Electrode material surface three-dimensional morphology reconstruction method and device and electronic equipment

Abstract

This invention discloses a method, apparatus, and electronic device for reconstructing the three-dimensional morphology of an electrode material surface. The method includes: determining the total depth confidence index corresponding to multiple candidate depth values ​​in the (t-1)th iteration; determining the depth probability value of a corresponding pixel in the tth iteration for each of the candidate depth values ​​based on the total depth confidence index and pixel sharpness; determining the iterative depth value of the corresponding pixel based on the corresponding depth probability value, until a target condition is met, to obtain the target depth value; combining the target depth values ​​corresponding to multiple pixels to obtain a depth map; and reconstructing the three-dimensional morphology of the electrode material surface based on the depth map. This invention solves the technical problem in related technologies where inaccurate determination of the depth value of the electrode material surface leads to low accuracy in the three-dimensional morphology reconstruction of the electrode material surface.

Description

Technical Field

[0001] This invention relates to the field of batteries, and more specifically, to a method, apparatus, and electronic device for reconstructing the three-dimensional morphology of an electrode material surface. Background Technology

[0002] The surface microstructure of electrode materials, such as porosity, roughness, and particle distribution, has a decisive impact on their electrochemical performance (e.g., capacity, rate performance, and cycle life). Accurately and rapidly acquiring their three-dimensional morphology is crucial for material development and performance optimization. In related technologies, there is a technical problem in reconstructing the three-dimensional morphology of electrode material surfaces: inaccurate determination of the surface depth value leads to low accuracy in the reconstructed three-dimensional morphology.

[0003] There is currently no effective solution to the above problems. Summary of the Invention

[0004] This invention provides a method, apparatus, and electronic device for reconstructing the three-dimensional morphology of an electrode material surface, in order to at least solve the technical problem in the related art where the depth value of the electrode material surface is not accurately determined, resulting in low accuracy of the three-dimensional morphology reconstruction of the electrode material surface.

[0005] According to one aspect of the present invention, a method for reconstructing the three-dimensional morphology of an electrode material surface is provided, comprising: acquiring an image sequence of the electrode material; for a plurality of pixels in the image sequence, determining the pixel sharpness corresponding to the corresponding pixel at a plurality of candidate depth values; according to the execution order of multiple iterations, for the t-th iteration, determining the total depth confidence index corresponding to the plurality of candidate depth values ​​at the (t-1)-th iteration; based on the total depth confidence index corresponding to the plurality of candidate depth values ​​at the (t-1)-th iteration, and the pixel sharpness corresponding to the corresponding pixel at the plurality of candidate depth values, determining the pixel sharpness at the plurality of candidate depth values ​​at the t-th iteration. The depth probability values ​​corresponding to the selected depth values ​​are determined, where the total depth confidence index represents the total confidence level that the pixel sharpness of the multiple pixels at the corresponding candidate depth values ​​is a predetermined sharpness, and t is an integer greater than or equal to 1. Based on the depth probability values ​​corresponding to the pixels at the multiple candidate depth values ​​in the t-th iteration, the iterative depth value of the corresponding pixels is determined until the target condition is met, and the target depth value of the corresponding pixels is obtained, wherein the target condition includes the completion of multiple iterations. The target depth values ​​corresponding to the multiple pixels are combined to obtain a depth map. Based on the depth map, the three-dimensional morphology of the electrode material surface is reconstructed.

[0006] Optionally, determining the pixel sharpness of the corresponding pixel at multiple candidate depth values ​​includes: determining the color information entropy and grayscale gradient variance of the corresponding pixel at multiple candidate depth values, wherein the color information entropy is used to quantify the complexity of the color distribution of the corresponding pixel at the corresponding candidate depth value, and the grayscale gradient variance is used to quantify the fluctuation of the grayscale change of the corresponding pixel at the corresponding candidate depth value; and determining the pixel sharpness of the corresponding pixel at multiple candidate depth values ​​based on the color information entropy and grayscale gradient variance of the corresponding pixel at multiple candidate depth values.

[0007] Optionally, determining the color information entropy of the corresponding pixel at each of the multiple candidate depth values ​​includes: for any one of the multiple candidate depth values, dividing a target region from the target color image corresponding to the corresponding pixel, wherein the image sequence includes the target color image; determining multiple first region points within the target region, wherein the multiple first region points are pixels within the target region; and determining the color information entropy of the corresponding pixel at the given candidate depth value based on the color parameters corresponding to the multiple first region points within the target region.

[0008] Optionally, determining the color information entropy of a corresponding pixel at any candidate depth value based on the color parameters corresponding to the multiple first region points within the target region includes: determining the color component distribution corresponding to multiple color channels within the target region based on the color parameters corresponding to the multiple first region points within the target region, wherein the corresponding color component distribution is the distribution of the multiple first region points under different color components under the corresponding color channels; determining the probability of the occurrence of the corresponding first region point under the color components corresponding to the multiple color channels within the target region based on the color component distribution corresponding to the multiple color channels; and determining the color information entropy of the corresponding pixel at any candidate depth value based on the probability of the occurrence of the corresponding first region point under the color components corresponding to the multiple color channels within the target region.

[0009] Optionally, the step of dividing the target region from the target color image corresponding to the corresponding pixel point includes: dividing the target region from the target color image corresponding to the corresponding pixel point with the pixel position of the corresponding pixel point as the center.

[0010] Optionally, determining the gray-level gradient variance of the corresponding pixel at multiple candidate depth values ​​includes: for any one of the multiple candidate depth values, dividing a gray-level region from the target gray-level image corresponding to the corresponding pixel, wherein the image sequence includes the target gray-level image; and determining the gray-level gradient variance of the corresponding pixel at the any one candidate depth value based on the gray-level values ​​corresponding to multiple second region points in the gray-level region.

[0011] Optionally, determining the pixel sharpness of a corresponding pixel at multiple candidate depth values ​​includes: constructing an objective function, wherein the objective function includes a first sub-function and a second sub-function, the first sub-function being used to quantize the sharpness of a pixel at a corresponding candidate depth value in a color image scene, and the second sub-function being used to quantize the sharpness of a pixel at a corresponding candidate depth value in a grayscale image scene, the objective function being a function aimed at maximizing the total sharpness of a pixel at a corresponding candidate depth value; and determining the pixel sharpness of a corresponding pixel at multiple candidate depth values ​​based on the objective function.

[0012] According to one aspect of the present invention, a three-dimensional morphology reconstruction apparatus for an electrode material surface is provided, comprising: an acquisition module for acquiring an image sequence of the electrode material; a first determination module for determining, for a plurality of pixels in the image sequence, the pixel sharpness corresponding to a corresponding pixel at a plurality of candidate depth values; and a second determination module for determining, according to the execution order of multiple iterations, for the t-th iteration, a total depth confidence index corresponding to a plurality of candidate depth values ​​at the (t-1)-th iteration; and, based on the total depth confidence index corresponding to a plurality of candidate depth values ​​at the (t-1)-th iteration and the pixel sharpness corresponding to the corresponding pixel at the plurality of candidate depth values, determining, at the t-th iteration, the pixel sharpness corresponding to a corresponding pixel at a plurality of candidate depth values ​​at the plurality of candidate depth values. The following modules are used to determine the depth probability values ​​corresponding to the pixels at the selected depth values: a first module determines the depth probability value of the corresponding pixel at the selected depth values, and a second module determines the depth probability value of the corresponding pixel at the selected depth values, and so on, until a target condition is met to obtain the target depth value of the corresponding pixel. The target condition includes the completion of multiple iterations. A third module combines the target depth values ​​corresponding to the pixels to obtain a depth map. A fifth module reconstructs the three-dimensional morphology of the electrode material surface based on the depth map.

[0013] According to one aspect of the present invention, an electronic device is provided, comprising: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to execute the instructions to implement the three-dimensional morphology reconstruction method for electrode material surface as described in any of the preceding embodiments.

[0014] According to one aspect of the present invention, a computer-readable storage medium is provided, wherein when the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform the three-dimensional morphology reconstruction method for the electrode material surface described in any of the preceding claims.

[0015] In this embodiment of the invention, by acquiring an image sequence of the electrode material and determining the pixel sharpness of each pixel at multiple candidate depth values, and following the execution order of multiple iterations, each iteration combines the total depth confidence index of the previous iteration with the pixel sharpness at each candidate depth. This allows for a comprehensive consideration of historical depth confidence and current pixel sharpness features, accurately quantifying the probability that each candidate depth becomes the optimal depth, and thus gradually and precisely determining the target depth value of each pixel. This achieves high-precision reconstruction of the three-dimensional morphology of the electrode material surface, thereby solving the technical problem in related technologies where the determination of the electrode material surface depth value is inaccurate, resulting in low accuracy in the reconstruction of the three-dimensional morphology of the electrode material surface. Attached Figure Description

[0016] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0017] Figure 1 This is a flowchart of a method for reconstructing the three-dimensional morphology of an electrode material surface according to an embodiment of the present invention;

[0018] Figure 2 This is a schematic diagram of the three-dimensional morphology reconstruction process of the electrode material surface in an optional embodiment of the present invention;

[0019] Figure 3 This is a schematic diagram of the structure of a multi-focus image sequence acquisition system in an optional embodiment of the present invention;

[0020] Figure 4 This is a sample image of a multi-focus image sequence of a lithium battery electrode surface in an optional embodiment of the present invention;

[0021] Figure 5 This is a comparison chart of the effects of the focusing evaluation function in an optional embodiment of the present invention;

[0022] Figure 6This is a schematic diagram comparing the surface reconstruction effect of electrode materials in an optional embodiment of the present invention;

[0023] Figure 7 This is a structural block diagram of an electrode material surface three-dimensional morphology reconstruction device according to an embodiment of the present invention. Detailed Implementation

[0024] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0025] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0026] Example 1

[0027] According to an embodiment of the present invention, an embodiment of a method for reconstructing the three-dimensional morphology of an electrode material surface is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0028] Figure 1 This is a flowchart of a method for reconstructing the three-dimensional morphology of an electrode material surface according to an embodiment of the present invention, such as... Figure 1 As shown, the method includes the following steps:

[0029] S102, acquire the image sequence of the electrode material;

[0030] Specifically, acquiring the image sequence of the electrode material includes: placing the electrode material on a displacement stage, controlling the displacement stage to move at equal intervals along the optical axis, and acquiring a microscopic image at each step position, thereby obtaining an image sequence covering the undulation range of the electrode material surface. .in, Indicates pixel index, Indicates the first Zhang electrode material surface image, Image sequence number The total number of images is [value], and the size of a single image in the image sequence is [value]. , Indicates the length of the image. This represents the width of the image. The higher the number of images, the higher the accuracy of the reconstructed depth map, but the lower the efficiency of the algorithm. To achieve a balance between reconstruction accuracy and efficiency, the number of images is approximately between 60 and 150. The image sequence is characterized by having the same field of view size and angle, but with continuously varying focus.

[0031] Optionally, an image acquisition system can be used to acquire multi-focus microscopic image sequences to obtain image sequences of the electrode material. The image acquisition system includes an optical microscope, a displacement stage (specifically a precision Z-axis displacement stage), and a computer control system. Additionally, it may include:

[0032] Mechanical base: The rigid support for the entire system, ensuring the stability of the equipment.

[0033] Motion guide rail: A vertical linear guide rail that provides a high-precision Z-axis motion path for the imaging components.

[0034] Stepper motor: Drives the imaging component to move up and down along the motion guide rail to achieve image acquisition at different focus depths. Stepper control can accurately position each candidate depth.

[0035] Tilt adjustment device: Installed below the stage, it can finely adjust the tilt angle of the sample to ensure that the sample surface is perpendicular to the imaging optical axis.

[0036] Stage: A platform for placing the sample to be tested, with attitude correction achieved through a tilt adjustment device.

[0037] Camera: The core device for image acquisition, used to capture sample images at different depths of focus, specifically including charge-coupled device (CCD) / complementary metal-oxide-semiconductor (CMOS) cameras.

[0038] Lens assembly: Optical magnification and focusing components, working in conjunction with the Z-axis movement of a stepper motor to achieve clear imaging at different depths.

[0039] Magnifying lens: Further improves imaging resolution and is used to observe the fine structure of samples.

[0040] Coaxial light source: Illuminates the sample along the optical axis, which is suitable for highlighting surface smoothness and edge details and reducing shadow interference.

[0041] Ring light source: Arranged around the magnifying lens, it provides uniform ring illumination, enhances the surface texture and three-dimensionality of the sample, and is suitable for imaging needs of samples of different materials.

[0042] Host computer: Communicates with the camera and stepper motor, is responsible for controlling the movement of the stepper motor, triggering image acquisition, and performing subsequent processing on the acquired multi-focus image sequence (such as depth reconstruction and sharpness evaluation).

[0043] Optionally, when acquiring image sequences of electrode materials, a ring light source is installed around the lens group acquiring the image sequences to improve field of view clarity and overall imaging quality by optimizing illumination.

[0044] S104, For multiple pixels in an image sequence, determine the pixel sharpness of the corresponding pixel under multiple candidate depth values;

[0045] Specifically, this includes: the depth distribution range corresponding to multiple candidate depth values. The pixel sharpness of each pixel (i.e., the corresponding pixel) under multiple candidate depth values. , represented as:

[0046]

[0047] in, Indicates pixel index, Indicates the depth value to be selected. .

[0048] After determining the pixel sharpness corresponding to each pixel at multiple candidate depth values, a normalization process is performed using the following formula:

[0049]

[0050] in, This represents the pixel sharpness of the corresponding pixel after normalization at the selected depth value z. is a constant used to prevent division by zero and avoid zero values; and Represents pixel index The extreme values ​​of pixel sharpness under multiple candidate depth values.

