CN114897868A - Pole piece defect identification and model training method and device and electronic equipment - Google Patents
Pole piece defect identification and model training method and device and electronic equipment Download PDFInfo
- Publication number
- CN114897868A CN114897868A CN202210607956.2A CN202210607956A CN114897868A CN 114897868 A CN114897868 A CN 114897868A CN 202210607956 A CN202210607956 A CN 202210607956A CN 114897868 A CN114897868 A CN 114897868A
- Authority
- CN
- China
- Prior art keywords
- pole piece
- model
- defect
- image
- training
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
- G06V10/443—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components by matching or filtering
- G06V10/449—Biologically inspired filters, e.g. difference of Gaussians [DoG] or Gabor filters
- G06V10/451—Biologically inspired filters, e.g. difference of Gaussians [DoG] or Gabor filters with interaction between the filter responses, e.g. cortical complex cells
- G06V10/454—Integrating the filters into a hierarchical structure, e.g. convolutional neural networks [CNN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30148—Semiconductor; IC; Wafer
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
- Y02E60/10—Energy storage using batteries
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Multimedia (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Biodiversity & Conservation Biology (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The application provides a pole piece defect identification and model training method, a pole piece defect identification and model training device and electronic equipment, and the method comprises the following steps: inputting an original image of a pole piece into a trained pole piece defect identification model to obtain an identification result output by the pole piece defect identification model, wherein the identification result comprises a defect image; the pole piece defect identification model comprises an area submodel and a defect identification submodel; the area sub-model is used for extracting a pole piece area in the pole piece original image to obtain a pole piece area image; and the defect identifier model is used for extracting the defects of the pole piece region image to obtain a defect image. According to the method and the device, the original image of the pole piece is input into the pole piece defect identification model, the identification of the defect pole piece is realized through the model, the speed of processing the picture based on the model is far higher than the speed of processing the picture by a conventional image processing method, and the efficiency of identifying the pole piece defect is improved.
Description
Technical Field
The application relates to the field of model training, in particular to a pole piece defect identification and model training method and device and electronic equipment.
Background
At present, the detection of the defects of the die-cut material is generally to enhance and filter the pole piece image, then to carry out binarization processing on the pole piece image, and to realize the identification of the bad products of the die-cut material through a specific algorithm. Nevertheless, this method enables the identification of defective products of the die-cut material. However, this recognition method requires many processing steps, and involves many human-induced settings for each step, which results in a problem of low detection efficiency.
Disclosure of Invention
In view of the above, an object of the present disclosure is to provide a pole piece defect identification and model training method, device, electronic device and readable storage medium. The detection efficiency of the die cutting material can be improved.
In a first aspect, an embodiment of the present application provides a pole piece defect identification method, including: inputting an original image of a pole piece into a trained pole piece defect identification model to obtain an identification result output by the pole piece defect identification model, wherein the identification result comprises a defect image; the pole piece defect identification model comprises an area submodel and a defect identification submodel; the area sub-model is used for extracting a pole piece area in the pole piece original image to obtain a pole piece area image; and the defect identifier model is used for extracting the defects of the pole piece region image to obtain a defect image.
In the implementation process, the original image of the pole piece is input into the pole piece defect identification model, the image is identified and processed through the trained model, the identification result of the original image of the pole piece is obtained, and whether the pole piece corresponding to the original image of the pole piece has defects or not can be judged based on the result. Because the speed of processing the image by the model is far faster than the speed of processing the image by the image processing device, the efficiency of pole piece defect identification is improved by processing the original pole piece image by the model.
With reference to the first aspect, an embodiment of the present application provides a first possible implementation manner of the first aspect, where: after obtaining the recognition result output by the pole piece defect recognition model, the method further comprises: extracting a defect area in the defect image; and judging whether the pole piece corresponding to the defect image is a defect pole piece or not according to the defect attribute of the defect area.
In the implementation process, some small defects can be considered as negligible defects in the actual defect identification, and the pole piece corresponding to the defect image cannot be considered as the defective pole piece because the small defects exist in the defect image. Therefore, after the pole piece original image is identified through the defect identification model, the defect image is output, and the defect area is further judged according to the attribute of the defect area to obtain the defect pole piece, so that the accuracy of the defect pole piece identification is improved.
In a second aspect, an embodiment of the present application further provides a pole piece defect identification model training method, including: inputting a plurality of pole piece original images marked with characteristic regions into a pre-training model for pre-training to obtain a pre-trained region sub-model; carrying out region feature extraction on the plurality of pole piece original images by using the pre-trained region sub-model to obtain pole piece region images; and inputting the image to be recognized into a model to be trained for training to obtain a trained pole piece defect recognition sub-model, wherein the image to be recognized is an image obtained after defect labeling is carried out on the pole piece region image.
In the implementation process, the area sub-model training is carried out through a plurality of pole piece original images marked with the characteristic areas, and then the training of the pole piece defect identification sub-model is carried out through the pole piece area images marked with the defects. The pole piece defect identification model is divided into two submodels, pole piece region extraction and defect extraction are respectively carried out, each submodel is obtained through specific characteristic training, the accuracy of specific characteristic extraction is high, and the accuracy of the pole piece defect model is improved by extracting different regions through the submodels.
In combination with the second aspect, the present embodiments provide a first possible implementation manner of the second aspect, where: the method for pre-training the pole piece original images with the characteristic regions to be trained in the pre-training model to obtain the pre-trained region sub-model comprises the following steps: obtaining a plurality of target images, wherein the target images are obtained by labeling characteristic regions of the pole piece original images; inputting a plurality of pole piece original images and a plurality of target images into the pre-training model for pre-training to obtain a pre-trained area sub-model; the pre-training model is obtained by constructing a deep learning network under a TensorFlow framework by a Python model, and the region sub-model is used for extracting a pole piece region in the pole piece original image.
In the implementation process, the target image and the pole piece original image are input into a pre-training model for pre-training, so that the pre-training model is subjected to model training through the target image and the pole piece original image to form a region sub-model for identifying the pole piece region. The sub-model in the region is obtained through training, the pole piece region in the original image of the pole piece can be extracted, so that the pole piece region is only identified when the pole piece defect is identified, the identification range is reduced, the influence of other region identification on the identification result of the pole piece is prevented, and the identification accuracy is improved while the identification efficiency of the pole piece defect identification model is improved.
In combination with the first possible implementation manner of the second aspect, the present embodiments provide a second possible implementation manner of the second aspect, where: the acquiring a plurality of target images includes: generating a JSON file according to the plurality of pole piece original images, wherein the JSON file comprises a plurality of marked pole piece original images obtained after feature region marking is carried out on the plurality of pole piece original images; analyzing the JSON file to obtain a plurality of marked pole piece original images; carrying out binarization processing on the marked pole piece original image to obtain a binarized image; and processing the binary image through OpenCV to obtain a target image.
In the implementation process, the JSON file is generated according to the original image of the pole piece, and the JSON file has the advantages of convenience in transmission, conversion and the like because the JSON is a lightweight data exchange format. The JSON file is generated from the pole piece original image, so that the transmission of the pole piece original image is facilitated, and the training efficiency of the regional sub-model is improved. In addition, the marked original image is processed through binarization and OpenCV, and the binarization processing ensures that the marked original image does not relate to multi-level values of pixels any more, so that the marked original image is simplified, and the transmission of subsequent processing of the image is facilitated. The OpenCV further processes the binary image, so that the definition and the accuracy of the target image are improved.
With reference to the second possible implementation manner of the second aspect, an embodiment of the present application provides a third possible implementation manner of the second aspect, where the performing, by using the pre-trained region model, region feature extraction on multiple original pole piece images to obtain a pole piece region image includes: inputting a plurality of pole piece original images into an area sub-model; and carrying out regional characteristic extraction on the pole piece original image through the regional sub-model to obtain a pole piece regional image.
In the implementation process, when the pole piece defect recognition submodel is trained, the trained area submodel is directly used for extracting pole piece area images, so that the work of pole piece area labeling is reduced, and the training efficiency of the pole piece defect recognition submodel is improved.
In a third aspect, an embodiment of the present application further provides a pole piece defect identification device, including: an identification module: the pole piece defect recognition method comprises the steps of inputting an original pole piece image into a trained pole piece defect recognition model to obtain a recognition result output by the pole piece defect recognition model, wherein the recognition result comprises an area image and a defect image; the pole piece defect identification model comprises an area submodel and a defect identification submodel; the area sub-model is used for extracting a pole piece area in the pole piece original image to obtain a pole piece area image; and the defect identifier model is used for extracting the defects of the pole piece region image to obtain a defect image.
In a fourth aspect, an embodiment of the present application further provides a pole piece defect recognition model training device, including: a pre-training module: the device comprises a pre-training model, a pre-training model and a pre-training area sub-model, wherein the pre-training model is used for pre-training a plurality of pole piece original images with characteristic areas to be trained in the pre-training model to obtain the pre-trained area sub-model; an extraction module: the pre-trained area sub-model is used for extracting the area characteristics of the plurality of pole piece original images to obtain pole piece area images; a labeling module: the model training is carried out on the image to be recognized to obtain a trained pole piece defect recognition sub-model, and the image to be recognized is an image obtained after defect marking is carried out on the pole piece region image.
In a fifth aspect, an embodiment of the present application further provides an electronic device, including: a processor, a memory storing machine-readable instructions executable by the processor, the machine-readable instructions, when executed by the processor, performing the steps of the method of the first aspect described above, or any possible implementation of the first aspect, when the electronic device is run.
In a sixth aspect, this embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, performs the steps of the method in the first aspect, or any one of the first aspect, the second aspect, or any one of the possible implementations of the second aspect.
In order to make the aforementioned objects, features and advantages of the present application more comprehensible, embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained from the drawings without inventive effort.
Fig. 1 is a flowchart of a pole piece defect identification method according to an embodiment of the present disclosure;
fig. 2 is a flowchart of a pole piece defect recognition model training method provided in the embodiment of the present application;
fig. 3 is a schematic functional block diagram of a pole piece defect identification apparatus according to an embodiment of the present disclosure;
fig. 4 is a functional module schematic diagram of a pole piece defect recognition model training device provided in the embodiment of the present application;
fig. 5 is a block diagram of an electronic device according to an embodiment of the present application.
Detailed Description
The technical solution in the embodiments of the present application will be described below with reference to the drawings in the embodiments of the present application.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures. Meanwhile, in the description of the present application, the terms "first", "second", and the like are used only for distinguishing the description, and are not to be construed as indicating or implying relative importance.
At present, with the rapid development of intellectualization and industrialization, industries such as electronic equipment, electrical equipment, precise instruments, electronic communication and the like also enter a rapid development stage. The die cutting material is widely applied to the industries, and gradually occupies an important role in each industry, and the market thereof is rapidly developed.
As die-cutting materials play a crucial role in various industries, the throughput of die-cutting materials is gradually increased, and the quality of the die-cutting materials affects the quality of the whole equipment. Therefore, how to improve the efficiency and accuracy of identifying the defects of the die-cutting material becomes an urgent problem to be solved in the detection of the die-cutting material.
The inventor of the application discovers that the defect identification efficiency of the die-cutting material is improved in the process of: the image recognition model has a processing speed on the image which is far higher than that of a common image processing device. Generally, the time for processing one image by a general image processing apparatus is about 400ms, and the time for processing one image by an image recognition model is about 100 ms. In view of the above, the inventor of the present application provides a pole piece defect identification method, which greatly improves the efficiency of pole piece defect identification by inputting an original pole piece image into a trained pole piece defect identification model for identification and identifying the defect of a pole piece according to an identification result.
The pole piece defect identification method disclosed by the embodiment of the application can be used for detecting die-cutting materials, detecting chips, detecting electronic screens and detecting glass materials, but is not limited to the detection of the die-cutting materials, the detection of the chips, the detection of the electronic screens and the detection of the glass materials. The pole piece defect identification model training method can be used for training different defect identification models to realize defect identification of various objects to be detected.
In order to facilitate understanding of the present embodiment, a pole piece defect identification method provided in the embodiments of the present application is first described in detail.
Please refer to fig. 1, which is a flowchart illustrating a pole piece defect identification method according to an embodiment of the present disclosure. The specific process shown in FIG. 1 will be described in detail below.
The pole piece defect identification model comprises a region submodel and a defect identification submodel; the area sub-model is used for extracting a pole piece area in an original pole piece image to obtain a pole piece area image; the defect identification submodel is used for extracting defects of the pole piece area image to obtain a defect image.
It will be appreciated that the region sub-model and defect identifier sub-model may be part of both of the pole piece defect identification models. In one case, the pole piece defect identification model is a generic term for the area sub-model and the defect identification sub-model, which may be two separate models.
The original image of the pole piece here is an image of the pole piece acquired by an image acquisition device, and the original image of the pole piece may include an image of the pole piece, an image of another article in the area where the pole piece is located, and an image of the area itself.
The identification result comprises a defect image and a pole piece area image.
After the pole piece defect identification model identifies the defects of the original images of the pole pieces, for the pole pieces without the defects, the defect identification model can output the images of the pole piece areas, and can also directly identify the original images of the next pole piece without outputting the identification results. For the pole piece with the defect, the defect identification model can only output a defect image, and can also output images of non-defect areas in the defect image and the pole piece area image.
In the implementation process, the original image of the pole piece is input into the pole piece defect identification model, the image is identified and processed through the trained model, the identification result of the original image of the pole piece is obtained, and whether the pole piece corresponding to the original image of the pole piece has defects or not can be judged based on the result. Because the speed of processing the image by the model is far faster than the speed of processing the image by the image processing device, the efficiency of pole piece defect identification is improved by processing the original pole piece image by the model.
In one possible implementation, after step 201, the method further includes: and extracting a defect area in the defect image, and judging whether the pole piece corresponding to the defect image is a defect pole piece according to the defect attribute of the defect area.
The defect area is an area to which a defect in a defect image belongs, and one or more defect areas may exist in one defect image. The defect attribute may be an area of the defect region, a length of the defect region, a width of the defect region, a shape of the defect region, and the like.
In the actual pole piece defect judgment process, some defect areas in the defect image are small, and the defect can be considered as a negligible defect. In order to ensure the accuracy of pole piece defect identification, further judgment is usually performed on a defect image. For example, the determining whether the pole piece corresponding to the defect image is the defective pole piece according to the defect attribute of the defect area herein may include: comparing the area of the defect region with a preset area, and if the area of the defect region is larger than the preset area, taking the pole piece corresponding to the defect image as a defect pole piece; or comparing the length of the defect area with a preset length, and if the length of the defect area is greater than the preset length, taking the pole piece corresponding to the defect image as a defect pole piece; or comparing the width of the defect area with a preset width, and if the width of the defect area is greater than the preset width, the pole piece corresponding to the defect image is a defect pole piece. It can be understood that the above method for judging whether the pole piece corresponding to the defect image is the defective pole piece according to the defect attribute of the defect area is only an example, and the specific judgment mode may be adjusted according to the actual situation, and the application is not particularly limited.
In the implementation process, some small defects can be considered as negligible defects in the actual defect identification, and the pole piece corresponding to the defect image cannot be considered as the defective pole piece because the small defects exist in the defect image. Therefore, after the pole piece original image is identified through the defect identification model, the defect image is output, and the defect area is further judged according to the attribute of the defect area to obtain the defect pole piece, so that the accuracy of the defect pole piece identification is improved.
Please refer to fig. 2, which is a flowchart illustrating a pole piece defect recognition model training method according to an embodiment of the present disclosure. The specific process shown in fig. 2 will be described in detail below.
The pre-training model here may be an image classification model, an image processing model, etc. For example, Le Net model, Alex Net model, VGG model, etc.
As can be appreciated, prior to step 301, the method further comprises: and marking the characteristic region of the original image of the pole piece. Optionally, the characteristic region labeling of the original image of the pole piece may be performed manually or by image processing software. The specific labeling method can be selected according to actual conditions, and the application is not particularly limited.
And 302, performing area feature extraction on the multiple pole piece original images by using the pre-trained area sub-model to obtain pole piece area images.
The method includes the steps that a plurality of pole piece original images are input into a pre-trained area sub-model, the area sub-model processes the pole piece original images, pole piece area images only containing pole piece areas can be extracted from the pole piece original images, and the pole piece areas can be marked in the pole piece original images.
And 303, inputting the image to be recognized into the model to be trained for training to obtain a trained pole piece defect recognition sub-model.
The image to be identified is the image of the pole piece region image after defect marking.
It will be appreciated that the model to be trained may be an image classification model, an image processing model, or the like. For example, Le Net model, Alex Net model, VGG model, etc. The model to be trained and the pre-training model can be the same model or different models.
If the mode of processing the pole piece original image of the area sub-model is as follows: if the pole piece region is labeled in the original pole piece image, step 302 can be omitted for the pole piece defect identification model training method.
In one embodiment, the pole piece defect recognition model training method further includes: inputting a plurality of pole piece original images marked with characteristic regions into a pre-training model for pre-training to obtain a pre-trained region sub-model; and inputting the pole piece original image marked with the defects and the characteristic regions into a model to be trained for training to obtain a trained pole piece defect identification sub-model.
In some embodiments, the step of obtaining the pre-trained region sub-model and the step of obtaining the trained pole piece defect recognition sub-model may be performed simultaneously.
In the implementation process, the area sub-model training is carried out through a plurality of pole piece original images marked with the characteristic areas, and then the training of the pole piece defect identification sub-model is carried out through the pole piece area images marked with the defects. The pole piece defect identification model is divided into two submodels, pole piece region extraction and defect extraction are respectively carried out, each submodel is obtained through specific characteristic training, the accuracy of specific characteristic extraction is high, and the accuracy of the pole piece defect model is improved by adopting different submodels to extract different regions.
In one possible implementation, step 301 includes: acquiring a plurality of target images; and inputting the plurality of pole piece original images and the plurality of target images into a pre-training model for pre-training to obtain a pre-trained region sub-model.
The pre-training model is obtained by constructing a deep learning network under a TensorFlow framework by a Python model, and the region sub-model is used for extracting a pole piece region in an original pole piece image.
The target image is obtained by labeling the characteristic region of the original image of the pole piece.
It can be understood that the original image of the pole piece may be input into a source picture corresponding to a pre-trained model, and the target image may be input into a marked picture corresponding to the pre-trained model.
In the implementation process, the target image and the pole piece original image are input into a pre-training model for pre-training, so that the pre-training model is subjected to model training through the target image and the pole piece original image to form a region sub-model for identifying the pole piece region. The sub-model in the region is obtained through training, the pole piece region in the original image of the pole piece can be extracted, so that the pole piece region is only identified when the pole piece defect is identified, the identification range is reduced, the influence of other region identification on the identification result of the pole piece is prevented, and the identification accuracy is improved while the identification efficiency of the pole piece defect identification model is improved.
In one possible implementation, acquiring a plurality of target images includes: generating a JSON file according to the original images of the pole pieces; analyzing the JSON file to obtain a plurality of marked pole piece original images; carrying out binarization processing on the marked pole piece original image to obtain a binarized image; and processing the binary image through OpenCV to obtain a target image.
The JSON file comprises a plurality of marked pole piece original images obtained after feature region marking is carried out on the pole piece original images. The JSON file can be analyzed through analysis modes such as manual analysis, Gson analysis and FastJson analysis.
As can be appreciated, generating a JSON file from a plurality of pole piece original images includes: and marking the characteristic regions of the multiple pole piece original images, storing the marked pole piece original images into a JSON format, and packaging one or more marked pole piece original images in the JSON format to form a JSON file. The marking of the characteristic region can be carried out by an image marking tool. For example, Labelme, VOTT, Labellmy, Vatic, and the like.
The OpenCV here can perform processing such as scaling, cropping, and edge correction on the binarized image.
In the implementation process, the JSON file is generated according to the original image of the pole piece, and the JSON file has the advantages of convenience in transmission, conversion and the like because the JSON is a lightweight data exchange format. The JSON file is generated from the pole piece original image, so that the transmission of the pole piece original image is facilitated, and the training efficiency of the regional sub-model is improved. In addition, the marked original image is processed through binarization and OpenCV, and the binarization processing ensures that the marked original image does not relate to multi-level values of pixels any more, so that the marked original image is simplified, and the transmission of subsequent processing of the image is facilitated. The OpenCV further processes the binary image to improve the definition and accuracy of the target image.
In one possible implementation, step 302 includes: inputting a plurality of pole piece original images into an area sub-model; and carrying out regional characteristic extraction on the original image of the pole piece through a regional sub-model to obtain a regional image of the pole piece.
Optionally, the pole piece original image may be a pole piece original image when performing area sub-model training, or may not be a pole piece original image when performing area sub-model training.
The pole piece region image may be a single image including only the pole piece region, or may be an image obtained by labeling the pole piece region on the original pole piece image.
In the implementation process, when the pole piece defect recognition submodel is trained, the trained area submodel is directly used for extracting pole piece area images, so that the work of pole piece area labeling is reduced, and the training efficiency of the pole piece defect recognition submodel is improved.
Based on the same application concept, the embodiment of the present application further provides a pole piece defect identification device corresponding to the pole piece defect identification method, and since the principle of solving the problem of the device in the embodiment of the present application is similar to that of the embodiment of the pole piece defect identification method, the implementation of the device in the embodiment of the present application can refer to the description in the embodiment of the above method, and repeated details are omitted.
Please refer to fig. 3, which is a schematic diagram of a functional module of a pole piece defect identification apparatus according to an embodiment of the present disclosure. Each module in the pole piece defect identification apparatus in this embodiment is used to execute each step in the above method embodiments. The pole piece defect identification device comprises an identification module 401; wherein the content of the first and second substances,
the recognition module 401 is configured to input the original pole piece image into the trained pole piece defect recognition model, and obtain a recognition result output by the pole piece defect recognition model, where the recognition result includes a defect image; the pole piece defect identification model comprises a region submodel and a defect identification submodel; the area sub-model is used for extracting a pole piece area in the pole piece original image to obtain a pole piece area image; the defect identifier model is used for extracting the defects of the pole piece region image to obtain a defect image.
In a possible implementation manner, the pole piece defect identification device further comprises an interpretation module; wherein the interpretation module is to: extracting a defect area in the defect image; and judging whether the pole piece corresponding to the defect image is a defect pole piece or not according to the defect attribute of the defect area.
Based on the same application concept, a pole piece defect recognition model training device corresponding to the pole piece defect recognition model training method is further provided in the embodiment of the present application, and as the principle of solving the problem of the device in the embodiment of the present application is similar to that of the embodiment of the pole piece defect recognition model training method, the implementation of the device in the embodiment of the present application can refer to the description in the embodiment of the method, and repeated details are omitted.
Please refer to fig. 4, which is a schematic diagram of a functional module of a pole piece defect recognition model training apparatus according to an embodiment of the present disclosure. Each module in the pole piece defect recognition model training device in this embodiment is used to execute each step in the above method embodiments. The pole piece defect recognition model training device comprises a pre-training module 501, an extraction module 502 and a labeling module 503; wherein the content of the first and second substances,
the pre-training module 501 is configured to put a plurality of pole piece original images with feature regions to be trained into a pre-training model for pre-training, so as to obtain a pre-trained region sub-model.
The extraction module 502 is configured to perform area feature extraction on the multiple pole piece original images by using the pre-trained area sub-model to obtain pole piece area images.
The labeling module 503 is configured to perform model training on an image to be recognized to obtain a trained sub-model for pole piece defect recognition, where the image to be recognized is an image obtained by performing defect labeling on an image in a pole piece region.
In a possible implementation, the pre-training module 501 is further configured to: acquiring a plurality of target images, wherein the target images are obtained by labeling characteristic regions of original polar plate images; inputting a plurality of pole piece original images and a plurality of target images into a pre-training model for pre-training to obtain a pre-trained area sub-model; the pre-training model is obtained by constructing a deep learning network under a TensorFlow framework by a Python model, and the region sub-model is used for extracting a pole piece region in a pole piece original image.
In a possible implementation, the pre-training module 501 is specifically configured to: generating a JSON file according to the plurality of pole piece original images, wherein the JSON file comprises a plurality of marked pole piece original images obtained after the characteristic regions of the plurality of pole piece original images are marked; analyzing the JSON file to obtain a plurality of marked pole piece original images; carrying out binarization processing on the marked pole piece original image to obtain a binarized image; and processing the binary image through OpenCV to obtain a target image.
In a possible implementation, the extracting module 502 is further configured to: inputting a plurality of pole piece original images into an area sub-model; and carrying out regional characteristic extraction on the original image of the pole piece through a regional sub-model to obtain a regional image of the pole piece.
In order to facilitate understanding of the present embodiment, the following describes in detail an electronic device for executing the pole piece defect identification method and the pole piece defect identification model training method disclosed in the embodiments of the present application.
Fig. 5 is a block diagram of an electronic device. The electronic device 100 may include a memory 111, a memory controller 112, a processor 113, a peripheral interface 114, an input-output unit 115, and a display unit 116. It will be understood by those skilled in the art that the structure shown in fig. 5 is merely an illustration and is not intended to limit the structure of the electronic device 100. For example, electronic device 100 may also include more or fewer components than shown in FIG. 5, or have a different configuration than shown in FIG. 5.
The above-mentioned elements of the memory 111, the memory controller 112, the processor 113, the peripheral interface 114, the input/output unit 115 and the display unit 116 are electrically connected to each other directly or indirectly, so as to implement data transmission or interaction. For example, the components may be electrically connected to each other via one or more communication buses or signal lines. The processor 113 is used to execute the executable modules stored in the memory.
The Memory 111 may be, but is not limited to, a Random Access Memory (RAM), a Read Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Read-Only Memory (EPROM), an electrically Erasable Read-Only Memory (EEPROM), and the like. The memory 111 is configured to store a program, and the processor 113 executes the program after receiving an execution instruction, and the method executed by the electronic device 100 defined by the process disclosed in any embodiment of the present application may be applied to the processor 113, or implemented by the processor 113.
The memory 111 can be used for storing pole piece original images, pole piece region images, pole piece defect identification models, region sub-models, defect identification sub-models and other information.
The processor 113 may be an integrated circuit chip having signal processing capability. The Processor 113 may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Network Processor (NP), and the like; the Integrated Circuit may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component. The various methods, steps, and logic blocks disclosed in the embodiments of the present application may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
It can be understood that the processor 113 may be configured to pre-train a pre-training model, or perform to-be-trained on the to-be-trained model, and may be configured to perform feature region labeling on an original image of a pole piece, perform defect labeling on an image of the pole piece region, and the like.
The peripheral interface 114 couples various input/output devices to the processor 113 and memory 111. In some embodiments, the peripheral interface 114, the processor 113, and the memory controller 112 may be implemented in a single chip. In other examples, they may be implemented separately from the individual chips. Illustratively, the peripheral interface 114 may be used to interface with an image capture device to input the raw image of the pole piece captured by the image capture device into a pole piece defect identification model or a region sub-model.
The input/output unit 115 is used to provide input data to the user. The input/output unit 115 may be, but is not limited to, a mouse, a keyboard, and the like.
The display unit 116 provides an interactive interface (e.g., a user operation interface) between the electronic device 100 and the user or is used for displaying image data to the user for reference. In this embodiment, the display unit may be a liquid crystal display or a touch display. In the case of a touch display, the display can be a capacitive touch screen or a resistive touch screen, which supports single-point and multi-point touch operations. The support of single-point and multi-point touch operations means that the touch display can sense touch operations simultaneously generated from one or more positions on the touch display, and the sensed touch operations are sent to the processor for calculation and processing.
It can be understood that the above labeling of the characteristic region of the original image of the pole piece and the labeling of the defect of the original image of the pole piece can be performed manually, and a user performs operations such as clicking and sliding on the image displayed on the display unit 116 to implement the labeling of the characteristic region or the labeling of the defect.
The electronic device 100 in this embodiment may be configured to perform each step in each method provided in this embodiment.
Furthermore, an embodiment of the present application further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the computer program performs the steps of the methods described in the above method embodiments.
The computer program product of each method provided in the embodiments of the present application includes a computer-readable storage medium storing a program code, where instructions included in the program code may be used to execute steps of each method described in the above method embodiments, which may be specifically referred to in the above method embodiments, and details are not described here again.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method can be implemented in other ways. The apparatus embodiments described above are merely illustrative, and for example, the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
In addition, functional modules in the embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes. It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The above description is only a preferred embodiment of the present application and is not intended to limit the present application, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application. It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.
The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present application, and shall be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (10)
1. A pole piece defect identification method is characterized by comprising the following steps:
inputting an original image of a pole piece into a trained pole piece defect identification model to obtain an identification result output by the pole piece defect identification model, wherein the identification result comprises a defect image;
the pole piece defect identification model comprises an area submodel and a defect identification submodel; the area sub-model is used for extracting a pole piece area in the pole piece original image to obtain a pole piece area image; and the defect identifier model is used for extracting the defects of the pole piece region image to obtain a defect image.
2. The method according to claim 1, wherein after obtaining the recognition result output by the pole piece defect recognition model, the method further comprises:
extracting a defect area in the defect image;
and judging whether the pole piece corresponding to the defect image is a defect pole piece or not according to the defect attribute of the defect area.
3. A pole piece defect recognition model training method is characterized by comprising the following steps:
inputting a plurality of pole piece original images marked with characteristic regions into a pre-training model for pre-training to obtain a pre-trained region sub-model;
extracting the regional characteristics of the plurality of pole piece original images by using the pre-trained regional submodel to obtain pole piece regional images;
and inputting the image to be recognized into a model to be trained for training to obtain a trained pole piece defect recognition sub-model, wherein the image to be recognized is an image obtained after defect labeling is carried out on the pole piece region image.
4. The method of claim 3, wherein the pre-training of the pole piece raw images with the feature regions to be trained in the pre-training model to obtain the pre-trained region submodel comprises:
obtaining a plurality of target images, wherein the target images are obtained by labeling characteristic regions of the pole piece original images;
inputting a plurality of pole piece original images and a plurality of target images into the pre-training model for pre-training to obtain a pre-trained area sub-model;
the pre-training model is obtained by constructing a deep learning network under a TensorFlow framework by a Python model, and the region sub-model is used for extracting a pole piece region in the pole piece original image.
5. The method of claim 4, wherein said acquiring a plurality of target images comprises:
generating a JSON file according to the plurality of pole piece original images, wherein the JSON file comprises a plurality of marked pole piece original images obtained after feature region marking is carried out on the plurality of pole piece original images;
analyzing the JSON file to obtain a plurality of marked pole piece original images;
carrying out binarization processing on the marked pole piece original image to obtain a binarized image;
and processing the binary image through OpenCV to obtain a target image.
6. The method according to claim 3, wherein the performing region feature extraction on the plurality of pole piece original images by using the pre-trained region model to obtain pole piece region images comprises:
inputting a plurality of pole piece original images into an area sub-model;
and carrying out regional characteristic extraction on the pole piece original image through the regional sub-model to obtain a pole piece regional image.
7. A pole piece defect identification device, comprising:
an identification module: the pole piece defect recognition method comprises the steps of inputting an original pole piece image into a trained pole piece defect recognition model to obtain a recognition result output by the pole piece defect recognition model, wherein the recognition result comprises an area image and a defect image;
the pole piece defect identification model comprises an area submodel and a defect identification submodel; the area sub-model is used for extracting a pole piece area in the pole piece original image to obtain a pole piece area image; and the defect identifier model is used for extracting the defects of the pole piece region image to obtain a defect image.
8. The utility model provides a pole piece defect recognition model trainer which characterized in that includes:
a pre-training module: the device comprises a pre-training model, a pre-training model and a pre-training area sub-model, wherein the pre-training model is used for pre-training a plurality of pole piece original images with characteristic areas to be trained in the pre-training model to obtain the pre-trained area sub-model;
an extraction module: the pre-trained area sub-model is used for extracting the area characteristics of the plurality of pole piece original images to obtain pole piece area images;
a labeling module: the model training is carried out on the image to be recognized to obtain a trained pole piece defect recognition sub-model, and the image to be recognized is an image obtained after defect marking is carried out on the pole piece region image.
9. An electronic device, comprising: a processor, a memory storing machine-readable instructions executable by the processor, the machine-readable instructions when executed by the processor performing the steps of the method of any of claims 1 to 6 when the electronic device is run.
10. A computer-readable storage medium, having stored thereon a computer program which, when being executed by a processor, is adapted to carry out the steps of the method according to any one of claims 1 to 6.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210607956.2A CN114897868A (en) | 2022-05-31 | 2022-05-31 | Pole piece defect identification and model training method and device and electronic equipment |
| PCT/CN2022/140037 WO2023231380A1 (en) | 2022-05-31 | 2022-12-19 | Electrode plate defect recognition method and apparatus, and electrode plate defect recognition model training method and apparatus, and electronic device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210607956.2A CN114897868A (en) | 2022-05-31 | 2022-05-31 | Pole piece defect identification and model training method and device and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114897868A true CN114897868A (en) | 2022-08-12 |
Family
ID=82725113
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210607956.2A Pending CN114897868A (en) | 2022-05-31 | 2022-05-31 | Pole piece defect identification and model training method and device and electronic equipment |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN114897868A (en) |
| WO (1) | WO2023231380A1 (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116168030A (en) * | 2023-04-25 | 2023-05-26 | 宁德时代新能源科技股份有限公司 | Pole piece defect detection method and device, electronic equipment and storage medium |
| WO2023231380A1 (en) * | 2022-05-31 | 2023-12-07 | 广东利元亨智能装备股份有限公司 | Electrode plate defect recognition method and apparatus, and electrode plate defect recognition model training method and apparatus, and electronic device |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109598721B (en) * | 2018-12-10 | 2021-08-31 | 广州市易鸿智能装备有限公司 | Defect detection method and device for battery pole piece, detection equipment and storage medium |
| EP3739513A1 (en) * | 2019-05-13 | 2020-11-18 | Fujitsu Limited | Surface defect identification method and apparatus |
| CN114037645A (en) * | 2020-07-20 | 2022-02-11 | 耿晋 | Coating defect detection method and device for pole piece, electronic equipment and readable medium |
| CN113870208A (en) * | 2021-09-22 | 2021-12-31 | 上海联麓半导体技术有限公司 | Semiconductor image processing method, semiconductor image processing apparatus, computer device, and storage medium |
| CN114897868A (en) * | 2022-05-31 | 2022-08-12 | 广东利元亨智能装备股份有限公司 | Pole piece defect identification and model training method and device and electronic equipment |
-
2022
- 2022-05-31 CN CN202210607956.2A patent/CN114897868A/en active Pending
- 2022-12-19 WO PCT/CN2022/140037 patent/WO2023231380A1/en unknown
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023231380A1 (en) * | 2022-05-31 | 2023-12-07 | 广东利元亨智能装备股份有限公司 | Electrode plate defect recognition method and apparatus, and electrode plate defect recognition model training method and apparatus, and electronic device |
| CN116168030A (en) * | 2023-04-25 | 2023-05-26 | 宁德时代新能源科技股份有限公司 | Pole piece defect detection method and device, electronic equipment and storage medium |
| CN116168030B (en) * | 2023-04-25 | 2023-11-14 | 宁德时代新能源科技股份有限公司 | Pole piece defect detection method and device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2023231380A1 (en) | 2023-12-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110826416B (en) | Bathroom ceramic surface defect detection method and device based on deep learning | |
| US11295435B2 (en) | Product defect detection method, device and system | |
| CN109726643B (en) | Method and device for identifying table information in image, electronic equipment and storage medium | |
| US11741593B2 (en) | Product defect detection method, device and system | |
| US11748873B2 (en) | Product defect detection method, device and system | |
| CN114897868A (en) | Pole piece defect identification and model training method and device and electronic equipment | |
| US10489687B2 (en) | Classification method, classification module and computer program product using the same | |
| JP2024509411A (en) | Defect detection method, device and system | |
| CN109961040B (en) | Identity card area positioning method and device, computer equipment and storage medium | |
| JP2018005640A (en) | Classifying unit generation device, image inspection device, and program | |
| CN108108731B (en) | Text detection method and device based on synthetic data | |
| CN111680750B (en) | Image recognition method, device and equipment | |
| WO2015115274A1 (en) | Object assessment device, image sensor, and object assessment method | |
| CN111310826B (en) | Method and device for detecting labeling abnormality of sample set and electronic equipment | |
| CN111242899B (en) | Image-based flaw detection method and computer-readable storage medium | |
| CN113222913B (en) | Circuit board defect detection positioning method, device and storage medium | |
| CN116542975B (en) | Defect classification method, device, equipment and medium for glass panel | |
| WO2017107533A1 (en) | Electronic component sample labeling method and device | |
| CN112767354A (en) | Defect detection method, device and equipment based on image segmentation and storage medium | |
| CN117392042A (en) | Defect detection method, defect detection apparatus, and storage medium | |
| Schneider | Real time circle detection by simplified Hough transform on smartphones | |
| CN116381053A (en) | Ultrasonic detection method and system for welding metal materials | |
| JP7364639B2 (en) | Processing of digitized writing | |
| Rao et al. | MTESSERACT: An Application for Form Recognition in Courier Services | |
| CN114596243A (en) | Defect detection method, device, equipment and computer readable storage medium |
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 |