CN115063421B - Pole piece region detection method, system and device, medium and defect detection method - Google Patents
Pole piece region detection method, system and device, medium and defect detection method Download PDFInfo
- Publication number
- CN115063421B CN115063421B CN202210981487.0A CN202210981487A CN115063421B CN 115063421 B CN115063421 B CN 115063421B CN 202210981487 A CN202210981487 A CN 202210981487A CN 115063421 B CN115063421 B CN 115063421B
- Authority
- CN
- China
- Prior art keywords
- image
- pole piece
- obtaining
- contour
- battery pole
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20004—Adaptive image processing
-
- 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]
-
- 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
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P70/00—Climate change mitigation technologies in the production process for final industrial or consumer products
- Y02P70/50—Manufacturing or production processes characterised by the final manufactured product
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a pole piece region detection method, a pole piece region detection system, a pole piece region detection device, a medium and a defect detection method, and relates to the field of lithium ion batteries, wherein a machine vision method is adopted in the invention, so that battery pole piece regions in different pictures to be detected can be selected in a self-adaptive frame mode, and the influence of burrs possibly existing on the edges of the battery pole pieces on the frame selection of the battery pole piece regions can be removed through histogram statistics; meanwhile, the detection result obtained by detecting the image to be detected through deep learning and the battery pole piece region are subjected to intersection operation, so that the over-detection phenomenon caused by the fact that noise in the background is recognized as a defect in the battery region due to factors such as industrial fluctuation, machine station difference and illumination in the production process in the deep learning process is effectively solved.
Description
Technical Field
The invention relates to the field of lithium ion batteries, in particular to a pole piece region detection method, a pole piece region detection system, a pole piece region detection device, a pole piece region detection medium and a defect detection method.
Background
In recent years, with the continuous development of various electric devices, people have an increasing demand for lithium ion batteries, and the existing lithium ion batteries generally comprise laminated batteries and wound batteries, and the quality of battery pole pieces of the laminated batteries or the wound batteries has a serious influence on the quality of the manufactured lithium ion batteries. Therefore, it is very important to accurately detect the defects of the battery pole piece in the manufacturing process of the battery pole piece. The existing detection of the defects of the battery pole piece is usually realized through deep learning, however, the color and background difference of a battery pole piece region in a shot picture can be smaller due to factors such as industrial fluctuation, machine station difference and illumination in the industrial manufacturing process, so that the problem of over-detection of the defects of the battery pole piece by detecting the noise of the background region into the defects of the battery pole piece can be generated when the defects of the battery pole piece are detected by using the deep learning, the battery pole piece is different from the defect detection of electronic components, the structure is simple, the labeling frame can not be enlarged by modifying the label of training data, and the problem of over-detection of the defects by the deep learning is avoided through structural information around the electronic components or a mode of enlarging a training data set.
Disclosure of Invention
In order to accurately obtain the area of the battery pole piece and distinguish the area of the battery pole piece from the background area of the battery pole piece, the invention provides a pole piece area detection method, which comprises the following steps:
obtaining an image to be detected, and converting the image to be detected into a gray image;
calculating a binarization threshold value according to the gray level image to obtain a binarization image;
filtering the binary image to obtain a de-noised image;
extracting the contours of the de-noised images to obtain a plurality of contour regions, and screening the contour regions to obtain a screening result;
and calculating according to the screening result to obtain the area coordinates of the battery pole piece.
The pole piece region detection method provided by the invention has the following principle: obtaining an image to be detected, carrying out graying processing on the image to be detected to obtain a grayscale image, then obtaining a corresponding binaryzation threshold value according to the grayscale image, separating a battery pole piece region from a background region in the grayscale image according to the binaryzation threshold value, then sequentially carrying out filtering processing, contour extraction and contour screening on the binaryzation image, removing the influence of various noises in the binaryzation image, obtaining a contour corresponding to the battery pole piece region, and finally calculating the contour corresponding to the battery pole piece region to obtain a coordinate of the battery pole piece region.
The method comprises the following steps of calculating different binarization threshold values in a self-adaptive manner according to different gray level images because the influence of factors such as industrial fluctuation, machine station difference and illumination possibly generated in the production process on the image to be detected is random, and calculating the binarization threshold values according to the gray level images to obtain the binarization image, wherein the following steps are included:
cutting the gray level image, reserving the central part of the gray level image, and obtaining a standard gray level image;
calculating the gray average value of the standard gray image, wherein the gray average value is the binarization threshold value, and the specific calculation mode is as follows:
wherein thres represents a binarization threshold, n represents the number of pixels in the standard gray-scale map,
representing the coordinates in the standard gray scale map as
The gray value corresponding to the pixel of (a);
and segmenting the pixels of the gray level image according to the binarization threshold value to obtain a binarization image.
The method comprises the following steps of calculating a plurality of contours of a binary image, wherein the contours obtained by contour extraction of the binary image are a set of a plurality of points, and the step of calculating the coordinates of a battery pole piece region according to the screening result comprises the following steps:
obtaining a first contour point set corresponding to the screening result;
obtaining coordinate extreme values of a plurality of points in the first contour point set, wherein the coordinate extreme values comprise a first abscissa maximum value, a first abscissa minimum value, a first ordinate maximum value and a first ordinate minimum value;
and obtaining the area coordinates of the battery pole piece according to the coordinate extreme value.
Furthermore, since the identification result obtained after the contour identification of the binarized image may include a contour corresponding to a battery pole piece region, a contour corresponding to a defect on the battery pole piece, and a contour corresponding to noise in an image background region, the identification result obtained by the contour identification needs to be screened, and in order to facilitate realization of the screening condition and ensure that the screening condition can accurately distinguish the identification result obtained by the contour identification, the screening of the plurality of contour regions is performed to obtain the screening result, the screening condition is the area of the plurality of contour regions, and the contour region with the largest area in the plurality of contour regions is the screening result.
Further, since the plurality of contour regions are irregular figures formed by surrounding a plurality of contour regions, in order to accurately obtain the areas of the plurality of contour regions and then screen the plurality of contour regions according to the areas to obtain the screening result, the pole piece region detection method calculates the areas of the plurality of contour regions by using a green formula, and the specific calculation method is as follows:
wherein S represents the area of the contour region, Q represents a fitting function corresponding to the contour in the contour region,
the partial derivative of the function Q in the x direction is shown, P represents the fitting function corresponding to the outline of the contour region,
representing the partial derivative of the function P in the y-direction,
represent
In the x-direction and y-directionThe double differential of (a) is obtained,
representing the differential of the function P in the x-direction,
representing the differential of the function Q in the y-direction.
Further, since burrs may exist at the edges of the battery pole piece, in order to eliminate the influence of the burrs on the extracted coordinates of the contour region of the battery pole piece, the obtaining of the coordinates of the contour region of the battery pole piece according to the extreme coordinate value further includes the following steps:
obtaining image size data of the binary image, wherein the image size data comprises a first image height and a first image width;
obtaining image size data of the first contour point set according to the coordinate extreme value, wherein the image size data comprises a second image height and a second image width;
judging the heights of the first image and the second image, if the height of the first image is greater than the height of the second image, namely the image to be detected comprises the upper edge and/or the lower edge of a battery pole piece, the influence of burrs possibly generated on the upper edge and/or the lower edge of the battery pole piece on the detection of the battery pole piece area needs to be eliminated, so that the first contour point set is processed through the histogram statistics to obtain the maximum value of a second vertical coordinate and the minimum value of the second vertical coordinate, and the coordinate extreme value is updated;
judging the sizes of the first image width and the second image width, if the first image width is larger than the second image width, namely the image to be detected comprises the left edge and/or the right edge of the battery pole piece, the influence of burrs possibly generated on the left edge and/or the right edge of the battery pole piece on the detection of the battery pole piece area needs to be eliminated, so that the first contour point set is processed through the histogram statistics to obtain a second abscissa maximum value and a second abscissa minimum value, and the coordinate extreme value is updated;
and obtaining the area coordinates of the battery pole piece according to the updated coordinate extreme value.
In order to filter out noise that may occur in an obtained image and improve accuracy of image processing, filtering processing needs to be performed on a binarized line difference map to obtain a denoised image, where the filtering processing includes: and carrying out corrosion operation on the binary image to obtain a first picture, and then carrying out expansion operation on the first picture. The erosion operation can remove burrs, small points and small bridges in the image, the expansion operation can expand the image boundary to the outside, the erosion operation is firstly carried out on the image, then the expansion operation is carried out, the small object can be eliminated, the object is separated at the fine point, the boundary of the larger object is smoothened, and meanwhile the area of the image is not obviously changed, and the clear de-noising image is obtained.
In order to achieve the above object, the present invention provides a pole piece region detection system, comprising:
the image acquisition unit is used for acquiring an image to be detected and converting the image to be detected into a gray image;
the image processing unit is used for calculating a binarization threshold value according to the gray level image to obtain a binarization image, and filtering the binarization image to obtain a de-noised image;
and the region calculation unit is used for extracting the contours of the de-noised image to obtain a plurality of contour regions, screening the contour regions to obtain a screening result, and calculating according to the screening result to obtain the region coordinates of the battery pole piece.
In order to achieve the above object, the present invention provides a pole piece region detection apparatus, including a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor implements the steps of any one of the pole piece region detection methods when executing the computer program.
In order to achieve the above object, the present invention provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, the computer program implements the steps of any one of the above pole piece region detection methods.
In order to accurately obtain the area of the battery pole piece, distinguish the area of the battery pole piece from the background of the battery pole piece, and screen the detection result of the battery pole piece image through deep learning by the obtained area of the battery pole piece, thereby avoiding the over-detection problem possibly generated when the defect detection is carried out on the battery pole piece by the deep learning, the invention also provides a pole piece defect detection method, which comprises the following steps:
establishing a deep learning model, and obtaining a training set for training the deep learning model, wherein the training set comprises a plurality of battery pole piece images with defects and corresponding defect marks;
training the deep learning model through the training set to obtain a second deep learning model;
obtaining the area coordinates of the battery pole piece according to any one of the pole piece area detection methods;
detecting defects of the image to be detected according to the second deep learning model to obtain a plurality of defect coordinates;
and judging whether the plurality of defect coordinates are in the battery pole piece area coordinates, if so, determining that the battery pole piece has defects at the position corresponding to the defect coordinates.
The pole piece defect detection principle provided by the invention is as follows: according to any pole piece region detection method, adaptive threshold segmentation is carried out on an image through the traditional machine vision, the coordinates of the battery pole piece region are accurately found out, on the basis, intersection calculation is carried out on the pole piece region defects detected through deep learning and the coordinates of the battery pole piece region, and defect detection frames in the background can be filtered out, so that the over-detection problem of the A/D conversion system of the system frequently occurring in the detection of the battery pole piece defects is reduced, and the accuracy of battery pole piece defect identification is improved.
One or more technical schemes provided by the invention at least have the following technical effects or advantages: the pole piece region detection method provided by the invention can adaptively frame out the battery pole piece regions in different pictures to be detected, and can remove the influence of burrs possibly existing on the edges of the battery pole piece on the frame selection of the battery pole piece regions through histogram statistics; meanwhile, the pole piece defect detection method provided by the invention can select the battery pole piece region through the pole piece region detection method frame, and performs intersection operation on the detection result obtained by detecting the image to be detected through deep learning and the battery pole piece region, so that the over-detection phenomenon caused by recognizing noise in the background as the defect in the battery region due to factors such as industrial fluctuation, machine station difference and illumination in the production process in the deep learning process is avoided.
Drawings
The accompanying drawings, which are included to provide a further understanding of the embodiments of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention;
FIG. 1 is a schematic view of a detection process of a pole piece region in the present invention;
FIG. 2 is a schematic diagram of the defect detection process of the pole piece of the present invention;
FIG. 3 is a schematic view of a pole piece area detection system of the present invention;
FIG. 4 is a partial schematic view of the edge burr of the pole piece of the present invention.
Detailed Description
In order that the above objects, features and advantages of the present invention can be more clearly understood, a more particular description of the invention will be rendered by reference to the appended drawings. It should be noted that the embodiments and features of the embodiments of the present invention may be combined with each other without conflicting with each other.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, however, the present invention may be practiced in other ways than those specifically described and thus the scope of the present invention is not limited by the specific embodiments disclosed below.
Example one
Referring to fig. 1, a first embodiment of the present invention provides a pole piece region detection method, where the pole piece region detection method includes the following steps:
obtaining an image to be detected, and converting the image to be detected into a gray image;
calculating a binarization threshold value according to the gray level image to obtain a binarization image;
filtering the binary image to obtain a de-noised image;
extracting the contours of the de-noised images to obtain a plurality of contour regions, and screening the contour regions to obtain a screening result;
and calculating according to the screening result to obtain the area coordinates of the battery pole piece.
The contour extraction of the image is performed to find shape boundary information in the image, and the contour extraction of the denoised image may be implemented by a Canny contour extraction algorithm, a threshold segmentation contour extraction algorithm, a connected region algorithm, an ant colony algorithm, or the like, and the specific algorithm is not specifically limited in this embodiment.
In order to filter out noise that may occur in an obtained image and improve accuracy of image processing, filtering processing needs to be performed on a binarized line difference map to obtain a denoised image, where the specific method of filtering and denoising may be median filtering, mean filtering, gaussian filtering, or bilateral filtering, and in this embodiment, the filtering processing includes: and carrying out corrosion operation on the binary image to obtain a first picture, and then carrying out expansion operation on the first picture. The erosion operation can remove burrs, small points and small bridges in the image, the expansion operation can expand the image boundary to the outside, the erosion operation is firstly carried out on the image, then the expansion operation is carried out, the small object can be eliminated, the object is separated at the fine point, the boundary of the larger object is smoothened, and meanwhile the area of the image is not obviously changed, and the clear de-noising image is obtained.
In the embodiment, the image to be detected can be obtained by a CCD camera or a CMOS camera, and the image pixel obtained by the CCD camera has a simple structure, so that the requirement of the acquisition definition of the image under the industrial condition can be met, and the CCD camera is preferably used for acquiring the image to be detected.
In this embodiment, the calculating a binarization threshold according to the grayscale image to obtain a binarized image includes the following steps:
cutting the gray image, and reserving the central part of the gray image to obtain a standard gray image;
calculating the gray average value of the standard gray image, wherein the gray average value is the binarization threshold value, and the specific calculation mode is as follows:
wherein thres represents a binarization threshold, n represents the number of pixels in the standard gray-scale map,
representing the coordinates in the standard gray scale map as
The gray value corresponding to the pixel of (a);
and segmenting the pixels of the gray level image according to the binarization threshold value to obtain a binarization image.
Specifically, the method for segmenting the pixels of the gray image according to the binarization threshold value is to white a pole piece region in the gray image and black other regions; recording the gray value of each pixel point in the gray image as
And processing each pixel point according to the binarization threshold value by using the following formula to obtain a binarization image:
wherein, the first and the second end of the pipe are connected with each other,
representing the binarized image, thres representing a binarization threshold,
and expressing the gray value corresponding to each pixel point in the gray image, wherein x and y represent the coordinate corresponding to each pixel point in the gray image.
In this embodiment, since the plurality of contours obtained by extracting the contours of the binarized image are a set of a plurality of points, the step of calculating the coordinates of the battery pole piece region according to the screening result includes the following steps:
obtaining a first contour point set corresponding to the screening result;
obtaining the coordinate extreme values of a plurality of points in the first contour point set, including the maximum value of the first abscissa
First minimum value of abscissa
Maximum value of first ordinate
And a first ordinate minimum
；
Obtaining the coordinates of the battery pole piece region according to the coordinate extreme value, specifically, the coordinates of the four corners of the battery pole piece region are respectively
、
、
、
. The plurality of contour areas comprise contour areas corresponding to the battery pole pieces, contour areas corresponding to defects possibly existing in the battery pole pieces and wheels possibly existing in the background imagesIn the embodiment, the plurality of contour regions are screened to obtain a screening result, the screening basis is the areas of the plurality of contour regions, and the contour region with the largest area in the plurality of contour regions is the screening result; because the plurality of contour regions are irregular figures formed by surrounding a plurality of contour regions, in order to accurately obtain the areas of the plurality of contour regions and screen the plurality of contour regions according to the areas to obtain screening results, the pole piece region detection method calculates the areas of the plurality of contour regions by a Green formula, and the specific calculation mode is as follows:
wherein S represents the area of the contour region, Q represents a fitting function corresponding to the contour in the contour region,
the partial derivative of the function Q in the x direction is shown, P represents the fitting function corresponding to the outline of the contour region,
representing the partial derivative of the function P in the y-direction,
to represent
Double differentiation in the x-direction and y-direction,
representing the differential of the function P in the x-direction,
representing the differential of the function Q in the y-direction.
As shown in fig. 4, a white area in the drawing is a battery pole piece area, and it can be observed that burrs may exist at the edge of the battery pole piece, so that obtaining coordinates of the battery pole piece area according to the coordinate extreme value further includes the following steps:
obtaining image size data of the binary image, wherein the image size data comprises a first image height and a first image width;
obtaining image size data of the first contour point set according to the coordinate extreme value, wherein the image size data comprises a second image height and a second image width;
judging the height of the first image and the height of the second image, if the height of the first image is larger than the height of the second image, processing the first contour point set through histogram statistics to obtain a second vertical coordinate maximum value and a second vertical coordinate minimum value, and updating the coordinate extreme value;
judging the sizes of the first image width and the second image width, if the first image width is larger than the second image width, processing the first contour point set through histogram statistics to obtain a second abscissa maximum value and a second abscissa minimum value, and updating the coordinate extreme value;
and obtaining the area coordinates of the battery pole piece according to the updated coordinate extreme value.
Specifically, the processing the first contour point set through the histogram statistics to obtain a second maximum value and a second minimum value of a vertical coordinate includes the following steps:
obtaining a first maximum value of ordinate
And a first ordinate minimum
；
Calculating to obtain a longitudinal coordinate intermediate line
；
Dividing the first contour point set into a second contour point set and a third contour point set according to the ordinate middle line;
respectively carrying out histogram statistics on the vertical coordinates of a plurality of points in the second profile point set and the third profile point set, and respectively obtaining the vertical coordinate corresponding to the maximum value in the statistical values, namely the maximum value of the second vertical coordinate
And a second ordinate minimum
；
Specifically, the processing the first contour point set through the histogram statistics to obtain a second abscissa maximum value and a second abscissa minimum value includes the following steps:
obtaining a first maximum value of abscissa
And a first abscissa minimum value
；
Calculating to obtain a middle line of the abscissa
；
Dividing the first contour point set into a fourth contour point set and a fifth contour point set according to the abscissa middle line;
respectively carrying out histogram statistics on the horizontal coordinates of a plurality of points in the fourth contour point set and the fifth contour point set, and respectively obtaining the horizontal coordinate corresponding to the maximum value in the statistics, namely the maximum value of the second horizontal coordinate
And a second abscissa minimum value
。
It is understood that when the first image height is greater than the second image height, but the first image width is less than or equal to the second image width, the four angular coordinates of the battery pole piece region are respectively
、
、
、
；
It is understood that when the first image height is less than or equal to the second image height, but the first image width is greater than the second image width, the four angular coordinates of the battery pole piece region are respectively
、
、
、
；
It is understood that when the first image height is greater than the second image height and the first image width is greater than the second image width, the four corner coordinates of the battery pole piece region are respectively
、
、
、
；
It is understood that when the first image height is less than or equal to the second image height and the first image width is less than or equal to the second image width, the four corner coordinates of the battery pole piece region are respectively
、
、
、
。
Example two
Referring to fig. 2, a second embodiment of the present invention provides a pole piece defect detection method, which is implemented on the basis of the first embodiment, and the pole piece defect detection method includes the following steps:
establishing a deep learning model, and obtaining a training set for training the deep learning model, wherein the training set comprises a plurality of battery pole piece images with defects and corresponding defect marks;
training the deep learning model through the training set to obtain a second deep learning model;
obtaining an image to be detected, and processing the image to be detected according to the pole piece region detection method provided by the embodiment I to obtain the region coordinates of the battery pole piece;
detecting defects of the image to be detected according to the second deep learning model to obtain a plurality of defect coordinates;
and judging whether the plurality of defect coordinates are in the battery pole piece area coordinates, if so, determining that the battery pole piece has defects at the position corresponding to the defect coordinates.
In this embodiment, the deep learning model may be a convolutional neural network model (CNN), a recurrent neural network model (RNN), a Stacked Automatic Encoder (SAE), or a deep belief network model (DBN), and various deep learning models derived from the deep learning model, which is not specifically limited herein.
EXAMPLE III
The third embodiment of the invention provides a pole piece region detection device, which comprises a memory, a processor and a computer program which is stored in the memory and can run on the processor, wherein the step of the pole piece region detection method is realized when the processor executes the computer program.
Example four
The fourth embodiment of the present invention provides a computer-readable storage medium, where a computer program is stored, and when the computer program is executed by a processor, the steps of the pole piece region detection method are implemented.
The processor may be a Central Processing Unit (CPU), other general purpose processor, a digital signal processor (digital signal processor), an Application Specific Integrated Circuit (Application Specific Integrated Circuit), an off-the-shelf programmable gate array (Field programmable gate array) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory can be used for storing the computer program and/or the module, and the processor can realize various functions of the pole piece region detection device in the invention by operating or executing the data stored in the memory. The memory may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function (such as a sound playing function, an image playing function, etc.), and the like. Further, the memory may include high speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card, a secure digital card, a flash memory card, at least one magnetic disk storage device, a flash memory device, or other volatile solid state storage device.
The pole piece region detection device, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, all or part of the flow in the method of the embodiments described above can be realized by the present invention, and can also be stored in a computer readable storage medium through a computer program, and when the computer program is executed by a processor, the steps of the method embodiments described above can be realized. Wherein the computer program comprises computer program code, an object code form, an executable file or some intermediate form, etc. The computer readable medium may include: any entity or device capable of carrying said computer program code, a recording medium, a usb-disk, a removable hard disk, a magnetic disk, an optical disk, a computer memory, a read-only memory, a random access memory, a point carrier signal, a telecommunications signal, a software distribution medium, etc. It should be noted that the computer readable medium may contain content that is appropriately increased or decreased as required by legislation and patent practice in the jurisdiction.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (9)
1. The pole piece region detection method is characterized by comprising the following steps:
obtaining an image to be detected, and converting the image to be detected into a gray image;
calculating a binarization threshold value according to the gray level image to obtain a binarization image;
filtering the binary image to obtain a de-noised image;
extracting the contours of the de-noised images to obtain a plurality of contour regions, and screening the contour regions to obtain a screening result;
calculating according to the screening result to obtain the area coordinates of the battery pole piece;
the method for obtaining the area coordinates of the battery pole piece through calculation according to the screening result comprises the following steps:
obtaining a first contour point set corresponding to the screening result;
obtaining coordinate extreme values of a plurality of points in the first contour point set, wherein the coordinate extreme values comprise a first abscissa maximum value, a first abscissa minimum value, a first ordinate maximum value and a first ordinate minimum value;
obtaining the area coordinates of the battery pole piece according to the coordinate extreme value;
the step of obtaining the coordinates of the battery pole piece region according to the coordinate extreme value further comprises the following steps: obtaining image size data of the binary image, wherein the image size data comprises a first image height and a first image width;
obtaining image size data of the first contour point set according to the coordinate extreme value, wherein the image size data comprises a second image height and a second image width;
judging the height of the first image and the height of the second image, if the height of the first image is larger than the height of the second image, processing the first contour point set through histogram statistics to obtain a second vertical coordinate maximum value and a second vertical coordinate minimum value, and updating the coordinate extreme value;
judging the sizes of the first image width and the second image width, if the first image width is larger than the second image width, processing the first contour point set through histogram statistics to obtain a second abscissa maximum value and a second abscissa minimum value, and updating the coordinate extreme value;
and obtaining the area coordinates of the battery pole piece according to the updated coordinate extreme value.
2. The pole piece region detection method according to claim 1, wherein the calculating a binarization threshold value according to the gray level image to obtain a binarization image comprises the following steps:
cutting the gray image, and reserving the central part of the gray image to obtain a standard gray image;
calculating the gray average value of the standard gray image, wherein the gray average value is the binarization threshold value, and the specific calculation mode is as follows:
wherein thres represents a binarization threshold, n represents the number of pixels in the standard gray-scale map,
representing the coordinates in the standard gray scale map as
The gray value corresponding to the pixel of (a);
and segmenting the pixels of the gray level image according to the binarization threshold value to obtain a binarization image.
3. The pole piece region detection method of claim 1, wherein in the screening of the plurality of contour regions to obtain the screening result, the screening condition is the area of the plurality of contour regions, and the contour region with the largest area in the plurality of contour regions is the screening result.
4. The pole piece region detection method of claim 3, wherein the pole piece region detection method calculates the areas of the plurality of contour regions by a Green formula in a specific calculation manner:
wherein S represents the area of the contour region, Q represents a fitting function corresponding to the contour in the contour region,
the partial derivative of the function Q in the x direction is shown, P is a fitting function corresponding to the outline of the outline region,
representing the partial derivative of the function P in the y-direction,
represent
Double differentiation in the x-direction and y-direction,
representing the differential of the function P in the x-direction,
representing the differential of the function Q in the y direction.
5. The pole piece region detection method of claim 1, wherein the filtering process comprises: and carrying out corrosion operation on the binary image to obtain a first picture, and then carrying out expansion operation on the first picture to obtain the de-noised image.
6. The pole piece defect detection method is characterized by comprising the following steps:
establishing a deep learning model, and obtaining a training set for training the deep learning model, wherein the training set comprises a plurality of battery pole piece images with defects and corresponding defect marks;
training the deep learning model through the training set to obtain a second deep learning model;
obtaining an image to be detected, and processing the image to be detected according to the pole piece region detection method of any one of claims 1-5 to obtain the coordinates of the battery pole piece region;
detecting defects of the image to be detected according to the second deep learning model to obtain a plurality of defect coordinates;
and judging whether the plurality of defect coordinates are in the battery pole piece area coordinates, if so, determining that the battery pole piece has defects at the position corresponding to the defect coordinates.
7. Pole piece area detection system, characterized in that the system comprises:
the image acquisition unit is used for acquiring an image to be detected and converting the image to be detected into a gray image;
the image processing unit is used for calculating a binarization threshold value according to the gray level image to obtain a binarization image, and filtering the binarization image to obtain a de-noised image;
the region calculation unit is used for extracting the contours of the de-noised image to obtain a plurality of contour regions, screening the contour regions to obtain screening results, and calculating according to the screening results to obtain the coordinates of the battery pole piece region;
the method for obtaining the battery pole piece region coordinates by the region calculation unit according to the screening result comprises the following steps:
obtaining a first contour point set corresponding to the screening result and coordinate extreme values of a plurality of points in the first contour point set, wherein the coordinate extreme values comprise a first abscissa maximum value, a first abscissa minimum value, a first ordinate maximum value and a first ordinate minimum value, and obtaining a battery pole piece area coordinate according to the coordinate extreme values;
the step of obtaining the coordinates of the battery pole piece region according to the coordinate extreme value further comprises the following steps: obtaining image size data of the binary image, wherein the image size data comprises a first image height and a first image width;
obtaining image size data of the first contour point set according to the coordinate extreme value, wherein the image size data comprises a second image height and a second image width;
judging the height of the first image and the height of the second image, if the height of the first image is larger than the height of the second image, processing the first contour point set through histogram statistics to obtain a second vertical coordinate maximum value and a second vertical coordinate minimum value, and updating the coordinate extreme value;
judging the sizes of the first image width and the second image width, if the first image width is larger than the second image width, processing the first contour point set through histogram statistics to obtain a second abscissa maximum value and a second abscissa minimum value, and updating the coordinate extreme value;
and obtaining the area coordinates of the battery pole piece according to the updated coordinate extreme value.
8. Pole piece region detection apparatus comprising a memory, a processor and a computer program stored in said memory and executable on said processor, wherein said processor implements the steps of the pole piece region detection method according to any one of claims 1 to 5 when executing said computer program.
9. A computer-readable storage medium, in which a computer program is stored, which, when being executed by a processor, carries out the steps of the pole piece region detection method according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210981487.0A CN115063421B (en) | 2022-08-16 | 2022-08-16 | Pole piece region detection method, system and device, medium and defect detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210981487.0A CN115063421B (en) | 2022-08-16 | 2022-08-16 | Pole piece region detection method, system and device, medium and defect detection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115063421A CN115063421A (en) | 2022-09-16 |
| CN115063421B true CN115063421B (en) | 2022-10-28 |
Family
ID=83207646
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210981487.0A Active CN115063421B (en) | 2022-08-16 | 2022-08-16 | Pole piece region detection method, system and device, medium and defect detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115063421B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115829922B (en) * | 2022-09-23 | 2024-06-04 | 正泰新能科技股份有限公司 | Method, device, equipment and medium for detecting spacing of battery pieces |
| CN115797254B (en) * | 2022-09-29 | 2023-11-10 | 宁德时代新能源科技股份有限公司 | Pole piece defect detection method, device, computer equipment and storage medium |
| WO2024087179A1 (en) * | 2022-10-28 | 2024-05-02 | 宁德时代新能源科技股份有限公司 | Electrode sheet testing method, electrode sheet testing apparatus, and terminal |
| CN115797314B (en) * | 2022-12-16 | 2024-04-12 | 哈尔滨耐是智能科技有限公司 | Method, system, equipment and storage medium for detecting surface defects of parts |
| CN116416268B (en) * | 2023-06-09 | 2023-08-18 | 浙江双元科技股份有限公司 | Method and device for detecting edge position of lithium battery pole piece based on recursion dichotomy |
| CN117173156B (en) * | 2023-10-23 | 2024-02-20 | 杭州百子尖科技股份有限公司 | Pole piece burr detection method, device, equipment and medium based on machine vision |
| CN117635615B (en) * | 2024-01-26 | 2024-06-25 | 深圳市常丰激光刀模有限公司 | Defect detection method and system for realizing punching die based on deep learning |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103793712A (en) * | 2014-02-19 | 2014-05-14 | 华中科技大学 | Image recognition method and system based on edge geometric features |
| CN107341473A (en) * | 2017-07-04 | 2017-11-10 | 深圳市利众信息科技有限公司 | Palm characteristic recognition method, palm characteristic identificating equipment and storage medium |
| CN107895376A (en) * | 2017-12-11 | 2018-04-10 | 福州大学 | Based on the solar panel recognition methods for improving Canny operators and contour area threshold value |
| CN109949261A (en) * | 2017-12-15 | 2019-06-28 | 中科晶源微电子技术（北京）有限公司 | Handle method, graphic processing facility and the electronic equipment of figure |
| CN110020985A (en) * | 2019-04-12 | 2019-07-16 | 广西师范大学 | A kind of video-splicing system and method for Binocular robot |
| CN110264445A (en) * | 2019-05-30 | 2019-09-20 | 西安交通大学 | The screen printing of battery quality determining method of piecemeal template matching combining form processing |
| CN110653525A (en) * | 2019-08-23 | 2020-01-07 | 江苏理工学院 | Battery pole piece pre-welding positioning detection system and method |
| CN111398287A (en) * | 2019-11-29 | 2020-07-10 | 合肥国轩高科动力能源有限公司 | Battery pole piece scratch detection system and detection method |
| CN114519743A (en) * | 2022-02-25 | 2022-05-20 | 成都数联云算科技有限公司 | Panel defect area extraction method, device, equipment and storage medium |
| CN114764804A (en) * | 2022-06-16 | 2022-07-19 | 深圳新视智科技术有限公司 | Lithium battery pole piece defect detection method, device, equipment and storage medium |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110288511B (en) * | 2019-05-10 | 2023-04-07 | 台州宏达电力建设有限公司台州经济开发区运检分公司 | Minimum error splicing method and device based on double camera images and electronic equipment |
| CN114862817A (en) * | 2022-05-24 | 2022-08-05 | 成都数之联科技股份有限公司 | Circuit board golden finger area defect detection method, system, device and medium |
-
2022
- 2022-08-16 CN CN202210981487.0A patent/CN115063421B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103793712A (en) * | 2014-02-19 | 2014-05-14 | 华中科技大学 | Image recognition method and system based on edge geometric features |
| CN107341473A (en) * | 2017-07-04 | 2017-11-10 | 深圳市利众信息科技有限公司 | Palm characteristic recognition method, palm characteristic identificating equipment and storage medium |
| CN107895376A (en) * | 2017-12-11 | 2018-04-10 | 福州大学 | Based on the solar panel recognition methods for improving Canny operators and contour area threshold value |
| CN109949261A (en) * | 2017-12-15 | 2019-06-28 | 中科晶源微电子技术（北京）有限公司 | Handle method, graphic processing facility and the electronic equipment of figure |
| CN110020985A (en) * | 2019-04-12 | 2019-07-16 | 广西师范大学 | A kind of video-splicing system and method for Binocular robot |
| CN110264445A (en) * | 2019-05-30 | 2019-09-20 | 西安交通大学 | The screen printing of battery quality determining method of piecemeal template matching combining form processing |
| CN110653525A (en) * | 2019-08-23 | 2020-01-07 | 江苏理工学院 | Battery pole piece pre-welding positioning detection system and method |
| CN111398287A (en) * | 2019-11-29 | 2020-07-10 | 合肥国轩高科动力能源有限公司 | Battery pole piece scratch detection system and detection method |
| CN114519743A (en) * | 2022-02-25 | 2022-05-20 | 成都数联云算科技有限公司 | Panel defect area extraction method, device, equipment and storage medium |
| CN114764804A (en) * | 2022-06-16 | 2022-07-19 | 深圳新视智科技术有限公司 | Lithium battery pole piece defect detection method, device, equipment and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| Research on Defect Recognition of Lithium Battery Pole Piece Based on Deep Learning;Li Jiwei等;《2021 7th International Conference on Energy Materials and Environment Engineering (ICEMEE 2021)》;20210521;第261卷;1-5 * |
| 融合梯度信息和邻域点云分布的3D线特征提取与配准;缪永伟等;《中国科学：信息科学》;20211221;第51卷(第12期);第2069-2088页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115063421A (en) | 2022-09-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115063421B (en) | Pole piece region detection method, system and device, medium and defect detection method | |
| CN110992329B (en) | Product surface defect detection method, electronic equipment and readable storage medium | |
| CN111612781B (en) | Screen defect detection method and device and head-mounted display equipment | |
| CN108171104B (en) | Character detection method and device | |
| CN109961049B (en) | Cigarette brand identification method under complex scene | |
| CN107545239B (en) | Fake plate detection method based on license plate recognition and vehicle characteristic matching | |
| WO2021109697A1 (en) | Character segmentation method and apparatus, and computer-readable storage medium | |
| CN111738342B (en) | Pantograph foreign matter detection method, storage medium and computer equipment | |
| CN112819772B (en) | High-precision rapid pattern detection and recognition method | |
| CN107045634B (en) | Text positioning method based on maximum stable extremum region and stroke width | |
| CN111027546B (en) | Character segmentation method, device and computer readable storage medium | |
| CN115063430B (en) | Electric pipeline crack detection method based on image processing | |
| CN111047556B (en) | Strip steel surface defect detection method and device | |
| CN116542982B (en) | Departure judgment device defect detection method and device based on machine vision | |
| CN114495098B (en) | Diaxing algae cell statistical method and system based on microscope image | |
| CN114926387A (en) | Weld defect detection method and device based on background estimation and edge gradient suppression | |
| CN117094975A (en) | Method and device for detecting surface defects of steel and electronic equipment | |
| CN117152165A (en) | Photosensitive chip defect detection method and device, storage medium and electronic equipment | |
| CN108205678B (en) | Nameplate character recognition processing method containing bright spot interference | |
| CN115619775B (en) | Material counting method and device based on image recognition | |
| CN112184619A (en) | Metal part surface defect detection method based on deep learning | |
| CN114627113B (en) | Method, system, device and medium for detecting defects of printed circuit board | |
| CN116363097A (en) | Defect detection method and system for photovoltaic panel | |
| CN114067122B (en) | Two-stage binarization image processing method | |
| CN113538500B (en) | Image segmentation method and device, electronic equipment and 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |