SEI thickness obtaining method and device, storage medium and program product

Abstract

The invention provides an SEI thickness obtaining method and device, a storage medium and a program product. The SEI thickness obtaining method comprises the following steps: inputting an image containing an SEI region into a pre-trained instance segmentation model so as to identify the SEI region from the image through the instance segmentation model; obtaining a scale value and a scale straight line in the image so as to establish a corresponding relation between pixels in the image and physical dimensions; performing curve fitting on the contour of the SEI region identified from the image to obtain a contour curve of the SEI region; and obtaining the thickness of the SEI region based on the contour curve and the corresponding relationship between the pixel and the physical size. According to the invention, the efficiency and accuracy of SEI region thickness analysis can be improved.

Description

Technical Field

[0001] The present disclosure relates to the technical field of battery detection, and in particular to a method, device, storage medium, and program product for obtaining SEI thickness. Background Art

[0002] The SEI (Solid Electrolyte Interface) membrane is a crucial component of batteries. In existing technologies, the phase evolution of SEI membranes is primarily analyzed manually. The identification of the SEI in in-situ electron microscopy images and its thickness analysis rely on visual judgment and expert data analysis. However, a single in-situ electron microscopy experiment generates tens of thousands of images, making manual processing difficult. Furthermore, subjective interpretation leads to low accuracy, which can affect subsequent research. In existing technologies, SEI thickness analysis in images is inefficient and inaccurate. Summary of the Invention

[0003] A technical problem solved by the present disclosure is that in related technologies, the efficiency and accuracy of SEI thickness analysis in images are low.

[0004] According to one aspect of the present disclosure, a method for obtaining SEI thickness is provided, comprising: inputting an image containing a solid electrolyte interface SEI region into a pre-trained instance segmentation model to identify the SEI region from the image through the instance segmentation model; obtaining scale values ​​and scale lines in the image to establish a correspondence between pixels in the image and physical dimensions; performing curve fitting on the contour of the SEI region identified from the image to obtain a contour curve of the SEI region; and obtaining the thickness of the SEI region based on the contour curve and the correspondence between the pixels and the physical dimensions.

[0005] In this technical solution, a pre-trained instance segmentation model is used to identify SEI regions in images. A scale value and a straight line are then used to establish a correspondence between pixels and physical dimensions. After fitting the outline of the identified SEI region to generate a contour curve, the SEI region thickness is determined based on the contour curve and the correspondence between pixels and physical dimensions. Compared to manual methods, this method improves the efficiency of SEI thickness analysis and, because it eliminates the need for subjective human judgment, improves the accuracy of SEI thickness analysis.

[0006] In some embodiments, the contour curve includes a first curve corresponding to a first edge of the SEI region and a second curve corresponding to a second edge of the SEI region. The first edge is an edge of the SEI region extending along the length of the SEI region, and the second edge is an edge of the SEI region opposite the first edge. Using the first and second curves in the contour curve facilitates obtaining the thickness of the SEI region and improves the accuracy of SEI region thickness analysis.

[0007] In some embodiments, obtaining the thickness of the SEI region based on the contour curve and the correspondence between the pixels and the physical dimensions includes: obtaining a tangent line at a location on the first edge of the SEI region based on the first curve; obtaining a normal line of the first curve at the location based on the tangent line and the location, wherein the normal line intersects with the second curve to form an intersection point; and obtaining a distance between the location point and the intersection point based on the location point, the intersection point, and the correspondence between the pixels and the physical dimensions, wherein the distance represents the thickness of the SEI region at the location point. In this embodiment, the tangent line and the normal line at a location on the first edge of the SEI region can be obtained based on the first curve, thereby obtaining the intersection point of the normal line with the second curve. Furthermore, based on the location point, the intersection point, and the correspondence between the pixels and the physical dimensions, the distance between the location point and the intersection point can be obtained. This can improve the accuracy of SEI region thickness analysis.

[0008] In some embodiments, obtaining the distance between the location point and the intersection point based on the location point, the intersection point, and the correspondence between the pixels and the physical size includes: obtaining the number of pixels between the location point and the intersection point along the direction of the normal; and calculating the distance between the location point and the intersection point based on the correspondence between the pixels and the physical size and the number of pixels between the location point and the intersection point. In this embodiment, by obtaining the number of pixels between the location point and the intersection point, the distance between the location point and the intersection point can be conveniently and accurately calculated based on the number of pixels and the correspondence between pixels and the physical size, thereby improving the accuracy of SEI region thickness analysis.

[0009] In some embodiments, the SEI thickness determination method further includes: determining multiple thicknesses of the SEI region at multiple locations on the first edge of the SEI region; and calculating an average of the multiple thicknesses as the average SEI region thickness. In this embodiment, by determining multiple thickness values ​​at multiple locations and calculating the average of the multiple thicknesses as the average SEI region thickness, the accuracy of SEI region thickness analysis can be further improved.

[0010] In some embodiments, obtaining a scale value and a scale line in the image to establish a correspondence between pixels in the image and physical dimensions includes: identifying the scale value in the image using a text recognition library; processing the image using an edge detection algorithm to obtain the edges of the scale line in the image; processing the scale line in the image based on the edges of the scale line to obtain the number of pixels in a single row corresponding to the scale line; and calculating the physical dimension corresponding to each pixel based on the scale value and the number of pixels in a single row corresponding to the scale line to establish a correspondence between pixels in the image and physical dimensions. This establishes a correspondence between pixels and physical dimensions, thereby facilitating subsequent accurate determination of the SEI region thickness, which can improve the accuracy of SEI region thickness analysis.

[0011] In some embodiments, the SEI thickness acquisition method further includes: prior to inputting the image into the instance segmentation model, training the instance segmentation model using an image pre-labeled with the SEI region. In this embodiment, training the instance segmentation model using an image pre-labeled with the SEI region enables the instance segmentation model to identify SEI regions in the image, thereby improving the accuracy of SEI region identification.

[0012] In some embodiments, training the instance segmentation model using images pre-labeled with SEI regions includes: using a labeling tool to label SEI regions in multiple images to obtain multiple images pre-labeled with SEI regions; and inputting the multiple images pre-labeled with SEI regions into the instance segmentation model to train the instance segmentation model. This enables the instance segmentation model to identify SEI regions in images, thereby improving the accuracy of SEI region identification.

[0013] In some embodiments, the instance segmentation model is a large instance segmentation model built using a Transformer architecture. Using the large instance segmentation model built using a Transformer architecture to identify SEI regions can improve the accuracy of SEI region identification.

[0014] According to another aspect of the present disclosure, a device for obtaining SEI thickness is provided, comprising: an input unit for inputting an image containing a solid electrolyte interface (SEI) region into a pre-trained instance segmentation model, so that the SEI region can be identified from the image using the instance segmentation model; a first obtaining unit for obtaining scale values ​​and scale lines in the image to establish a correspondence between pixels in the image and physical dimensions; a fitting unit for performing curve fitting on the contour of the SEI region identified from the image to obtain a contour curve of the SEI region; and a second obtaining unit for obtaining the thickness of the SEI region based on the contour curve and the correspondence between pixels and physical dimensions. This SEI thickness obtaining device can improve the efficiency and accuracy of SEI region thickness analysis.

[0015] According to another aspect of the present disclosure, a device for obtaining SEI thickness is provided, comprising: a memory; and a processor coupled to the memory, wherein the processor is configured to execute the aforementioned method for obtaining SEI thickness based on instructions stored in the memory.

[0016] According to another aspect of the present disclosure, a computer-readable storage medium is provided, on which computer instructions are stored. When the computer instructions are executed by a processor, the method for obtaining the SEI thickness as described above is implemented.

[0017] According to another aspect of the present disclosure, a computer program product is provided. The computer program product includes a computer program or instructions. When the computer program or instructions are executed by a processor, the method for obtaining the SEI thickness as described above is implemented.

[0018] Other features and advantages of the present disclosure will become apparent from the following detailed description of exemplary embodiments of the present disclosure with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0020] The present disclosure can be more clearly understood from the following detailed description with reference to the accompanying drawings, in which: Figure 1 is a flow chart illustrating a method for obtaining SEI thickness according to some embodiments of the present disclosure; Figure 2 is a TEM (Transmission Electron Microscope) image showing a SEI region according to some embodiments of the present disclosure; Figure 3 1 is a TEM image showing a SEI region according to other embodiments of the present disclosure; Figure 4 1 is a TEM image showing a SEI region according to other embodiments of the present disclosure; Figure 5 is a partial diagram showing a TEM image containing a SEI region according to some embodiments of the present disclosure; Figure 6 is a block diagram schematically illustrating a structure of a device for obtaining SEI thickness according to some embodiments of the present disclosure; Figure 7 is a block diagram schematically illustrating a structure of an apparatus for obtaining SEI thickness according to other embodiments of the present disclosure; Figure 8 is a block diagram schematically illustrating a structure of an apparatus for obtaining SEI thickness according to other embodiments of the present disclosure; Figure 9 is a histogram showing the thickness distribution of the SEI region according to some embodiments of the present disclosure.

