can be further increased, and the accuracy of subsequent core and core box segmentation is ensured.
It should be understood that: due to the fact that the size of the rock core box is large, accurate segmentation can be achieved without special detailed features, and therefore the target fusion feature image is the first fusion feature image and the second fusion feature image. Namely: fused feature images of the P4 and P5 layers.
In order to avoid inaccuracy of the proportional correspondence construction in step S102 due to human negligence or other external factors, which may result in inaccuracy of the obtained actual size of the core, in some embodiments of the present invention, as shown in fig. 2, the target segmentation model further includes a partition segmentation module, the object prediction category further includes a partition, and the object prediction image pixel size further includes a partition prediction image pixel size; then, as shown in fig. 3, the method for determining the actual size of the core further comprises:
s301, determining a partition in an image to be partitioned and a partition predicted image pixel size based on a partition partitioning module;
s302, acquiring the actual size of the partition, and constructing a proportional association relation between the pixel size of the partition predicted image and the actual size of the partition;
s303, determining the core verification size according to the core predicted image pixel size and the proportional correlation;
and S304, verifying the reliability of the actual size of the rock core according to the rock core verification size.
Because the size of the partition plate is also fixed, the core verification size is obtained through the proportional correlation relationship between the partition plate predicted image pixel size and the actual size of the partition plate, the actual size of the core can be verified, and the reliability of the actual size of the core is ensured.
It should be understood that: the actual size of the rock core can also be obtained through the proportional correlation relationship and the predicted image pixel size of the rock core, the verification size of the rock core can be obtained through the proportional correspondence relationship, and the actual size of the rock core can be verified, which is not described in detail herein.
In an embodiment of the present invention, as shown in fig. 4, step S304 includes:
s401, judging whether the difference value between the rock core verification size and the actual rock core size is smaller than a threshold difference value;
s402, if the difference value between the rock core verification size and the actual rock core size is smaller than a threshold difference value, the actual rock core size is reliable;
and S403, if the difference value between the rock core verification size and the actual rock core size is larger than or equal to the threshold difference value, the actual rock core size is unreliable.
It should be noted that: the threshold difference may be adjusted according to actual conditions, and details are not described herein.
In some embodiments of the present invention, as shown in fig. 5, the segmenting the image to be segmented based on the target segmentation model in step S101, and determining the pixel size of the object prediction image in the image to be segmented comprises:
s501, inputting an image to be segmented into a target segmentation model to obtain a plurality of core prediction frames;
s502, screening the plurality of core prediction frames based on a Non-Maximum Suppression (NMS) method to obtain a plurality of target core prediction frames, and determining the pixel size of a core prediction image according to the plurality of target core prediction frames.
According to the invention, the plurality of core prediction frames are screened by a non-maximum value inhibition method, redundant core prediction frames can be removed, and the accuracy and speed of obtaining the pixel size of the core prediction image are further improved.
In a specific embodiment of the present invention, the non-maximum suppression method specifically includes: acquiring an overlapping area of the core prediction frame and the real core frame, determining the total area of the core prediction frame, namely the real core frame, judging whether the ratio of the overlapping area to the total area is greater than a threshold value, if so, determining that the core prediction frame is the target core prediction frame, and if not, deleting the core prediction frame.
It should be understood that: the threshold may be adjusted according to actual conditions, and is not specifically limited herein.
In some embodiments of the present invention, as shown in fig. 6, constructing the object segmentation model in step S101 includes:
s601, acquiring a real core box image, and performing enhancement processing on the real core box image to obtain an enhanced core box image;
s602, marking the real core box image and the enhanced core box image to obtain a core box image sample set;
s603, constructing an initial segmentation model, and training the initial segmentation model based on the core box image sample set to obtain a target segmentation model.
The enhancement processing in step S601 includes, but is not limited to, processing such as illumination change, texture change, rotation change, and gray scale change.
According to the embodiment of the invention, the number of sample images in the core box image sample set can be increased by performing enhancement processing on the real core box image, and all conditions such as core texture, illumination and the like which may occur in practice can be simulated, so that the generalization performance of the target segmentation model can be increased.
In an embodiment of the present invention, step S603 specifically includes: dividing the core box image sample set into a training set, a verification set and a test set according to a preset proportion, training an initial segmentation model by using the training set, optimizing network parameters of the initial segmentation model by using the verification set, and performing supervised learning on the initial segmentation model. And testing the initial segmentation model by using the test set, and evaluating the performance of the trained initial segmentation model based on preset evaluation indexes until a target segmentation model with the performance meeting the requirements of the initial segmentation model is obtained.
The preset evaluation indexes include, but are not limited to, accuracy, recall rate and intersection ratio of the initial segmentation model.
