CN112907498A - Pore identification method, device, equipment and storage medium - Google Patents
Pore identification method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN112907498A CN112907498A CN201911130002.1A CN201911130002A CN112907498A CN 112907498 A CN112907498 A CN 112907498A CN 201911130002 A CN201911130002 A CN 201911130002A CN 112907498 A CN112907498 A CN 112907498A
- Authority
- CN
- China
- Prior art keywords
- pore
- metallographic
- identification
- image
- metallographic image
- 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.)
- Granted
Links
- 239000011148 porous material Substances 0.000 title claims abstract description 224
- 238000000034 method Methods 0.000 title claims abstract description 58
- 238000003860 storage Methods 0.000 title claims abstract description 17
- 238000012549 training Methods 0.000 claims abstract description 56
- 239000002131 composite material Substances 0.000 claims abstract description 26
- 230000008569 process Effects 0.000 claims abstract description 21
- 238000012545 processing Methods 0.000 claims description 17
- 230000009467 reduction Effects 0.000 claims description 15
- 238000007906 compression Methods 0.000 claims description 14
- 230000006835 compression Effects 0.000 claims description 11
- 238000007689 inspection Methods 0.000 claims description 4
- 239000011800 void material Substances 0.000 claims description 3
- 238000004590 computer program Methods 0.000 claims description 2
- 238000000605 extraction Methods 0.000 claims description 2
- 230000000694 effects Effects 0.000 abstract description 5
- 230000006870 function Effects 0.000 description 7
- 238000005070 sampling Methods 0.000 description 6
- 230000007547 defect Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000002372 labelling Methods 0.000 description 4
- 238000001914 filtration Methods 0.000 description 3
- 239000000463 material Substances 0.000 description 3
- 229920000049 Carbon (fiber) Polymers 0.000 description 2
- 239000004917 carbon fiber Substances 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000007667 floating Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 239000012535 impurity Substances 0.000 description 2
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 description 2
- 239000011347 resin Substances 0.000 description 2
- 229920005989 resin Polymers 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 1
- 239000010931 gold Substances 0.000 description 1
- 229910052737 gold Inorganic materials 0.000 description 1
- 238000000227 grinding Methods 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000010587 phase diagram Methods 0.000 description 1
- 238000005498 polishing Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000001303 quality assessment method Methods 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 238000011946 reduction process Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
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
- G06T7/0004—Industrial image inspection
-
- 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
- G06T2207/10056—Microscopic image
- G06T2207/10061—Microscopic image from scanning electron microscope
-
- 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
- G06T2207/20081—Training; Learning
-
- 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
- G06T2207/20084—Artificial neural networks [ANN]
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The embodiment of the invention discloses a pore identification method, a pore identification device, pore identification equipment and a storage medium, wherein the method comprises the following steps: acquiring a metallographic image of the composite material; inputting the metallographic image into a pre-trained pore recognition model; in the training process, the pore identification model determines the pore edge pixel points of the training samples by carrying out bilinear interpolation on the compressed training samples; and determining the pore identification result of the metallographic image according to the output result of the pore identification model. By operating the technical scheme provided by the request, the problems that the pore counting work is time-consuming and labor-consuming and the counting error is large can be solved, and the effects of improving the efficiency and the accuracy of pore identification are achieved.
Description
Technical Field
The present invention relates to computer technologies, and in particular, to a method, an apparatus, a device, and a storage medium for identifying a hole.
Background
The pores of the composite material are gas pore type defects distributed in a composite material laminated structure resin layer and a carbon fiber layer, the diameters of the pores are generally distributed in the range of 1-100 mu m, and the pores are the most common internal quality defects in composite material products. A large amount of literature and test data show that the pores have a remarkable influence on the interlaminar shear strength of the composite material product. Therefore, the pore detection is an important component of the nondestructive detection, and plays a very positive and important role in the internal quality assessment and quality improvement of the composite material and the continuous improvement of the product process.
At this stage, the pores of the composite material are usually obtained by manual identification or image analysis.