[0051] in, The formula for determining it is:

[0052]

[0053] The formula for determining it is:

[0054]

[0055] in, This represents the total number of depth values.

[0056] Optionally, determining the pixel sharpness of a corresponding pixel at multiple candidate depth values ​​includes: determining the color information entropy and grayscale gradient variance of the corresponding pixel at multiple candidate depth values, wherein the color information entropy is used to quantify the complexity of the color distribution of the corresponding pixel at the corresponding candidate depth value, and the grayscale gradient variance is used to quantify the fluctuation of the grayscale change of the corresponding pixel at the corresponding candidate depth value; and determining the pixel sharpness of the corresponding pixel at multiple candidate depth values ​​based on the color information entropy and grayscale gradient variance of the corresponding pixel at multiple candidate depth values.

[0057] By first determining the color information entropy and grayscale gradient variance of a pixel at each candidate depth value, the complexity of color distribution and the degree of grayscale variation fluctuation are quantified respectively. This effectively compensates for the inability of grayscale gradient to represent texture in complex scenes such as low texture, few textures, or single color. Robust feature representation is achieved by utilizing the richness of color information entropy in color distribution. At the same time, by combining grayscale gradient variance to capture the fluctuation of grayscale variation, complementary coupling of color and grayscale dimensions is achieved. This enables a comprehensive multi-dimensional representation of the imaging sharpness features of pixels at different candidate depths, ensuring more accurate and comprehensive quantification of pixel sharpness and providing a reliable feature basis for subsequent determination of the optimal depth value of a pixel based on sharpness.

[0058] For example, in cases with high color richness and low texture, grayscale gradient variance is difficult to capture effective texture changes, while color information entropy can clearly distinguish the imaging differences at different depths through the complexity of color distribution, avoiding the failure of sharpness judgment due to missing texture. In cases with low color richness and high texture, color information entropy is not sensitive to color changes, while grayscale gradient variance can accurately characterize the degree of texture sharpness through the amplitude of grayscale fluctuations, making up for the deficiency of a single color system in being unable to identify textures in monotonous color scenes.

[0059] Optionally, the color information entropy of the corresponding pixel at multiple candidate depth values ​​is determined, including:

[0060] A1. For any one of the multiple candidate depth values, take the pixel position of the corresponding pixel as the center, and divide the target region from the target color image corresponding to the corresponding pixel. The image sequence includes the target color image.

[0061] Specifically, this includes arranging the acquired image sequences parallel and at equal intervals along the Z-axis to form a three-dimensional data cube (hereinafter referred to as the cube) covering the height range of the sample. Any pixel within this cube can be represented as... ,in, Represents pixel coordinates, Indicates pixel index, Indicates the image sequence number. , , , Image width, Image height, The total number of images; the number of images in the image sequence. Pixels on an image Draw a circle of size centered on . The rectangular window is the target area. The window size of this rectangular window can only be an integer, and is generally an odd number.

[0062] A2, determine multiple first region points within the target area, where the multiple first region points are pixels within the target area;

[0063] Among them, multiple first region points within the target area constitute a pixel set. , represented as:

[0064]

[0065] in, Represents the first image in the image sequence. The pixel coordinates in the image are The pixels; Indicates the length and / or width of the target area.

[0066] A3. Based on the color parameters corresponding to multiple first area points within the target area, determine the color component distribution corresponding to multiple color channels within the target area. The corresponding color component distribution is the distribution of multiple first area points under different color components under the corresponding color channel.

[0067] When the color component distribution can be represented by a histogram, specifically including: for each color channel of the same image in the image sequence. Statistical target area The color components on each color channel are equal to The number of pixels is used to obtain the histogram. , .

[0068] A4, based on the color component distribution corresponding to multiple color channels, determines the probability of the first region point appearing under the color components corresponding to multiple color channels in the target area, using the following formula:

[0069]

[0070] in, Indicates color channel Corresponding color components The probability of the first point appearing (i.e., the color probability). .

[0071] A5 determines the color information entropy of the corresponding pixel at any candidate depth value based on the probability of the first region point appearing under the color components corresponding to multiple color channels in the target region.

[0072] Specifically, this includes: for each color channel Calculate the color channels within the rectangular window The corresponding color information entropy is given by the formula:

[0073]

[0074] in, Color channels within a rectangular window The corresponding color information entropy, when hour, Defined as 0.

[0075] Furthermore, determine the color information entropy of the corresponding pixel at any candidate depth value. The formula is:

[0076]

[0077] in, The color information entropy corresponding to the red channel; The color information entropy corresponding to the green channel; This represents the color information entropy corresponding to the blue channel.

[0078] By dividing the target area into W×W regions centered on the pixel, noise interference from single-pixel color information can be avoided, and more statistically significant local color features can be obtained. By extracting the color component distribution of each color channel within the target area and calculating the corresponding probability, the local color distribution can be transformed into a quantifiable probability statistical form. By calculating the color information entropy of each channel based on probability and summing them, the imaging clarity of the pixel at the corresponding candidate depth can be accurately quantified from the perspective of color distribution complexity.

[0079] Optionally, the variance of the grayscale gradient of the corresponding pixel at multiple candidate depth values ​​is determined, including:

[0080] B1, for any one of the multiple candidate depth values, divide the gray area from the target gray image corresponding to the corresponding pixel, wherein the image sequence includes the target gray image;

[0081] Specifically, this includes: calculating the gradient matrix of each image in the order of the image sequence. :

[0082]

[0083]

[0084]

[0085] in, and These represent the horizontal and vertical gradients of each image in the image sequence, respectively. Represents grayscale value; and These represent the horizontal and vertical gradient convolution kernels, respectively.

[0086] Gradient matrix Divided according to a third dimension (e.g., the depth dimension of an image sequence) A planar matrix, with pixels on each planar matrix. Centered on, delineate a scale of [missing information]. The window is used as a grayscale area.

[0087] B2, based on the gray values ​​corresponding to multiple points in the second region within the gray area, determines the gray-level gradient variance of the corresponding pixel at any candidate depth value, using the following formula:

[0088]

[0089] in, This represents the variance of the grayscale gradient. This is the gradient average.

[0090] The value can be determined using the following formula:

[0091]

[0092] By dividing the grayscale region from the target grayscale image corresponding to the pixel, the local grayscale feature range can be obtained around the corresponding pixel, avoiding the random interference of single pixel grayscale values ​​and ensuring the effectiveness of grayscale feature statistics. Based on the grayscale value of the second region point within the grayscale region, the grayscale gradient variance can be calculated, which can accurately quantify the degree of fluctuation of the grayscale change of the pixel at the corresponding candidate depth, intuitively reflecting the clear features of grayscale texture, and providing a reliable grayscale dimension quantification basis for multi-dimensional evaluation of pixel clarity.

[0093] Optionally, determining the pixel sharpness of a corresponding pixel at multiple candidate depth values ​​includes: constructing an objective function, wherein the objective function includes a first sub-function and a second sub-function, the first sub-function being used to quantize the pixel sharpness at the corresponding candidate depth value in a color image scene, the second sub-function being used to quantize the pixel sharpness at the corresponding candidate depth value in a grayscale image scene, and the objective function being a function that aims to maximize the total sharpness of the pixel at the corresponding candidate depth value; and determining the pixel sharpness of a corresponding pixel at multiple candidate depth values ​​based on the objective function.

[0094] Wherein, objective function for:

[0095]

[0096] in, Used to quantify the total sharpness of a pixel at the corresponding candidate depth value; This is the first sub-function, used to quantize the sharpness of pixels at the corresponding candidate depth values ​​in a color image scene; This is the second sub-function, used to quantize the sharpness of pixels at the corresponding candidate depth values ​​in grayscale image scenarios; These are adaptive weighting coefficients.

[0097] By constructing an objective function that maximizes the total sharpness of pixels at the selected depth, we can accurately quantify the coupling relationship between color information entropy and grayscale gradient variance. This enables adaptive coupling and adaptation of sharpness features in color and grayscale dimensions, thereby helping to accurately integrate sharpness features in different dimensions and maximize the restoration of the true sharpness of pixels. This makes the quantification results of pixel sharpness more consistent with the actual imaging state.

[0098] S106, following the execution order of multiple iterations, for the t-th iteration, determine the total depth confidence index corresponding to the multiple candidate depth values ​​under the (t-1)-th iteration; based on the total depth confidence index corresponding to the multiple candidate depth values ​​under the (t-1)-th iteration, and the pixel sharpness corresponding to the corresponding pixel under the multiple candidate depth values, determine the depth probability value corresponding to the corresponding pixel under the multiple candidate depth values ​​under the t-th iteration, where the corresponding total depth confidence index represents the total confidence level that the pixel sharpness of the multiple pixels under the corresponding candidate depth value is a predetermined sharpness, and t is an integer greater than or equal to 1;

[0099] S108, based on the depth probability values ​​of the corresponding pixel under multiple candidate depth values ​​in the t-th iteration, determine the iterative depth value of the corresponding pixel until the target condition is reached, and obtain the target depth value of the corresponding pixel. The target condition includes the completion of multiple iterations.

[0100] Specifically, this includes: initializing the total depth confidence index for each of the multiple candidate depth values ​​in the 0th iteration (i.e., before the start of the iteration) to... The formula is:

[0101]

[0102] in, This is the initial cardinality (specifically, it can be the initial pheromone cardinality). .

[0103] And / or, will It is simply initialized to the constant 1.

[0104] The total depth confidence index can be represented by a pheromone matrix.

[0105] The process of determining the total depth confidence index corresponding to multiple candidate depth values ​​includes:

[0106] After determining the target depth value for each pixel under the corresponding candidate depth value, update the depth confidence index corresponding to the candidate depth value of each pixel in the current iteration, using the following formula:

[0107]

[0108] in, It is the local volatility coefficient, which controls the decay rate of the depth confidence index; Pixel Index The next pixel is in the first The candidate depth value in the next iteration The corresponding depth confidence index.

[0109] After determining the target depth value for each candidate depth value for all pixels to complete the depth selection, the total depth confidence index is updated.

[0110]

[0111] in, For all pixels at the 1st The candidate depth value in the next iteration The corresponding total depth confidence index; The global volatility coefficient controls the decay rate of the total depth confidence index.

[0112] For each pixel index The pixel is selected only for the target depth value. Additional depth confidence index:

[0113]

[0114] in, Pixel Index The next pixel is in the first Target depth value in the next iteration An additional deep confidence index is added; The gain coefficient controls the strength of the depth confidence index supplementation in a single iteration; For clarity; This is a smoothing term used to quantize the pixel index. The pixel below the target depth value The smoothness of the neighborhood.

[0115]

[0116] in, It is a set of neighborhood offsets. It is an indicator function; it is 1 when the condition is true and 0 when the condition is false. Neighboring pixels relative to pixel index The offset of the pixel; These are the smoothing weighting coefficients.

[0117] The final candidate depth value in the t-th iteration is Total Depth Confidence Index at Time for:

[0118]

[0119] Similarly, the total depth confidence index corresponding to multiple candidate depth values ​​in the (t-1)th iteration is also described, and will not be repeated here.

[0120] Based on the total depth confidence index corresponding to multiple candidate depth values ​​in the (t-1)th iteration, and the pixel sharpness corresponding to the corresponding pixel under multiple candidate depth values, the depth probability value corresponding to the corresponding pixel under multiple candidate depth values ​​in the tth iteration is determined, using the following formula:

[0121]

[0122] in, The probability of choosing the depth value z; The weighting of historical pixel sharpness influence; This indicates the weight of the current pixel sharpness. Indicates a depth index. This represents the total number of depth values.

[0123] Until the target condition is met, the target depth value of the corresponding pixel is obtained. .

[0124] The target conditions include:

[0125] Until the maximum number of iterations is reached, that is, until multiple iterations are completed;

[0126] Depth change rate Less than the threshold :

[0127]

[0128] Based on the above, the depth selection is transformed into a global optimization problem. Through iterative updates, the pixel depth selection is constrained by neighborhood and historical information, and will not be easily skewed by single-frame noise, thus smoothing out local noise interference.

[0129] S110, combine the target depth values ​​corresponding to multiple pixels to obtain a depth map;

[0130] S112, based on the depth mapping map, performs three-dimensional morphological reconstruction of the electrode material surface.

[0131] Specifically, the target depth values ​​corresponding to multiple pixels are combined to obtain a reconstructed depth map of the electrode material surface. For this depth map Median filtering is performed to eliminate isolated noise points, resulting in the final optimized depth map. .

[0132] Through the above steps S102-S112, by acquiring the image sequence of the electrode material and determining the pixel clarity of each pixel at multiple candidate depth values, and following the execution order of multiple iterations, each iteration combines the total depth confidence index of the previous iteration with the pixel clarity of each pixel at each candidate depth. This allows for a comprehensive consideration of historical depth confidence and current pixel clarity features, accurately quantifying the probability that each candidate depth becomes the optimal depth, and thus gradually and precisely determining the target depth value of each pixel. This achieves high-precision reconstruction of the three-dimensional morphology of the electrode material surface, thereby solving the technical problem in related technologies where the determination of the electrode material surface depth value is inaccurate, resulting in low accuracy of the three-dimensional morphology reconstruction of the electrode material surface.

[0133] Based on the above embodiments and optional embodiments, an optional implementation method is provided, which is described in detail below.

[0134] In related technologies, the surface microstructure of electrode materials, such as porosity, roughness, and particle distribution, has a decisive influence on their electrochemical performance (e.g., capacity, rate performance, and cycle life). Accurately and rapidly acquiring their three-dimensional morphology is crucial for material development and performance optimization. However, in related technologies, there is a technical problem in reconstructing the three-dimensional morphology of electrode material surfaces: inaccurate determination of the surface depth value leads to low accuracy in the reconstructed three-dimensional morphology.

[0135] There is currently no effective solution to the above problems.

[0136] For example, the traditional Shape From Focus (SFF) method quantifies image sharpness using a focus measure (FM) function and then simply selects the depth corresponding to the peak of the FM curve as the height value. However, when the electrode material surface is usually low in contrast and sparse in texture, the FM curve becomes too flat or has multiple local extrema, making peak search difficult and height estimation error large, resulting in inaccurate peak localization. Furthermore, imaging noise can seriously interfere with the calculation of FM, especially in areas with low signal-to-noise ratio, which may lead to erroneous peaks. In addition, traditional methods process each pixel independently, ignoring the physical constraints of the continuity and smoothness of the surface itself, resulting in a large number of isolated noise points and discontinuous jumps in the reconstructed surface. Therefore, when reconstructing the three-dimensional morphology of the electrode material surface, there is a technical problem of inaccurate determination of the depth value of the electrode material surface, resulting in low accuracy of the three-dimensional morphology reconstruction of the electrode material surface.

[0137] In view of this, an optional embodiment of the present invention provides a method for reconstructing the three-dimensional morphology of an electrode material surface, which can effectively solve the above-mentioned technical problems.

[0138] Figure 2 This is a schematic diagram of the three-dimensional morphology reconstruction process of the electrode material surface in an optional embodiment of the present invention, as shown below. Figure 2 As shown, using a lithium battery as an example, a detailed description follows.

[0139] S1, Multifocus microscopic image sequence acquisition:

[0140] Figure 3 This is a schematic diagram of the structure of a multi-focus image sequence acquisition system in an optional embodiment of the present invention, such as... Figure 3 As shown, a multi-focus image sequence acquisition system (i.e., an image acquisition system) is constructed, consisting of an optical microscope, a displacement stage (such as a precision Z-axis displacement stage), and a computer control system. It also includes:

[0141] Mechanical base: The rigid support for the entire system, ensuring the stability of the equipment.

[0142] Motion guide rail: A vertical linear guide rail that provides a high-precision Z-axis motion path for the imaging components.

[0143] Stepper motor: Drives the imaging component to move up and down along the motion guide rail to achieve image acquisition at different focus depths. Stepper control can accurately position each candidate depth.

[0144] Tilt adjustment device: Installed below the stage, it can finely adjust the tilt angle of the sample to ensure that the sample surface is perpendicular to the imaging optical axis.

[0145] Stage: A platform for placing the sample to be tested, with attitude correction achieved through a tilt adjustment device.

[0146] Camera: The core device for image acquisition, used to capture sample images at different depths of focus, specifically including charge-coupled device (CCD) / complementary metal-oxide-semiconductor (CMOS) cameras.

[0147] Lens assembly: Optical magnification and focusing components, working in conjunction with the Z-axis movement of a stepper motor to achieve clear imaging at different depths.

[0148] Magnifying lens: Further improves imaging resolution and is used to observe the fine structure of samples.

[0149] Coaxial light source: Illuminates the sample along the optical axis, which is suitable for highlighting surface smoothness and edge details and reducing shadow interference.

[0150] Ring light source: Arranged around the magnifying lens, it provides uniform ring illumination, enhances the surface texture and three-dimensionality of the sample, and is suitable for imaging needs of samples of different materials.

[0151] The electrode material sample is placed on a displacement stage, and the stage is moved at equal intervals along the optical axis under computer control. A microscopic image is acquired at each step position, thus obtaining a series of image sequences covering the entire surface undulation of the sample. ,in, Indicates pixel index, Indicates the first Zhang electrode material surface image, Image sequence number The total number of images is [value], and the size of a single image in the image sequence is [value]. , Indicates the length of the image. This represents the width of the image. A higher number of images results in a more accurate reconstructed depth map, but it also decreases the algorithm's efficiency. Therefore, to achieve a balance between reconstruction accuracy and efficiency, the number of images is generally between 60 and 150. The image sequence is characterized by a similar field of view and angle, but with continuously varying focus. Figure 4 These are sample images of a multi-focus image sequence of a lithium battery electrode surface in an optional embodiment of the present invention, such as... Figure 4 As shown.

[0152] S2, Construct the comprehensive focused evaluation function (i.e., the objective function):

[0153] A focus evaluation function based on color image information entropy and a comprehensive focus evaluation function based on grayscale image gradient variance are proposed. Then, adaptive weighting coefficients are calculated based on the local statistical characteristics of the image to construct a weighted fusion comprehensive focus evaluation function. Grayscale images are advantageous for evaluating pixel sharpness based on the richness and sharpness of texture details, while color images can evaluate pixel sharpness based on the characteristic of color softening when out of focus. Combining the two yields more accurate focus evaluation results with higher robustness, specifically including:

[0154] S21, the acquired image sequence Arranged parallel to each other and at equal intervals along the Z-axis, a three-dimensional data cube covering the height range of the sample is formed. Any pixel in this cube can be represented as... ,in, Represents pixel coordinates, Indicates pixel index, Indicates the image sequence number. , , , Image width, Image height, This represents the total number of images.

[0155] S22, with the first image in the image sequence Pixels on an image Draw a circle of size centered on . The rectangular window is the target region. The window size of this rectangle can only be an integer, and is generally an odd number. The set of pixels within the rectangular window is:

[0156]

[0157] in, Represents the first image in the image sequence. The pixel coordinates in the image are The pixels; Indicates the length and / or width of the target area.

[0158] For each color channel of the same image in the image sequence Statistical target area The color components on each color channel are equal to The number of pixels is used to obtain the histogram. , .

[0159] For each color channel And for each pixel level (i.e., color component), determine the color probability. :

[0160]

[0161] in, Indicates color channel Corresponding color components The probability of the first point appearing (i.e., the color probability). .

[0162] For each color channel Calculate its information entropy within that window:

[0163]

[0164] in, Color channels within a rectangular window The corresponding color information entropy, when hour, Defined as 0.

[0165] Calculate the local color information entropy of the image to construct a focusing evaluation function based on color information entropy (i.e., the first sub-function):

[0166]

[0167] in, The color information entropy corresponding to the red channel; The color information entropy corresponding to the green channel; This represents the color information entropy corresponding to the blue channel.

[0168] S23, Calculate the gradient matrix of each image in the order of the image sequence. :

[0169]

[0170]

[0171]

[0172] in, and These represent the horizontal and vertical gradients of each image in the image sequence, respectively. Represents grayscale value; and These represent the horizontal and vertical gradient convolution kernels, respectively.