Specifically, the accuracy is:
wherein, P is the accuracy; TP is the number of core prediction boxes with the intersection ratio greater than or equal to 0.5; FP is the number of core prediction boxes with the intersection ratio less than 0.5;
the recall ratio is:
in the formula, FN is the number of actual core frames that are not detected.
In order to better implement the method for determining the actual size of the core in the embodiment of the present invention, on the basis of the method for determining the actual size of the core, correspondingly, an embodiment of the present invention further provides a device for determining the actual size of the core, as shown in fig. 7, the device 700 for determining the actual size of the core includes:
the image segmentation unit 701 is used for constructing a target segmentation model, segmenting an image to be segmented based on the target segmentation model, and determining an object prediction category and an object prediction image pixel size in the image to be segmented, wherein the object prediction category comprises a rock core and a rock core box, and the object prediction image pixel size comprises the rock core prediction image pixel size and the rock core box prediction image pixel size;
a size correspondence construction unit 702, configured to obtain an actual size of the core box, and construct a proportional correspondence between a predicted image pixel size of the core box and the actual size of the core box;
and a core actual size determining unit 703, configured to determine the core actual size according to the core predicted image pixel size and the proportional correspondence.
The apparatus 700 for determining the actual size of the core provided in the foregoing embodiment may implement the technical solutions described in the foregoing embodiments of the method for determining the actual size of the core, and the specific implementation principles of the modules or units may refer to the corresponding contents in the foregoing embodiments of the method for determining the actual size of the core, and are not described herein again.
As shown in fig. 8, the present invention also provides an electronic device 800. The electronic device 800 includes a processor 801, a memory 802, and a display 803. Fig. 8 shows only some of the components of the electronic device 800, but it is to be understood that not all of the shown components are required to be implemented, and that more or fewer components may be implemented instead.
The processor 801 may be, in some embodiments, a Central Processing Unit (CPU), microprocessor or other data Processing chip for executing program code stored in the memory 802 or Processing data, such as the core sizing method of the present invention.
In some embodiments, processor 801 may be a single server or a group of servers. The server groups may be centralized or distributed. In some embodiments, the processor 801 may be local or remote. In some embodiments, processor 801 may be implemented in a cloud platform. In an embodiment, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an intra-site, a multi-cloud, and the like, or any combination thereof.
The memory 802 may be an internal storage unit of the electronic device 800 in some embodiments, such as a hard disk or memory of the electronic device 800. The memory 802 may also be an external storage device of the electronic device 800 in other embodiments, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), etc., provided on the electronic device 800.
Further, the memory 802 may also include both internal storage units and external storage devices of the electronic device 800. The memory 802 is used for storing application software and various data installed in the electronic device 800.
The display 803 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch panel, or the like in some embodiments. The display 803 is used to display information at the electronic device 800 as well as to display a visual user interface. The components 801 and 803 of the electronic device 800 communicate with each other via a system bus.
In one embodiment, when the processor 801 executes the core physical sizing program in the memory 802, the following steps may be implemented:
constructing a target segmentation model, segmenting an image to be segmented based on the target segmentation model, and determining an object prediction category and an object prediction image pixel size in the image to be segmented, wherein the object prediction category comprises a rock core and a rock core box, and the object prediction image pixel size comprises the rock core prediction image pixel size and the rock core box prediction image pixel size;
acquiring the actual size of the core box, and constructing the proportional corresponding relation between the predicted image pixel size of the core box and the actual size of the core box;
and determining the actual size of the rock core according to the pixel size of the rock core predicted image and the proportional corresponding relation.
It should be understood that: the processor 801, when executing the core physical dimension determination program in the memory 802, may perform other functions in addition to the above functions, see in particular the description of the corresponding method embodiments above.
Further, the type of the electronic device 800 is not particularly limited in the embodiment of the present invention, and the electronic device 800 may be a portable electronic device such as a mobile phone, a tablet computer, a Personal Digital Assistant (PDA), a wearable device, and a laptop computer (laptop). Exemplary embodiments of portable electronic devices include, but are not limited to, portable electronic devices that carry an IOS, android, microsoft, or other operating system. The portable electronic device may also be other portable electronic devices such as laptop computers (laptop) with touch sensitive surfaces (e.g., touch panels) and the like. It should also be understood that in other embodiments of the present invention, the electronic device 800 may not be a portable electronic device, but may be a desktop computer having a touch-sensitive surface (e.g., a touch panel).
Accordingly, the present application also provides a computer-readable storage medium, which is used for storing a computer-readable program or instruction, and when the program or instruction is executed by a processor, the steps or functions of the method for determining the actual size of the core provided by the above method embodiments can be implemented.