[0021] It should be understood that the size of each part shown in the drawings is not drawn according to the actual proportional relationship.In addition, the same or similar reference numerals represent the same or similar components. DETAILED DESCRIPTION

[0022] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. The description of the exemplary embodiments is merely illustrative and is in no way intended to limit the present disclosure, its application, or use. The present disclosure can be implemented in many different forms and is not limited to the embodiments described herein. These embodiments are provided to make the present disclosure thorough and complete and to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that unless otherwise specifically stated, the relative arrangement of parts and steps, the composition of materials, numerical expressions, and numerical values ​​set forth in these embodiments should be interpreted as being merely exemplary and not as limiting.

[0023] The terms "first", "second" and similar terms used in this disclosure do not indicate any order, quantity or importance, but are only used to distinguish different parts. The terms "include" or "comprises" and similar terms mean that the elements before the term include the elements listed after the term, and do not exclude the possibility of also including other elements. The terms "upper", "lower", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0024] In the present disclosure, when a specific device is described as being located between a first device and a second device, an intervening device may or may not be present between the specific device and the first device or the second device. When a specific device is described as being connected to another device, the specific device may be directly connected to the other device without an intervening device, or may be not directly connected to the other device but with an intervening device.

[0025] All terms (including technical or scientific terms) used in this disclosure have the same meaning as understood by one of ordinary skill in the art to which this disclosure belongs, unless otherwise specifically defined. It should also be understood that terms defined in, for example, commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant art, and should not be interpreted in an idealized or highly formal sense, unless explicitly defined as such herein.

[0026] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered part of the specification.

[0027] Figure 1 FIG. 1 is a flow chart illustrating a method for obtaining SEI thickness according to some embodiments of the present disclosure. The SEI thickness obtaining method may be performed by an SEI thickness obtaining device. Figure 1 As shown, the method for obtaining SEI thickness includes steps S102 to S108.

[0028] In step S102 , the image containing the SEI region is input into a pre-trained instance segmentation model to identify the SEI region from the image through the instance segmentation model.

[0029] For example, a TEM image containing SEI regions can be input into a pre-trained instance segmentation model to identify SEI regions (or SEI layers) and non-SEI regions from the image through the instance segmentation model. Here, non-SEI regions are regions other than SEI regions in the image. For example, Figure 2 or Figure 3 As shown, the image includes a SEI region 21 and a non-SEI region 22. The SEI region 21 includes a first edge 211 and a second edge 212 extending in the longitudinal direction of the SEI region. Figure 2 or Figure 3 As shown, the image also includes scale values ​​32 and a scale line 31 .

[0030] Here, the instance segmentation model is a machine learning model known to those skilled in the art. For example, the instance segmentation model is a large instance segmentation model built using the Transformer architecture. Using a large instance segmentation model built using the Transformer architecture to identify SEI regions can improve the accuracy of SEI region identification.

[0031] In step S104, the scale value and the scale line in the image are obtained to establish a corresponding relationship between the pixels in the image and the physical size.

[0032] In some embodiments, step S104 may include: using a text recognition library to identify the scale value in the image. Figure 4 As shown, the scale value 32 (eg, 20 nm (nanometers)) can be recognized using the text recognition library EasyOCR.

[0033] The step S104 may also include: processing the image using an edge detection algorithm to obtain the edge of the ruler line in the image. Figure 4 As shown, the Canny edge detection algorithm is used to process the image to obtain the edge of the scale line 31 in the image, and the edge is composed of pixels.

[0034] This step S104 may further include: processing the scale line in the image based on the edge of the scale line to obtain the number of single-row pixels corresponding to the scale line. For example, the scale line in the image may be processed using a Hough transform detection method to obtain pixels in the region where the scale line is located. The pixels in the region where the scale line is located may include one or more rows of pixels. A single row of pixels (i.e., a row of pixels) is obtained from the one or more rows of pixels, and the number of pixels in the single row is then obtained.

[0035] This step S104 may further include calculating the physical size corresponding to each pixel based on the scale value and the number of pixels in a single row corresponding to the scale line, thereby establishing a correspondence between pixels in the image and the physical size. Here, the ratio of the scale value to the number of pixels in a single row is calculated, and this ratio is the physical size corresponding to each pixel, thereby establishing a correspondence between pixels and physical size.

[0036] By obtaining the scale value and scale line in the image, a correspondence between the pixels in the image and the physical dimensions can be established. This facilitates the subsequent accurate determination of the SEI region thickness, thereby improving the accuracy of SEI region thickness analysis.

[0037] In this method, the ruler values ​​are identified using a text recognition library, and the ruler lines are detected using an edge detection algorithm and Hough transform. This allows the identification of length markings (e.g., "20 nm" or "200 nm"). The conversion relationship between the ruler pixel length and the actual physical size is automatically calculated, and all input images are scaled to a uniform physical size based on the ruler information (e.g., 1 pixel = 0.5 nm). This method produces a uniform output resolution, reducing size variations caused by different devices and magnifications, achieving image size normalization, and improving the comparability of measurement results.

[0038] In related technologies, SEI thickness analysis uses inconsistent image scales due to the use of different magnifications (such as ×100k, ×300k, ×500k, etc.) when acquiring electron microscope images. This makes it impossible to directly perform unified SEI thickness identification and measurement on images of different magnifications.

[0039] In the method of the disclosed embodiment, text recognition, edge detection, and scale detection algorithms are used to precisely locate and identify ruler text and graduations, accurately extracting scale information at varying image quality levels. By establishing a correspondence between pixels and physical dimensions (as a pixel-to-physical dimension conversion model), TEM images acquired with different devices and magnifications are unified into a standard coordinate system, minimizing image size.

[0040] In step S106 , curve fitting is performed on the contour of the SEI region identified from the image to obtain a contour curve of the SEI region.

[0041] For example, a spline function can be used to perform curve fitting on the contour of the SEI region to obtain the contour curve of the SEI region. Figure 5 As shown, the contour curve includes a first curve 41 corresponding to the first edge of the SEI region and a second curve 42 corresponding to the second edge of the SEI region. Figure 5 Only a portion of first curve 41 and a portion of second curve 42 are shown. As previously described, the first edge (e.g., first edge 211) is an edge of the SEI region extending along the length of the SEI region, and the second edge (e.g., second edge 212) is the edge of the SEI region opposite the first edge. Using the first and second curves in this profile curve facilitates the determination of SEI region thickness and improves the accuracy of SEI region thickness analysis.

[0042] In step S108 , the thickness of the SEI region is obtained based on the contour curve and the correspondence between pixels and physical dimensions.

[0043] In some embodiments, step S108 may include: obtaining a tangent line at a position point on a first edge of the SEI region based on the first curve; obtaining a normal line of the first curve at the position point based on the tangent line and the position point, wherein the normal line intersects with the second curve to form an intersection point; and obtaining a distance between the position point and the intersection point based on the correspondence between the position point, the intersection point, and the pixel and the physical size, wherein the distance is the thickness of the SEI region at the position point.

[0044] For example, Figure 5 As shown, based on the first curve 41, a tangent line 51 at a position point 411 of the first edge of the SEI region is obtained. Here, the position point 411 is also a point on the first curve 41. Then, based on the tangent line 51 and the position point 411, a normal line 61 of the first curve 41 at the position point 411 can be obtained, wherein the normal line 61 intersects with the second curve 42 to form an intersection point 421. In addition, Figure 5 Also shown in FIG. 5 is a tangent line 52 of the second curve 42 at the intersection point 421. Figure 5 As shown, based on the correspondence between position point 411, intersection point 421, and pixels and physical dimensions, the distance d between position point 411 and intersection point 421 is obtained, where this distance d is the thickness of the SEI region at the position point. Here, the thickness of the SEI region at the position point is the dimension of the SEI region at that position point and extending along the normal direction. In other words, the SEI thickness is the perpendicular distance between the two edges of the SEI layer along the long direction.

[0045] In this embodiment, the tangent and normal at a position point on the first edge of the SEI region can be obtained based on the first curve, thereby obtaining the intersection of the normal and the second curve, and then based on the position point, the intersection point, and the correspondence between the pixel and the physical size, the distance between the position point and the intersection point is obtained, which can improve the accuracy of the SEI region thickness analysis.

[0046] In some embodiments, obtaining the distance between a location point and an intersection point based on the correspondence between the location point, the intersection point, and the pixels and physical dimensions includes: obtaining the number of pixels between the location point and the intersection point along a normal direction; and calculating the distance between the location point and the intersection point based on the correspondence between the pixels and the physical dimensions and the number of pixels between the location point and the intersection point. In other words, after curve fitting the SEI region contour and establishing a reference tangent equation, a pixel-level scan is performed along the normal direction of the tangent line of the SEI region contour to obtain the vertical distance between two edges (or boundaries) of the SEI layer, which serves as the thickness of the SEI region.

[0047] In the above embodiment, by obtaining the number of pixels between the position point and the intersection point, the distance between the position point and the intersection point can be conveniently and accurately calculated based on the number of pixels and the correspondence between pixels and physical sizes (that is, the size of each pixel), thereby improving the accuracy of the thickness analysis of the SEI region.

[0048] In some embodiments, multiple position points can be determined near the above-mentioned position point, and corresponding tangents and normals can be obtained at each position point to obtain multiple normals. The average value of the slopes of the multiple normals is calculated as the average slope, and the average value of the intercepts of the multiple normals is calculated as the average intercept. A new normal is determined using the average slope and the average intercept. The normal is used as the average normal of the above-mentioned multiple normals, and the thickness of the SEI region determined along the direction of the average normal is used as the thickness at the above-mentioned position point and its nearby positions. The thickness is more accurate.

[0049] Thus, a method for obtaining SEI thickness according to some embodiments of the present disclosure is provided. The method includes: inputting an image containing a solid electrolyte interface (SEI) region into a pre-trained instance segmentation model to identify the SEI region from the image using the instance segmentation model; obtaining scale values ​​and scale lines in the image to establish a correspondence between pixels in the image and physical dimensions; performing curve fitting on the contour of the SEI region identified from the image to obtain a contour curve of the SEI region; and obtaining the thickness of the SEI region based on the contour curve and the correspondence between pixels and physical dimensions. In this method, the image is identified using a pre-trained instance segmentation model to identify the SEI region from the image, and the correspondence between pixels in the image and physical dimensions is established by obtaining scale values ​​and scale lines. After fitting the contour of the identified SEI region to obtain a contour curve, the thickness of the SEI region is obtained based on the contour curve and the correspondence between pixels and physical dimensions. Compared to manual methods, this method can improve the efficiency of SEI region thickness analysis, and because this method does not involve manual subjective judgment, it can also improve the accuracy of SEI region thickness analysis. For example, this method can accurately measure the thickness of curved SEI.

[0050] In the disclosed embodiments, precise segmentation of the SEI region in TEM images and an automated scale recognition algorithm are used to normalize image size and quantify SEI thickness parameters. Compared to manual analysis, this method can shorten the processing time for a single image, for example, from 10 minutes to 1 second. It also supports batch processing, accelerating the study of the dynamic evolution of the SEI and enabling the establishment of SEI models and the identification of evolution mechanisms.

[0051] In some cases, the above method can, to a certain extent, solve the problem of cross-sectional SEI measurement, enabling automated pixel-level scanning of the entire area. For example, a fully automated measurement process can achieve over 1,000 measurement points per micron of SEI length, increasing data volume by 2 to 3 orders of magnitude compared to manual single-point measurement. In some embodiments, the thickness distribution map automatically generated by the system includes multi-dimensional parameters: average thickness (for example, accurate to 0.1 nm), thickness variance (reflecting uniformity), maximum / minimum values ​​(to identify abnormal areas), and gradient distribution, providing a comprehensive quantitative basis for SEI performance evaluation.

[0052] In some embodiments, the SEI thickness determination method may further include: determining multiple thicknesses of the SEI region at multiple locations on the first edge of the SEI region; and calculating an average of the multiple thicknesses as the average SEI region thickness. Specifically, multiple locations may be selected and their corresponding thicknesses calculated to obtain multiple SEI region thicknesses. The average of the multiple thicknesses may then be calculated as the average SEI region thickness, further improving the accuracy of SEI region thickness analysis.

[0053] Here, the system can automatically perform a large number of normal line measurements. For example, it can perform measurements at 1000 measurement points (i.e., the locations described above) (e.g., with a 1nm interval between adjacent measurement points), thus obtaining 1000 thickness data points. By statistically eliminating outliers, it can output statistical results such as thickness distribution histograms. Through full-area pixel-level measurement and data statistics, a thickness distribution feature map can be established.

[0054] For example, Figure 9 The thickness distribution histograms of two SEIs (SEI1 and SEI2) are shown, so that the thickness of the two SEIs can be compared and the SEI can be analyzed accordingly.

[0055] In some embodiments, the SEI thickness acquisition method may further include: prior to inputting the image into the instance segmentation model, training the instance segmentation model using an image with pre-labeled SEI regions. In this embodiment, training the instance segmentation model using an image with pre-labeled SEI regions enables the instance segmentation model to identify SEI regions in the image, thereby improving the accuracy of SEI region identification.

[0056] In some embodiments, training an instance segmentation model using images with pre-labeled SEI regions includes: labeling SEI regions in multiple images using a labeling tool to obtain multiple images with pre-labeled SEI regions; and inputting the multiple images with pre-labeled SEI regions into the instance segmentation model to train the instance segmentation model.

[0057] For example, to perform high-resolution transmission electron microscopy data acquisition and annotation, a transmission electron microscope can be used to obtain high-resolution TEM images, and the SEI area can be annotated using labelme software to accurately define the SEI / electrode interface; SEI annotation data can be provided to the machine learning model for training to distinguish between SEI areas and non-SEI areas.

[0058] For example, you can only annotate key frames (for example, select an image randomly from every 100 images) and use semi-automatic annotation tools to annotate the SEI areas in the image to facilitate subsequent training operations. For another example, you can analyze high-resolution transmission electron microscopy data and manually select data for SEI annotation. For example, Figure 2 As shown, the red boxes are manually annotated SEI regions. This establishes an automated recognition system for intelligent SEI identification. We introduce a large instance segmentation model built on the Transformer architecture, leveraging its superior feature extraction and generalization capabilities to identify SEI regions. SEI-annotated data is provided to train a machine learning model to distinguish between SEI and non-SEI regions. Figure 3 The green area in the figure is the SEI area marked by the trained instance segmentation model.

[0059] Generally speaking, in situ liquid electrochemical experiments are difficult and high-quality in situ electron microscopy data are relatively scarce. However, in the above training method, a high-precision instance segmentation model of the SEI region can be obtained by fine-tuning using a small amount of labeled data (for example, about 50 to 100 images) through transfer learning technology.

[0060] Through the above training process, an instance segmentation model for efficient and intelligent SEI identification can be obtained. Through the key frame annotation strategy, the annotation quality is improved while the manual annotation workload is reduced by more than 90%, achieving high-precision SEI identification with small samples, and the segmentation accuracy can reach more than 95%.

[0061] Figure 6 FIG. 1 is a block diagram schematically illustrating a structure of an apparatus for obtaining SEI thickness according to some embodiments of the present disclosure. Figure 6 As shown, the SEI thickness obtaining device includes: an input unit 602 , a first obtaining unit 604 , a fitting unit 606 and a second obtaining unit 608 .

[0062] The input unit 602 is used to input an image containing a solid electrolyte interface SEI region into a pre-trained instance segmentation model, so as to identify the SEI region from the image through the instance segmentation model.

[0063] The first obtaining unit 604 is used to obtain the scale value and the scale line in the image to establish a corresponding relationship between the pixels in the image and the physical size.

[0064] The fitting unit 606 is configured to perform curve fitting on the contour of the SEI region identified from the image to obtain a contour curve of the SEI region.

[0065] The second obtaining unit 608 is configured to obtain the thickness of the SEI region based on the contour curve and the correspondence between pixels and physical dimensions.

[0066] In this way, a device for obtaining SEI thickness is provided, which can improve the efficiency and accuracy of SEI region thickness analysis.

[0067] In some embodiments, the contour curve includes a first curve corresponding to a first edge of the SEI region and a second curve corresponding to a second edge of the SEI region, the first edge being an edge of the SEI region extending along the long direction of the SEI region, and the second edge being an edge of the SEI region opposite to the first edge.

[0068] In some embodiments, the second obtaining unit 608 is used to: obtain a tangent line at a position point on the first edge of the SEI region based on the first curve; obtain a normal line of the first curve at the position point based on the tangent line and the position point, wherein the normal line intersects with the second curve to form an intersection point; and obtain a distance between the position point and the intersection point based on the correspondence between the position point, the intersection point, and the pixel and the physical size, wherein the distance is the thickness of the SEI region at the position point.

[0069] In some embodiments, the second obtaining unit 608 is used to: obtain the number of pixels between the position point and the intersection point along the direction of the normal; and calculate the distance between the position point and the intersection point based on the correspondence between the pixels and the physical size and the number of pixels between the position point and the intersection point.

[0070] In some embodiments, the second obtaining unit 608 is further configured to: obtain multiple thicknesses of the SEI region at multiple locations based on multiple locations at the first edge of the SEI region; and calculate an average value of the multiple thicknesses as the average thickness of the SEI region.

[0071] In some embodiments, the first obtaining unit 604 is used to: use a text recognition library to identify the scale value in the image; use an edge detection algorithm to process the image to obtain the edge of the scale line in the image; based on the edge of the scale line, process the scale line in the image to obtain the number of single-row pixels corresponding to the scale line; and calculate the physical size corresponding to each pixel based on the scale value and the number of single-row pixels corresponding to the scale line to establish a correspondence between pixels in the image and physical sizes.

[0072] In some embodiments, the SEI thickness obtaining device further includes: a training unit, configured to train an instance segmentation model using images with SEI regions pre-labeled.

[0073] In some embodiments, the training unit is used to use a labeling tool to label SEI regions in multiple images to obtain multiple images with pre-labeled SEI regions, and input the multiple images with pre-labeled SEI regions into the instance segmentation model to train the instance segmentation model.

[0074] In some embodiments, the instance segmentation model is a large instance segmentation model built using a Transformer architecture.

[0075] In the above-mentioned methods and devices of the disclosed embodiments, intelligent image processing methods are used to precisely segment the SEI region, and scale recognition is used to normalize the image size. Full-area pixel-level measurement and data statistics are used to establish a thickness distribution feature map, providing a quantitative analysis tool for battery material interface research. Combining artificial intelligence with electron microscopy image processing technology and developing machine learning for data analysis provides accurate and efficient data analysis tools, which can accelerate the study of SEI dynamic evolution, enable SEI modeling, and determine evolution mechanisms.

[0076] Figure 7 Schematically illustrates a block diagram of a device for obtaining SEI thickness according to some other embodiments of the present disclosure. The device for obtaining SEI thickness includes a memory 710 and a processor 720. The memory 710 can be a disk, a flash memory or any other non-volatile storage medium. Figure 1 The instructions in the corresponding embodiment.

[0077] The processor 720 is coupled to the memory 710 and can be implemented as one or more integrated circuits, such as a microprocessor or a microcontroller. The processor 720 is used to execute instructions stored in the memory, which can improve the efficiency and accuracy of SEI region thickness analysis.

[0078] In one embodiment, it is also possible to Figure 8 As shown, the SEI thickness obtaining device 700 includes a memory 710 and a processor 720. The processor 720 is coupled to the memory 710 via a BUS 730. The SEI thickness obtaining device 700 can also be connected to an external storage device 750 via a storage interface 740 to access external data, and can also be connected to a network or another computer system (not shown) via a network interface 760, which will not be described in detail here.

[0079] In this embodiment, the efficiency and accuracy of SEI region thickness analysis can be improved by storing data instructions in a memory and then processing the instructions through a processor.

[0080] In another embodiment, the present disclosure further provides a computer-readable storage medium (eg, a non-transitory computer-readable storage medium) having computer program instructions stored thereon. When the instructions are executed by a processor, the computer-readable storage medium implements the following operations: Figure 1 The steps of the method in the corresponding embodiment. Those skilled in the art will understand that the embodiments of the present disclosure can be provided as methods, apparatuses, or computer program products. Therefore, the present disclosure can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present disclosure can take the form of a computer program product implemented on one or more computer-usable non-transitory storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0081] The present disclosure is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present disclosure. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0082] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0083] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0084] In some embodiments of the present disclosure, a computer program product is further provided. The computer program product includes a computer program or instructions. When the computer program or instructions are executed by a processor, the method for obtaining the SEI thickness as described above is implemented.

[0085] In some embodiments of the present disclosure, a computer program is further provided, comprising: instructions, which, when executed by a processor, cause the processor to perform the above-mentioned method for obtaining the SEI thickness.

[0086] Thus far, various embodiments of the present disclosure have been described in detail. To avoid obscuring the concept of the present disclosure, some details known in the art have not been described. Based on the above description, those skilled in the art can fully understand how to implement the technical solutions disclosed herein.

[0087] Although some specific embodiments of the present disclosure have been described in detail through examples, those skilled in the art will understand that the above examples are for illustration only and are not intended to limit the scope of the present disclosure. Those skilled in the art will understand that the above embodiments may be modified or some technical features may be replaced with equivalents without departing from the scope and spirit of the present disclosure. The scope of the present disclosure is defined by the appended claims.

Claims

1. A method for obtaining SEI thickness, characterized in that: include: Inputting an image containing a solid electrolyte interface (SEI) region into a pre-trained instance segmentation model to identify the SEI region from the image using the instance segmentation model, wherein the instance segmentation model is a large instance segmentation model constructed using a Transformer architecture; Obtaining a scale value and a scale line in the image to establish a corresponding relationship between pixels in the image and physical dimensions; performing curve fitting on the contour of the SEI region identified from the image to obtain a contour curve of the SEI region; and Obtaining the thickness of the SEI region based on the contour curve and the correspondence between the pixels and the physical size; The contour curve includes a first curve corresponding to a first edge of the SEI region and a second curve corresponding to a second edge of the SEI region, the first edge being an edge of the SEI region extending along a longitudinal direction of the SEI region, and the second edge being an edge of the SEI region opposite to the first edge; Based on the contour curve and the correspondence between the pixels and the physical size, the thickness of the SEI region is obtained, including: based on the first curve, obtaining a tangent at a position point on the first edge of the SEI region; based on the tangent and the position point, obtaining a normal of the first curve at the position point, wherein the normal intersects with the second curve to form an intersection; along the direction of the normal, obtaining the number of pixels between the position point and the intersection; and according to the correspondence between the pixels and the physical size and the number of pixels between the position point and the intersection, calculating the distance between the position point and the intersection, wherein the distance is the thickness of the SEI region at the position point.

2. The method for obtaining SEI thickness according to claim 1, wherein: Also includes: Based on a plurality of position points at a first edge of the SEI region, obtaining a plurality of thicknesses of the SEI region at the plurality of position points; and The average value of the plurality of thicknesses is calculated as the average thickness of the SEI region.

3. The method for obtaining SEI thickness according to claim 1, wherein: Obtaining a scale value and a scale line in the image to establish a corresponding relationship between pixels in the image and physical dimensions includes: Recognizing the scale value in the image using a text recognition library; Processing the image using an edge detection algorithm to obtain the edge of the ruler line in the image; processing the scale line in the image based on an edge of the scale line to obtain the number of pixels in a single row corresponding to the scale line; and Based on the scale value and the number of pixels in a single row corresponding to the scale line, the physical size corresponding to each pixel is calculated to establish a corresponding relationship between the pixels in the image and the physical size.

4. The method for obtaining SEI thickness according to claim 1, wherein: Also includes: Before inputting the image into the instance segmentation model, the instance segmentation model is trained using the image with the SEI region pre-labeled.

5. The method for obtaining SEI thickness according to claim 4, wherein: Training the instance segmentation model using an image pre-labeled with the SEI region includes: Annotating the SEI regions in the plurality of images using an annotation tool to obtain a plurality of images in which the SEI regions are pre-annotated; and A plurality of images in which the SEI region is pre-labeled are input into the instance segmentation model to train the instance segmentation model.

6. A device for obtaining SEI thickness, characterized in that: include: An input unit is configured to input an image containing a solid electrolyte interface (SEI) region into a pre-trained instance segmentation model, so as to identify the SEI region from the image using the instance segmentation model, wherein the instance segmentation model is a large instance segmentation model constructed using a Transformer architecture; A first obtaining unit is used to obtain a scale value and a scale line in the image to establish a corresponding relationship between pixels in the image and physical dimensions; a fitting unit, configured to perform curve fitting on the contour of the SEI region identified from the image to obtain a contour curve of the SEI region; and A second obtaining unit is configured to obtain a thickness of the SEI region based on the contour curve and the correspondence between the pixels and the physical size; The contour curve includes a first curve corresponding to a first edge of the SEI region and a second curve corresponding to a second edge of the SEI region, the first edge being an edge of the SEI region extending along a longitudinal direction of the SEI region, and the second edge being an edge of the SEI region opposite to the first edge; The second obtaining unit is configured to: obtain a tangent line at a position point on a first edge of the SEI region based on the first curve; obtain a normal line of the first curve at the position point based on the tangent line and the position point, wherein the normal line intersects with the second curve to form an intersection point; and obtain a distance between the position point and the intersection point based on the position point, the intersection point, and a correspondence between the pixel and the physical size, wherein the distance is a thickness of the SEI region at the position point; The second obtaining unit is used to: obtain the number of pixels between the position point and the intersection point along the direction of the normal; and calculate the distance between the position point and the intersection point based on the correspondence between the pixels and the physical size and the number of pixels between the position point and the intersection point.

7. A device for obtaining SEI thickness, characterized in that: include: Memory; as well as A processor coupled to the memory, wherein the processor is configured to execute the SEI thickness obtaining method according to any one of claims 1 to 5 based on instructions stored in the memory.

8. A computer-readable storage medium, characterized in that Computer instructions are stored thereon, and when the computer instructions are executed by a processor, the method for obtaining the SEI thickness according to any one of claims 1 to 5 is implemented.

9. A computer program product, characterized in that The computer program product includes a computer program or instructions, and when the computer program or instructions are executed by a processor, the method for obtaining SEI thickness according to any one of claims 1 to 5 is implemented.

Images