[0173] Gradient matrix Divided according to a third dimension (e.g., the depth dimension of an image sequence) A planar matrix, with pixels on each planar matrix. Centered on, delineate a scale of [missing information]. The window is used as a grayscale area.

[0174] Pixels are calculated within the grayscale window. The local gradient variance constitutes the focusing evaluation function (i.e., the second sub-function) based on the gray-level gradient variance:

[0175]

[0176] in, This represents the variance of the grayscale gradient. It is the gradient average;

[0177] The value can be determined using the following formula:

[0178]

[0179] S24, Construct the objective function, that is, adjust the focus weights according to the color and texture complexity of the scene image to obtain the comprehensive focus evaluation function (i.e., the objective function):

[0180]

[0181] in, Used to quantify the total sharpness of a pixel at the corresponding candidate depth value; This is the first sub-function, used to quantize the sharpness of pixels at the corresponding candidate depth values ​​in a color image scene; This is the second sub-function, used to quantize the sharpness of pixels at the corresponding candidate depth values ​​in grayscale image scenarios; These are adaptive weighting coefficients.

[0182] Figure 5 This is a comparison chart of the effects of the focusing evaluation function in an optional embodiment of the present invention, such as... Figure 5 As shown, the horizontal axis represents the image number (corresponding to different depths of focus), and the vertical axis represents the normalized focus evaluation score (a higher score indicates a sharper image). Wherein:

[0183] The orange asterisk line illustrates the effect of the second-order differential focusing evaluation function (SML) based on the Laplace operator;

[0184] The yellow squares illustrate the effect of the gradient-focused evaluation function (Tenengrad);

[0185] The purple cross line illustrates the effect of the Gray Scale Statistical Evaluation Function (GLV);

[0186] The green triangles illustrate the effect of the frequency domain focusing evaluation function (DCT);

[0187] The light blue circled lines illustrate the effect of the Focused Evaluation Function (SWAV) based on wavelet transform;

[0188] The dark blue diamond lines illustrate the effect of the focused evaluation function (Bre4d_var) based on four-dimensional gradient variance;

[0189] The red solid line illustrates the intended effect of the objective function (Proposed).

[0190] As can be seen, the adaptive coupling of color information entropy and gray-level gradient variance based on the objective function forms a sharp single peak at the focus position (approximately image number 25), with less side lobes and noise interference, a steeper peak, and more accurate sharpness discrimination.

[0191] S3, each pixel The problem of finding the optimal focusing position in a depth sequence is modeled as a multi-stage decision-making process, with the depth direction z as the stage and the comprehensive focusing evaluation function value as the stage. As a stage benefit, the ant colony algorithm is used for global optimization to determine each pixel. The corresponding optimal focusing depth (i.e., the target depth value) is obtained by turning depth selection into a global optimization problem. Through pheromone iterative updates, the pixel depth selection is constrained by neighborhood and historical information, and will not be easily skewed by single-frame noise, thus smoothing out local noise interference.

[0192] S31, regarding depth distribution range The pixel sharpness of each pixel (i.e., the corresponding pixel) under multiple candidate depth values. (That is, focusing on the evaluation value), that is, indexing each pixel. For each index z, calculate the focus evaluation value to obtain the focus evaluation matrix, which is represented as:

[0193]

[0194] in, Indicates pixel index, Indicates the depth value to be selected. .

[0195] S32, after determining the pixel sharpness corresponding to the corresponding pixel point under multiple candidate depth values, normalization processing is also performed, that is, for each pixel index... Normalize its depth-direction focus evaluation curve, and then calculate the normalized heuristic information:

[0196]

[0197] in, is a constant used to prevent division by zero and avoid zero values; and Represents pixel index The extreme values ​​of pixel sharpness at multiple candidate depth values ​​(i.e., the extreme values ​​of focus evaluation values ​​at all depths).

[0198] in, The formula for determining it is:

[0199]

[0200] The formula for determining it is:

[0201]

[0202] in, This represents the total number of depth values.

[0203] S33, Initialize the pheromone matrix (That is, initialize the total depth confidence index corresponding to multiple candidate depth values ​​respectively):

[0204]

[0205] in, This is the initial cardinality (specifically, it can be the initial pheromone cardinality). .

[0206] And / or, will It is simply initialized to the constant 1.

[0207] S34, for iteration number t=1 to the maximum iteration number For each pixel Choose depth z based on probability:

[0208]

[0209] in, The probability of choosing the depth value z; The influence weight of historical pixel sharpness (i.e., pheromone factor). This represents the influence weight of the current pixel sharpness (i.e., the heuristic factor). Indicates a depth index. This represents the total number of depth values.

[0210] Record the results of the selection .

[0211] S35: After the target depth value under the corresponding candidate depth value is determined for each pixel, local pheromone update is performed immediately:

[0212]

[0213] in, It is the local volatility coefficient, which controls the decay rate of the depth confidence index (i.e., pheromone concentration). Pixel Index The next pixel is in the first The candidate depth value in the next iteration The corresponding depth confidence index (i.e., pheromone concentration).

[0214] After determining the target depth value for each candidate depth value for all pixels to complete the depth selection, the total depth confidence index is updated (i.e., the global pheromone is updated).

[0215]

[0216] in, For all pixels at the 1st The candidate depth value in the next iteration The corresponding total depth confidence index (i.e., global pheromone concentration); The global volatility coefficient controls the decay rate of the total depth confidence index.

[0217] For each pixel index The pixel is selected only for the target depth value. The additional depth confidence index (i.e., the additional pheromone concentration):

[0218]

[0219] in, Pixel Index The next pixel is in the first Target depth value in the next iteration The additional deep confidence index (i.e., the additional pheromone concentration) is added. The gain coefficient (i.e., the pheromone constant) controls the strength of the depth confidence index supplementation in a single iteration; For clarity; This is a smoothing term used to quantize the pixel index. The pixel below the target depth value The smoothness of the neighborhood.

[0220]

[0221] in, It is the neighborhood offset set (i.e., the neighborhood window scale). It is an indicator function; it is 1 when the condition is true and 0 when the condition is false. Neighboring pixels relative to pixel index The offset of the pixel; These are the smoothing weighting coefficients.

[0222] The final global pheromone update is as follows:

[0223]

[0224] in, It is the total depth confidence index (i.e., global pheromone) when the candidate depth value is z in the updated t-th iteration.

[0225] S36, Repeat steps S32 to S35 until the maximum number of iterations or depth change rate is reached. Less than the threshold :

[0226]

[0227] Terminate the iteration.