Those skilled in the art will appreciate that all or part of the flow of the method implementing the above embodiments may be implemented by instructing relevant hardware (such as a processor, a controller, etc.) by a computer program, and the computer program may be stored in a computer readable storage medium. The computer readable storage medium is a magnetic disk, an optical disk, a read-only memory or a random access memory.
The method and the device for determining the actual size of the core provided by the invention are described in detail, a specific example is applied in the text to explain the principle and the implementation mode of the invention, and the description of the example is only used for helping to understand the method and the core idea of the invention; meanwhile, for those skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.
Claims (10)
1. A method for determining the actual size of a core, comprising:
constructing a target segmentation model, segmenting an image to be segmented based on the target segmentation model, and determining an object prediction category and an object prediction image pixel size in the image to be segmented, wherein the object prediction category comprises a rock core and a rock core box, and the object prediction image pixel size comprises the rock core prediction image pixel size and the rock core box prediction image pixel size;
acquiring the actual size of a core box, and constructing the proportional corresponding relation between the predicted image pixel size of the core box and the actual size of the core box;
and determining the actual size of the rock core according to the pixel size of the rock core predicted image and the proportional corresponding relation.
2. The method for determining the actual size of the core according to claim 1, wherein the target segmentation model comprises a feature extraction module, a feature fusion module, a core box segmentation module and a core segmentation module;
the feature extraction module is used for extracting image features in the image to be segmented to obtain a plurality of feature extraction images;
the feature fusion layer is used for extracting detail features in the feature extraction images to obtain a plurality of detail feature images, and fusing the detail feature images and the extraction images to obtain a plurality of fusion feature images;
the core box segmentation module is used for determining at least two target fusion feature images from the plurality of fusion feature images and determining the core box and the core box predicted image pixel size based on the at least two target fusion feature determination images;
the core segmentation module is to determine the core and the core predicted image pixel size based on the plurality of fused feature images.
3. The method of determining the actual size of a core according to claim 2, wherein the feature extraction module comprises a first feature extraction layer, a second feature extraction layer, a third feature extraction layer, a fourth feature extraction layer, and a fifth feature extraction layer;
the first feature extraction layer is used for extracting image features in the image to be segmented to obtain a first feature extraction image;
the second feature extraction layer is used for carrying out downsampling on the first feature extraction image to obtain a second feature extraction image;
the third feature extraction layer is used for performing downsampling on the second feature extraction image to obtain a third feature extraction image;
the fourth feature extraction layer is used for down-sampling the third feature extraction image to obtain a fourth feature extraction image;
the fifth feature extraction layer is configured to perform downsampling on the fourth feature extraction image to obtain a fifth feature extraction image.
4. The method of claim 3, wherein the feature fusion module comprises a first feature fusion layer, a second feature fusion layer, a third feature fusion layer, and a fourth feature fusion layer;
the first feature fusion layer is used for performing convolution on the fifth feature extraction image to obtain a first fusion feature image;
the second feature fusion layer is used for performing convolution on the fourth feature extraction image to obtain a first convolution image, performing up-sampling on the first fusion feature image to obtain a first up-sampling image, and fusing the first convolution image and the first up-sampling image to obtain a second fusion feature image;
the third feature fusion layer is used for convolving the third feature extraction image to obtain a second convolution image, upsampling the second fusion feature image to obtain a second upsampled image, and fusing the second convolution image and the second upsampled image to obtain a third fusion feature image;
the fourth feature fusion layer is used for convolving the second feature extraction image to obtain a third convolution image, upsampling the third fusion feature image to obtain a third upsampled image, and fusing the third convolution image and the third upsampled image to obtain a fourth fusion feature image.
5. The method of determining the actual size of a core according to claim 4, wherein the target fused feature image is the first fused feature image and the second fused feature image.
6. The method of determining the actual size of a core according to claim 1, wherein said target segmentation model further comprises a partition segmentation module, said object prediction class further comprises partitions, said object prediction image pixel size further comprises partition prediction image pixel size; the method for determining the actual size of the rock core further comprises the following steps:
determining a partition and a partition predicted image pixel size in the image to be segmented based on the partition segmentation module;
acquiring the actual size of the partition, and constructing the proportional association relationship between the pixel size of the partition predicted image and the actual size of the partition;
determining the core verification size according to the core predicted image pixel size and the proportional correlation;
and verifying the reliability of the actual size of the rock core according to the rock core verification size.
7. The method of determining the actual size of a core according to claim 6, wherein said verifying the reliability of the actual size of the core according to the core verification size comprises:
judging whether the difference value between the rock core verification size and the actual size of the rock core is smaller than a threshold difference value or not;
if the difference value between the core verification size and the actual core size is smaller than the threshold difference value, the actual core size is reliable;
and if the difference value between the core verification size and the actual core size is larger than or equal to the threshold difference value, the actual core size is unreliable.
8. The method of claim 1, wherein the constructing the object segmentation model comprises:
acquiring a real core box image, and performing enhancement processing on the real core box image to obtain an enhanced core box image;
marking the real core box image and the enhanced core box image to obtain a core box image sample set;
and constructing an initial segmentation model, and training the initial segmentation model based on the core box image sample set to obtain the target segmentation model.
9. The method for determining the actual size of the core according to claim 1, wherein the segmenting the image to be segmented based on the target segmentation model, and determining the pixel size of the image predicted by the object in the image to be segmented comprises:
inputting the image to be segmented into the target segmentation model to obtain a plurality of core prediction frames;
and screening the plurality of core prediction frames based on a non-maximum value inhibition method to obtain a plurality of target core prediction frames, and determining the pixel size of the core prediction image according to the plurality of target core prediction frames.
10. An apparatus for determining the actual size of a core, comprising:
the image segmentation unit is used for constructing a target segmentation model, segmenting an image to be segmented based on the target segmentation model, and determining an object prediction category and an object prediction image pixel size in the image to be segmented, wherein the object prediction category comprises a rock core and a rock core box, and the object prediction image pixel size comprises the rock core prediction image pixel size and the rock core box prediction image pixel size;
the size corresponding relation construction unit is used for acquiring the actual size of the core box and constructing the proportional corresponding relation between the predicted image pixel size of the core box and the actual size of the core box;
and the actual core size determining unit is used for determining the actual core size according to the predicted image pixel size of the core box and the proportional corresponding relation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210578421.7A CN114841988A (en) | 2022-05-24 | 2022-05-24 | Method and device for determining actual size of rock core |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210578421.7A CN114841988A (en) | 2022-05-24 | 2022-05-24 | Method and device for determining actual size of rock core |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114841988A true CN114841988A (en) | 2022-08-02 |
Family
ID=82573034
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210578421.7A Pending CN114841988A (en) | 2022-05-24 | 2022-05-24 | Method and device for determining actual size of rock core |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114841988A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117611829A (en) * | 2024-01-24 | 2024-02-27 | 三一重型装备有限公司 | Underground ore image segmentation method and device, operation machine and electronic equipment |
-
2022
- 2022-05-24 CN CN202210578421.7A patent/CN114841988A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117611829A (en) * | 2024-01-24 | 2024-02-27 | 三一重型装备有限公司 | Underground ore image segmentation method and device, operation machine and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11048018B2 (en) | Systems, methods, and computer-readable media for modeling complex wellbores in field-scale reservoir simulation | |
| EP3397833B1 (en) | Machine learning for production prediction | |
| US9922142B2 (en) | Systems and methods for subsurface reservoir simulation | |
| US11775353B2 (en) | Mapping workloads to cloud infrastructure | |
| CN109155064A (en) | Automate core description | |
| CN114841988A (en) | Method and device for determining actual size of rock core | |
| CN116168172B (en) | Shale oil gas dessert prediction method, device, equipment and storage medium | |
| WO2017058738A1 (en) | Network based simulation workflows | |
| US11227372B2 (en) | Geological imaging and inversion using object storage | |
| Cheng et al. | Digital Documentation and Data Management for Offshore Drilling | |
| US20200334404A1 (en) | Computer implemented method for manipulating a numerical model of a 3d domain | |
| Szyndel et al. | Implementing A Hardware Agnostic Commercial Black-oil Reservoir Simulator | |
| CA3027113C (en) | Structural volume segmentation | |
| Park et al. | Multioutput Image Classification to Support Postearthquake Reconnaissance | |
| Nemushchenko et al. | Implementation of Vendor-Independent Stochastic Inversion for Improving Quality and Efficiency of Well Placement on the Field of Novatek Company | |
| US20220092617A1 (en) | Rapid region wide production forecasting | |
| US20220381127A1 (en) | Sidetrack well parameter identification based on simulations related to an existing physical well | |
| EP3513033B1 (en) | Integrated hydrocarbon fluid distribution modeling | |
| US10890065B2 (en) | Well infiltration area calculation using logging while drilling data | |
| CN114842161A (en) | Fracture structure modeling method and system based on fracture vector field | |
| US20200401468A1 (en) | Visualization infrastructure for web applications | |
| WO2023154312A1 (en) | Quantification of expressive experimental semi-variogram ranges uncertainties | |
| WO2021163054A1 (en) | Optimizing reservoir simulation runtime and storage | |
| CN117745964A (en) | Parameter optimization method for disaster-bearing body three-dimensional model | |
| Firincioglu et al. | Soft Computing Algorithms Accelerate and Improve the History-Matching Process: Elk Hills, California—29R Reservoir |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |