Control Method, Device and System for Rating Carbide Distribution in Bars

Abstract

The present application provides a control method, device and system for the carbide distribution rating of bars, belonging to the technical field of metal material detection. The system includes: an electron microscopy subsystem, a server and a control device; the method includes: obtaining a rating field-of-view image of a target bar sample collected by the electron microscopy subsystem; performing sample area extraction and sample boundary determination processing on the rating field-of-view image, and generating an acquisition control signal for the electron microscopy subsystem according to the boundary determination result; after the acquisition of the rating field-of-view image is completed, performing stitching processing on multiple frames of rating field-of-view images; extracting features from the stitched image, and comparing the feature extraction result with the carbide standard to determine the bar carbide distribution rating result. The present application can realize the automatic control of the acquisition, processing, feature extraction and rating of the rating field-of-view image in the whole process of the bar carbide distribution rating, and improve the evaluation accuracy and the stability of the rating standard.

Description

Technical Field

[0001] This application relates to the technical field of metal material testing, and particularly to a control method, device and system for the carbide distribution rating of bars. Background Art

[0002] Metallographic inspection in the iron and steel metallurgy industry is one of the important bases for the quality of steel products, and also one of the important bases for solving problems in the production process. The detection of the distribution of banded carbides is an important index in metallographic structure detection. In the traditional metallographic inspection of the distribution of banded carbides, laboratory testers mainly observe, identify and analyze the state and distribution of banded carbides in special bars through vertical microscopes and horizontal microscopes, so as to judge the quality of metal materials.

[0003] In the traditional detection method, laboratory testers obtain the state and distribution of banded carbides by observing under a microscope and complete the rating work relying on personal experience. The existing detection scheme has problems of poor accuracy and stability in metallographic analysis rating. Summary of the Invention

[0004] Embodiments of this application provide a control method, device and system for the carbide distribution rating of bars to solve the problems of poor accuracy and stability in metallographic analysis rating in the existing rating scheme for the carbide distribution of bars.

[0005] In a first aspect, embodiments of this application provide a control method for the carbide distribution rating of bars. The carbide distribution rating system of the bars includes: an electron microscopy subsystem, a server and a control device; the method includes:

[0006] Obtain a rating field-of-view image of a target bar sample collected by the electron microscopy subsystem;

[0007] Perform sample area extraction and sample boundary determination processing on the rating field-of-view image, and generate an acquisition control signal for the electron microscopy subsystem according to the boundary determination result; wherein, the acquisition control signal is used to control the electron microscopy subsystem to adjust the detection area of the target bar sample to obtain multiple frames of rating field-of-view images of the target bar sample;

[0008] After the acquisition of the rating field-of-view images is completed, perform stitching processing on multiple frames of rating field-of-view images;

[0009] Extract features from the stitched image, and compare the feature extraction result with the carbide standard to determine the rating result of the bar carbide distribution.

[0010] In a possible implementation manner, the performing sample area extraction and sample boundary determination processing on the rating field-of-view image includes:

[0011] Call the region extraction algorithm and the boundary determination algorithm;

[0012] Extract the sample region from the rating field of view image based on the region extraction algorithm, and determine the sample boundary of the extracted image based on the boundary determination algorithm;

[0013] Among them, the region extraction algorithm is a segmentation algorithm, which realizes pixel-level recognition and extraction of the cross-section of the target bar sample in the rating field of view image by segmenting the rating field of view image into multi-grid images; the boundary determination algorithm determines the sample boundary based on multiple row vectors, multiple column vectors and the effective region threshold of the rating field of view image.

[0014] In a possible implementation, the stitching process for multiple frames of rating field of view images includes:

[0015] Call the image stitching algorithm;

[0016] Stitch multiple frames of rating field of view images based on the image stitching algorithm;

[0017] Among them, the image stitching algorithm stitches images based on the acquisition order of multiple frames of rating field of view images.

[0018] In a possible implementation, the feature extraction of the stitched image includes:

[0019] Call the trained special bar carbide detection algorithm;

[0020] Extract features from the stitched image based on the special bar carbide detection algorithm;

[0021] Among them, the special bar carbide detection algorithm is trained based on the Feature Pyramid Network (FPN) and the Pyramid Attention Network (PAN).

[0022] In a possible implementation, before obtaining the rating field of view image of the target bar sample collected by the electron microscopy subsystem, it further includes:

[0023] Obtain the rating start signal of the target bar sample; among them, the rating start signal is set input information or scanning information corresponding to the target bar sample;

[0024] Generate a start control signal for the electron microscopy subsystem according to the rating start signal, and send the start control signal to the electron microscopy subsystem;

[0025] Obtain the temporary field-of-view image collected by the electron microscopy subsystem in response to the start control signal, and generate an initial acquisition control signal when the target bar sample is included in the temporary field-of-view image and the target bar sample is placed at the target position.

[0026] In a possible implementation manner, after determining the rating result of the carbide distribution of the bar by comparing the feature extraction result with the carbide standard, it further includes:

[0027] Generate rating record information according to the target bar sample batch information, the rating field-of-view image, and the rating result of the bar carbide distribution;

[0028] Send the evaluation record information to the server for storage.

[0029] In a possible implementation manner, the method further includes:

[0030] Obtain a query instruction for historical rating record information, and send the query instruction to the server;

[0031] Obtain the historical rating record query result fed back by the server, and display the rating query result; wherein, the rating query result includes one or more of the bar sample batch information, magnification, number of strip particle bands, width of strip particle bands, and bar carbide distribution grade.

[0032] In a second aspect, an embodiment of the present application provides a control device for rating the carbide distribution of bars, including:

[0033] An acquisition module, configured to acquire a rating field-of-view image of a target bar sample collected by the electron microscopy subsystem;

[0034] A control module, configured to perform sample area extraction and sample boundary determination processing on the rating field-of-view image, and generate an acquisition control signal for the electron microscopy subsystem according to the boundary determination result; wherein, the acquisition control signal is used to control the electron microscopy subsystem to adjust the detection area of the target bar sample to acquire multiple frames of the rating field-of-view image of the target bar sample;

[0035] An image stitching module, configured to perform stitching processing on multiple frames of rating field-of-view images after the acquisition of the rating field-of-view images is completed;

[0036] An image feature extraction module, configured to extract features from the stitched image;

[0037] A rating module, configured to compare the feature extraction result with the carbide standard to determine the rating result of the bar carbide distribution.

[0038] In a possible implementation manner, the control module is specifically configured to:

[0039] Call a region extraction algorithm and a boundary determination algorithm;

[0040] Extract a sample region from the rating field-of-view image based on the region extraction algorithm, and determine the sample boundary of the extracted image based on the boundary determination algorithm;

[0041] Wherein, the region extraction algorithm is a segmentation algorithm, and the pixel-level recognition and extraction of the cross-section of the target bar sample in the rating field-of-view image is realized by segmenting the rating field-of-view image into multi-grid images; the boundary determination algorithm determines the sample boundary based on multiple row vectors, multiple column vectors and an effective region threshold of the rating field-of-view image.

[0042] In a possible implementation manner, the image stitching module is specifically configured to:

[0043] Call an image stitching algorithm;

[0044] Perform stitching processing on multiple frames of rating field-of-view images based on the image stitching algorithm;

[0045] Wherein, the image stitching algorithm performs image stitching based on the acquisition order of multiple frames of rating field-of-view images.

[0046] In a possible implementation manner, the image feature extraction module is specifically configured to:

[0047] Call a trained special bar carbide detection algorithm;

[0048] Extract features from the stitched image based on the special bar carbide detection algorithm;

[0049] Wherein, the special bar carbide detection algorithm is trained based on FPN and PAN.

[0050] In a possible implementation manner,

[0051] The obtaining module is further configured to obtain a rating start signal of the target bar sample before obtaining the rating field-of-view image of the target bar sample collected by the electron microscopy subsystem; wherein, the rating start signal is set input information or scanning information corresponding to the target bar sample;

[0052] The control module is further configured to generate a start control signal for the electron microscopy subsystem according to the rating start signal, and send the start control signal to the electron microscopy subsystem;

[0053] The obtaining module is further configured to obtain a temporary field-of-view image collected by the electron microscopy subsystem in response to the start control signal;

[0054] The control module is further configured to generate an initial acquisition control signal when the target bar sample is included in the temporary field-of-view image and the target bar sample is placed at the target position.

[0055] In a possible implementation manner, after determining the bar carbide distribution rating result by comparing the feature extraction result with the carbide standard, the generating module is further configured to generate rating record information according to the target bar sample batch information, the rating field-of-view image, and the bar carbide distribution rating result;

[0056] The control device for bar carbide distribution rating further includes: a sending module, configured to send the evaluation record information to the server for storage.

[0057] In a possible implementation manner, the obtaining module is further configured to obtain a query instruction for historical rating record information;

[0058] The sending module is further configured to send the query instruction to the server;

[0059] The obtaining module is further configured to obtain a historical rating record query result fed back by the server;

[0060] The control device for bar carbide distribution rating further includes: a display module, configured to display a rating query result; wherein, the rating query result includes one or more of bar sample batch information, magnification, number of strip particle bands, strip particle band width, and bar carbide distribution grade.

[0061] In a third aspect, an embodiment of the present application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the method described in the first aspect or any possible implementation manner of the first aspect are implemented.

[0062] In a fourth aspect, an embodiment of the present application provides a system for bar carbide distribution rating, including: an electron microscopy subsystem, a server, and a control device;

[0063] Wherein, the electron microscopy subsystem includes: an electron microscope, a stage module, a driving motor, and a controller; wherein, the driving motor is configured to drive the stage module to move under the control of the controller to adjust the detection area of the electron microscope eyepiece for the target bar sample; the electron microscope is configured to collect a field-of-view image under the control of the controller;

[0064] The server is communicatively connected to the control device and is used to store the rating field-of-view images of the target bar sample and the rating results of the carbide distribution of the bar.

[0065] The control device is used to execute the steps of the method described in the first aspect or any possible implementation manner of the first aspect above.

[0066] In a fifth aspect, an embodiment of the present application provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the steps of the method described in the first aspect or any possible implementation manner of the first aspect above are implemented.

[0067] An embodiment of the present application provides a control method, device, and system for rating the carbide distribution of bars. By obtaining the rating field-of-view images of the target bar sample collected by the electron microscopy subsystem; performing sample area extraction and sample boundary determination processing on the rating field-of-view images, and generating an acquisition control signal for the electron microscopy subsystem according to the boundary determination result, so as to realize the automatic control of the electron microscopy subsystem by the control device based on the rating field-of-view images of the target bar sample collected by the electron microscopy subsystem, realize precise control of adjusting the detection area of the target bar sample, thereby improving the stitching quality of multiple frames of rating field-of-view images, and further improving the accuracy of rating the carbide distribution of bars. After the acquisition of the rating field-of-view images is completed, perform stitching processing on multiple frames of rating field-of-view images, extract features from the stitched images, and compare the feature extraction results with the carbide standard to determine the rating results of the carbide distribution of the bars. Perform the rating of the carbide distribution of the bars based on the quantified carbide standard, and improve the stability of the rating results of the carbide distribution of the bars. The embodiment of the present application can realize the automatic control of the acquisition, processing, feature extraction, and rating of the rating field-of-view images in the whole process of rating the carbide distribution of the bars, and improve the evaluation accuracy and the stability of the rating standard. Description of the Drawings

[0068] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for use in the embodiments or the description of the prior art. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0069] Figure 1 is an application scenario diagram of a control method for rating the carbide distribution of bars provided by an embodiment of the present application;

[0070] Figure 2 is an implementation flowchart of a control method for rating the carbide distribution of bars provided by an embodiment of the present application;

[0071] Figure 3 It is a flowchart of the implementation of a control method for the carbide distribution rating of bars provided in another embodiment of the present application;

[0072] Figure 4 It is a schematic structural diagram of a control device for the carbide distribution rating of bars provided in an embodiment of the present application;

[0073] Figure 5 It is a schematic structural diagram of an electronic device provided in an embodiment of the present application. Detailed implementation manners

[0074] In the following description, for the purpose of illustration rather than limitation, specific details such as specific system architectures and technologies are presented in order to thoroughly understand the embodiments of the present application. However, those skilled in the art should clearly understand that the present application can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present application.

[0075] The terms "first", "second", etc. in the specification, claims, and above-mentioned drawings of the embodiments of the present application are used to distinguish similar objects and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so as to implement the embodiments of the present application described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion.

[0076] Unless otherwise stated, the term "plurality" means two or more. The character " / " indicates that the objects before and after are in an "or" relationship. For example, A / B means: A or B. The term "and / or" is an associative relationship describing an object and indicates that three relationships can exist. For example, A and / or B means: A or B, or, the three relationships of A and B.

[0077] The terms used in this application are only for describing embodiments and do not limit the claims. As used in the description of the embodiments and the claims, unless the context clearly indicates otherwise, the singular forms "a", "an", and "the" are intended to also include the plural forms. Similarly, the term "and / or" as used in this application refers to any and all possible combinations of one or more of the associated listed items. Additionally, when used in this application, the term "comprise" and its variants "comprises" and / or "comprising" etc. mean the presence of the stated features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or groupings of these. Without further limitation, an element defined by the statement "comprising one..." does not exclude the presence of additional identical elements in the process, method, or device comprising the element.