[0228] Figure 6 This is a comparative schematic diagram of the surface reconstruction effect of the electrode material in an optional embodiment of the present invention, such as... Figure 6 As shown, (a) is a schematic diagram of the traditional 3D reconstruction effect based on focus morphology restoration, and (b) is a schematic diagram of the reconstruction effect using the above-mentioned electrode material surface 3D morphology reconstruction method. In the traditional 3D reconstruction method based on focus morphology restoration, the focus value of each pixel is usually calculated sequentially within a parallel window according to the image sequence, and the peak position of the focus curve is found by interpolation fitting to determine the depth. In contrast, the above-mentioned iterative optimization replaces the focus position calculation: within a set depth range, each pixel is randomly assigned a depth value. Based on the law of large numbers and image continuity, some pixels will always be assigned the correct depth value. Subsequent processing uses an ant colony algorithm iterative method to provide the depth information corresponding to these correct pixels to the remaining pixels until all pixels obtain a depth value that matches their own.

[0229] S4, Surface 3D morphology restoration:

[0230] By combining the target depth values ​​corresponding to multiple pixels, a reconstructed depth map of the electrode material surface is obtained. For this depth map Median filtering is performed to eliminate isolated noise points, resulting in a reconstructed depth map. Reconstructing the depth map The micro-morphology shown is the micro-3D morphology of the electrode material surface.

[0231] The above optional implementation methods can achieve at least the following beneficial effects:

[0232] (1) Compared with related technologies, the present invention obtains the image sequence of the electrode material and determines the pixel clarity of each pixel under multiple candidate depth values. According to the execution order of multiple iterations, each iteration combines the total depth confidence index of the previous iteration and the pixel clarity of each pixel under each candidate depth. It can comprehensively consider the historical depth confidence level and the current pixel clarity features, accurately quantify the probability that each candidate depth becomes the optimal depth, and then gradually and accurately determine the target depth value of each pixel, so as to realize the high-precision reconstruction of the three-dimensional morphology of the electrode material surface. This solves the technical problem in related technologies that the depth value of the electrode material surface is not accurately determined when reconstructing the three-dimensional morphology of the electrode material surface, resulting in low accuracy of the reconstruction of the three-dimensional morphology of the electrode material surface.

[0233] (2) Compared with related technologies, this invention first determines the color information entropy and grayscale gradient variance of the pixel at each candidate depth value, and quantifies the complexity of color distribution and the degree of grayscale variation fluctuation, respectively. It can effectively make up for the deficiency of grayscale gradient in representing texture in complex scenes such as low texture, few texture or single color, and realize robust feature representation by utilizing the richness of color information entropy in color distribution. At the same time, it combines grayscale gradient variance to capture the fluctuation of grayscale variation, realizes complementary coupling of color and grayscale dimensions, and can comprehensively represent the imaging clarity features of the pixel at different candidate depths from multiple dimensions, ensuring that the quantification of pixel clarity is more accurate and comprehensive, and providing a reliable feature basis for subsequent determination of the optimal depth value of the pixel based on clarity.

[0234] (3) Compared with related technologies, the present invention can avoid noise interference of single pixel color information by dividing the target area into W×W with the pixel as the center, and obtain more statistically significant local color features; by extracting the color component distribution of each color channel in the target area and calculating the corresponding probability, the local color distribution can be transformed into a quantifiable probability statistical form; by calculating the color information entropy of each channel based on probability and summing them, the imaging clarity of the pixel at the corresponding candidate depth can be accurately quantified from the perspective of color distribution complexity.

[0235] (4) Compared with related technologies, the present invention can obtain the local gray-scale feature range around the corresponding pixel by dividing the gray-scale region from the target gray-scale image corresponding to the pixel, avoiding the random interference of single pixel gray-scale value and ensuring the effectiveness of gray-scale feature statistics; the gray-scale gradient variance is calculated based on the gray-scale value of the second region point in the gray-scale region, which can accurately quantify the degree of gray-scale change of the pixel at the corresponding candidate depth, intuitively reflect the clear features of gray-scale texture, and provide a reliable gray-scale dimension quantification basis for multi-dimensional evaluation of pixel clarity.

[0236] (5) Compared with related technologies, this invention constructs an objective function with the goal of maximizing the total sharpness of the pixel at the selected depth. This can accurately quantify the coupling relationship between color information entropy and grayscale gradient variance, and realize adaptive coupling and adaptation of the sharpness features of color and grayscale dimensions. This helps to accurately integrate the sharpness features of different dimensions and maximize the restoration of the true sharpness of the pixel, so that the quantification result of pixel sharpness is more in line with the actual imaging state.

[0237] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that the present invention is not limited to the described order of actions, because according to the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to the present invention.

[0238] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0239] Example 2

[0240] According to embodiments of the present invention, an apparatus for implementing the above-described method for reconstructing the three-dimensional morphology of an electrode material surface is also provided. Figure 7 This is a structural block diagram of a three-dimensional morphology reconstruction device for electrode material surface according to an embodiment of the present invention, such as... Figure 7 As shown, the device includes: an acquisition module 702, a first determination module 704, a second determination module 706, a third determination module 708, a fourth determination module 710, and a fifth determination module 712. The device will be described in detail below.

[0241] The acquisition module 702 is used to acquire image sequences of the electrode material;

[0242] The first determining module 704 is connected to the above-mentioned obtaining module 702 and is used to determine the pixel sharpness of the corresponding pixel under multiple candidate depth values ​​for multiple pixel points in the image sequence.

[0243] The second determining module 706, connected to the first determining module 704, is used to determine the total depth confidence index corresponding to multiple candidate depth values ​​in the (t-1)th iteration for the t-th iteration according to the execution order of multiple iterations; based on the total depth confidence index corresponding to multiple candidate depth values ​​in the (t-1)th iteration, and the pixel sharpness corresponding to the corresponding pixel in the multiple candidate depth values, determine the depth probability value corresponding to the corresponding pixel in the t-th iteration for the multiple candidate depth values, where the corresponding total depth confidence index represents the total confidence level that the pixel sharpness of multiple pixels in the corresponding candidate depth value is a predetermined sharpness, and t is an integer greater than or equal to 1;

[0244] The third determining module 708, connected to the second determining module 706, is used to determine the iterative depth value of the corresponding pixel based on the depth probability values ​​of the corresponding pixel under multiple candidate depth values ​​in the t-th iteration, until the target condition is reached, and obtain the target depth value of the corresponding pixel. The target condition includes the completion of multiple iterations.

[0245] The fourth determining module 710, connected to the third determining module 708, is used to combine the target depth values ​​corresponding to multiple pixels to obtain a depth mapping map.

[0246] The fifth determining module 712, connected to the fourth determining module 710, is used to reconstruct the three-dimensional morphology of the electrode material surface based on the depth mapping map.

[0247] It should be noted that the above-mentioned acquisition module 702, first determination module 704, second determination module 706, third determination module 708, fourth determination module 710, and fifth determination module 712 correspond to steps S102 to S112 in the method for reconstructing the three-dimensional morphology of the electrode material surface. The multiple modules and the corresponding steps are the same in terms of the instances and application scenarios implemented, but are not limited to the content disclosed in the above embodiment 1.

[0248] Example 3

[0249] According to another aspect of the present invention, an electronic device is also provided, comprising: a processor; and a memory for storing processor-executable instructions, wherein the processor is configured to execute instructions to implement the electrode material surface three-dimensional morphology reconstruction method of any of the above embodiments.

[0250] Example 4

[0251] According to another aspect of the present invention, a computer-readable storage medium is also provided, which, when the instructions in the computer-readable storage medium are executed by a processor of an electronic device, enables the electronic device to perform the three-dimensional morphology reconstruction method for the electrode material surface described above.

[0252] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0253] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0254] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0255] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0256] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0257] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0258] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for reconstructing the three-dimensional morphology of an electrode material surface, characterized in that, include: Acquire image sequences of electrode materials; For multiple pixels in the image sequence, determine the pixel sharpness of each pixel under multiple candidate depth values; According to the execution order of multiple iterations, for the t-th iteration in the multiple iterations, the total depth confidence index corresponding to the multiple candidate depth values ​​under the (t-1)-th iteration is determined respectively; Based on the total depth confidence index corresponding to multiple candidate depth values ​​in the (t-1)th iteration, and the pixel sharpness corresponding to the corresponding pixel point under the multiple candidate depth values, the depth probability value corresponding to the corresponding pixel point under the multiple candidate depth values ​​in the tth iteration is determined, wherein the corresponding total depth confidence index represents the total confidence level that the pixel sharpness of the multiple pixels point under the corresponding candidate depth value is a predetermined sharpness, and t is an integer greater than or equal to 1; Based on the depth probability values ​​of the corresponding pixel under the multiple candidate depth values ​​in the t-th iteration, the iterative depth value of the corresponding pixel is determined until the target condition is reached, and the target depth value of the corresponding pixel is obtained. The target condition includes the completion of multiple iterations. The target depth values ​​corresponding to the multiple pixels are combined to obtain a depth mapping map; Based on the depth mapping map, the three-dimensional morphology of the electrode material surface is reconstructed.

2. The method according to claim 1, characterized in that, Determining the pixel sharpness of the corresponding pixel at multiple candidate depth values ​​includes: Determine the color information entropy and gray-level gradient variance of the corresponding pixel at multiple candidate depth values, wherein the color information entropy is used to quantify the complexity of the color distribution of the corresponding pixel at the corresponding candidate depth value, and the gray-level gradient variance is used to quantify the degree of fluctuation of the gray-level change of the corresponding pixel at the corresponding candidate depth value. Based on the color information entropy and grayscale gradient variance of the corresponding pixel at multiple candidate depth values, the pixel sharpness of the corresponding pixel at multiple candidate depth values ​​is determined.

3. The method according to claim 2, characterized in that, Determining the color information entropy corresponding to the corresponding pixel at multiple candidate depth values ​​includes: For any one of the multiple candidate depth values, a target region is delineated from the target color image corresponding to the corresponding pixel, wherein the image sequence includes the target color image; Determine a plurality of first region points within the target region, wherein the plurality of first region points are pixels within the target region; Based on the color parameters corresponding to multiple first region points within the target area, the color information entropy of the corresponding pixel point under any candidate depth value is determined.

4. The method according to claim 3, characterized in that, The step of determining the color information entropy of a corresponding pixel at any candidate depth value based on the color parameters corresponding to multiple first region points within the target region includes: Based on the color parameters corresponding to multiple first area points within the target area, the color component distribution corresponding to multiple color channels within the target area is determined, wherein the corresponding color component distribution is the distribution of multiple first area points under different color components under the corresponding color channel. Based on the color component distribution corresponding to the multiple color channels, the probability of the first region point appearing under the color components corresponding to the multiple color channels in the target region is determined. Based on the probability of the first region point appearing under the color components corresponding to multiple color channels in the target region, the color information entropy of the corresponding pixel point under any candidate depth value is determined.

5. The method according to claim 3, characterized in that, The step of dividing the target region from the target color image corresponding to the corresponding pixel includes: Using the pixel position of the corresponding pixel as the center, the target region is divided from the target color image corresponding to the corresponding pixel.

6. The method according to claim 1, characterized in that, Determine the grayscale gradient variance of the corresponding pixel at multiple candidate depth values, including: For any one of the multiple candidate depth values, a grayscale region is divided from the target grayscale image corresponding to the corresponding pixel, wherein the image sequence includes the target grayscale image; Based on the gray values ​​corresponding to multiple second region points in the gray region, the gray gradient variance of the corresponding pixel point under any candidate depth value is determined.

7. The method according to any one of claims 1 to 6, characterized in that, Determining the pixel sharpness of the corresponding pixel at multiple candidate depth values ​​includes: Construct an objective function, wherein the objective function includes a first sub-function and a second sub-function. The first sub-function is used to quantize the sharpness of a pixel at a corresponding candidate depth value in a color image scene, and the second sub-function is used to quantize the sharpness of a pixel at a corresponding candidate depth value in a grayscale image scene. The objective function is a function that aims to maximize the total sharpness of the pixel at the corresponding candidate depth value. Based on the objective function, the pixel sharpness of each pixel is determined for each of the multiple candidate depth values.

8. A device for reconstructing the three-dimensional morphology of an electrode material surface, characterized in that, include: The acquisition module is used to acquire image sequences of the electrode material; The first determining module is used to determine the pixel sharpness of a given pixel at multiple candidate depth values ​​for a given number of pixels in the image sequence. The second determining module is used to determine the total depth confidence index corresponding to the multiple candidate depth values ​​in the (t-1)th iteration for the tth iteration in the execution order of the multiple iterations. Based on the total depth confidence index corresponding to multiple candidate depth values ​​in the (t-1)th iteration, and the pixel sharpness corresponding to the corresponding pixel point under the multiple candidate depth values, the depth probability value corresponding to the corresponding pixel point under the multiple candidate depth values ​​in the tth iteration is determined, wherein the corresponding total depth confidence index represents the total confidence level that the pixel sharpness of the multiple pixels point under the corresponding candidate depth value is a predetermined sharpness, and t is an integer greater than or equal to 1; The third determining module is used to determine the iterative depth value of the corresponding pixel based on the depth probability values ​​of the corresponding pixel under the multiple candidate depth values ​​in the t-th iteration, until the target condition is reached, and obtain the target depth value of the corresponding pixel, wherein the target condition includes the completion of multiple iterations. The fourth determining module is used to combine the target depth values ​​corresponding to the plurality of pixels respectively to obtain a depth mapping map; The fifth determining module is used to reconstruct the three-dimensional morphology of the electrode material surface based on the depth mapping map.

9. An electronic device, characterized in that, include: processor; Memory used to store the processor's executable instructions; The processor is configured to execute the instructions to implement the method for reconstructing the three-dimensional morphology of the electrode material surface as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, When the instructions in the computer-readable storage medium are executed by the processor of the electronic device, the electronic device is able to perform the method for reconstructing the three-dimensional morphology of the electrode material surface as described in any one of claims 1 to 7.

Images