CN114037645A - Coating defect detection method and device for pole piece, electronic equipment and readable medium - Google Patents
Coating defect detection method and device for pole piece, electronic equipment and readable medium Download PDFInfo
- Publication number
- CN114037645A CN114037645A CN202010699517.XA CN202010699517A CN114037645A CN 114037645 A CN114037645 A CN 114037645A CN 202010699517 A CN202010699517 A CN 202010699517A CN 114037645 A CN114037645 A CN 114037645A
- Authority
- CN
- China
- Prior art keywords
- defect
- coating
- pole piece
- defect detection
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- 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
- G06T7/001—Industrial image inspection using an image reference approach
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
-
- 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]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- Signal Processing (AREA)
- Probability & Statistics with Applications (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Image Analysis (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The disclosure relates to a pole piece coating defect detection method and device, electronic equipment and a computer readable medium. Coating the electrode plate; acquiring a pole piece image subjected to coating treatment by a visual sensing device; judging whether the pole piece image has defects or not through a coating defect detection model, wherein the coating defect detection model is generated through deep learning neural network model training; and generating a defect detection result when the pole piece image has a defect. The method can detect the defects of the pole piece of the lithium battery with high coverage rate, and can meet the requirements of the detection speed and the defect identification rate of the pole piece even in the wide and high-speed production scene of the pole piece of the lithium battery.
Description
Technical Field
The disclosure relates to the field of application of lithium ion batteries, in particular to a method and a device for detecting coating defects of a pole piece, electronic equipment and a computer readable medium.
Background
The lithium ion battery is a mainstream development direction for power batteries and energy storage applications in new energy industries due to mature technology and excellent performance. The main energy storage element of the battery is a battery pole piece, i.e. a current collector metal foil coated with a thin layer of electrode material. At present, coating is the main mode of large-scale production and processing of pole pieces. The coating equipment and control equipment require extremely high precision to ensure accurate coating coefficients. Meanwhile, the base material is ensured to be smooth in running and stable in tension, and phenomena of running sliding, serious shaking and wrinkling are avoided.
However, during the high-speed production process, the phenomena of unstable slurry properties, mechanical abnormality, abnormal control and the like occur, so that coating defects such as coating bubbles, shrinkage cavities, scratches, stripes, thick edges and the like are caused. Meanwhile, in the drying process of the coating downstream, due to the instability of the temperature field and the airflow field, drying defects such as wrinkling, cracking, binder drifting and the like can also occur. Due to the high price of the battery paste, production defects can waste a large amount of materials and production energy, or long-term production stoppage, and the economic loss is huge. If the defective pole piece is not detected and installed in the battery, the problems of poor battery consistency, capacity attenuation, low battery life and the like can be caused if the defective pole piece is light, and the problems of internal micro short circuit of the battery can be caused if the defective pole piece is heavy, so that the battery expands, smokes and even explodes. Therefore, in the coating process, the defect detection with high coverage rate is carried out on the pole piece of the lithium battery, whether the coating defect exists or not is judged, the type and the position of the defect are determined, and the corresponding treatment is carried out on the defect area subsequently, so that the method is an indispensable important step in the production of the lithium battery.
The above information disclosed in this background section is only for enhancement of understanding of the background of the disclosure and therefore it may contain information that does not constitute prior art that is already known to a person of ordinary skill in the art.
Disclosure of Invention
In view of this, the present disclosure provides a method and an apparatus for detecting a coating defect of a pole piece, an electronic device, and a computer readable medium, which can perform a high coverage defect detection on a pole piece of a lithium battery, and can meet requirements of detection speed and defect identification rate of the pole piece even in a wide-width high-speed production scenario of the lithium battery pole piece.
Additional features and advantages of the disclosure will be set forth in the detailed description which follows, or in part will be obvious from the description, or may be learned by practice of the disclosure.
According to an aspect of the disclosure, a method for detecting coating defects of a pole piece is provided, which includes: coating the electrode plate; acquiring a pole piece image subjected to coating treatment by a visual sensing device; judging whether the pole piece image has defects or not through a coating defect detection model, wherein the coating defect detection model is generated through deep learning neural network model training; and generating a defect detection result when the pole piece image has a defect.
In an exemplary embodiment of the present disclosure, further comprising: extracting a defect area image from the defect detection result; determining the defect type of the defect area image through the coating defect detection model; and when the defect type is an unknown type, performing transfer learning processing of the coating defect detection model.
In an exemplary embodiment of the present disclosure, further comprising: marking the area with the defects in the historical pole piece image to generate a defect image set; generating a normal image set through the historical pole piece images without defects; and training a deep learning neural network model through the defect image set and the normal image set to generate the coating defect detection model.
In an exemplary embodiment of the present disclosure, acquiring an image of a pole piece that has undergone a coating process by a vision sensing device includes: and acquiring the pole piece image subjected to coating treatment by the aid of the parallel linear array cameras in the visual sensing device.
In an exemplary embodiment of the present disclosure, further comprising: and determining hardware parameters of the visual sensing device according to the width, transmission speed, measurement precision and pixel parameters of a linear array camera sensor of the pole piece to be measured.
In an exemplary embodiment of the present disclosure, determining the defect type of the defect region image by the coating defect detection model includes: the coating defect detection model determines the defect type of the defect region image based on a manifold learning method.
In an exemplary embodiment of the present disclosure, the coating defect detection model determines the defect type of the defect region image based on a manifold learning method, including: comparing the defect area image to a plurality of samples of known types of defect area images to determine the defect type.