The manual identification is very huge because of the collected metallographic pictures, the pore counting work is time-consuming and labor-consuming, and the results counted by different personnel may have differences, so that the efficiency is very low. The image analyzer identifies that the influence of human factors is small, but the influence of the metallographic surface quality is easy to be received, and the statistical error is large.
Disclosure of Invention
The embodiment of the invention provides a pore identification method, a pore identification device, pore identification equipment and a storage medium, and aims to achieve the effect of improving the efficiency and accuracy of pore identification.
In a first aspect, an embodiment of the present invention provides a method for identifying a pore, including:
acquiring a metallographic image of the composite material;
inputting the metallographic image into a pre-trained pore recognition model; in the training process, the pore identification model determines the pore edge pixel points of the training samples by carrying out bilinear interpolation on the compressed training samples;
and determining the pore identification result of the metallographic image according to the output result of the pore identification model.
In a second aspect, embodiments of the present invention further provide a hole identification device, including:
the image acquisition module is used for acquiring a metallographic image of the composite material;
the model input module is used for inputting the metallographic image into a pre-trained pore recognition model; in the training process, the pore identification model determines the pore edge pixel points of the training samples by carrying out bilinear interpolation on the compressed training samples;
and the identification result determining module is used for determining the pore identification result of the metallographic image according to the output result of the pore identification model.
In a third aspect, an embodiment of the present invention further provides an apparatus, where the apparatus includes:
one or more processors;
a storage device for storing one or more programs,
when executed by the one or more processors, cause the one or more processors to implement the pore identification method as described above.
In a fourth aspect, the present invention further provides a computer-readable storage medium, on which a computer program is stored, which when executed by a processor implements the pore identification method as described above.
According to the embodiment of the invention, a metallographic image of the composite material is obtained; inputting the metallographic image into a pre-trained pore recognition model; in the training process, the pore identification model determines the pore edge pixel points of the training samples by carrying out bilinear interpolation on the compressed training samples; and determining the pore identification result of the metallographic image according to the output result of the pore identification model. By adopting the technical scheme provided by the invention, the problems of time and labor waste and large statistical error in pore statistics work are solved, and the effects of improving the efficiency and accuracy of pore identification are realized.
Drawings
Fig. 1 is a flowchart of a void identification method according to an embodiment of the present invention;
FIG. 2 is a flowchart illustrating a training process of a pore recognition model according to a second embodiment of the present invention;
fig. 3 is a schematic structural diagram of a pore identification device according to a third embodiment of the present invention;
fig. 4 is a schematic structural diagram of an apparatus according to a fourth embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.
Example one
Fig. 1 is a flowchart of a method for identifying pores in a metallographic image of a composite material according to an embodiment of the present invention, where the method is applicable to identifying pores in the metallographic image of the composite material, and the method can be performed by a pore identification apparatus according to an embodiment of the present invention, and the apparatus can be implemented by software and/or hardware. Referring to fig. 1, the present embodiment provides a method for identifying a hole, including:
and S110, acquiring a metallographic image of the composite material.
The composite material is a new material formed by artificially optimizing and combining material components with different properties by adopting a material preparation technology, and is applied to the fields of aviation, aerospace, high-end automobiles and the like. The metallographic phase of the composite material is a section obtained by cutting, inlaying, grinding, polishing and the like of the composite material.
The metallographic microscope can be used for observing the metallographic section and obtaining a metallographic image, and the same composite material can obtain at least one metallographic image.
S120, inputting the metallographic image into a pre-trained pore recognition model; and in the training process, the pore identification model determines the pore edge pixel points of the training samples by carrying out bilinear interpolation on the compressed training samples.
The pores of the composite material are gas pore type defects distributed in a composite laminated structure resin layer and a carbon fiber layer, the diameter of the pores is generally in the range of 1-100 mu m, and the pores are internal quality defects in composite material parts. The pore identification model is used to identify a pore part in the input metallographic image, and may be a machine learning model such as a convolutional neural network, which is not limited in this embodiment.
In the training process of the pore identification model, the metallographic image is compressed, and the coordinates of the sampling points are usually floating point numbers in the compression process, so that the pixel value of the sampling points needs to be obtained by using an interpolation method after compression. At the moment, the pixel value of the point is obtained by common calculation of four adjacent pixel values at the periphery in a bilinear interpolation mode.