[0078] In this application, what each embodiment focuses on can be the differences from other embodiments, and the same or similar parts among the embodiments can be referred to each other. For the methods, products, etc. disclosed in the embodiments, if they correspond to the method part disclosed in the embodiments, the relevant parts can refer to the description of the method part.

[0079] In the traditional rating scheme for carbide distribution in bars, there are the following problems:

[0080] On the one hand, laboratory testers obtain the state and distribution of banded carbides by observing under a microscope and complete the rating work relying on personal experience, without a unified metallographic detection standard.

[0081] On the other hand, laboratory testers have a high work intensity, with dry eyes and being prone to fatigue, and are likely to cause inaccurate metallographic analysis and rating due to factors such as fatigue, mood, and working state.

[0082] Due to the above factors, the accuracy and stability of metallographic analysis rating are not strong.

[0083] The main objective of the present invention is to provide a special bar carbide distribution rating system based on artificial intelligence technology to solve problems in the traditional metallographic detection process, such as excessive dependence on human experience in detection results, many influencing factors, difficult preservation of process data, and difficult traceability of the process. By means of informatization, sample images and video information in the process of special bar carbide detection are recorded, and through intelligent algorithms, rapid determination of sample grades is realized, reducing the work intensity of front-line testers and improving the efficiency of metallographic detection, the accuracy and stability of metallographic analysis rating.

[0084] Figure 1It is an application scenario diagram of a control method for rating the carbide distribution of bars provided by an embodiment of the present application. As Figure 1 shown, the bar carbide distribution rating system includes: an electron microscopy subsystem, a server, and a control device.

[0085] Among them, the electron microscopy subsystem is connected to the control device, and is used to complete the metallographic inspection of the target bar sample under the control of the control device, and send the acquired field-of-view image of the electron microscope eyepiece to the control device, and the control device side completes image processing, bar carbide distribution rating, and display of information such as the field-of-view image and the rating result.

[0086] The server is connected to the control device through a wireless communication network, and is used to store information such as the rated field-of-view image of the bar sample and the bar carbide distribution rating result, so as to facilitate the subsequent traceability of the rating result and the rating process.

[0087] In the specific implementation process, the control device or an image acquisition device externally connected to the control device records the video of the bar carbide distribution rating site, and sends the on-site video to the server side for storage, realizing the recording and storage of the whole process data of the special bar carbide rating, and ensuring that the rating process is traceable, quantifiable, and retrievable.

[0088] Optionally, a metallographic microscope is used in the electron microscopy subsystem to improve the detection accuracy. When used in the bar carbide distribution rating system, it is transmitted through a wired connection. The transmission methods are mainly the interaction of data among the metallographic microscope, the control device, and the server; the interaction of video data and picture extraction between the metallographic microscope and the control device; the interaction of data between the server and the control device, such as: algorithm call or rating record data transmission, etc.

[0089] In the embodiment of the present application, the relevant description of the field-of-view image collected by the electron microscopy subsystem is also the field-of-view image of the electron microscope eyepiece in the electron microscopy subsystem.

[0090] To make the purpose, technical solution, and advantages of the present application clearer, the following will be described through specific embodiments in conjunction with the accompanying drawings.

[0091] Figure 2 It is a flowchart of the implementation of a control method for rating the carbide distribution of bars provided by an embodiment of the present application. As Figure 2 shown, the method includes the following steps:

[0092] S201, obtain the rated field-of-view image of the target bar sample collected by the electron microscopy subsystem.

[0093] In this embodiment, the execution subject of the control method for the carbide distribution rating of bars is a control device. The control device is a terminal device such as a desktop computer, a laptop computer, or a tablet computer, which mainly realizes the interaction between laboratory testers and the system. For example, it includes functions such as viewing the real-time rating status, viewing historical rating records, and scanning barcodes.

[0094] Among them, the target bar sample is the sample material of special bars to be inspected or rated after steel production. By randomly intercepting the special bars, samples of a set volume size (usually the size of a fingernail) are obtained, and the cross-section is polished to ensure that the cross-section is flat and smooth. In this embodiment, the target bar sample collected in a random manner is a local material of the corresponding bar, and its sampling randomness ensures that it can reflect the overall carbide distribution of the corresponding bar. Therefore, the carbide distribution rating result of the target bar sample is the carbide distribution rating result of the corresponding bar.

[0095] In the specific implementation process, to improve the accuracy of the carbide distribution rating of bars, it is necessary to randomly obtain samples from the special bars multiple times, that is, the number of target bar samples for the same special bar is multiple. Correspondingly, the samples of the same special bar need to be uniformly packed into a sample bag, and a QR code, a barcode, or an electronic tag, etc. is posted. Among them, the QR code, the barcode, or the electronic tag stores at least the batch number of the special bar. In addition, other information of the special bar other than the batch number is also stored, such as production time, production line, and other information.

[0096] In addition, the rating field-of-view image is the field-of-view image collected by the eyepiece of the electron microscope, and the rating field-of-view image contains at least the image information of the target bar sample. When the target bar sample is not placed in place, the eyepiece of the electron microscope will also collect the images of the edge of the target bar sample and non-target bar samples. Optionally, the proportion of the image area of the target bar sample in the rating field-of-view image is greater than a set ratio value. Optionally, the set value is 70-90. Optionally, the set value is 70%, 80%, or 90%. Among them, the large proportion of the image of the target bar sample in the rating field-of-view image is of reference value for analyzing the carbide distribution and rating of the bar. Therefore, the rating field-of-view images are screened based on the proportion of the image area of the target bar sample, reducing the number of irrelevant image samples, and mainly performing image processing and data analysis on the images with a large proportion of the image of the target bar sample, improving the efficiency of the carbide distribution rating of the bar.

[0097] S202. Extract the sample area and determine the sample boundary of the rating field-of-view image, and generate an acquisition control signal for the electron microscopy subsystem according to the boundary determination result; the acquisition control signal is used to control the electron microscopy subsystem to adjust the detection area of the target bar sample to obtain multiple frames of rating field-of-view images of the target bar sample.

[0098] In this embodiment, to achieve the carbide distribution rating of the target bar sample based on the rating field-of-view images, multiple frames of rating field-of-view images are required. The field-of-view image of the complete target bar sample is obtained by stitching the multiple frames of rating field-of-view images. In other possible implementation manners, to ensure the quality of the obtained rating field-of-view images for the same detection area, repeated sampling is performed a set number of times for the same detection area, and the optimal image or the average value of multiple frames of images in a certain detection area is used as the rating field-of-view image representing the detection area. Optionally, the set number of times is 3 to 5 times, that is, 3 to 5 frames of field-of-view images are sampled for the same detection area, and the rating field-of-view image of the detection area is determined based on the 3 to 5 frames of sampled field-of-view images. In one possible implementation manner, the set number of times is 3 times, 4 times, or 5 times.

[0099] In addition, the eyepiece of the electron microscope will magnify and display the image. Therefore, slight movement when manually moving the specimen slide or the stage will cause a large difference in the field-of-view image area in the eyepiece, and the field-of-view images before and after adjustment cannot be effectively connected. In this embodiment, by performing sample area extraction and sample boundary determination processing on the rating field-of-view images, and generating an acquisition control signal for the electron microscopy subsystem according to the boundary determination result, the automation control of the electron microscopy subsystem, especially the movement of the specimen slide or the stage, is realized, the fine control of the detection area adjustment of the target bar sample is improved, thereby improving the stitching quality of multiple frames of rating field-of-view images, and further improving the accuracy of the bar carbide distribution rating.

[0100] S203. After the acquisition of the rating field-of-view images is completed, perform stitching processing on the multiple frames of rating field-of-view images.

[0101] Among them, when single image sampling is performed for the detection area of a certain target bar sample, the complete image of the cross-section of the target bar sample is obtained by stitching the rating field-of-view images corresponding to each detection area.

[0102] When multiple repeated image samplings are performed for the detection area of a certain target bar sample, preprocessing is respectively performed on the multiple frames of sampled images in each monitoring area, the image obtained after preprocessing is used as the rating field-of-view image of the detection area, and stitching processing is performed based on the rating field-of-view images of each detection area to obtain the complete image of the cross-section of the target bar sample.

[0103] S204. Extract features from the stitched image, and compare the feature extraction result with the carbide standard to determine the bar carbide distribution rating result.

[0104] It is necessary to adjust the hardness and strength of steel for specific applications. For special steel with high hardness requirements, tempering treatment etc. is usually required before use to adjust mechanical properties. During this process, carbon will precipitate in the form of carbides. The content and size of carbides will affect the microstructure of the material. Among them, the spliced image contains other components besides carbides. To achieve the rating of carbide distribution in the bar, it is necessary to extract the features of carbides to facilitate the determination of the size and content of carbides.

[0105] In addition, to solve the problem of poor rating stability caused by relying on personal experience in the manual rating of carbide distribution in bars, the embodiment of this application provides a quantitative carbide standard, and comprehensively controls the automatic and precise acquisition of the rating field image and the artificial intelligence-based image processing, so as to improve the stability of the carbide distribution rating of bars as a whole.

[0106] In this embodiment, by obtaining the rating field image of the target bar sample collected by the electron microscopy subsystem; performing sample area extraction and sample boundary determination processing on the rating field image, and generating an acquisition control signal for the electron microscopy subsystem according to the boundary determination result, so as to realize the automatic control of the electron microscopy subsystem based on the rating field image of the target bar sample collected by the electron microscopy subsystem, realize the precise control of the detection area of the target bar sample, thereby improving the splicing quality of multiple frames of rating field images, and further improving the accuracy of the carbide distribution rating of bars. After the acquisition of the rating field image is completed, perform splicing processing on multiple frames of rating field images, perform feature extraction on the spliced image, and compare the feature extraction result with the carbide standard to determine the carbide distribution rating result of the bar, and perform the carbide distribution rating of the bar based on the quantitative carbide standard to improve the stability of the carbide distribution rating result of the bar. The embodiment of this application can realize the automatic control of the acquisition, processing, feature extraction and rating of the rating field image in the whole process of the carbide distribution rating of the bar, and improve the evaluation accuracy and the stability of the rating standard.

[0107] In different embodiments, there are various ways to perform sample area extraction, sample boundary determination, splicing processing and feature extraction on the image in steps S202, S203 and S204.

[0108] In a possible implementation manner, first perform sample boundary determination on the image, and then, after the sample boundary determination is completed, perform sample area extraction based on the sample boundary determination result, which can improve the sample area extraction efficiency.

[0109] In a possible implementation manner, the processing of sample area extraction and sample boundary determination on the rating field image in step S202 includes:

[0110] Call the area extraction algorithm and the boundary determination algorithm;

[0111] The sample area of the rating field image is extracted based on the area extraction algorithm, and the sample boundary of the extracted image is determined based on the boundary determination algorithm;

[0112] Among them, the area extraction algorithm is a segmentation algorithm, which realizes the pixel-level recognition and extraction of the cross-section of the target bar sample in the rating field image by dividing the rating field image into multi-grid images; the boundary determination algorithm determines the sample boundary based on multiple row vectors, multiple column vectors and the effective area threshold.

[0113] Among them, performing boundary judgment after the sample area extraction is completed helps to improve the accuracy of multi-frame rating field images. Optionally, the area extraction algorithm and the boundary determination algorithm are stored on the server side or in the system of the control device, which is convenient for quick calling when performing sample area extraction and sample boundary determination processing on the rating field image.

[0114] In the specific implementation process, when controlling the electron microscopy subsystem to adjust the detection area of the target bar sample, that is, when controlling the movement of the slide or the stage, the field of view of the detected target bar sample is automatically recognized, and the cross-section of the sample is pixel-level recognized and extracted, and the quality of the extracted image is judged to ensure that the field of view picture is clear.

[0115] Optionally, the area extraction algorithm is the yolov5-seg segmentation algorithm, which divides the input image into multiple grids and predicts multiple bounding boxes for each grid. Each bounding box contains information such as the location, category and confidence of the target. The confidence and category probability of each bounding box are:

[0116]

[0117]

[0118] Among them, is the confidence of the th bounding box, is the confidence that the bounding box contains the target, represents the category of the target contained in the bounding box, represents the range of the bounding box, and Pr represents the probability that the target belongs to the detected sample category under the condition of the given bounding box . In the embodiments of the present application, the detected sample categories include the cross-section image of the target bar sample and the non-sample cross-section image. Then the probability that the target belongs to the detected sample category is the probability that the image information contained in a certain grid based on the yolov5-seg segmentation algorithm is the cross-section image information of the target bar sample, or the probability of non-sample cross-section image information.

[0119] On this basis, a segmentation head is added to perform pixel-level segmentation on the image. Specifically, yolov5-seg can be expressed as:

[0120]

[0121] Among them, represents the segmentation head of yolov5-seg, which can predict whether the image information corresponding to each pixel belongs to the cross-sectional image of the target bar sample to be measured according to the feature map.

[0122] In addition, during the process of controlling the electron microscopy subsystem to adjust the detection area of the target bar sample, through the boundary determination algorithm, the boundary of the target bar sample to be detected is automatically judged. When the algorithm detects the sample boundary, the automatic frame extraction action of the image stops, the region extraction algorithm stops, and the electron microscopy subsystem stops adjusting the detection area of the target bar sample, that is, the glass slide or the stage stops moving. The execution process of the boundary determination algorithm is as follows:

[0123] Randomly select 3 row vectors by row from the multi-frame rated field-of-view images to be detected 、 、 , and randomly select 3 column vectors by column 、 、 , calculate the mean vectors of the 3 row vectors and column vectors 、 , set the effective area threshold , select and less than coordinates, and obtain the coordinates of the effective area , 、 、 、 respectively represent the positions of the left, right, upper, and lower four boundaries of the distance matrix boundary:

[0124]

[0125] Among them, when the mean vector of the row vector is between the coordinates of the effective area and corresponding set values, the upper or lower boundary of the sample is not detected, otherwise, the sample boundary is detected. Similarly, when the mean vector of the column vector is between the coordinates of the effective area and corresponding set values, the left or right boundary of the sample is not detected, otherwise, the sample boundary is detected.

[0126] In this embodiment, based on the segmentation algorithm as the region extraction algorithm, pixel-level recognition and extraction of the cross-section of the target bar sample in the rating field of view image are realized, the image extraction accuracy is improved, and the boundary determination algorithm is integrated. Based on multiple row vectors, multiple column vectors and the effective region threshold, the sample boundary is determined, avoiding deviations in the single boundary information acquisition caused by the unstable movement of the stage or light refraction, improving the accuracy of the control of the electron microscopy subsystem, efficiently controlling the electron microscopy subsystem to adjust the detection area of the target bar sample, and optimizing the sampling result of the rating field of view image of the target bar sample.

[0127] In a possible implementation manner, the stitching process of multiple frames of rating field of view images in step S203 includes:

[0128] Invoking the image stitching algorithm;

[0129] Stitching multiple frames of rating field of view images based on the image stitching algorithm;

[0130] Among them, the image stitching algorithm stitches images based on the acquisition order of multiple frames of rating field of view images.

[0131] Optionally, the image stitching algorithm is stored on the server side or in the system of the control device, which is convenient for quick invocation when stitching the rating field of view images.

[0132] After multiple acquisitions of the current sample field of view pictures are completed, by adopting enhancement algorithms, noise reduction algorithms, etc., pictures that are blurred, unclear, and of low quality are automatically screened, and the picture with the best clarity is retained. The overall image is stitched according to the picture acquisition order to obtain a complete detection image of the strip carbide distribution of special bars. The image can be expressed as:

[0133]

[0134] In a possible implementation manner, the feature extraction of the stitched image in step S204 includes:

[0135] Invoking the trained detection algorithm for strip carbides of special bars;

[0136] Performing feature extraction on the stitched image based on the detection algorithm for strip carbides of special bars;

[0137] Among them, the detection algorithm for strip carbides of special bars is trained based on FPN and PAN.

[0138] Among them, optionally, the image stitching algorithm is stored on the server side or in the system of the control device, which is convenient for quick invocation when stitching the rating field of view images.

[0139] In this embodiment, the special bar-shaped carbide detection algorithm adopts a method combining FPN and PAN. In the FPN+PAN structure, FPN is used to generate feature maps of different resolutions, and PAN is used to fuse these feature maps to obtain a more accurate feature representation. The feature map fused by FPN+PAN can be expressed as:

[0140]

[0141] Among them, 、 、 and respectively represent the feature maps of different resolutions obtained by fusing through PAN. Specifically, for the feature map of each resolution, it can be calculated through the following formula:

[0142]

[0143]

[0144]

[0145]

[0146] Among them, 、 、 、 respectively represent the feature maps of different resolutions generated by FPN, 、 、 、 respectively represent the feature maps of different resolutions extracted by PAN, and PAN represents the function of feature fusion by the attention mechanism. By combining FPN and PAN, a more accurate multi-scale feature representation can be obtained, improving the performance of object detection. A fully connected network is added later to solve for the distribution level of the strip-shaped carbide.

[0147] According to the above special bar-shaped carbide detection algorithm, the spliced special bar-shaped carbide distribution image is compared with the carbide standard library to quickly and accurately identify feature information such as the number of strip-shaped carbides and the bandwidth of strip-shaped particles, and calculate the accurate grading information.

[0148] The foregoing embodiment mainly introduces the control process of obtaining the rating field-of-view image of the target bar sample and the processing process of the rating field-of-view image. In the specific implementation process, before sampling the rating field-of-view image, it is necessary to ensure that the placement position of the target bar sample meets the requirement that the electron microscopy subsystem can collect the image of the target bar sample and the proportion in the field-of-view image is greater than the set proportion value.

[0149] In a possible implementation, before obtaining the rating field image of the target bar sample collected by the electron microscopy subsystem in step S201, it further includes:

[0150] Obtaining a rating start signal for the target bar sample; wherein, the rating start signal is set input information or scanning information corresponding to the target bar sample;

[0151] Generating a start control signal for the electron microscopy subsystem according to the rating start signal, and sending the start control signal to the electron microscopy subsystem;

[0152] Obtaining a temporary field image collected by the electron microscopy subsystem in response to the start control signal, and generating an initial acquisition control signal when the target bar sample is included in the temporary field image and the target bar sample is placed at the target position.

[0153] Among them, in different embodiments, the methods of starting the rating process of the target bar sample are different.

[0154] In a possible implementation, the rating start signal is generated in response to a relevant laboratory tester triggering the start button through a control device, triggering the start of a virtual button or option, or in response to a corresponding voice command.

[0155] In another possible implementation, the rating start signal is generated in response to the action of collecting the basic information of the target bar sample. For example: the relevant laboratory tester inputs the required data items through the basic information collection interface of the target bar sample in the relevant rating system, and generates a rating start signal in response to the completion of the input of the required data items, or the relevant laboratory tester scans the barcode, QR code or electronic tag corresponding to the target bar sample and storing the basic information of the target bar sample, and generates a rating start signal in response to reading the basic information of the target bar sample.

[0156] Generally, before and after the laboratory tester performs the operation related to generating the rating start signal, the target bar sample will be placed on the stage of the electron microscope. When it is not placed at the set position, for example: the image of the target bar sample cannot be collected in the electron microscope eyepiece or the proportion of the image of the target bar sample collected is less than the set ratio, the field object collected by the current eyepiece has little reference value for the rating of the carbide distribution of the bar, and the image at this time is not used as the rating field image, that is, it does not participate in subsequent operations such as image stitching and feature extraction. In addition, to adjust the target bar sample to the set position, the electron microscopy system can still be automatically adjusted through the control device to complete the initial adjustment before the rating field image is collected.

[0157] Among them, in this embodiment, during the initial adjustment process before collecting the rated field of view image, the control device is used to obtain the temporary field of view image collected by the electron microscopy subsystem, determine whether there is an image of the target bar sample and the position of the target bar sample in the field of view, and determine the adjustment direction of the slide or stage of the electron microscopy subsystem based on the position of the target bar sample in the field of view. Correspondingly, the control signal in the initial adjustment stage is different from the acquisition control signal in the rated field of view image, that is, the initial acquisition control signal.

[0158] Optionally, the difference between the initial acquisition control signal and the acquisition control signal for the rated field of view image is reflected in the different image sampling frequencies and the adjustment value sizes of the detection regions corresponding to the fields of view. For example, when there is no target bar sample in the temporary field of view image collected by the electron microscopy subsystem, the adjustment value is larger and greater than the adjustment value in the acquisition control signal for the rated field of view image to quickly adjust the target bar sample to the proper position.

[0159] In this embodiment, after obtaining the rating start signal of the target bar sample, a start control signal is generated to control the electron microscopy subsystem to start and collect a temporary field of view image, so as to determine whether the target bar sample is placed in place based on the temporary field of view image. When the target bar sample is placed in place, control the electron microscopy subsystem to start collecting the rated field of view image; otherwise, perform initial acquisition control on the electron microscopy subsystem to adjust the target bar sample to the proper position.

[0160] In a possible implementation manner, after step S204 of comparing the feature extraction result with the carbide standard to determine the bar carbide distribution rating result, it further includes:

[0161] Generate rating record information according to the target bar sample batch information, the rated field of view image, and the bar carbide distribution rating result;

[0162] Send the evaluation record information to the server for storage.

[0163] In the specific implementation process, after the rating of the bar carbide distribution is completed, the functions such as determining the adjustment process of the bar smelting process and determining the use of the bar can be realized by using the bar carbide distribution rating result. When specifically using the bar carbide distribution rating result to realize different functions, historical rating results need to be obtained. Therefore, for the convenience of historical rating traceability query, the bar carbide distribution rating system will also store the bar carbide distribution rating result and the corresponding rated field of view image.

[0164] In addition, the bar carbide distribution rating system will perform ratings on various different bar samples according to specific requirements. For the convenience of accurately querying the historical rating results of a certain bar sample, rating record information is generated by integrating the target bar sample batch information, the rated field of view image, and the bar carbide distribution rating result.

[0165] In this embodiment, after the rating is completed, rating record information is generated by integrating the batch information of the target bar sample, the rating field of view image, and the rating result of the carbide distribution of the bar, which is convenient for giving guidance information during the generation or use of the bar. At the same time, the evaluation record information is sent to the server for storage to reduce the data storage pressure of the control device.

[0166] In a possible implementation, the method further includes:

[0167] Obtaining a query instruction for historical rating record information and sending the query instruction to the server;

[0168] Obtaining the historical rating record query result fed back by the server and displaying the rating query result; wherein, the rating query result includes one or more of the batch information of the bar sample, the magnification, the number of strip particle bands, the width of the strip particle bands, and the carbide distribution grade of the bar.

[0169] In this embodiment, the control terminal interacts with the server to complete the query and interactive display of the historical rating record information. Among them, based on the batch information of the bar sample, one or more of the magnification, the number of strip particle bands, the width of the strip particle bands, and the carbide distribution grade of the bar can be queried.

[0170] Figure 3 It is a flowchart of the implementation of the control method for the carbide distribution rating of bars provided in another embodiment of the present application. As Figure 3 shown, the method includes the following steps:

[0171] S301, the grading starts;

[0172] Among them, the grading starts specifically based on the grading start signal. Optionally, the grading start signal is generated in response to the relevant laboratory testing personnel triggering the start button through the control device, triggering the start of the virtual button or option, in response to the corresponding voice instruction, in response to the completion of the input of the required data item, or in response to reading the basic information of the target bar sample.

[0173] In a specific embodiment, after the steel production is completed with special bars, a bar sample about the size of a fingernail is randomly intercepted, and the cross-section is polished to ensure that the cross-section is flat and smooth, and then uniformly packed into a sample bag, and a two-dimensional code of the special bar batch number is pasted. The bar sample is placed on the stage of the metallographic electron microscope, and the metallographic electron microscope is connected to the tablet computer. The two-dimensional code of the special bar batch number on the sample bag is scanned by the tablet computer, and the rating starts.

[0174] S302, the target bar sample is placed on the stage, ensuring that the smooth surface faces the observation hole;

[0175] S303, call the algorithm service;

[0176] In a specific embodiment, the algorithm is stored in the algorithm server. Calling the algorithm service means calling the image recognition algorithm from the algorithm server to determine whether the target bar sample is placed in place based on this algorithm.

[0177] S304, determine whether the placement is in place;

[0178] Among them, being placed in place means that the proportion of the target bar sample image in the field image collected by the electron microscope is greater than the set proportion value. The proportion of the target bar sample image is determined by the image recognition algorithm in step S303.

[0179] S305, the microscope controller works to slowly push the target bar sample;

[0180] The control device generates a microscope motion control signal according to the image recognition result. The microscope controller is connected to the driving motor and controls the movement of the slide or the stage according to the microscope motion control signal to slowly push the target bar sample to complete the image acquisition of the complete cross-section of the target bar sample.

[0181] S306, call the sample area extraction algorithm;

[0182] S307, call the sample boundary determination algorithm;

[0183] Among them, the sample area extraction algorithm and the sample boundary determination algorithm are the algorithms in the foregoing embodiments.

[0184] S308, the sample boundary is detected;

[0185] S309, the microscope controller stops working;

[0186] In the specific implementation process, the specific form of the sample boundary is determined by the sample cross-section shape. For example: for circular and rectangular samples, the sample boundaries are different. For circular samples, the boundary is an arc curve boundary, and for rectangular samples, the boundary is a straight line boundary.

[0187] In step S308, detecting the sample boundary specifically means detecting the set sample boundary. Usually, rectangular samples are used for the rating of the carbide distribution in the bar. Taking rectangular samples as an example, in the foregoing steps S301 - S305, the control starts the acquisition of the rating field image from a certain corner of the rectangular target bar sample and completes the acquisition of the rating field image when the diagonal is detected. When two sides intersect in the rating field image, the microscope controller is controlled to stop working and end the acquisition of the rating field image.

[0188] S310, call the picture stitching algorithm;

[0189] S311, call the detection algorithm for the distribution of strip carbides in special bar materials;

[0190] S312, determine the rating result of the carbide distribution in the bar material;

[0191] Among them, during the process of determining the detection and grading result, obtain the carbide standard, and determine the rating result of the carbide distribution in the bar material based on the comparison between the operation result of the detection algorithm for the distribution of strip carbides in special bar materials and the carbide standard.

[0192] S313, the grading ends.

[0193] Furthermore, to realize the traceability query of the historical distribution rating result, after step S313, it also includes: generating rating record information according to the batch information of the target bar material sample, the rating field-of-view image, and the rating result of the carbide distribution in the bar material; sending the evaluation record information to the server for storage.

[0194] It should be understood that the magnitudes of the sequence numbers of the steps in the above embodiments do not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0195] The following is the device embodiment of the present application. For the details not described in detail therein, reference can be made to the corresponding method embodiment above.

[0196] Figure 4 It is a schematic structural diagram of a control device for rating the carbide distribution in bar materials provided by an embodiment of the present application. As Figure 4 shown, for the convenience of description, only the parts related to the embodiments of the present application are shown. As Figure 4 shown, the device includes:

[0197] An acquisition module 401, configured to acquire the rating field-of-view image of the target bar material sample collected by the electron microscopy subsystem;

[0198] A control module 402, configured to perform sample area extraction and sample boundary determination processing on the rating field-of-view image, and generate an acquisition control signal for the electron microscopy subsystem according to the boundary determination result; wherein, the acquisition control signal is used to control the electron microscopy subsystem to adjust the detection area of the target bar material sample to obtain multiple frames of rating field-of-view images of the target bar material sample;

[0199] An image stitching module 403, configured to perform stitching processing on multiple frames of rating field-of-view images after the acquisition of the rating field-of-view images ends;

[0200] An image feature extraction module 404, configured to extract features from the stitched image;

[0201] A rating module 405, which is used to compare the feature extraction result with the carbide standard to determine the rating result of the carbide distribution of the bar.

[0202] In a possible implementation manner, the control module 402 is specifically configured to:

[0203] Call the region extraction algorithm and the boundary determination algorithm;

[0204] Based on the region extraction algorithm, perform sample region extraction on the rating field of view image, and based on the boundary determination algorithm, perform sample boundary determination on the extracted image;

[0205] Among them, the region extraction algorithm is a segmentation algorithm, which realizes pixel-level recognition and extraction of the cross-section of the target bar sample in the rating field of view image by segmenting the rating field of view image into multi-grid images; the boundary determination algorithm performs sample boundary determination based on multiple row vectors, multiple column vectors, and the effective region threshold.

[0206] In a possible implementation manner, the image stitching module 403 is specifically configured to:

[0207] Call the image stitching algorithm;

[0208] Based on the image stitching algorithm, perform stitching processing on multiple frames of rating field of view images;

[0209] Among them, the image stitching algorithm performs image stitching based on the acquisition order of multiple frames of rating field of view images.

[0210] In a possible implementation manner, the image feature extraction module 404 is specifically configured to:

[0211] Call the trained special bar strip carbide detection algorithm;

[0212] Based on the special bar strip carbide detection algorithm, perform feature extraction on the stitched image;

[0213] Among them, the special bar strip carbide detection algorithm is trained based on FPN and PAN.

[0214] In a possible implementation manner,

[0215] The acquisition module 401 is further configured to obtain a rating start signal of the target bar sample before obtaining the rating field of view image of the target bar sample collected by the electron microscopy subsystem; among them, the rating start signal is a set input information or the scanning information corresponding to the target bar sample;

[0216] The control module 402 is further configured to generate a start control signal for the electron microscopy subsystem according to the rating start signal, and send the start control signal to the electron microscopy subsystem;

[0217] The acquisition module 401 is further configured to acquire a temporary field-of-view image collected by the electron microscope subsystem in response to the start control signal;

[0218] The control module 402 is further configured to generate an initial acquisition control signal when the target bar sample is included in the temporary field-of-view image and the target bar sample is placed at the target position.

[0219] In a possible implementation manner, the generation module is further configured to generate rating record information according to the target bar sample batch information, the rating field-of-view image, and the bar carbide distribution rating result after determining the bar carbide distribution rating result by comparing the feature extraction result with the carbide standard;

[0220] The control device for bar carbide distribution rating further includes: a sending module, configured to send the evaluation record information to the server for storage.

[0221] In a possible implementation manner, the acquisition module 401 is further configured to acquire a query instruction for historical rating record information;

[0222] The sending module is further configured to send the query instruction to the server;

[0223] The acquisition module 401 is further configured to acquire the historical rating record query result fed back by the server;

[0224] The control device for bar carbide distribution rating further includes: a display module, configured to display the rating query result; wherein, the rating query result includes one or more of the bar sample batch information, magnification, number of strip particle bands, strip particle band width, and bar carbide distribution grade.

[0225] In this embodiment, a rated field-of-view image of a target bar sample collected by an electron microscopy subsystem is obtained; the sample area of the rated field-of-view image is extracted and the sample boundary is determined, and an acquisition control signal for the electron microscopy subsystem is generated according to the boundary determination result, so as to realize the automatic control of the electron microscopy subsystem by the control device based on the rated field-of-view image of the target bar sample collected by the electron microscopy subsystem, realize the precise control of the detection area of the target bar sample, thereby improving the stitching quality of multiple frames of rated field-of-view images, and further improving the accuracy of the carbide distribution rating of the bar. After the acquisition of the rated field-of-view images is completed, the multiple frames of rated field-of-view images are stitched, features are extracted from the stitched image, and the feature extraction results are compared with the carbide standard to determine the carbide distribution rating result of the bar. The carbide distribution rating of the bar is carried out based on the quantified carbide standard, and the stability of the carbide distribution rating result of the bar is improved. The embodiment of the present application can realize the automatic control of the acquisition, processing, feature extraction and rating of the rated field-of-view images in the whole process of the carbide distribution rating of the bar, and improve the evaluation accuracy and the stability of the rating standard.

[0226] Figure 5 is a schematic structural diagram of an electronic device provided by an embodiment of the present application. As Figure 5 shown, the electronic device 5 of this embodiment includes: a processor 50, a memory 51, and a computer program 52 stored in the memory 51 and executable on the processor 50. When the processor 50 executes the computer program 52, the steps in the above-mentioned various control method embodiments for the carbide distribution rating of the bar are implemented, such as Figure 1 the steps S201 to S204 shown. Alternatively, when the processor 50 executes the computer program 52, the functions of each module / unit in the above-mentioned device embodiments are implemented, such as Figure 4 the functions of the modules 401 to 405 shown.

[0227] Exemplarily, the computer program 52 can be divided into one or more modules / units, and the one or more modules / units are stored in the memory 51 and executed by the processor 50 to complete the present application. The one or more modules / units can be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program 52 in the electronic device 5. For example, the computer program 52 can be divided into Figure 3 the modules 301 to 303 shown.

[0228] The electronic device 5 can be a computing device such as a desktop computer, a notebook, a palm computer, and a cloud server. The electronic device 5 may include, but is not limited to, a processor 50 and a memory 51. Those skilled in the art can understand, Figure 5This is only an example of the electronic device 5 and does not constitute a limitation on the electronic device 5. It may include more or fewer components than those shown in the figure, or combine certain components, or different components. For example, the electronic device may also include input / output devices, network access devices, a bus, etc.

[0229] The so-called processor 50 may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0230] The memory 51 may be an internal storage unit of the electronic device 5, such as the hard disk or memory of the electronic device 5. The memory 51 may also be an external storage device of the electronic device 5, such as a plug-in hard disk equipped on the electronic device 5, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. Further, the memory 51 may also include both the internal storage unit and the external storage device of the electronic device 5. The memory 51 is used to store the computer program and other programs and data required by the electronic device. The memory 51 may also be used to temporarily store data that has been output or is to be output.

[0231] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the above division of each functional unit and module is used as an example. In actual applications, the above functions can be allocated to different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiments can be integrated into a processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit. In addition, the specific names of each functional unit and module are only for the convenience of mutual distinction and do not limit the protection scope of this application. The specific working processes of the units and modules in the above system can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0232] In the above embodiments, the descriptions of the various embodiments have their own emphases. For the parts not detailed or recorded in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0233] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in this application can be implemented by electronic hardware, or by a combination of computer software and electronic hardware. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0234] In the embodiments provided in this application, it should be understood that the disclosed device / electronic device and method can be implemented in other ways. For example, the device / electronic device embodiments described above are only illustrative. For example, the division of the above modules or units is only a logical function division. In actual implementation, there can be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection to each other can be through some interfaces. The indirect coupling or communication connection of the device or unit can be in an electrical, mechanical or other form.

[0235] The unit described as a separate component may or may not be physically separated, and the component displayed as a unit may or may not be a physical unit, that is, it can be located in one place, or can be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0236] In addition, the functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit. The above-mentioned integrated units may be implemented in the form of hardware or in the form of software functional units.

[0237] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it may be stored in a computer-readable storage medium. Based on such an understanding, all or part of the processes in the above-mentioned method embodiments of the present application may also be completed by instructing relevant hardware through a computer program. The computer program may be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-mentioned various control method embodiments for rating the carbide distribution of bars can be implemented. Among them, the computer program includes computer program code, and the computer program code may be in the form of source code, object code, executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0238] The above-mentioned embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A control method for the carbide distribution rating of bars, characterized in that The bar carbide distribution rating system includes: an electron microscopy subsystem, a server, and a control device; the method includes: Obtaining a rating field-of-view image of a target bar sample collected by the electron microscopy subsystem; Performing sample region extraction and sample boundary determination processing on the rating field-of-view image, and generating an acquisition control signal for the electron microscopy subsystem according to the boundary determination result; wherein, the acquisition control signal is used to control the electron microscopy subsystem to adjust the detection region of the target bar sample to obtain multiple frames of rating field-of-view images of the target bar sample; After the acquisition of the rating field-of-view images is completed, performing stitching processing on multiple frames of rating field-of-view images; Performing feature extraction on the stitched image, and comparing the feature extraction result with the carbide standard to determine the bar carbide distribution rating result.

2. The control method according to claim 1, wherein The performing sample region extraction and sample boundary determination processing on the rating field-of-view image includes: Invoking a region extraction algorithm and a boundary determination algorithm; Performing sample region extraction on the rating field-of-view image based on the region extraction algorithm, and performing sample boundary determination on the extracted image based on the boundary determination algorithm; Wherein, the region extraction algorithm is a segmentation algorithm, which realizes pixel-level recognition and extraction of the cross-section of the target bar sample in the rating field-of-view image by dividing the rating field-of-view image into multiple grid images; the boundary determination algorithm performs sample boundary determination based on multiple row vectors, multiple column vectors, and an effective region threshold of the rating field-of-view image.

3. The control method according to claim 1, wherein The performing stitching processing on multiple frames of rating field-of-view images includes: Invoking an image stitching algorithm; Performing stitching processing on multiple frames of rating field-of-view images based on the image stitching algorithm; Wherein, the image stitching algorithm performs image stitching based on the acquisition order of multiple frames of rating field-of-view images.

4. The control method according to claim 1, wherein The performing feature extraction on the stitched image includes: Invoking a trained special bar strip carbide detection algorithm; Performing feature extraction on the stitched image based on the special bar strip carbide detection algorithm; Wherein, the special bar strip carbide detection algorithm is trained based on a Feature Pyramid Network (FPN) and a Pyramid Attention Network (PAN).

5. The control method according to claim 1, wherein Before the obtaining a rating field-of-view image of a target bar sample collected by the electron microscopy subsystem, it further includes: Obtaining a rating start signal for the target bar sample; wherein, the rating start signal is set input information or scanning information corresponding to the target bar sample; Generating a start control signal for the electron microscopy subsystem according to the rating start signal, and sending the start control signal to the electron microscopy subsystem; Obtaining a temporary field-of-view image collected by the electron microscopy subsystem in response to the start control signal, and generating an initial acquisition control signal when the target bar sample is included in the temporary field-of-view image and the target bar sample is placed at the target position.

6. The control method according to claim 1, wherein After the comparing the feature extraction result with the carbide standard to determine the bar carbide distribution rating result, it further includes: Generating rating record information according to the batch information of the target bar sample, the rating field-of-view image, and the bar carbide distribution rating result; Send the rating record information to the server for storage.

7. The control method according to claim 6, wherein It further includes: Obtain a query instruction for historical rating record information and send the query instruction to the server; Obtain the historical rating record query result fed back by the server and display the rating query result; wherein, the rating query result includes one or more of the bar sample batch information, magnification, number of strip particle bands, width of strip particle bands, and bar carbide distribution grade.

8. A control device for the carbide distribution rating of bars, characterized in that It includes: An acquisition module, configured to acquire a rating field-of-view image of a target bar sample collected by an electron microscopy subsystem; A control module, configured to perform sample area extraction and sample boundary determination processing on the rating field-of-view image, and generate an acquisition control signal for the electron microscopy subsystem according to the boundary determination result; wherein, the acquisition control signal is used to control the electron microscopy subsystem to adjust the detection area of the target bar sample to obtain multiple frames of the rating field-of-view image of the target bar sample; An image stitching module, configured to perform stitching processing on multiple frames of rating field-of-view images after the collection of the rating field-of-view images is completed; An image feature extraction module, configured to extract features from the stitched image; A rating module, configured to compare the feature extraction result with the carbide standard to determine the bar carbide distribution rating result.

9. A system for rating the carbide distribution of bars, characterized in that, It includes: An electron microscopy subsystem, a server, and a control device; Wherein, the electron microscopy subsystem includes: an electron microscope, a sample stage module, a driving motor, and a controller; wherein, the driving motor is used to drive the sample stage module to move under the control of the controller to adjust the detection area of the electron microscope eyepiece for the target bar sample; the electron microscope is used to collect a field-of-view image under the control of the controller; The server is communicatively connected to the control device and is used to store the rating field-of-view image of the target bar sample and the bar carbide distribution rating result; The control device is used to execute the control method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7 above.

Images