In an exemplary embodiment of the present disclosure, comparing the defect area image to a sample of defect area images of known types to determine the defect type includes: extracting a first defect concentrated position in the defect area image; acquiring a plurality of second defect concentrated positions in a plurality of defect area image samples of known types; comparing the first defect concentration position with a plurality of second defect concentration positions respectively to generate a plurality of comparison distances; and determining the defect type based on the plurality of comparison distances.
In an exemplary embodiment of the present disclosure, when the defect type is an unknown type, the method includes: and when the comparison distances are all larger than a distance threshold value, determining that the defect type is an unknown type.
In an exemplary embodiment of the present disclosure, performing a transfer learning process of the coating defect detection model includes: carrying out transfer learning on the coating defect detection model based on the pole piece image with the defect type being an unknown type to generate an updated model parameter; and updating the coating defect detection model through the updated model parameters.
In an exemplary embodiment of the disclosure, training a deep learning neural network model through the set of defect images and the set of normal images to generate the coating defect detection model includes: training the input defect image set and the normal image set by a plurality of depth modules of a deep learning neural network model; determining a plurality of probability distributions corresponding to a plurality of defect types by a logistic regression activation function of the deep learning neural network model; determining a loss function from the plurality of probability distributions; and determining the coating defect detection model through the related parameters of the deep neural network model when the loss function is the minimum value.
In an exemplary embodiment of the present disclosure, determining the coating defect detection model by the relevant parameters of the deep neural network model when the loss function is a minimum value further includes: and obtaining the minimum value of the loss function through an optimization technology.
In an exemplary embodiment of the present disclosure, further comprising: generating a defect marking instruction according to the defect detection result; and/or generating alarm record information according to the defect detection result; and/or generating a device adjustment instruction according to the defect detection result.
In an exemplary embodiment of the present disclosure, further comprising: acquiring a defect detection result within a preset time; and generating a shutdown operation instruction when the coating number of the pole pieces with the defects in the defect detection result is greater than a threshold value.
According to an aspect of this disclosure, a coating defect detection device of pole piece is proposed, the device includes: the coating module is used for coating the pole piece; the image module is used for acquiring the pole piece image subjected to coating treatment through the visual sensing device; the detection module is used for judging whether the pole piece image has defects and the types of the defects through a coating defect detection model, wherein the coating defect detection model is generated through deep learning neural network model training; and the processing module is used for carrying out error processing when the pole piece image has defects.
In an exemplary embodiment of the present disclosure, further comprising: the transfer learning module is used for extracting a defect area image from the defect detection result; determining the defect type of the defect area image through the coating defect detection model; and when the defect type is an unknown type, performing transfer learning processing of the coating defect detection model.
In an exemplary embodiment of the present disclosure, further comprising: the model training module is used for marking the areas with the defects in the historical pole piece images to generate a defect image set; generating a normal image set through the historical pole piece images without defects; and training a deep learning neural network model through the defect image set and the normal image set to generate the coating defect detection model.
According to an aspect of the present disclosure, an electronic device is provided, the electronic device including: one or more processors; storage means for storing one or more programs; when executed by one or more processors, cause the one or more processors to implement a method as above.
According to an aspect of the disclosure, a computer-readable medium is proposed, on which a computer program is stored, which program, when being executed by a processor, carries out the method as above.
According to the method, the device, the electronic equipment and the computer readable medium for detecting the coating defects of the pole piece, the pole piece is coated; acquiring a pole piece image subjected to coating treatment by a visual sensing device; judging whether the pole piece image has defects or not through a coating defect detection model, wherein the coating defect detection model is generated through deep learning neural network model training; and a mode of generating a defect detection result when the pole piece image has defects can be used for carrying out high-coverage defect detection on the pole piece of the lithium battery, and the requirements of the detection speed and the defect identification rate of the pole piece can be met even in a wide high-speed production scene of the pole piece of the lithium battery.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.
Drawings
The above and other objects, features and advantages of the present disclosure will become more apparent by describing in detail exemplary embodiments thereof with reference to the attached drawings. The drawings described below are merely some embodiments of the present disclosure, and other drawings may be derived from those drawings by those of ordinary skill in the art without inventive effort.
Fig. 1 is a system block diagram illustrating a pole piece coating defect detection method and apparatus according to an exemplary embodiment.
Fig. 2 is a system block diagram illustrating a pole piece coating defect detection method and apparatus according to another exemplary embodiment.
FIG. 3 is a flow chart illustrating a method of coating defect detection of a pole piece according to an exemplary embodiment.
FIG. 4 is a flow chart illustrating a method of coating defect detection of a pole piece according to another exemplary embodiment.
FIG. 5 is a flow chart illustrating a method of coating defect detection of a pole piece according to another exemplary embodiment.
FIG. 6 is a schematic diagram illustrating a method of coating defect detection of a pole piece, according to an exemplary embodiment.
Fig. 7 is a diagram of anode material collected by a production line and an example of a detection result according to an example embodiment.
FIG. 8 is a flow chart illustrating a method of coating defect detection of a pole piece according to another exemplary embodiment.
Fig. 9 is a flowchart algorithm classification example illustrating generation of three types of negative electrode material defects according to an exemplary embodiment.
Fig. 10 is a block diagram illustrating a coating defect detection apparatus of a pole piece according to an exemplary embodiment.
Fig. 11 is a block diagram illustrating a coating defect detecting apparatus of a pole piece according to another exemplary embodiment.
FIG. 12 is a block diagram illustrating an electronic device in accordance with an example embodiment.
FIG. 13 is a block diagram illustrating a computer-readable medium in accordance with an example embodiment.
Detailed Description
Example embodiments will now be described more fully with reference to the accompanying drawings. Example embodiments may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of example embodiments to those skilled in the art. The same reference numerals denote the same or similar parts in the drawings, and thus, a repetitive description thereof will be omitted.
Furthermore, the described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of embodiments of the disclosure. One skilled in the relevant art will recognize, however, that the subject matter of the present disclosure can be practiced without one or more of the specific details, or with other methods, components, devices, steps, and so forth. In other instances, well-known methods, devices, implementations, or operations have not been shown or described in detail to avoid obscuring aspects of the disclosure.
The block diagrams shown in the figures are functional entities only and do not necessarily correspond to physically separate entities. I.e. these functional entities may be implemented in the form of software, or in one or more hardware modules or integrated circuits, or in different networks and/or processor means and/or microcontroller means.
The flow charts shown in the drawings are merely illustrative and do not necessarily include all of the contents and operations/steps, nor do they necessarily have to be performed in the order described. For example, some operations/steps may be decomposed, and some operations/steps may be combined or partially combined, so that the actual execution sequence may be changed according to the actual situation.
It will be understood that, although the terms first, second, third, etc. may be used herein to describe various components, these components should not be limited by these terms. These terms are used to distinguish one element from another. Thus, a first component discussed below may be termed a second component without departing from the teachings of the disclosed concept. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.
It is to be understood by those skilled in the art that the drawings are merely schematic representations of exemplary embodiments, and that the blocks or processes shown in the drawings are not necessarily required to practice the present disclosure and are, therefore, not intended to limit the scope of the present disclosure.
The inventor of the present disclosure finds that, through years of technology accumulation and development, the defect detection of lithium battery coating is gradually observed by human eyes at first, manually screened, and transited to an automatic stage of scanning a detected object by adopting a visual technology, processing an image in real time and analyzing defect types, so as to realize online detection of the surface of a pole piece. Compared with the traditional human eye recognition, the machine vision has the advantages of high repeatability, all-weather service, high accuracy and the like.
However, in the currently applied visual defect detection technology, each defect needs to be precisely defined before production and installation, the range and the characteristic value of the defect are strictly defined, and the monitoring system can process the acquired image through a preset value and perform defect identification. Therefore, the conventional identification system cannot identify the defect which is not consistent with the preset value and cannot identify the defect which is not defined in the system, and the conventional method has high requirements on the calculation amount and the calculation speed, so that the missing identification and the wrong identification in the wide-width high-speed production process can be caused, and the requirement that the overall production defect identification rate is increased increasingly cannot be met.
In view of this, the present disclosure provides a method and an apparatus for detecting a coating defect of a pole piece, an electronic device, and a computer readable medium. The coating defect identification does not need to define defect characteristic values, and the ability of active defect identification is mastered through learning and training of defect samples. In addition to providing defect monitoring, alignment deviation monitoring, width error monitoring, and speed monitoring for general visual inspection, the system may also be capable of automatically adding new defect types according to artificial intelligence determinations after system training is completed.
It should be noted that the method, the apparatus, the electronic device and the computer readable medium for detecting coating defects of the electrode plate in the disclosure can be applied to other production links of the lithium battery, such as defect monitoring in slitting and die cutting steps, and can also be applied to defect monitoring in coating production of other thin film materials, which is not limited in the disclosure. The method, the apparatus, the electronic device and the computer readable medium for detecting coating defects of a pole piece in the present disclosure are described in detail below with reference to specific embodiments.
Fig. 1 is a system block diagram illustrating a pole piece coating defect detection method, apparatus, electronic device and computer readable medium according to an example embodiment.
As shown in fig. 1, the system architecture 10 may include visual sensing devices 101, 102, 103, a network 104, and a server 105. The network 104 is used to provide a medium for communication links between the vision sensing devices 101, 102, 103 and the server 105. Network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.
The visual sensing devices 101, 102, 103 may be used to interact with a server 105 over a network 104 to receive or send messages or the like. The visual sensing devices 101, 102, 103 may have a communication client application installed thereon.
The visual sensing devices 101, 102, 103 may be various electronic devices with real-time image acquisition functionality including, but not limited to, video cameras, array cameras, and the like.
The server 105 may be a server providing various services, such as a background management server providing support for pole piece pictures acquired by the visual sensing devices 101, 102, 103. The background management server can analyze and process the received pole piece image, and feed back the processing result (for example, judging whether the pole piece has defects, defect types, and the like) to the manager.
The server 105 may, for example, control the relevant device to perform coating processing on the pole piece; the server 105 can acquire the pole piece image subjected to the coating process, for example, through the vision sensing device 101 (or 102, 103); the server 105 may determine whether there is a defect in the pole piece image, for example, by using a coating defect detection model generated by deep learning neural network model training; the server 105 may generate defect detection results, for example, when there is a defect in the pole piece image.
The server 105 may also extract a defect area image, for example, from the defect detection results; the server 105 may also determine the defect type of the defect region image, for example, by coating a defect detection model; the server 105 may also perform a transfer learning process of coating the defect detection model, for example, when the defect type is an unknown type.
The server 105 may also mark areas with defects in the historical pole piece images to generate a set of defect images, for example; the server 105 may also generate a set of normal images, for example, from the historical pole piece images without defects; and the server 105 may also train the deep learning neural network model to generate a coating defect detection model, e.g., through the set of defect images and the set of normal images.
The server 105 may be a server of one entity, and may also be composed of a plurality of servers, for example, a part of the server 105 may be, for example, a neural network detection module in the present disclosure, and is responsible for receiving image information; a portion of the server 105 may be, for example, a neural network learning module in the present disclosure, configured to analyze and classify the defects detected by the neural network detection module, and determine whether to initiate transfer learning of the neural network, and the like. It should be noted that the method for detecting the coating defect of the pole piece provided by the embodiment of the present disclosure may be executed by the server 105, and accordingly, the coating defect detection of the pole piece may be set in the server 105. And the means for acquiring real-time images of the pole pieces are located in the visual sensing means 101, 102, 103.
Fig. 2 is a system block diagram of a pole piece coating defect detection method and apparatus according to another exemplary embodiment.
As shown in fig. 2, the system architecture 20 may include a vision sensing device 201, a neural network detection module 201, a neural network learning module 203, a deep learning parameter model 204, and a defect processing module 205. The visual sensor 201 can be composed of linear array cameras connected in parallel, images acquired by the visual sensor are transmitted to the neural network detection module 202 in real time, and the hardware configuration of the visual sensor 201 can be determined by parameters such as the width of a base material to be detected, the production speed, the number of pixel points of the camera image sensor, the required measurement precision and the like. The neural network detection module 202 is configured to receive the image information, read neural network parameters from the deep learning model 204, process the acquired image information, detect defects and types, and determine whether to coordinate other modules for post-processing. The parameters of the neural network may include preprocessing threshold, weight, network hyper-parameter, etc. The neural network learning module 203 is used for analyzing and classifying the defects detected by the neural network detection module 202 and judging whether to start the transfer learning of the neural network.
FIG. 3 is a flow chart illustrating a method of coating defect detection of a pole piece according to an exemplary embodiment. The method 30 for detecting coating defects of a pole piece at least includes steps S302 to S308.
As shown in fig. 3, in S302, the pole piece is subjected to coating processing. Coating is the main mode of large-scale production and processing of the pole piece. In order to accurately and uniformly coat the electrode material on the surface of the current collector, the electrode active material, a solvent, a conductive agent, an adhesive and the like can be fully mixed and stirred firstly in the production process, so that the stable property of the slurry is ensured, no sedimentation occurs, and the viscosity, the solid content and the like are not changed. The coating device and the control device perform coating treatment on the pole piece.
In S304, the pole piece image on which the coating process has been performed is acquired by the visual sensing device. The pole piece images that have been subjected to the coating process can be acquired, for example, by parallel line cameras in a visual sensing device. The visual sensing device is used for transmitting the images acquired in real time to the background server for subsequent processing.
In one embodiment, further comprising: and determining hardware parameters of the visual sensing device according to the width, transmission speed, measurement precision and pixel parameters of the linear array camera sensor of the pole piece to be detected. More specifically, the measurement accuracy can be determined by parameters such as the number of pixel points of the camera sensor, the measurement accuracy requirement of the pole piece and the like.
In S306, whether there is a defect in the pole piece image is determined by a coating defect detection model, where the coating defect detection model is generated by deep learning neural network model training.
The coating defect detection model is used for receiving the pole piece image, acquiring the trained neural network parameters, processing the pole piece image according to the parameters and judging whether the pole piece image has defects or not. When the pole piece has detection defects, the defect type can be output.
The specific contents of training the deep learning neural network model to generate the coating defect detection model will be described in the embodiment corresponding to fig. 5.
In S308, a defect detection result is generated when there is a defect in the pole piece image. The defect detection result may include information related to the code of the defective pole piece, the position of the defect, the type of the defect, and the like.
In one embodiment, further comprising: generating a defect marking instruction according to a defect detection result; and/or generating alarm record information according to the defect detection result; and/or generating a device adjustment instruction according to the defect detection result. And after the acquired pole piece image is input into the coating defect detection model, if no defect is detected, the next image is continuously acquired. When a defect is detected, the defective portion may be first marked. And then determining whether to generate an alarm record or an equipment adjusting instruction according to the subsequent defect condition. The adjustment instructions may include: and instructions such as adjustment, alarm, recording, stopping and the like are sent to production equipment.
In one embodiment, further comprising: acquiring a defect detection result within a preset time; and generating a shutdown operation instruction when the coating number of the pole pieces with the defects in the defect detection result is greater than a threshold value. After the defect is detected, the type and number of defects may be transmitted to a defect cache unit, which determines if shutdown is required based on the defect input over a period of time.
According to the method for detecting the coating defects of the pole piece, the pole piece image subjected to coating treatment is obtained through a visual sensing device; judging whether the pole piece image has defects or not through a coating defect detection model, wherein the coating defect detection model is generated through deep learning neural network model training; and a mode of generating a defect detection result when a defect exists in the pole piece image can be used for carrying out high-coverage defect detection on the pole piece of the lithium battery, and the requirements of the detection speed and the defect identification rate of the pole piece can be met even in a wide high-speed production scene of the pole piece of the lithium battery.
It should be clearly understood that this disclosure describes how to make and use particular examples, but the principles of this disclosure are not limited to any details of these examples. Rather, these principles can be applied to many other embodiments based on the teachings of the present disclosure.
FIG. 4 is a flow chart illustrating a method of coating defect detection of a pole piece according to another exemplary embodiment.
As shown in fig. 4, in S402, a defect area image is extracted from the defect detection result.
In S404, the defect type of the defective region image is determined by the coating defect detection model. Can include the following steps: the coating defect detection model determines the defect type of the defect region image based on a manifold learning method.
In one embodiment, the coating defect detection model determines the defect type of the defect region image based on a manifold learning method, including: the defect area image is compared to a plurality of samples of known types of defect area images to determine the type of defect.
In one embodiment, comparing the defect area image to a sample of defect area images of known type to determine the type of defect comprises: extracting a defect concentrated position in the defect area image, wherein the position can be defined as a first defect concentrated position; acquiring a plurality of defect area image samples of known types, and defining the marked defect positions on the defect area samples of known types as second defect concentrated positions; respectively placing the defect area image and the known type of defect area image in a plane coordinate (or other coordinates, but the disclosure is not limited thereto), respectively performing distance comparison on the first defect concentration position and the plurality of second defect concentration positions, and respectively generating a plurality of comparison distances; the defect region image of the known type closest to the comparison distance can be used as the defect most similar to the current defect image, and the defect type of the current image can be further determined.
In one specific embodiment, the manifold learning algorithm steps may be as follows (taking principal component analysis technique PCA as an example):
1. carrying out normalization preprocessing on the training images to obtain a training set:
X=(x1,x2,...,xm);
wherein X1 and X2 … … are normalized image data, and X is training set data.
2. Centering the image:
3. calculating a covariance matrix of the sample image:
4. solving the covariance matrix by using an SVD (singular value decomposition) method to obtain an eigenvalue and an eigenvector;
5. forming a matrix M by the feature vectors of the first two lines, and calculating MX to obtain a two-dimensional image;
6. classifying the samples in the two-dimensional space using a clustering method such as Fuzzy C Means;
7. the training set, validation set, and test set data are selected to ensure that each data set covers all sample classes.
In S406, when the defect type is unknown, the transfer learning process of the coating defect detection model is performed. The distance threshold may be empirically determined from historical data, and the defect type may be determined to be an unknown type, for example, when a plurality of comparison distances are each greater than the distance threshold.
In one embodiment, a transfer learning process of a coating defect detection model is performed, comprising: performing transfer learning on the coating defect detection model based on the pole piece image with the defect type being unknown to generate updated model parameters; and updating the coating defect detection model by updating the model parameters. In the deep learning, an original coating defect detection model is used as a starting point of a new model in a computer vision task and a natural language processing task, a newly acquired pole piece image of an unknown type is used as training data, and the coating defect detection model is trained again to acquire updated model parameters. Transfer learning can transfer learned powerful skills to related questions.
FIG. 5 is a flow chart illustrating a method of coating defect detection of a pole piece according to another exemplary embodiment. The flow shown in fig. 5 is a description of "generating a coating defect detection model by deep learning neural network model training".
As shown in fig. 5, in S502, the areas with defects in the historical pole piece images are marked to generate a set of defect images.
In S504, a normal image set is generated from the history pole piece images without defects.
In S506, the deep learning neural network model is trained through the defect image set and the normal image set to generate a coating defect detection model.
The input defect image set and normal image set may be trained, for example, by a plurality of depth modules of a deep learning neural network model; determining a plurality of probability distributions corresponding to a plurality of defect types by a logistic regression activation function of the deep learning neural network model; determining a loss function from the plurality of probability distributions; and determining a coating defect detection model through the related parameters of the deep neural network model when the loss function is the minimum value.
Determining a coating defect detection model through the relevant parameters of the deep neural network model when the loss function is the minimum value, and further comprising: the minimum of the loss function is obtained by the adammoptimizer optimization technique.
FIG. 6 is a schematic diagram illustrating a method of coating defect detection of a pole piece, according to an exemplary embodiment. Fig. 6 exemplarily illustrates the structure of the coating defect detection model, and the coating defect detection model is limited to be composed of four depth modules. Wherein each depth module comprises two convolutional layers (cov) and one pooling layer (pool). Finally, the features are classified by a full connectivity layer (Dense) and a logistic regression layer (Softmax).
The coating defect detection model comprises a training mode and a working mode, wherein in the training mode, one or more slave pole piece images are input into the coating defect detection model and can be divided into a damage group and a normal group, and the local images with damages in the damage group are marked by scratching. As an output, the network gets the probability distribution of each class by a Softmax activation function. This probability distribution can be calculated by cross entry as the error between the current classification and the gold standard is denoted as loss. Meanwhile, the training phase may use an adammoptimizer optimizer to minimize the loss value.
In one embodiment, the network hyper-parameter settings may be: the learning rate is 0.0003 and the filter window width is 3, 3. The training images can be divided into a lesion group and a normal group, wherein each image can generate labels of the training data in a manual type labeling mode.
Fig. 7 is a diagram of anode material collected by a production line and an example of a detection result according to an example embodiment. As can be seen from fig. 7, the shape and the appearance of the defect are clearly shown in the detection image, and the method in the disclosure can accurately acquire the defect appearance feature from the detection image.
FIG. 8 is a flow chart illustrating a method of coating defect detection of a pole piece according to another exemplary embodiment. Fig. 8 illustrates the overall process of the pole piece coating defect detection method by a specific application real-time example.
As shown in fig. 8, in S802, an image is acquired.
In S804, the coating defect detection model is input.
In S806, it is determined whether or not a defect exists.
In S808, defect marking is performed.
In S810, model training is performed again to update the coating defect detection model.
In S812, the defect is cached.
In S814, it is determined whether the number of defects exceeds a threshold.
In S816, a stop instruction is generated.
And inputting the collected pole piece image into a coating defect detection model. If no defect is detected, the next image is collected. When a defect is detected, the defective portion is first marked. And then transmitting the defect picture to a machine learning model for retraining. The machine learning model will determine whether or not migration learning needs to be continued. And using the parameters updated after learning for subsequent defect detection. While the defect type is transmitted to the defect cache unit. The unit determines whether shutdown processing is required based on defect input over a period of time.