In the actual identification process, each pixel point is detected, if the probability of other pixel points adjacent to the pixel point is high, for example, more than eighty percent, and the other pixel points also accord with the identification rule, the pixel point is judged to belong to a pore edge point, and therefore the pore edge pixel point is determined.
S130, determining a pore identification result of the metallographic image according to an output result of the pore identification model.
The output result of the pore identification model can be the range and/or position distribution of each pore of the pores on the golden image. The result of identifying the pores may be whether the pores exist on the gold phase diagram, or may be a specific type of the pores, which is not limited in this embodiment.
According to the technical scheme provided by the embodiment, a metallographic image of the composite material is taken; inputting the metallographic image into a pre-trained pore recognition model; in the training process, the pore identification model determines the pore edge pixel points of the training samples by carrying out bilinear interpolation on the compressed training samples; and determining the pore identification result of the metallographic image according to the output result of the pore identification model. The problems that the pore counting work is time-consuming and labor-consuming and the counting error is large are solved, and the effects of improving the efficiency and the accuracy of pore identification are achieved.
On the basis of the above technical solution, optionally, after determining the pore recognition result of the metallographic image according to the output result of the pore recognition model, the method further includes:
performing spot inspection on the pore identification result of the metallographic image;
and if the accuracy of the pore identification result of the metallographic image is lower than a preset reference value, secondarily marking the pores of the metallographic image, and updating the pore identification model by taking the metallographic image subjected to secondary marking as a training sample.
Wherein, the sampling inspection of the pore identification result can be carried out manually, and the pore identification result of the output phase image is inspected. And if the accuracy of the pore identification result of the metallographic image is not lower than the preset reference value, the pore identification model is continuously put into use.
In the identification process, an unpredictable pore type may occur, and the pore type is not generated in the previous training sample, so that the accuracy of the pore identification result is lower than a preset reference value, and therefore, when an error is found in the spot check, the wrong pore type is identified. And marking in a manual mode, and taking the manually marked image as a training sample so as to update the pore recognition model.
On the basis of the embodiment, the porosity identification result of the metallographic image is subjected to sampling inspection and secondary labeling, so that the reduction of the accuracy of the porosity identification result caused by the unpredictable porosity type is prevented, and the accuracy of the porosity identification and the application range of the porosity identification method are improved.
Example two
Fig. 2 is a flowchart illustrating a training process of a pore recognition model according to a second embodiment of the present invention. The technical scheme is supplementary explained aiming at the training process of a pore recognition model trained in advance. Compared with the scheme, the scheme is specifically optimized as a training process of a pore recognition model trained in advance, and comprises the following steps:
the training process of the pre-trained pore recognition model is as follows:
inputting a metallographic image sample, and performing noise reduction treatment on the metallographic image sample;
extracting a characteristic image block from the metallographic image sample subjected to noise reduction;
compressing the characteristic image blocks, and performing bilinear interpolation to obtain compressed characteristic image blocks;
carrying out pore edge point identification processing on the compressed feature image blocks to obtain pore edge pixel points of the compressed feature image blocks;
and determining a pore contour formed by the pore edge pixel points of the compressed characteristic image block by adopting a fitting function as a pore identification result.
Specifically, a training flowchart of the pore recognition model is shown in fig. 2:
s210, inputting a metallographic image sample, and performing noise reduction on the metallographic image sample.
The metallographic image sample can be a metallographic image collected in advance, and the metallographic image is used as a training sample.
The noise reduction process may be, but is not limited to, employing median filtering, maximum filtering, and/or minimum filtering algorithms, and the like.
And S220, extracting a characteristic image block from the metallographic image sample subjected to the noise reduction treatment.
The characteristic image block is extracted by selecting a pore candidate region from the image obtained by the noise reduction processing by using the information of texture, edge, color and the like in the image subjected to the noise reduction processing, wherein the pore candidate region is a range in which pores are approximately located, and parts which are possible to be pores are outlined in a mode of a square frame, a circle, an ellipse, an irregular polygon and the like to be used as regions needing further processing.
