Lithium battery diaphragm production process optimization method and system based on intelligent AI

Abstract

The invention relates to a lithium battery diaphragm production process optimization method and system based on intelligent AI. The method comprises the following steps: acquiring a quality inspection image of a lithium battery diaphragm and identifying a feature region in the quality inspection image; determining a central position point of the feature region; establishing an analysis reference line segment based on the center position point of the feature region; obtaining pixel point numerical values at the intersection points and constructing a screening domain by using the obtained pixel point numerical values; extracting a quality inspection image by using the obtained screening domain to obtain a distinguishing region; and calculating the ratio of the area of the distinguishing region to the area of the quality inspection image to obtain porosity data, comparing the porosity data with porosity reference data, and giving a process optimization direction according to a comparison result. According to the intelligent AI-based lithium battery diaphragm production process optimization method and system disclosed by the invention, the gaps in the lithium battery diaphragm are identified and the porosity data are calculated in combination with feature identification and secondary reconstruction modes, so that a data reference is provided for diaphragm process adjustment.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of image processing, in particular to a lithium battery diaphragm production process optimization method and system based on intelligent AI. BACKGROUND

[0002] The lithium battery diaphragm is one of the core inner components of the lithium ion battery, and plays a dual role of "safety barrier" and "ion channel". Its core function is to realize physical isolation between the positive and negative electrodes (to prevent short circuit), while allowing lithium ions to pass freely, ensuring normal electrochemical reaction.

[0003] The scanning electron microscope (SEM) is a core technical means for characterizing the surface morphology of the lithium battery diaphragm, and its working principle is based on the interaction between the high-energy electron beam and the sample surface. By collecting secondary electron or backscattered electron signals, a high-resolution microscopic morphology image is generated.

[0004] Pore rate directly affects the final battery quality, and generally requires 30%-60%, but this value will change under different battery performance requirements, and even be determined within a certain range, as both conductivity and strength need to be considered (the tensile strength of the diaphragm generally decreases by 20%-30% for every 10% increase in porosity).

[0005] The commonly used threshold segmentation method has a large error in processing low-contrast images, and even leads to a difference of 8% in the calculation results of porosity when the threshold deviation is 5%. At the same time, due to the existence of a certain degree of fuzziness in the gap edge, the calculation result difference will be further aggravated, and these differences will not be able to provide a positive reference for diaphragm process adjustment. SUMMARY

[0006] The application provides a lithium battery diaphragm production process optimization method and system based on intelligent AI, which combines feature recognition and secondary reconstruction to recognize the gap on the lithium battery diaphragm and calculate the porosity data, providing data reference for diaphragm process adjustment.

[0007] The above object of the application is achieved by the following technical solution: In a first aspect, the application provides a lithium battery diaphragm production process optimization method based on intelligent AI, comprising: Obtaining a quality inspection image of a lithium battery diaphragm and recognizing a feature area in the quality inspection image; Determining a center position point of the feature area; Establishing an analysis reference line segment based on the center position point of the feature area, and the analysis reference line segment has an intersection with the edge of the feature area; Obtaining a pixel point value at the intersection and using the obtained pixel point value to construct a screening domain; The obtained screening field is used to extract the quality inspection image to obtain a distinguished area; A ratio of an area of the distinguished area to an area of the quality inspection image is calculated to obtain porosity data; The porosity data is compared with reference porosity data, and a process optimization direction is given according to a comparison result.

[0008] In a possible implementation manner of the first aspect, when the feature area in the quality inspection image is identified, the method further includes: The quality inspection image is extracted to obtain a plurality of local features; A shape feature area of each local feature is calculated; The obtained plurality of local features are sorted according to the shape feature area; A first local feature in the sequence is taken as a background feature; In the sorting of the obtained plurality of local features, the shape feature area tends to decrease in the sequence.

[0009] In a possible implementation manner of the first aspect, for the remaining local features, the method further includes: An analysis reference line is established on the remaining local features, and the analysis reference line includes a horizontal analysis reference line and a vertical analysis reference line; An analysis curve is established using pixel points on the analysis reference line; The analysis curve is converted into a time domain for decomposition to obtain a plurality of position point groups, each position point group including two position points and two position point pixel values; All position point groups are combined and only the outermost two position points are retained; A dynamic value range is constructed using pixel values of the retained position points, and the quality inspection image is extracted using the dynamic value range.

[0010] In a possible implementation manner of the first aspect, when the dynamic value range is constructed using the pixel values of the retained position points, each position point pixel value is located at a middle point of an independent sub-dynamic value range.

[0011] In a possible implementation manner of the first aspect, identifying the feature area in the quality inspection image includes: Color and shape are used to identify in the quality inspection image to obtain a hole region; The hole region is classified, and the classification includes a surface hole region and a three-dimensional hole region, and the classification basis includes a color difference value and an area; The three-dimensional hole region is used as the feature area; The surface hole region is expanded in edge and used as the feature area.

[0012] In a possible implementation of the first aspect, when the three-dimensional hole region is used as the feature region, the method further includes: determining a transition region between two adjacent three-dimensional hole regions; connecting the outer contours of the two adjacent three-dimensional hole regions to maximize the area of the region composed of the two adjacent three-dimensional hole regions and the transition region between the two adjacent three-dimensional hole regions; adding the newly added edge of the composed region as the edge of the transition region; calculating whether the edge of the transition region is associated with the edges of the two adjacent three-dimensional hole regions, and merging the transition region and the two adjacent three-dimensional hole regions when the edge of the transition region is associated with the edges of the two adjacent three-dimensional hole regions.

[0013] In a possible implementation of the first aspect, using the edge-expanded surface hole region as the feature region includes: finding a highlight region near the surface hole region and determining the edge of the highlight region; adding the edge of the highlight region to the edge of the surface hole region to merge and maximize the area of the surface hole region.

[0014] In a second aspect, the present application provides a lithium battery separator production process optimization device based on intelligent AI, which includes: an image acquisition unit configured to acquire a quality inspection image of a lithium battery separator and identify a feature region in the quality inspection image; a first processing unit configured to determine a center position point of the feature region; a second processing unit configured to establish an analysis reference line segment based on the center position point of the feature region, and the analysis reference line segment has an intersection point with the edge of the feature region; a third processing unit configured to acquire a pixel point value at the intersection point and construct a screening domain using the obtained pixel point value; an extraction processing unit configured to extract the quality inspection image using the obtained screening domain to obtain a distinguished region; a calculation processing unit configured to calculate a ratio of the area of the distinguished region to the area of the quality inspection image to obtain a porosity data; a result output unit configured to compare the porosity data with a porosity reference data and give a process optimization direction according to a comparison result.

[0015] In a third aspect, the present application provides a lithium battery separator production process optimization system based on intelligent AI, which includes: one or more memories configured to store instructions; and One or more processors configured to invoke and run the instructions from the memory to perform the method as claimed in the first aspect and any possible implementation of the first aspect.

[0016] In a fourth aspect, a computer readable storage medium is provided, the computer readable storage medium comprising: A program which, when executed by a processor, causes the method as claimed in the first aspect and any possible implementation of the first aspect to be performed.

[0017] In a fifth aspect, a computer program product is provided, comprising program instructions which, when executed by a computing device, cause the method as claimed in the first aspect and any possible implementation of the first aspect to be performed.

[0018] In a sixth aspect, a chip system is provided, the chip system comprising a processor configured to implement the functions of the above aspects, for example, generating, receiving, sending, or processing data and / or information involved in the above methods.

[0019] The chip system can be composed of a chip, or can include a chip and other discrete devices.

[0020] In a possible design, the chip system further includes a memory, the memory configured to store necessary program instructions and data. The processor and the memory can be decoupled and disposed on different devices, connected through a wired or wireless manner, or the processor and the memory can be coupled on the same device.

[0021] The application has the following beneficial technical effects: The AI-based lithium battery separator production process optimization method and system disclosed in the application identifies the gap on the lithium battery separator and calculates the porosity data in combination with the feature recognition and secondary reconstruction mode. This mode can calculate the lithium battery separator porosity in the continuous production process, and give the calculation result, providing data reference for the separator process adjustment. BRIEF DESCRIPTION OF DRAWINGS

[0022] Figure 1 is a step flow schematic block diagram of an AI-based lithium battery separator production process optimization method provided by the application.

[0023] Figure 2 is a schematic diagram of a quality inspection image provided by the application.

[0024] Figure 3 is a schematic diagram of a feature area provided by the application.

[0025] Figure 4is a schematic diagram of establishing an analysis reference line provided by the present application.

[0026] Figure 5 is a schematic diagram of obtaining a position point using an analysis curve provided by the present application.

[0027] Figure 6 is a schematic diagram of a three-dimensional hole region and a surface hole region provided by the present application.

[0028] Figure 7 is a schematic diagram of a transition region edge and a highlight region edge provided by the present application. DETAILED DESCRIPTION

[0029] The technical solutions in the present application are further described in detail below with reference to the accompanying drawings.

[0030] The present application discloses a lithium battery separator production process optimization method based on intelligent AI, please refer to Figure 1 In some examples, the lithium battery separator production process optimization method based on intelligent AI disclosed by the present application includes the following steps: S101, obtaining a quality inspection image of a lithium battery separator and identifying a feature region in the quality inspection image; S102, determining a center position point of the feature region; S103, establishing an analysis reference line segment based on the center position point of the feature region, and the analysis reference line segment has an intersection with the edge of the feature region; S104, obtaining pixel point values at the intersection and using the obtained pixel point values to construct a screening domain; S105, using the obtained screening domain to extract the quality inspection image to obtain a distinguished region; S106, calculating the ratio of the area of the distinguished region to the area of the quality inspection image to obtain porosity data; S107, comparing the porosity data with porosity reference data and giving a process optimization direction according to the comparison result.

[0031] Overall, the lithium battery separator production process optimization method based on intelligent AI provided by the present application is to dynamically optimize the separator process based on porosity data. For example, first, a specific value or value range is calibrated, that is, the porosity reference data. When the actual detected porosity data is greater than or less than the porosity reference data, a process optimization direction is given. The role of the process optimization direction is to make the porosity data equal to the porosity reference data or within the required range of the porosity reference data.

[0032] Or it can be explained that the purpose of the present application is to detect whether the porosity data meets the requirements, and to give whether it is lower than the requirements or higher than the requirements.

[0033] Of course, when the porosity data meets the requirements, it is also necessary to combine parameters such as pore size distribution and tortuosity for comprehensive judgment, that is, the technical solution in the application is part of a judgment system, mainly responsible for solving the detection of porosity data.

[0034] The judgment system carried out in the above content is a neural network, or called intelligent AI, which is trained to give process optimization direction after analyzing the received data. In addition, the technical solution given in the application can also be part of an image processing model (intelligent AI), when the image is given to the image processing model (intelligent AI), the image processing model (intelligent AI) uses the technical solution to process the received image.

[0035] In step S101, first, the quality inspection image of the lithium battery diaphragm is acquired (as shown in Figure 2 The feature area in the quality inspection image refers to the hole on the lithium battery diaphragm, as shown in Figure 3 Then, in step S102, the center position point of the feature area is determined, and then in S103, the analysis reference line segment is established based on the center position point of the feature area.

[0036] For the analysis reference line segment, it is required that the analysis reference line segment has intersection with the edge of the feature area, of course, the extension of the analysis reference line segment also needs to be set with a termination condition, in the application, the analysis reference line segment is limited to 1.1-1.3 times of the average diameter of the hole on the lithium battery diaphragm.

[0037] Here, the specific calculation method of the average diameter of the hole is to calculate the ratio of the total area of the hole to the total number, and then calculate the average diameter of the hole according to the average area of the hole.

[0038] Here, the average diameter of the hole on the lithium battery diaphragm is obtained by mean value calculation after identifying the feature area in the quality inspection image, which is a dynamic value, not a fixed value.

[0039] Then, in step S104, the pixel point value at the intersection is acquired and the obtained pixel point value is used to construct the screening domain, that is, the actual pixel point value on the image is used to determine which values are used for screening.

[0040] The specific way of using the pixel point value to construct the screening domain is to take the pixel point value as a reference, and take the positive and negative error values as endpoints to construct the screening domain, for example, the pixel point value is N, the screening domain is [N-M, N+M], and M is a number greater than zero.

[0041] In step S105, the obtained screening domain is used to extract the quality inspection image to obtain a distinguishing area. Here, some curve segments are first obtained, and after merging the obtained curve segments, a surrounding area, i.e., the distinguishing area, can be obtained.

[0042] In some possible implementations, when there is a breakpoint in the obtained curve segment, the M value is made larger, and when there are more overlapping areas in the obtained curve segment, the M value is made smaller, and the M value is adjusted through dynamic changes.

[0043] In step S106, the ratio of the area of the distinguishing area to the area of the quality inspection image is calculated to obtain porosity data, and finally in step S107, the porosity data is compared with the porosity reference data, and a process optimization direction is given according to the comparison result (the process optimization direction is to make the porosity data equal to the porosity reference data or within the required range of the porosity reference data). This part has been stated in the foregoing content, and will not be repeated here.

[0044] There are difficulties in processing only using image recognition, which are as follows: The porosity of the lithium battery separator is various, including irregular polygons, narrow slits, branches, and the like, and there can be connected holes (through the thickness of the separator), closed holes (local depressions), and porosity clusters (densely distributed small holes). Image recognition needs to be modeled and manually labeled and trained, and the comprehensive coverage and adaptability are limited. The gray difference between the separator material and the porosity is low (for example, the base and the porosity of a polymer separator are both light in the SEM image, and only have a weak gray difference), or the porosity edge presents a "gradual transition" due to the roughness of the material itself (for example, the interlaced structure of a fibrous separator), which is not a clear boundary, resulting in "over-segmentation" or "under-segmentation" when image segmentation is performed. In a high-porosity separator, the porosities can be densely distributed or even overlap (in two-dimensional projection), forming a "connected region", and it is difficult to accurately separate adjacent porosities only through image recognition, resulting in counting errors or porosity measurement deviations.

[0045] In the present application, a secondary reconstruction method is used for processing. In this method, image recognition and reconstruction (the obtained screening domain is used to extract the quality inspection image) are combined, which can further clarify the edge position of the distinguishing area, and therefore more accurate calculation results can be obtained.

[0046] In some examples, when identifying the feature area in the quality inspection image, the following content is further added: S201, extracting the quality inspection image to obtain a plurality of local features; S202, calculating the shape feature area of each local feature; S203, sorting the obtained multiple local features according to shape feature areas; S204, taking the first local feature in the sequence as a background feature; In the sorting of the obtained multiple local features, the shape feature areas tend to decrease in the sequence.

[0047] The purpose of steps S201 to S204 is to determine the background color, and the specific method is to generate a dynamic numerical range, and then use the generated dynamic numerical range to extract the quality inspection image and finally determine the background color. The specific method of generating the dynamic numerical range is to first denoise the quality inspection image, such as Gaussian filtering and wavelet transform filtering.

[0048] After denoising, it is determined that the pixel points in the quality inspection image specifically have which colors.

[0049] In some possible implementations, it is also necessary to use mean filtering for processing, that is, the values of multiple pixel points (pixel blocks) are accumulated and then averaged, and then it is determined that the pixel points in the quality inspection image specifically have which colors.

[0050] The obtained values are used to extract multiple local features. For the obtained local features, the shape feature area of each local feature needs to be calculated, and then the obtained multiple local features are sorted according to the size, and it is required that the shape feature area values tend to decrease in the sequence.

[0051] The first local feature in the sequence is taken as the background feature. At this time, the contour of the feature can be obtained through the shape of the local feature, and then the area outside the background feature can be temporarily all included in the pore range.

[0052] In some possible implementations, the obtained values can also be expanded, for example, the pixel point value is N, the expansion range is [N-M, N+M], and the value of M is generally 2-4.

[0053] For the remaining local features, the following method is used for processing: An analysis reference line is established on the remaining local features, and the analysis reference line includes a horizontal analysis reference line and a vertical analysis reference line; An analysis curve is established using the pixel points on the analysis reference line; The analysis curve is converted into the time domain for decomposition to obtain multiple groups of position points, each group of position points including two position points and two position point pixel values; All position point groups are merged and only the two outermost position points are retained; The pixel values of the reserved position points are used to construct a dynamic value range, and the quality inspection image is extracted using the dynamic value range.

[0054] In the above manner, the partial edge points of the remaining local features are determined using the analysis reference line, as shown in Figure 4 The edge points have pixel values, and the resulting pixel values are used to construct a dynamic value range, which is a collection of some pixel values.

[0055] Converting the analysis curve into the time domain for decomposition means using the wavelet decomposition method to decompose the analysis curve, and the resulting sub-curves have start times (position points) and end times (position points), and each sub-curve corresponds to a set of position points.

[0056] The specific explanation of wavelet decomposition is as follows: Wavelet decomposition is a signal analysis method based on wavelet transform, which realizes multi-scale and multi-resolution analysis of signals by decomposing the original signal into different frequency components (from high frequency to low frequency) and preserving the time localization information of each component. Wavelet decomposition relies on the "mother wavelet" and the "daughter wavelet" after scaling and translation. The mother wavelet is a function with fast decay and zero mean, and in this application, Haar wavelet or Daubechies wavelet can be used.

[0057] Taking the Fourier fast transform as an example, the Fourier fast transform is to convert the analysis curve into the frequency domain for analysis, and the resulting sub-curves only have frequency domain characteristics but no time domain characteristics, that is, the resulting sub-curves have no start time and no end time. However, when using wavelet transform for decomposition, the resulting sub-curves have start times (position points) and end times (position points), and the position points can correspond to the light and dark changes on the picture.

[0058] Merging all sets of position points and retaining only the two outermost position points means placing all position points on an analysis reference line according to the corresponding positions obtained using wavelet decomposition, and then retaining only the two outermost position points.

[0059] Finally, the pixel values of the reserved position points are used to construct a dynamic value range, and the quality inspection image is extracted using the dynamic value range, which aims to re-extract the edges of the remaining local features by analyzing the obtained edges.

[0060] In some possible implementations, when constructing a dynamic value range using the pixel values of the reserved position points, each position point pixel value is located at the middle point of an independent sub-dynamic value range.

[0061] In some possible implementations, when the dynamic numerical value ranges overlap, the two dynamic numerical value ranges are merged.

[0062] When the dynamic numerical value ranges are constructed using the pixel values of the reserved position points and the quality inspection image is extracted using the dynamic numerical value ranges, a complete closed graph can be obtained, and of course, only some curve segments can be obtained, in which case the dynamic numerical value ranges need to be appropriately increased, or an analysis curve is constructed at this position and the above steps are performed.

[0063] The analysis curve and the position points are as shown in Figure 5 .

[0064] The present application also provides another way of identifying a feature region in a quality inspection image, which is as follows: Color and shape are used to identify the quality inspection image to obtain a hole region; The hole region is classified, and the classification includes a surface hole region and a three-dimensional hole region, and the classification basis includes a color difference value and an area; The three-dimensional hole region is used as a feature region; The surface hole region is used as a feature region after edge expansion.

[0065] After the closed graph is obtained, the closed graph can be used to perform range extraction on the quality inspection image, and at this time, the obtained content is a part of the quality inspection image, and then color and shape are used to identify the quality inspection image to obtain a hole region.

[0066] The color and shape here refer to content obtained through a large amount of statistics and manual annotation, but the disadvantage of this way is that only known types can be known, and position types cannot be obtained, for example, the color of the hole region is basically a known amount, but the shape type of the hole is complex and cannot be known in total.

[0067] After the hole region is obtained, the hole region is classified, and the classification includes a surface hole region and a three-dimensional hole region, and the classification basis includes a color difference value and an area. Specifically, the color difference value refers to the color difference between the hole region and the surrounding region, and the area is the actual area of the hole region. These two values are fixed set values, and when the color difference value is greater than the corresponding set value and the area is greater than the corresponding set value, the hole region is classified into a three-dimensional hole region, and vice versa, it is classified into a surface hole region, as shown in Figure 6 .

[0068] When the three-dimensional hole region is used as a feature region, the following way also needs to be used for processing: A transition region between two adjacent three-dimensional hole regions is determined; connecting the outer contours of two adjacent three-dimensional hole regions, so that the two adjacent three-dimensional hole regions and the transition region between the two adjacent three-dimensional hole regions form a region with the largest area; The newly added edge of the region is used as the edge of the transition region. The edge of the transition region is calculated to determine whether it is associated with the edges of the two adjacent three-dimensional hole regions. If there is an association, the transition region and the two adjacent three-dimensional hole regions are merged.

[0069] The purpose of this method is to determine whether the region between the two adjacent three-dimensional hole regions belongs to the three-dimensional hole region or the surface of the lithium battery separator.

[0070] Specifically, first, the outer contours of two adjacent three-dimensional hole regions are connected, so that the two adjacent three-dimensional hole regions and the transition region between the two adjacent three-dimensional hole regions form a region with the largest area, then the newly added edge of the region is used as the edge of the transition region, as shown in Figure 7 Finally, the edge of the transition region is calculated to determine whether it is associated with the edges of the two adjacent three-dimensional hole regions.

[0071] The association is calculated as follows: The pixels on the edge of the transition region and the pixels on the edge of a part of the adjacent three-dimensional hole region are used to form a sequence, and then the difference sequence or the second difference sequence of the sequence is calculated. Then, the position of the slope change point of the curve corresponding to the difference sequence or the second difference sequence and the connection point of the edge of the transition region and the edge of the adjacent part of the three-dimensional hole region are checked to determine whether they coincide.

[0072] If the positions do not coincide or the distance is greater than or equal to a certain value, it is considered that the edge (at least one) of the transition region is associated with the edges of the two adjacent three-dimensional hole regions, otherwise it is not associated.

[0073] In some examples, the surface hole region after edge expansion is used as a feature region as follows: A highlight region is found near the surface hole region and the edge of the highlight region is determined. The edge of the highlight region is added to the edge of the surface hole region to merge, so that the area of the surface hole region is maximized.

[0074] This is because the highlight region (with obvious reflection characteristics, white or off-white) generally has protruding edges, which are part of the surface hole region, so they are included in the range of the surface hole region.

[0075] The application also provides a lithium battery separator production process optimization device based on intelligent AI, comprising: an image acquisition unit configured to acquire a quality inspection image of a lithium battery separator and identify a feature region in the quality inspection image; a first processing unit configured to determine a center position point of the feature region; a second processing unit configured to establish an analysis reference line based on the center position point of the feature region, the analysis reference line having an intersection with an edge of the feature region; a third processing unit configured to acquire pixel point values at the intersection and construct a screening domain using the acquired pixel point values; an extraction processing unit configured to extract the quality inspection image using the constructed screening domain to obtain a distinguished region; a calculation processing unit configured to calculate a ratio of an area of the distinguished region to an area of the quality inspection image to obtain porosity data; a result output unit configured to compare the porosity data with porosity reference data and give a process optimization direction according to a comparison result.

[0076] Further, the identifying the feature region in the quality inspection image comprises: extracting the quality inspection image to obtain a plurality of local features; calculating a shape feature area of each of the local features; sorting the obtained plurality of local features according to the shape feature areas; taking a first local feature in the sorted sequence as a background feature; wherein, in the sorted sequence, the shape feature areas tend to decrease in value.

[0077] Further, for the remaining local features, further comprising: establishing an analysis reference line on the remaining local features, the analysis reference line including a horizontal analysis reference line and a vertical analysis reference line; establishing an analysis curve using pixel points on the analysis reference line; converting the analysis curve into a time domain for decomposition to obtain a plurality of position point groups, each position point group including two position points and pixel values of the two position points; merging all the position point groups and retaining only the outermost two position points; constructing a dynamic value range using the pixel values of the retained position points and extracting the quality inspection image using the dynamic value range.

[0078] Further, when the dynamic value range is constructed using the pixel values of the retained position points, each of the position point pixel values is located at a middle point of an independent sub-dynamic value range.

[0079] Further, the method further comprises: identifying the feature region in the quality inspection image using color and shape, obtaining a hole region; classifying the hole region, the classifying comprising a surface hole region and a three-dimensional hole region, the classifying being based on color difference and area; using the three-dimensional hole region as the feature region; using the surface hole region as the feature region after edge expansion.

[0080] Further, when using the three-dimensional hole region as the feature region, the method further comprises: determining a transition region between two adjacent three-dimensional hole regions; connecting the outer contours of the two adjacent three-dimensional hole regions to maximize the area of the region composed of the two adjacent three-dimensional hole regions and the transition region between the two adjacent three-dimensional hole regions; adding the newly added edge of the composed region as the edge of the transition region; calculating whether the edge of the transition region is associated with the edges of the two adjacent three-dimensional hole regions, and merging the transition region and the two adjacent three-dimensional hole regions when the edge of the transition region is associated with the edges of the two adjacent three-dimensional hole regions.

[0081] Further, using the surface hole region as the feature region after edge expansion comprises: finding a highlight region near the surface hole region and determining the edge of the highlight region; adding the edge of the highlight region to the edge of the surface hole region to merge, so as to maximize the area of the surface hole region.

[0082] In one example, the units in any of the above apparatuses can be one or more integrated circuits configured to implement one or more of the above methods, such as one or more application specific integrated circuits (ASICs), or one or more digital signal processors (DSPs), or one or more field programmable gate arrays (FPGAs), or a combination of at least two of these integrated circuit forms.

[0083] For example, when the units in the device can be implemented in the form of a processing element scheduler, the processing element can be a general-purpose processor, such as a central processing unit (CPU) or other processor that can call a program. For example, these units can be integrated together in the form of a system on a chip (SOC).

[0084] In this application, various objects / messages / information / equipment / network elements / systems / devices / actions / operations / processes / concepts, etc. may be named, and it can be understood that these specific names do not constitute a limitation on the related objects, and the names can be changed according to the scene, context or usage habits, etc. The technical meaning of the technical terms in this application should be mainly determined from the function and technical effect embodied / implemented in the technical scheme.

[0085] Those skilled in the art can clearly understand that, for the convenience and brevity of the description, the specific working process of the above-described system, device and unit can refer to the corresponding process in the foregoing method embodiments, which will not be repeated here.

[0086] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other form.

[0087] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, i.e. they can be located in one place or distributed on a plurality of network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment scheme.

[0088] Those of ordinary skill in the art can realize that the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical scheme. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0089] It should also be understood that in various embodiments of the present application, first, second, etc. are only used to represent that a plurality of objects are different. For example, the first time window and the second time window are only used to represent different time windows. The above-mentioned first, second, etc. should not have any effect on the time window itself, and should not limit the embodiments of the present application.

[0090] It should also be understood that in various embodiments of the present application, the terms and / or descriptions between different embodiments are consistent and can be referred to each other if there is no special description and logical conflict, and the technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationship.

[0091] The functions, if realized in the form of software function units and sold or used as independent products, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the parts of the prior art that make contributions or parts of the technical solutions can be embodied in the form of software products. The computer software product is stored in a computer readable storage medium, including a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in various embodiments of the present application. The aforementioned computer readable storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk and various program code storage media.

[0092] The present application also provides an intelligent AI-based lithium battery separator production process optimization system, which comprises: One or more memories for storing instructions; and One or more processors for calling and running the instructions from the memory, executing the method as described above.

[0093] The present application also provides a computer program product, which includes instructions that, when executed, cause the terminal device and the network device to perform operations corresponding to the terminal device and the network device of the method described above.

[0094] The present application also provides a chip system, which comprises a processor for realizing the functions involved in the above description, such as generating, receiving, sending, or processing the data and / or information involved in the above method.

[0095] The chip system can be composed of a chip, or can include a chip and other discrete devices.

[0096] The processor mentioned in any of the above embodiments can be a CPU, a microprocessor, an ASIC, or one or more integrated circuits for program execution to control the method of transmitting feedback information mentioned above.

[0097] In a possible design, the chip system further includes a memory, which is configured to store necessary program instructions and data. The processor and the memory can be decoupled and arranged on different devices respectively, and connected through wired or wireless means to support the chip system to implement various functions in the above embodiments. Alternatively, the processor and the memory can be coupled on the same device.

[0098] Optionally, the computer instructions are stored in the memory.

[0099] Optionally, the memory is a storage unit in the chip, such as a register, a cache, or the like. The memory can also be a storage unit outside the chip in the terminal, such as a ROM or another type of static storage device that can store static information and instructions, a RAM, or the like.

[0100] It can be understood that the memory in the present application can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories.

[0101] The non-volatile memory can be a ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically EPROM (EEPROM), or a flash memory.

[0102] The volatile memory can be a RAM, which is used as an external cache. There are many different types of RAM, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synch link DRAM (SLDRAM), and direct Rambus dynamic RAM (DRDRAM).

[0103] The embodiments of the present application are preferred embodiments of the present application, and are not intended to limit the protection scope of the present application, and thus: any equivalent changes made according to the structure, shape, principle of the present application should be covered within the protection scope of the present application.

Claims

1. A method for optimizing the production process of lithium battery separators based on intelligent AI, characterized in that, include: Acquire quality inspection images of lithium battery separators and identify feature regions in the quality inspection images; Determine the center point of the feature region; An analysis reference line segment is established based on the center point of the feature region, and the analysis reference line segment intersects with the edge of the feature region. Obtain the pixel values ​​at the intersection points and use the obtained pixel values ​​to construct the filtering area; The obtained filtering domain is used to extract the distinguishing regions from the quality inspection image; The porosity data is obtained by calculating the ratio of the area of ​​the distinguishing region to the area of ​​the quality inspection image; The porosity data is compared with the porosity reference data, and the direction of process optimization is given based on the comparison results.

2. The method for optimizing lithium battery separator production process based on intelligent AI according to claim 1, characterized in that, When identifying feature regions in quality inspection images, the process also includes: Multiple local features are extracted from the quality inspection image; Calculate the shape feature area of ​​each local feature; The obtained local features are sorted according to the area of ​​the shape features; Use the first local feature in the sequential sequence as the background feature; When sorting the obtained local features, the numerical value of the shape feature area tends to decrease in the sequential sequence.

3. The method for optimizing lithium battery separator production process based on intelligent AI according to claim 2, characterized in that, The remaining local features also include: Establish analysis reference lines on the remaining local features. The analysis reference lines include horizontal analysis reference lines and vertical analysis reference lines. Use the pixels on the analysis reference line to create the analysis curve; The analysis curve is decomposed in the time domain to obtain multiple sets of location points. Each set of location points includes two location points and two location point pixel values. All location point groups are merged, and only the two outermost location points are retained; A dynamic numerical range is constructed using the pixel values ​​of the retained location points, and this dynamic numerical range is then used to extract the quality inspection image.

4. The method for optimizing lithium battery separator production process based on intelligent AI according to claim 3, characterized in that, When constructing a dynamic numerical range using the pixel values ​​of the reserved location points, each location point pixel value is located at the midpoint of an independent sub-dynamic numerical range.

5. The method for optimizing lithium battery separator production process based on intelligent AI according to any one of claims 1 to 4, characterized in that, Its features are, Identifying feature regions in quality inspection images includes: Using color and shape, the hole areas are identified in the quality inspection image. The perforated areas are classified into surface perforated areas and three-dimensional perforated areas, based on color difference and area. Use the three-dimensional hole region as a feature region; The surface hole area is expanded at the edges and then used as a feature area.

6. The method for optimizing lithium battery separator production process based on intelligent AI according to claim 5, characterized in that, When using a three-dimensional hole region as a feature region, it also includes: Determine the transition region between two adjacent three-dimensional hole regions; Connect the outer contours of two adjacent three-dimensional hole regions to maximize the area of ​​the region formed by the transition region between the two adjacent three-dimensional hole regions. The newly added edges of the constituent regions are used as the edges of the transition regions; Calculate whether the edge of the transition region is related to the edges of the two adjacent three-dimensional hole regions. If there is a relationship, merge the transition region and the two adjacent three-dimensional hole regions.

7. The method for optimizing lithium battery separator production process based on intelligent AI according to claim 5, characterized in that, Using surface hole areas as feature areas after edge expansion includes: Locate the highlight areas near the surface pores and determine the edges of the highlight areas; The edges of the highlight areas are merged with the edges of the surface hole areas to maximize the area of ​​the surface hole areas.

8. A lithium battery separator production process optimization device based on intelligent AI, characterized in that, include: The image acquisition unit is used to acquire quality inspection images of lithium battery separators and identify feature regions in the quality inspection images. The first processing unit is used to determine the center location point of the feature region; The second processing unit is used to establish an analysis reference line segment based on the center point of the feature region, and the analysis reference line segment intersects with the edge of the feature region. The third processing unit is used to obtain the pixel values ​​at the intersection and use the obtained pixel values ​​to construct the filtering domain; The extraction processing unit is used to extract the quality inspection image using the obtained filtering domain to obtain the distinguishing region; The computational processing unit is used to calculate the ratio of the area of ​​the distinguishing region to the area of ​​the quality inspection image to obtain porosity data; The results output unit is used to compare porosity data with porosity reference data and provide directions for process optimization based on the comparison results.

9. A lithium battery separator production process optimization system based on intelligent AI, characterized in that, The system includes: One or more memories for storing instructions; and One or more processors are configured to retrieve and execute the instructions from the memory to perform the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes: The program, when run by the processor, executes the method as described in any one of claims 1 to 7.

Images