Fig. 9 is a flowchart algorithm classification example illustrating generation of three types of negative electrode material defects according to an exemplary embodiment. As shown, the square marks represent the distribution of bubble defects; the circular marks represent the distribution of scratch defects; the triangular marks represent the distribution of streak defects. The abscissa represents the first principal axis after principal component analysis. The ordinate shows the second principal axis after principal component analysis. As can be seen from the figure, principal component analysis can effectively distinguish the three types of defects. Therefore, the model training based on the manifold algorithm can be effectively carried out.
Those skilled in the art will appreciate that all or part of the steps implementing the above embodiments are implemented as computer programs executed by a CPU. When executed by the CPU, performs the functions defined by the above-described methods provided by the present disclosure. The program of (a) may be stored in a computer readable storage medium, which may be a read-only memory, a magnetic or optical disk, or the like.
Furthermore, it should be noted that the above-mentioned figures are only schematic illustrations of the processes involved in the methods according to exemplary embodiments of the present disclosure, and are not intended to be limiting. It will be readily understood that the processes shown in the above figures are not intended to indicate or limit the chronological order of the processes. In addition, it is also readily understood that these processes may be performed synchronously or asynchronously, e.g., in multiple modules.
The following are embodiments of the disclosed apparatus that may be used to perform embodiments of the disclosed methods. For details not disclosed in the embodiments of the apparatus of the present disclosure, refer to the embodiments of the method of the present disclosure.
Fig. 10 is a block diagram illustrating a coating defect detection apparatus of a pole piece according to an exemplary embodiment. As shown in fig. 10, the apparatus 100 for detecting coating defects of a pole piece includes: coating module 1002, image module 1004, detection module 1006, and processing module 1008.
The coating module 1002 is used for coating the pole piece;
the image module 1004 is used for acquiring the pole piece image subjected to coating processing through the visual sensing device; the pole piece images that have been subjected to the coating process can be acquired, for example, by parallel line cameras in a visual sensing device. The visual sensing device is used for transmitting the images acquired in real time to the background server for subsequent processing.
The detection module 1006 is configured to determine whether a defect and a type of the defect exist in the pole piece image through a coating defect detection model, where the coating defect detection model is generated through deep learning neural network model training; the coating defect detection model is used for receiving the pole piece image, acquiring the trained neural network parameters, processing the pole piece image according to the parameters and judging whether the pole piece image has defects or not. When the pole piece has detection defects, the defect type can be output.
The processing module 1008 is used for error handling when there is a defect in the pole piece image. The processing module 1008 is further configured to generate a defect marking instruction according to the defect detection result; and/or generating alarm record information according to the defect detection result; and/or generating a device adjustment instruction according to the defect detection result. The processing module 1008 is further configured to obtain a defect detection result within a predetermined time; and generating a shutdown operation instruction when the coating number of the pole pieces with the defects in the defect detection result is greater than a threshold value.
Fig. 11 is a block diagram illustrating a coating defect detecting apparatus of a pole piece according to another exemplary embodiment. As shown in fig. 11, the apparatus 110 for detecting coating defect of pole piece may further include, in addition to the apparatus 100 for detecting coating defect of pole piece: a migration learning module 1102 and a model training module 1104.
The transfer learning module 1102 is configured to extract a defect region image from the defect detection result; determining the defect type of the defect area image through a coating defect detection model; and when the defect type is an unknown type, performing transfer learning processing of the coating defect detection model.
The model training module 1104 is used for marking the defective areas in the historical pole piece images to generate a defective image set; generating a normal image set through the historical pole piece images without defects; and training the deep learning neural network model through the defect image set and the normal image set to generate a coating defect detection model.
According to the device for detecting the coating defects of the pole piece, disclosed by the invention, the pole piece image subjected to coating treatment is obtained through a visual sensing device; judging whether the pole piece image has defects or not through a coating defect detection model, wherein the coating defect detection model is generated through deep learning neural network model training; and a mode of generating a defect detection result when a defect exists in the pole piece image can be used for carrying out high-coverage defect detection on the pole piece of the lithium battery, and the requirements of the detection speed and the defect identification rate of the pole piece can be met even in a wide high-speed production scene of the pole piece of the lithium battery.
FIG. 12 is a block diagram illustrating an electronic device in accordance with an example embodiment.
An electronic device 1200 according to this embodiment of the disclosure is described below with reference to fig. 12. The electronic device 1200 shown in fig. 12 is only an example and should not bring any limitation to the functions and the scope of use of the embodiments of the present disclosure.
As shown in fig. 12, the electronic device 1200 is embodied in the form of a general purpose computing device. The components of the electronic device 1200 may include, but are not limited to: at least one processing unit 1210, at least one memory unit 1220, a bus 1230 connecting the various system components including the memory unit 1020 and the processing unit 1010, a display unit 1240, and the like.
Wherein the storage unit stores program codes, which can be executed by the processing unit 1210, so that the processing unit 1210 performs the steps according to various exemplary embodiments of the present disclosure described in the above-mentioned electronic prescription flow processing method section of this specification. For example, the processing unit 1210 may perform the steps as shown in fig. 3, 4, 5, 8.
The storage unit 1220 may include a readable medium in the form of a volatile memory unit, such as a random access memory unit (RAM)10201 and/or a cache memory unit 12202, and may further include a read only memory unit (ROM) 12203.
The memory unit 1220 may also include a program/utility 12204 having a set (at least one) of program modules 12205, such program modules 10205 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which, or some combination thereof, may comprise an implementation of a network environment.
The electronic device 1200 may also communicate with one or more external devices 1200' (e.g., keyboard, pointing device, bluetooth device, etc.), with one or more devices that enable a user to interact with the electronic device 1200, and/or with any devices (e.g., router, modem, etc.) that enable the electronic device 1200 to communicate with one or more other computing devices. Such communication may occur via input/output (I/O) interfaces 1250. Also, the electronic device 1200 may communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network such as the Internet) via the network adapter 1260. The network adapter 1260 may communicate with other modules of the electronic device 1200 via the bus 1230. It should be appreciated that although not shown, other hardware and/or software modules may be used in conjunction with the electronic device 1200, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, among others.
Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein may be implemented by software, or by software in combination with necessary hardware. Therefore, as shown in fig. 13, the technical solution according to the embodiment of the present disclosure may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (which may be a CD-ROM, a usb disk, a removable hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which may be a personal computer, a server, or a network device, etc.) to execute the above method according to the embodiment of the present disclosure.
The software product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
The computer readable storage medium may include a propagated data signal with readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A readable storage medium may also be any readable medium that is not a readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a readable storage medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
Program code for carrying out operations for the present disclosure may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computing device (e.g., through the internet using an internet service provider).
The computer readable medium carries one or more programs which, when executed by a device, cause the computer readable medium to perform the functions of: acquiring a pole piece image subjected to coating treatment by a visual sensing device; judging whether the pole piece image has defects or not through a coating defect detection model, wherein the coating defect detection model is generated through deep learning neural network model training; and generating a defect detection result when the pole piece image has a defect.
Those skilled in the art will appreciate that the modules described above may be distributed in the apparatus according to the description of the embodiments, or may be modified accordingly in one or more apparatuses unique from the embodiments. The modules of the above embodiments may be combined into one module, or further split into multiple sub-modules.
Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein may be implemented by software, or by software in combination with necessary hardware. Therefore, the technical solution according to the embodiments of the present disclosure may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (which may be a CD-ROM, a usb disk, a removable hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which may be a personal computer, a server, a mobile terminal, or a network device, etc.) to execute the method according to the embodiments of the present disclosure.
Exemplary embodiments of the present disclosure are specifically illustrated and described above. It is to be understood that the present disclosure is not limited to the precise arrangements, instrumentalities, or instrumentalities described herein; on the contrary, the disclosure is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.
Claims (19)
1. A method for detecting coating defects of a pole piece is characterized by comprising the following steps:
coating the electrode plate;
acquiring a pole piece image subjected to coating treatment by a visual sensing device;
judging whether the pole piece image has defects or not through a coating defect detection model, wherein the coating defect detection model is generated through deep learning neural network model training; and
and generating a defect detection result when the pole piece image has defects.
2. The method of claim 1, further comprising:
extracting a defect area image from the defect detection result;
determining the defect type of the defect area image through the coating defect detection model; and
and when the defect type is an unknown type, performing transfer learning processing of the coating defect detection model.
3. The method of claim 1, further comprising:
marking the area with the defects in the historical pole piece image to generate a defect image set;
generating a normal image set through the historical pole piece images without defects; and
training a deep learning neural network model through the defect image set and the normal image set to generate the coating defect detection model.
4. The method of claim 1, wherein acquiring the image of the coated pole piece by a visual sensing device comprises:
and acquiring the pole piece image subjected to coating treatment by the aid of the parallel linear array cameras in the visual sensing device.
5. The method of claim 4, further comprising:
and determining hardware parameters of the visual sensing device according to the width, transmission speed, measurement precision and pixel parameters of a linear array camera sensor of the pole piece to be measured.
6. The method of claim 2, wherein determining the defect type of the defect region image by the coating defect detection model comprises:
the coating defect detection model determines the defect type of the defect region image based on a manifold learning method.
7. The method of claim 6, wherein the coating defect detection model determines the defect type of the defect region image based on a manifold learning method, comprising:
comparing the defect area image to a plurality of samples of known types of defect area images to determine the defect type.
8. The method of claim 7, wherein comparing the defect area image to a sample of defect area images of known types to determine the defect type comprises:
extracting a first defect concentrated position in the defect area image;
acquiring a plurality of second defect concentrated positions in a plurality of defect area image samples of known types;
comparing the first defect concentration position with a plurality of second defect concentration positions respectively to generate a plurality of comparison distances; and
determining the defect type based on the plurality of comparison distances.
9. The method of claim 8, wherein when the defect type is an unknown type, comprising:
and when the comparison distances are all larger than a distance threshold value, determining that the defect type is an unknown type.
10. The method of claim 2, wherein performing a transfer learning process of the coating defect detection model comprises:
carrying out transfer learning on the coating defect detection model based on the pole piece image with the defect type being an unknown type to generate an updated model parameter; and
and updating the coating defect detection model through the updated model parameters.
11. The method of claim 3, wherein training a deep learning neural network model through the set of defect images and the set of normal images to generate the coating defect detection model comprises:
training the input defect image set and the normal image set by a plurality of depth modules of a deep learning neural network model;
determining a plurality of probability distributions corresponding to a plurality of defect types by a logistic regression activation function of the deep learning neural network model;
determining a loss function from the plurality of probability distributions;
and determining the coating defect detection model through the related parameters of the deep neural network model when the loss function is the minimum value.
12. The method of claim 11, wherein determining the coating defect detection model from parameters associated with the deep neural network model at which the loss function is a minimum further comprises:
and obtaining the minimum value of the loss function through an optimization technology.
13. The method of claim 1, further comprising:
generating a defect marking instruction according to the defect detection result; and/or
Generating alarm record information according to the defect detection result; and/or
And generating a device adjusting instruction according to the defect detection result.
14. The method of claim 1, further comprising:
acquiring a defect detection result within a preset time; and
and when the coating number of the pole pieces with the defects in the defect detection result is greater than a threshold value, generating a shutdown operation instruction.
15. A coating defect detection device of pole piece, characterized by includes:
the coating module is used for coating the pole piece;
the image module is used for acquiring the pole piece image subjected to coating treatment through the visual sensing device;
the detection module is used for judging whether the pole piece image has defects and the types of the defects through a coating defect detection model, wherein the coating defect detection model is generated through deep learning neural network model training; and
and the processing module is used for carrying out error processing when the pole piece image has defects.
16. The apparatus of claim 15, further comprising:
the transfer learning module is used for extracting a defect area image from the defect detection result; determining the defect type of the defect area image through the coating defect detection model; and when the defect type is an unknown type, performing transfer learning processing of the coating defect detection model.
17. The apparatus of claim 15, further comprising:
the model training module is used for marking the areas with the defects in the historical pole piece images to generate a defect image set; generating a normal image set through the historical pole piece images without defects; and training a deep learning neural network model through the defect image set and the normal image set to generate the coating defect detection model.
18. An electronic device, comprising:
one or more processors;
storage means for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement the method of any one of claims 1-14.
19. A computer-readable medium, on which a computer program is stored, which, when being executed by a processor, carries out the method according to any one of claims 1-14.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010699517.XA CN114037645A (en) | 2020-07-20 | 2020-07-20 | Coating defect detection method and device for pole piece, electronic equipment and readable medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010699517.XA CN114037645A (en) | 2020-07-20 | 2020-07-20 | Coating defect detection method and device for pole piece, electronic equipment and readable medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114037645A true CN114037645A (en) | 2022-02-11 |
Family
ID=80134049
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010699517.XA Pending CN114037645A (en) | 2020-07-20 | 2020-07-20 | Coating defect detection method and device for pole piece, electronic equipment and readable medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114037645A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114850057A (en) * | 2022-05-10 | 2022-08-05 | 苏州天准科技股份有限公司 | Electrode comprehensive detection equipment and detection method |
CN116645429A (en) * | 2023-07-25 | 2023-08-25 | 山东中胜涂料有限公司 | Visual-aided paint production sample color analysis and detection method |
CN116818799A (en) * | 2023-06-20 | 2023-09-29 | 广州宝立科技有限公司 | Intelligent cloth inspection method and device based on machine vision detection technology |
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 |
-
2020
- 2020-07-20 CN CN202010699517.XA patent/CN114037645A/en active Pending
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114850057A (en) * | 2022-05-10 | 2022-08-05 | 苏州天准科技股份有限公司 | Electrode comprehensive detection equipment and detection method |
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 |
CN116818799A (en) * | 2023-06-20 | 2023-09-29 | 广州宝立科技有限公司 | Intelligent cloth inspection method and device based on machine vision detection technology |
CN116645429A (en) * | 2023-07-25 | 2023-08-25 | 山东中胜涂料有限公司 | Visual-aided paint production sample color analysis and detection method |
CN116645429B (en) * | 2023-07-25 | 2023-10-20 | 山东中胜涂料有限公司 | Visual-aided paint production sample color analysis and detection method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN114037645A (en) | Coating defect detection method and device for pole piece, electronic equipment and readable medium | |
CN111598164B (en) | Method, device, electronic equipment and storage medium for identifying attribute of target object | |
CN110544231B (en) | Lithium battery electrode surface defect detection method based on background standardization and centralized compensation algorithm | |
CN111209958B (en) | Substation equipment detection method and device based on deep learning | |
CN113252701B (en) | Cloud edge cooperation-based power transmission line insulator self-explosion defect detection system and method | |
WO2024002187A1 (en) | Defect detection method, defect detection device, and storage medium | |
CN115131283A (en) | Defect detection and model training method, device, equipment and medium for target object | |
CN110472581A (en) | A kind of cell image analysis method based on deep learning | |
CN112419268A (en) | Method, device, equipment and medium for detecting image defects of power transmission line | |
CN113344862A (en) | Defect detection method, defect detection device, electronic equipment and storage medium | |
Hu et al. | Telescope performance real-time monitoring based on machine learning | |
CN115984158A (en) | Defect analysis method and device, electronic equipment and computer readable storage medium | |
EP4332915A1 (en) | Automated selection and model training for charged particle microscope imaging | |
CN115482436B (en) | Training method and device for image screening model and image screening method | |
CN116188445A (en) | Product surface defect detection and positioning method and device and terminal equipment | |
CN115937195A (en) | Surface defect detection method and device and computer equipment | |
KR20240064037A (en) | How to identify and characterize surface defects of objects and cracks in brake discs by artificial intelligence through fatigue testing | |
CN113487550B (en) | Target detection method and device based on improved activation function | |
CN116091416A (en) | Method and device for training assembly defect detection and change detection models of printed circuit board | |
CN112861670B (en) | Transmission line hardware detection method and system | |
Anilturk et al. | Automatic defect classification (ADC) solution using data-centric artificial intelligence (AI) for outgoing quality inspections in the semiconductor industry | |
CN114266941A (en) | Method for rapidly detecting annotation result data of image sample | |
Benbarrad et al. | On the Performance of Deep Learning in the Full Edge and the Full Cloud Architectures | |
CN112579429A (en) | Problem positioning method and device | |
CN113570566B (en) | Product appearance defect development cognition detection method and related device |
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 |