And S230, compressing the characteristic image block, and performing bilinear interpolation to obtain a compressed characteristic image block.
The compression processing is to compress the picture, and after the compression, a bilinear interpolation method is adopted to calculate and obtain a pixel value of a point of which the coordinate is a floating point number due to the compression. And the pixel points of the compressed characteristic image blocks correspond to the pixel points of the original image one by one.
S240, carrying out pore edge point identification processing on the compressed feature image block to obtain pore edge pixel points of the compressed feature image block.
And analyzing the pore region of the compressed characteristic image block by a pore recognition model to generate a recognition rule, such as a gray scale change rule, of the pixel points at the pore edge in the pore recognition model, so as to obtain the pixel points at the pore edge of the compressed characteristic image block.
And S250, determining a pore contour formed by the pore edge pixel points of the compressed characteristic image block by adopting a fitting function, and taking the pore contour as a pore identification result.
The fitting function is used for connecting the obtained pore edge pixel points into a line, so that the outline of the pore is formed. The compressed characteristic image block pore edge pixel points are reversely mapped back to the original image to form a pore outline in the original image; or acquiring the pore contour in the compressed feature image block, and then reversely mapping the pore contour back to the original image. After the pore identification result is obtained, the pore identification result can be compared with the mark direction of the metallographic image to determine whether the determined pore identification result is consistent with the mark information.
On the basis of the embodiment, the pore contour is recognized through the trained pore recognition model, so that the pore range on the metallographic image is obtained, and the efficiency and the accuracy of pore recognition are improved.
On the basis of the technical scheme, optionally, the metallographic image sample is a metallographic image sample with a pore mark;
correspondingly, after determining a pore contour formed by the pore edge pixel points of the compressed feature image block by adopting a fitting function as a pore identification result, the method further comprises the following steps:
comparing the pore identification result with the pore mark of the metallographic image sample to obtain a comparison result;
and if the comparison result meets the preset standard, determining that the training of the pore recognition model is finished.
The selection of the metallographic image sample with the pore mark can be divided into a typical metallographic sample and a special metallographic sample; typical metallographic specimens need to be sufficiently clear and free of significant adhering matter, and special metallographic specimens include impurities inside, scratches and adhering matter on the surface, and other unpredictable pore forms. And marking all pore areas on the two types of samples, and inputting the marked samples into a pore recognition model for training.
Comparing the result of the pore identification with the pore mark to obtain a comparison result;
and if the comparison result meets a preset standard, for example, the accuracy reaches more than ninety percent, determining that the training of the pore recognition model is finished. If the standard is not met, the parameters in the training process can be adjusted, so that the result meets the expected standard.
By marking the pores of the special metallographic sample, impurities, scratches and attachments are prevented from being mistakenly identified as pores; and comparing the result of the pore identification with the pore mark until the result meets the preset standard, thereby improving the accuracy of the pore identification.
On the basis of the above technical solution, optionally, the compression ratio of the compression processing is 1: 16.
The compression process accelerates the processing speed of the pore identification model, thereby improving the efficiency of pore identification.
EXAMPLE III
Fig. 3 is a schematic structural diagram of a pore identification device according to a third embodiment of the present invention. The device can be realized in a hardware and/or software mode, can execute the pore identification method provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method. As shown in fig. 3, the apparatus includes:
an image obtaining module 310, configured to obtain a metallographic image of the composite material;
the model input module 320 is used for inputting the metallographic image into a pre-trained pore recognition model; in the training process, the pore identification model determines the pore edge pixel points of the training samples by carrying out bilinear interpolation on the compressed training samples;
and the identification result determining module 330 is configured to determine a pore identification result of the metallographic image according to an output result of the pore identification model.
According to the technical scheme provided by the embodiment, a metallographic image of the composite material is taken; inputting the metallographic image into a pre-trained pore recognition model; in the training process, the pore identification model determines the pore edge pixel points of the training samples by carrying out bilinear interpolation on the compressed training samples; and determining the pore identification result of the metallographic image according to the output result of the pore identification model. The problems that the pore counting work is time-consuming and labor-consuming and the counting error is large are solved, and the effects of improving the efficiency and the accuracy of pore identification are achieved.
On the basis of the above technical solutions, optionally,
a model training module, the model training module comprising:
the sample noise reduction processing unit is used for inputting a metallographic image sample and performing noise reduction processing on the metallographic image sample;
the characteristic image block extraction unit is used for extracting a characteristic image block from the metallographic image sample subjected to noise reduction;
the characteristic image block compression unit is used for compressing the characteristic image blocks and performing bilinear interpolation to obtain compressed characteristic image blocks;
the pixel point acquisition unit is used for identifying and processing the pore edge points of the compressed characteristic image blocks to obtain the pore edge pixel points of the compressed characteristic image blocks;
and the recognition result determining unit is used for determining a pore contour formed by the pore edge pixel points of the compressed feature image block by adopting a fitting function as a pore recognition result.
On the basis of the technical schemes, optionally, the metallographic image sample is a metallographic image sample with a pore mark;
correspondingly, the device further comprises:
the comparison result acquisition module is used for comparing the pore identification result with the pore mark of the metallographic image sample after the identification result determination module to obtain a comparison result;
and the model training completion determining module is used for determining that the pore recognition model is trained completely if the comparison result meets the preset standard.
On the basis of the above technical solutions, optionally, the compression ratio of the compression processing is 1: 16.
On the basis of the above technical solutions, optionally, the apparatus further includes:
the identification result sampling module is used for sampling and inspecting the pore identification result of the metallographic image after the identification result determining module;
and the secondary labeling module is used for performing secondary labeling on the pores of the metallographic image if the accuracy of the pore identification result of the metallographic image is lower than a preset reference value after the identification result determining module, and updating the pore identification model by taking the metallographic image subjected to secondary labeling as a training sample.
Example four
Fig. 4 is a schematic structural diagram of an apparatus according to a fourth embodiment of the present invention, as shown in fig. 4, the apparatus includes a processor 40, a memory 41, an input device 42, and an output device 43; the number of processors 40 in the device may be one or more, and one processor 40 is taken as an example in fig. 4; the processor 40, the memory 41, the input means 42 and the output means 43 in the device may be connected by a bus or other means, as exemplified by the bus connection in fig. 4.
The memory 41 is a computer-readable storage medium for storing software programs, computer-executable programs, and modules, such as program instructions/modules corresponding to the pore identification method in the embodiments of the present invention. The processor 40 executes various functional applications of the device and data processing, i.e. implements the above-described pore identification method, by running software programs, instructions and modules stored in the memory 41.
The memory 41 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to the use of the terminal, and the like. Further, the memory 41 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some examples, memory 41 may further include memory located remotely from processor 40, which may be connected to the device over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
EXAMPLE five
Embodiments of the present invention also provide a storage medium containing computer-executable instructions which, when executed by a computer processor, perform a method of hole identification, the method comprising:
acquiring a metallographic image of the composite material;
inputting the metallographic image into a pre-trained pore recognition model; in the training process, the pore identification model determines the pore edge pixel points of the training samples by carrying out bilinear interpolation on the compressed training samples;
and determining the pore identification result of the metallographic image according to the output result of the pore identification model.
Of course, the storage medium provided by the embodiments of the present invention contains computer-executable instructions, and the computer-executable instructions are not limited to the operations of the method described above, and may also perform related operations in the pore identification method provided by any embodiments of the present invention.
From the above description of the embodiments, it is obvious for those skilled in the art that the present invention can be implemented by software and necessary general hardware, and certainly, can also be implemented by hardware, but the former is a better embodiment in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which may be stored in a computer-readable storage medium, such as a floppy disk, a Read-only memory (ROM), a Random Access Memory (RAM), a FLASH memory (FLASH), a hard disk or an optical disk of a computer, and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute the methods according to the embodiments of the present invention.
It should be noted that, in the embodiment of the above search apparatus, each included unit and module are merely divided according to functional logic, but are not limited to the above division as long as the corresponding functions can be implemented; in addition, specific names of the functional units are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present invention.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (10)
1. A method of identifying a void, comprising:
acquiring a metallographic image of the composite material;
inputting the metallographic image into a pre-trained pore recognition model; in the training process, the pore identification model determines the pore edge pixel points of the training samples by carrying out bilinear interpolation on the compressed training samples;
and determining the pore identification result of the metallographic image according to the output result of the pore identification model.
2. The method of claim 1, wherein the pre-trained pore recognition model is trained as follows:
inputting a metallographic image sample, and performing noise reduction treatment on the metallographic image sample;
extracting a characteristic image block from the metallographic image sample subjected to noise reduction;
compressing the characteristic image blocks, and performing bilinear interpolation to obtain compressed characteristic image blocks;
carrying out pore edge point identification processing on the compressed feature image blocks to obtain pore edge pixel points of the compressed feature image blocks;
and determining a pore contour formed by the pore edge pixel points of the compressed characteristic image block by adopting a fitting function as a pore identification result.
3. The method of claim 2, wherein the metallographic image sample is a pore-marked metallographic image sample;
correspondingly, after determining a pore contour formed by the pore edge pixel points of the compressed feature image block by adopting a fitting function as a pore identification result, the method further comprises the following steps:
comparing the pore identification result with the pore mark of the metallographic image sample to obtain a comparison result;
and if the comparison result meets the preset standard, determining that the training of the pore recognition model is finished.
4. The method of claim 2, wherein the compression ratio of the compression process is 1: 16.
5. The method of claim 1, wherein after determining a pore identification result for the metallographic image based on an output of the pore identification model, the method further comprises:
performing spot inspection on the pore identification result of the metallographic image;
and if the accuracy of the pore identification result of the metallographic image is lower than a preset reference value, secondarily marking the pores of the metallographic image, and updating the pore identification model by taking the metallographic image subjected to secondary marking as a training sample.
6. A void identification device, comprising:
the image acquisition module is used for acquiring a metallographic image of the composite material;
the model input module is used for inputting the metallographic image into a pre-trained pore recognition model; in the training process, the pore identification model determines the pore edge pixel points of the training samples by carrying out bilinear interpolation on the compressed training samples;
and the identification result determining module is used for determining the pore identification result of the metallographic image according to the output result of the pore identification model.
7. The apparatus of claim 6, wherein the model training module comprises:
the sample noise reduction processing unit is used for inputting a metallographic image sample and performing noise reduction processing on the metallographic image sample;
the characteristic image block extraction unit is used for extracting a characteristic image block from the metallographic image sample subjected to noise reduction;
the characteristic image block compression unit is used for compressing the characteristic image blocks and performing bilinear interpolation to obtain compressed characteristic image blocks;
the pixel point acquisition unit is used for identifying and processing the pore edge points of the compressed characteristic image blocks to obtain the pore edge pixel points of the compressed characteristic image blocks;
and the recognition result determining unit is used for determining a pore contour formed by the pore edge pixel points of the compressed feature image block by adopting a fitting function as a pore recognition result.
8. The apparatus of claim 7, wherein the metallographic image sample is a pore marked metallographic image sample;
correspondingly, the device further comprises:
the comparison result acquisition module is used for comparing the pore identification result with the pore mark of the metallographic image sample after the identification result determination module to obtain a comparison result;
and the model training completion determining module is used for determining that the pore recognition model is trained completely if the comparison result meets the preset standard.
9. An apparatus, characterized in that the apparatus comprises:
one or more processors;
a storage device for storing one or more programs,
when executed by the one or more processors, cause the one or more processors to implement the pore identification method of any one of claims 1-5.
10. A computer-readable storage medium, on which a computer program is stored, which program, when being executed by a processor, carries out the pore identification method according to any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911130002.1A CN112907498B (en) | 2019-11-18 | 2019-11-18 | Pore identification method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911130002.1A CN112907498B (en) | 2019-11-18 | 2019-11-18 | Pore identification method, device, equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112907498A true CN112907498A (en) | 2021-06-04 |
| CN112907498B CN112907498B (en) | 2024-04-30 |
Family
ID=76103113
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911130002.1A Active CN112907498B (en) | 2019-11-18 | 2019-11-18 | Pore identification method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112907498B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115131334A (en) * | 2022-07-21 | 2022-09-30 | 北京汉飞航空科技有限公司 | Aviation engine pinhole type identification and automatic sequencing method based on machine learning |
| CN117808856A (en) * | 2023-12-28 | 2024-04-02 | 中国人民解放军陆军装甲兵学院士官学校 | Component strength optimization method based on artificial intelligence |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110055205A1 (en) * | 2009-08-11 | 2011-03-03 | Terrence Stanley David Scott | Data processing system for manufacturing quality inspection management system |
| DE102013210864A1 (en) * | 2012-06-19 | 2013-12-19 | GM Global Technology Operations, LLC (n.d. Ges. d. Staates Delaware) | Method for determining pore size and pore distribution in metal casting, involves grouping measured pores size and spaces in cluster for generating measuring value for pores size or pores distribution in sample on cluster-to-cluster basis |
| CN104766095A (en) * | 2015-04-16 | 2015-07-08 | 成都汇智远景科技有限公司 | Mobile terminal image identification method |
| CN105699341A (en) * | 2014-11-28 | 2016-06-22 | 中国科学院金属研究所 | A fluorescence metallographic measuring method for pores of a metal material |
| CN108416748A (en) * | 2018-02-26 | 2018-08-17 | 阿博茨德（北京）科技有限公司 | The image pre-processing method and device of JPEG compression document |
| WO2018166114A1 (en) * | 2017-03-13 | 2018-09-20 | 平安科技（深圳）有限公司 | Picture identification method and system, electronic device, and medium |
| CN108647588A (en) * | 2018-04-24 | 2018-10-12 | 广州绿怡信息科技有限公司 | Goods categories recognition methods, device, computer equipment and storage medium |
| CN109409428A (en) * | 2018-10-25 | 2019-03-01 | 北京木业邦科技有限公司 | Training method, device and the electronic equipment of plank identification and plank identification model |
| CN109615630A (en) * | 2019-01-18 | 2019-04-12 | 东北大学 | Semi-continuous casting alusil alloy Analysis on Microstructure method based on image processing techniques |
| CN109829483A (en) * | 2019-01-07 | 2019-05-31 | 鲁班嫡系机器人（深圳）有限公司 | Defect recognition model training method, device, computer equipment and storage medium |
| CN110110661A (en) * | 2019-05-07 | 2019-08-09 | 西南石油大学 | A kind of rock image porosity type recognition methods based on unet segmentation |
| CN110308151A (en) * | 2019-07-22 | 2019-10-08 | 重庆大学 | A kind of surface of workpiece defect identification method and device based on machine vision |
| CN110348461A (en) * | 2019-07-05 | 2019-10-18 | 江苏海事职业技术学院 | A kind of Surface Flaw feature extracting method |
-
2019
- 2019-11-18 CN CN201911130002.1A patent/CN112907498B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110055205A1 (en) * | 2009-08-11 | 2011-03-03 | Terrence Stanley David Scott | Data processing system for manufacturing quality inspection management system |
| DE102013210864A1 (en) * | 2012-06-19 | 2013-12-19 | GM Global Technology Operations, LLC (n.d. Ges. d. Staates Delaware) | Method for determining pore size and pore distribution in metal casting, involves grouping measured pores size and spaces in cluster for generating measuring value for pores size or pores distribution in sample on cluster-to-cluster basis |
| CN105699341A (en) * | 2014-11-28 | 2016-06-22 | 中国科学院金属研究所 | A fluorescence metallographic measuring method for pores of a metal material |
| CN104766095A (en) * | 2015-04-16 | 2015-07-08 | 成都汇智远景科技有限公司 | Mobile terminal image identification method |
| WO2018166114A1 (en) * | 2017-03-13 | 2018-09-20 | 平安科技（深圳）有限公司 | Picture identification method and system, electronic device, and medium |
| CN108416748A (en) * | 2018-02-26 | 2018-08-17 | 阿博茨德（北京）科技有限公司 | The image pre-processing method and device of JPEG compression document |
| CN108647588A (en) * | 2018-04-24 | 2018-10-12 | 广州绿怡信息科技有限公司 | Goods categories recognition methods, device, computer equipment and storage medium |
| CN109409428A (en) * | 2018-10-25 | 2019-03-01 | 北京木业邦科技有限公司 | Training method, device and the electronic equipment of plank identification and plank identification model |
| CN109829483A (en) * | 2019-01-07 | 2019-05-31 | 鲁班嫡系机器人（深圳）有限公司 | Defect recognition model training method, device, computer equipment and storage medium |
| CN109615630A (en) * | 2019-01-18 | 2019-04-12 | 东北大学 | Semi-continuous casting alusil alloy Analysis on Microstructure method based on image processing techniques |
| CN110110661A (en) * | 2019-05-07 | 2019-08-09 | 西南石油大学 | A kind of rock image porosity type recognition methods based on unet segmentation |
| CN110348461A (en) * | 2019-07-05 | 2019-10-18 | 江苏海事职业技术学院 | A kind of Surface Flaw feature extracting method |
| CN110308151A (en) * | 2019-07-22 | 2019-10-08 | 重庆大学 | A kind of surface of workpiece defect identification method and device based on machine vision |
Non-Patent Citations (1)
| Title |
|---|
| 高志强等: "深度学习从入门到实战", 中国铁道出版社, pages: 66 - 67 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115131334A (en) * | 2022-07-21 | 2022-09-30 | 北京汉飞航空科技有限公司 | Aviation engine pinhole type identification and automatic sequencing method based on machine learning |
| CN115131334B (en) * | 2022-07-21 | 2023-07-25 | 北京汉飞航空科技有限公司 | Machine learning-based aerial small hole type identification and automatic sequencing method |
| CN117808856A (en) * | 2023-12-28 | 2024-04-02 | 中国人民解放军陆军装甲兵学院士官学校 | Component strength optimization method based on artificial intelligence |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112907498B (en) | 2024-04-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109087274B (en) | Electronic device defect detection method and device based on multi-dimensional fusion and semantic segmentation | |
| EP3937128A2 (en) | Image defect detection method and apparatus, electronic device, storage medium and product | |
| CN108009591B (en) | Contact network key component identification method based on deep learning | |
| CN108682008B (en) | A kind of Leukocyte Image clarity evaluation method and device | |
| CN108520514B (en) | Consistency detection method for electronic elements of printed circuit board based on computer vision | |
| Sharifzadeh et al. | Detection of steel defect using the image processing algorithms | |
| CN106296689B (en) | Flaw detection method, system and device | |
| CN107345916B (en) | Plane appearance detection method based on fixed contour | |
| CN112907498B (en) | Pore identification method, device, equipment and storage medium | |
| CN112784835A (en) | Method and device for identifying authenticity of circular seal, electronic equipment and storage medium | |
| CN115684176A (en) | Online visual inspection system for film surface defects | |
| CN111860176B (en) | Non-metal inclusion full-view-field quantitative statistical distribution characterization method | |
| CN113284094A (en) | Method, device, storage medium and equipment for acquiring defect information of glass substrate | |
| CN111028207B (en) | Button flaw detection method based on instant-universal feature extraction network | |
| CN111879972A (en) | Workpiece surface defect detection method and system based on SSD network model | |
| Tian et al. | Corrosion identification of fittings based on computer vision | |
| CN115266728A (en) | Method for identifying inclusion defects on surface of cold-rolled strip steel | |
| CN108182676B (en) | Sperm fragmentation rate detection method, device and equipment and readable storage medium | |
| CN109102486B (en) | Surface defect detection method and device based on machine learning | |
| CN113421235A (en) | Cigarette positioning device and method based on deep learning | |
| CN113298857A (en) | Bearing defect detection method based on neural network fusion strategy | |
| CN112967224A (en) | Electronic circuit board detection system, method and medium based on artificial intelligence | |
| CN112967226A (en) | High-reflectivity product detection system, method and medium based on artificial intelligence | |
| CN109272480B (en) | Detection method for boundary of immiscible solution | |
| CN116381053A (en) | Ultrasonic detection method and system for welding metal materials |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |