Image processing method, and method and system for detecting alignment degree of wound battery cell

Abstract

The invention discloses an image processing method and a wound cell alignment degree detection method and system, and relates to the technical field of cell detection, and the image processing method comprises the steps: obtaining an initial wound cell image; performing noise reduction on the initial winding battery cell image to obtain a first winding battery cell image; logarithmic transformation is carried out on the first winding battery cell image to adjust the brightness value of the first winding battery cell image, and a second winding battery cell image is obtained; performing edge enhancement on the second winding cell image to obtain a third winding cell image; performing expansion and Gaussian filtering on the third winding cell image to obtain a fourth winding cell image; and performing contrast adjustment on the fourth winding cell image to obtain a final winding cell image. According to the invention, under the condition that the number of pole piece layers contained in the wound battery cell is large, the collected original wound battery cell image is converted into a clearer wound battery cell image through a series of image processing means, so that the detection accuracy of the pole piece alignment degree of the wound battery cell is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of battery cell detection, and in particular to an image processing method, a method and a system for detecting the alignment degree of wound battery cells. Background Art

[0002] During the production process of battery cells, it is necessary to detect the alignment degree of the battery cell electrode sheets. Currently, X-ray detection equipment is usually used to perform non-destructive detection on battery cells. However, when the number of layers of the battery cell electrode sheets is large, the battery cell images collected by the X-ray detection equipment have defects such as being blurred, the electrode sheet features being unclear, and background interference between the electrode sheets, resulting in poor effects when directly predicting the electrode sheet features of the battery cell image by a deep learning model later. The endpoints of some electrode sheets cannot be predicted, and the alignment degree of the battery cell electrode sheets cannot be accurately detected. Summary of the Invention

[0003] The present invention provides an image processing method, a method and a system for detecting the alignment degree of wound battery cells to solve one or more technical problems existing in the prior art, and at least provide a beneficial choice or create conditions.

[0004] In a first aspect, an image processing method is provided, and the method includes:

[0005] Obtain an initial image of a wound battery cell;

[0006] Perform noise reduction on the initial image of the wound battery cell to obtain a first image of the wound battery cell;

[0007] Perform logarithmic transformation on the first image of the wound battery cell to adjust its brightness value to obtain a second image of the wound battery cell;

[0008] Perform edge enhancement on the second image of the wound battery cell to obtain a third image of the wound battery cell;

[0009] Perform dilation and Gaussian filtering on the third image of the wound battery cell to obtain a fourth image of the wound battery cell;

[0010] Perform contrast adjustment on the fourth image of the wound battery cell to obtain a final image of the wound battery cell.

[0011] Further, the obtaining of the initial image of the wound battery cell includes:

[0012] Obtain a plurality of images of wound battery cells collected by an X-ray detection device, and then determine the average gray value of all pixel points in each image of the wound battery cell;

[0013] Select, from the plurality of images of the wound battery cells, the image of the wound battery cell whose average gray value exceeds a preset gray threshold as the initial image of the wound battery cell.

[0014] Further, the noise reduction of the initial wound cell image to obtain the first wound cell image includes:

[0015] Converting the initial wound cell image into a grayscale image;

[0016] Performing convolution operation on the grayscale image using a mean filter convolution kernel of size 3×3 to obtain the first wound cell image after noise reduction.

[0017] Further, the edge enhancement of the second wound cell image to obtain the third wound cell image includes:

[0018] Performing convolution operation on the second wound cell image in the y-axis direction using a preset enhancement matrix of size 3×3 to obtain the third wound cell image after edge enhancement.

[0019] Further, the preset enhancement matrix is: kernel = (-9, -5, -9; 0, 1, 0; 9, 15, 9).

[0020] Further, the dilation and Gaussian filtering of the third wound cell image to obtain the fourth wound cell image includes:

[0021] Performing dilation operation on the third wound cell image in the vertical direction to obtain the fourth initial wound cell image;

[0022] Performing Gaussian filtering operation on the fourth initial wound cell image in the horizontal direction to obtain a preliminary filtered image;

[0023] Converting the preliminary filtered image into a 32-bit single-channel image and performing Gaussian filtering operation on it to obtain a secondary filtered image;

[0024] Dividing the preliminary filtered image by the secondary filtered image to obtain a ratio image;

[0025] Converting the ratio image into an 8-bit single-channel image and using it as the fourth wound cell image.

[0026] Further, the contrast adjustment of the fourth wound cell image to obtain the final wound cell image includes:

[0027] Performing power-law transformation on the grayscale value of each pixel point in the fourth wound cell image to obtain the final wound cell image with increased overall contrast.

[0028] In a second aspect, a method for detecting the alignment degree of wound cells is provided, and the method includes:

[0029] Collecting images of wound cells through an x-ray detection device to obtain a plurality of wound cell images;

[0030] Process the multiple wound battery cell images using the image processing method described in the first aspect to obtain the final wound battery cell image;

[0031] Judge the qualification of the alignment degree of each pole piece feature included in the final wound battery cell image.

[0032] In a third aspect, an image processing system includes:

[0033] A first module for acquiring an initial wound battery cell image;

[0034] A second module for denoising the initial wound battery cell image to obtain a first wound battery cell image;

[0035] A third module for performing logarithmic transformation on the first wound battery cell image to adjust its brightness value to obtain a second wound battery cell image;

[0036] A fourth module for performing edge enhancement on the second wound battery cell image to obtain a third wound battery cell image;

[0037] A fifth module for performing dilation and Gaussian filtering on the third wound battery cell image to obtain a fourth wound battery cell image;

[0038] A sixth module for adjusting the contrast of the fourth wound battery cell image to obtain the final wound battery cell image.

[0039] In a fourth aspect, a system for detecting the alignment degree of wound battery cells is provided, and the system includes:

[0040] An acquisition module for acquiring multiple wound battery cell images by an x-ray detection device;

[0041] A processing module for processing the multiple wound battery cell images using the image processing method described in the first aspect to obtain the final wound battery cell image;

[0042] A judgment module for judging the qualification of the alignment degree of each pole piece feature included in the final wound battery cell image.

[0043] The present invention has at least the following beneficial effects: the image is deblurred by convolution operation based on a mean filter convolution kernel, then the brightness of the image is adjusted by logarithmic transformation operation, and then convolution operation in the y-axis direction is performed on the image based on a preset enhancement matrix, making the tab feature contained in the image more obvious and reducing the noise existing inside during the image brightness adjustment to avoid affecting the image processing effect; subsequently, dilation operation in the vertical direction, Gaussian filtering operation in the horizontal direction, and Gaussian filtering operation after converting to a single-channel image are further performed on the image, making the connection relationship of the tab features contained in the image more clearly presented and making the noise existing inside the image smoother to reduce interference; when the number of tab layers contained in the wound battery cell is large, a clearer image of the wound battery cell can be obtained through the above image processing method, which helps to improve the detection accuracy of the tab alignment degree of the wound battery cell. Description of the Drawings

[0044] The drawings are used to provide a further understanding of the technical solutions of the present invention, and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the technical solutions of the present invention and do not constitute a limitation to the technical solutions of the present invention.

[0045] Figure 1 is a schematic flowchart of an image processing method in an embodiment of the present invention;

[0046] Figure 2 is an initial image of a wound battery cell in an embodiment of the present invention;

[0047] Figure 3 is a grayscale image after region cropping in an embodiment of the present invention;

[0048] Figure 4 is a first image of a wound battery cell after noise reduction in an embodiment of the present invention;

[0049] Figure 5 is a second image of a wound battery cell after increasing the overall brightness in an embodiment of the present invention;

[0050] Figure 6 is a third image of a wound battery cell after edge enhancement in an embodiment of the present invention;

[0051] Figure 7 is a fourth initial image of a wound battery cell after dilation in an embodiment of the present invention;

[0052] Figure 8 is a preliminary filtered image in an embodiment of the present invention;

[0053] Figure 9 is a fourth image of a wound battery cell in an embodiment of the present invention;

[0054] Figure 10 It is the final wound battery cell image after increasing the overall contrast in the embodiment of the present invention;

[0055] Figure 11 It is the final wound battery cell image after area splicing restoration in the embodiment of the present invention;

[0056] Figure 12 It is a schematic flowchart of a method for detecting the alignment degree of wound battery cells in the embodiment of the present invention;

[0057] Figure 13 It is a schematic diagram of the composition of an image processing system in the embodiment of the present invention;

[0058] Figure 14 It is a schematic diagram of the composition of a system for detecting the alignment degree of wound battery cells in the embodiment of the present invention;

[0059] Figure 15 It is a schematic diagram of the hardware structure of a computer device in the embodiment of the present invention. Detailed implementation manners

[0060] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0061] It should be noted that although functional module division is performed in the system schematic diagram and the logical sequence is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order from the module division in the system or the sequence in the flowchart. Terms such as "first", "second", "third", "fourth", etc. in the specification of this application and the above-mentioned accompanying drawings are used to distinguish similar objects, and do not necessarily need to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of this application described here can be implemented in an order other than those illustrated or described here. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not necessarily need to be limited to those steps or units clearly listed, but may include other steps or units inherent to these processes, methods, products or devices that are not clearly listed.

[0062] First, some terms involved in the present invention are explained as follows:

[0063] Image mean filtering is the simplest linear filtering algorithm. By giving a template to the target pixel on the original image, the template includes its surrounding adjacent pixels (an 8-pixel area centered around the target pixel forms a filtering template, excluding the target pixel itself), and then replacing the original pixel value with the average value of all pixels in the template. In other words, each pixel value of the output image of mean filtering is the weighted average of the surrounding M×M pixel values.

[0064] Image dilation is a commonly used method in image morphology operations. It can expand the boundaries of objects, making the objects larger. It can also be used to filter out small holes, smooth image boundaries, and connect objects in the image. The basic principle is to perform a local maximum operation on the image, which is generally achieved by sliding a filter (also known as a structuring element). Each pixel of the filter is compared with the corresponding pixel of the image, and then the central pixel value of the filter is set to the maximum pixel value in the filter.

[0065] Image Gaussian filtering is a linear smoothing filtering algorithm, suitable for removing Gaussian noise in images. It can be understood as a process of weighted averaging the entire image. By scanning each pixel point in the image with a template (also known as a convolution kernel, mask), the gray value obtained after weighted averaging of each pixel point and other pixel points in the neighborhood is determined using the template and replaces the gray value of the central pixel point of the template.

[0066] In the prior art, an existing x-ray detection device is usually directly used for non-destructive testing of the battery cell. When the number of layers of the battery cell electrode is small, the x-ray detection device has good detection effects. However, when the number of layers of the battery cell electrode is large, due to defects such as blurred battery cell images collected by the x-ray detection device, unclear electrode features, and background interference between electrodes, the detection and analysis of the battery cell image directly are significantly ineffective. Therefore, the present invention proposes to perform additional clarification processing on the battery cell image collected by the x-ray detection device and then hand it over to the x-ray detection device for continued detection and analysis to improve the final detection accuracy. The specific implementation is as described below.

[0067] Please refer to Figure 1 , Figure 1 which is a schematic flowchart of an image processing method provided by an embodiment of the present invention. The image processing method includes the following:

[0068] S110. Obtain an initial wound battery cell image;

[0069] S120. Denoise the initial wound battery cell image to obtain a first wound battery cell image;

[0070] S130. Perform a logarithmic transformation on the first wound battery cell image to adjust its brightness value, obtaining a second wound battery cell image;

[0071] S140. Perform edge enhancement on the second wound battery cell image, obtaining a third wound battery cell image;

[0072] S150. Perform dilation and Gaussian filtering on the third wound battery cell image, obtaining a fourth wound battery cell image;

[0073] S160. Perform contrast adjustment on the fourth wound battery cell image, obtaining the final wound battery cell image.

[0074] In some embodiments, the specific implementation process of the above step S110 includes the following steps:

[0075] S111. Obtain a plurality of wound battery cell images, which are obtained by technicians in advance through on-site acquisition of wound battery cells using existing x-ray detection equipment, and each of the wound battery cell images is a 16-bit image;

[0076] It should be noted that the x-ray detection equipment is a high-precision detection equipment integrating advanced computer vision technology and deep learning algorithms. When it is necessary to detect the internal structure of a certain material, only the material to be detected needs to be placed in the set detection area, and the internal structure of the material to be detected is scanned with high precision by adjusting the position and parameters of the x-ray source, and then the scanned image is analyzed through deep learning algorithms such as convolutional neural networks;

[0077] S112. For each wound battery cell image in the plurality of wound battery cell images, obtain the gray value of each pixel point in the wound battery cell image, and then calculate the average gray value of all pixel points in the wound battery cell image;

[0078] S113. Compare and judge each of the multiple average gray values associated with the plurality of wound battery cell images according to a preset gray threshold, so as to screen out the wound battery cell images whose average gray values exceed the preset gray threshold and use them as initial wound battery cell images, as shown in Figure 2 shown.

[0079] It should be noted that after performing the above step S110, the initial wound battery cell image is preferably preliminarily judged by a manual detection method. If the separator lines of the poles in the initial wound battery cell image show a horizontal parallel distribution (such as Figure 2When the above-mentioned situation occurs (as shown in the figure), directly execute the following step S120; if the dividing lines of the pole pieces in the initial wound battery cell image are longitudinally parallel, rotate the entire initial wound battery cell image by 90 degrees or 270 degrees so that the rotated initial wound battery cell image is horizontally parallel as a whole (as shown in Figure 2 the figure) to facilitate subsequent image processing, and then continue to execute the following step S120.

[0080] In some embodiments, the specific implementation process of the above step S120 includes the following steps:

[0081] S121. Convert the initial wound battery cell image saved in RGB format into a grayscale image;

[0082] S122. Obtain a mean filter convolution kernel with a size of 3×3, specifically as follows:

[0083]

[0084] S123. Perform convolution operation on the grayscale image through the mean filter convolution kernel to obtain a first denoised wound battery cell image.

[0085] In some embodiments, the specific implementation of the above step S130 is: perform a logarithmic operation on the grayscale value of each pixel point in the first wound battery cell image to obtain a second wound battery cell image with increased overall brightness. The mathematical expression used in this implementation process is: y i = log(x + β i ), where x is a given adjustment parameter, generally taking a value of 1, β i is the grayscale value of the i-th pixel point in the first wound battery cell image, and y i is the grayscale value adjustment result corresponding to the grayscale value β i , also known as the grayscale value of the i-th pixel point in the second wound battery cell image.

[0086] In some embodiments, the specific implementation process of the above step S140 includes the following steps:

[0087] S141. Obtain a preset enhancement matrix with a size of 3×3, specifically as follows:

[0088]

[0089] S142. Perform convolution operation on the second wound battery cell image in the y-axis direction through the preset enhancement matrix to obtain a third wound battery cell image with enhanced edges.

[0090] Since the first wound cell image is a 16-bit image with a relatively low overall brightness value, the details of the dark regions in the first wound cell image can be enhanced by performing the above step S130, and the edge features of the second wound cell image can be enhanced by performing the above step S140, which is beneficial to the accurate extraction and analysis of the features of each pole piece in the final image.

[0091] In some embodiments, the specific implementation process of the above step S150 includes the following steps:

[0092] S151. Perform a dilation operation on the third wound cell image in the vertical direction to obtain a dilated fourth initial wound cell image;

[0093] S152. Perform a Gaussian filtering operation on the fourth initial wound cell image in the horizontal direction to obtain a preliminary filtered image;

[0094] S153. Convert the preliminary filtered image into a 32-bit single-channel image;

[0095] S154. Perform a Gaussian filtering operation on the 32-bit single-channel image to obtain a second filtered image;

[0096] S155. Divide the preliminary filtered image by the second filtered image, that is, perform integer division on the gray values of every two pixel points with a position correspondence relationship in the preliminary filtered image and the second filtered image to obtain a ratio image;

[0097] S156. Convert the ratio image into an 8-bit single-channel image and use it as the fourth wound cell image.

[0098] By performing the above step S151, the gap between the positive and negative pole pieces in the third wound cell image can be increased, thereby alleviating the pole piece adhesion phenomenon that appears in the third wound cell image. Then, by performing the above steps S152 to S156, the boundary between the positive and negative pole pieces in the fourth initial wound cell image can be enhanced, which is beneficial to the accurate extraction and analysis of the features of each pole piece in the final image.

[0099] In some embodiments, the specific implementation manner of the above step S160 is: perform a power-law transformation on the gray value of each pixel point in the fourth wound cell image to obtain a final wound cell image with increased overall contrast.

[0100] In some preferred embodiments of the present invention, the above steps S120 to the above steps S160 can be executed relying on OpenCV (Open Source Computer Vision Library, a cross-platform computer vision library). It can be understood that at least the erode function, threshold function, findcontours function, sort function, boundingRect function, cvtColor function, blur function, filter2D function, dilate function, GaussianBlur function, convertTo function, and cv2.power function set in the OpenCV are used for execution. Here, a more specific description of the specific implementation process of the above steps S120 to the above steps S160 is given, including the following steps:

[0101] (1) Use the cvtColor function to convert the initial wound cell image to a grayscale image.

[0102] (2) Use the erode function to perform erosion operation on the grayscale image to obtain a first grayscale image.

[0103] (3) Use the threshold function to perform binarization operation on the first grayscale image to obtain a second grayscale image.

[0104] (4) Use the findcontours function to find and frame each contour included in the second grayscale image to obtain a third grayscale image.

[0105] (5) Use the sort function to sort each contour included in the third grayscale image in descending order of area, and then only retain the framing of the largest contour included in the third grayscale image.

[0106] (6) Use the boundingRect function to draw a rectangular frame for the largest contour included in the current third grayscale image to obtain a fourth grayscale image.

[0107] (7) Perform segmentation processing on the fourth grayscale image along the drawn rectangular frame to obtain a first image and a second image without changing the pixel position information. The first image is the image of the region of interest defined inside the rectangular frame, as Figure 3 shown. The second image is the non-region-of-interest image. The first image and the second image are spliced to obtain the fourth grayscale image.

[0108] (8) Based on the mean filter convolution kernel, use the blur function to perform convolution operation on the first image to obtain a first wound cell image, as Figure 4As shown, it can be seen that the first wound battery cell image can achieve a certain effect of deblurring and denoising compared to the first image.

[0109] (9) Perform a logarithmic operation on the grayscale value of each pixel point in the first wound battery cell image to obtain a second wound battery cell image, as Figure 5 shown. It can be seen that the second wound battery cell image can achieve the effect of enhancing the details in the dark part compared to the first wound battery cell image.

[0110] (10) Based on the preset enhancement matrix, use the filter2D function to perform a convolution operation on the second wound battery cell image in the y-axis direction to obtain a third wound battery cell image, as Figure 6 shown. It can be seen that the third wound battery cell image can achieve the effect of enhancing the edge features of each pole piece compared to the second wound battery cell image.

[0111] (11) Use the dilate function to perform a dilation operation on the third wound battery cell image in the vertical direction to obtain a fourth initial wound battery cell image, as Figure 7 shown. It can be seen that the fourth initial wound battery cell image can achieve the effect of increasing the gap between the positive and negative pole pieces to prevent adhesion between the pole pieces compared to the third wound battery cell image.

[0112] (12) Use the GaussianBlur function to perform a Gaussian filtering operation on the fourth initial wound battery cell image in the horizontal direction to obtain a preliminary filtered image, as Figure 8 shown. It can be seen that the preliminary filtered image can achieve the effect of clearer contrast between the positive and negative pole pieces compared to the fourth initial wound battery cell image.

[0113] (13) Use the convertTo function to convert the preliminary filtered image to obtain a 32-bit single-channel image.

[0114] (14) Use the GaussianBlur function to perform a Gaussian filtering operation on the 32-bit single-channel image in the horizontal and vertical directions to obtain a secondarily filtered image.

[0115] (15) Divide the preliminary filtered image by the secondarily filtered image to obtain a ratio image.

[0116] (16) Use the convertTo function to convert the ratio image to obtain an 8-bit single-channel image and use it as the fourth wound battery cell image, as Figure 9 shown. It can be seen that the fourth wound battery cell image can achieve the effect of removing the shadow interference around each pole piece compared to the preliminary filtered image.

[0117] (17) The power-law transformation is performed on the gray value of each pixel point in the fourth wound battery cell image by using the cv2.power function to obtain the fifth wound battery cell image. As Figure 10 shown, it can be seen that the fifth wound battery cell image can achieve the effect of overall contrast enhancement compared with the fourth wound battery cell image.

[0118] (18) The fifth wound battery cell image and the second image are stitched according to the pixel position information to obtain a complete image, which is used as the final wound battery cell image. As Figure 11 shown.

[0119] It should be noted that the purpose of additionally performing the above steps (2) to (7) is to reduce the image resolution input to the above step (8) by cropping the region of interest, thereby effectively reducing the image processing time consumed by the entire detection method. Correspondingly, the purpose of performing the above step (18) is to improve the resolution of the fifth wound battery cell image, that is, to keep the resolution of the final wound battery cell image unchanged from the initial wound battery cell image, and at the same time, it is beneficial to accurately extract and analyze each pole piece feature in the final wound battery cell image.

[0120] Of course, when there is no need to consider reducing the image processing time, the above steps (2) to (7) can be not executed, and the grayscale image output by the above step (1) can be directly used as the image input to the above step (8), and the above step (18) does not need to be executed subsequently, and the fifth wound battery cell image output by the above step (17) can be directly used as the final wound battery cell image.

[0121] In the embodiment of the present invention, the image is deblurred by a convolution operation based on a mean filter convolution kernel, then the brightness of the image is adjusted by a logarithmic transformation operation, and then a convolution operation in the y-axis direction is performed on the image based on a preset enhancement matrix, so that the pole piece features contained in the image are more obvious, and the noise existing inside the image during the image brightness adjustment is reduced to avoid affecting the image processing effect. Subsequently, the image is further subjected to a dilation operation in the vertical direction, a Gaussian filtering operation in the horizontal direction, and a Gaussian filtering operation after being converted into a single-channel image, so that the pole piece features contained in the image present a clearer connection relationship, and the noise existing inside the image is smoother to reduce interference. When the number of pole piece layers contained in the wound battery cell is large, a clearer wound battery cell image can be obtained through the above series of image processing means, thereby helping to improve the detection accuracy of the pole piece alignment degree of the wound battery cell.

[0122] Please refer to Figure 12 , Figure 121 is a flow chart of a method for detecting the alignment of a wound battery cell according to an embodiment of the present invention. The method for detecting the alignment of a wound battery cell comprises the following steps:

[0123] S210, controlling the x-ray detection equipment to collect images of the wound battery cells to obtain multiple images of the wound battery cells;

[0124] S20, using the above-mentioned image processing method to process the multiple wound battery cell images to obtain a final wound battery cell image;

[0125] S230, judging the conformity of the alignment of each pole piece feature included in the final wound battery cell image.

[0126] In some embodiments, the specific implementation process of the above step S230 includes the following steps:

[0127] S231, using the convolutional neural network of the x-ray detection device to extract pole piece features from the final wound battery cell image to obtain the vertex positions of N pole pieces, where N is the number of pole pieces included in the final wound battery cell image, and N is a positive integer greater than 1;

[0128] It should be noted that the convolutional neural network can be, but is not limited to, an existing Hourglass neural network, and the Hourglass neural network is originally trained using multiple different wound battery cell images with known alignment detection results;

[0129] S232, according to the vertex positions of the N pole pieces, calculate the corresponding N-1 pole piece vertex spacings, wherein the i-th pole piece vertex spacing is the vertex spacing between the i-th pole piece and the adjacent i+1-th pole piece among the N pole pieces, and i is a positive integer greater than or equal to 1 and less than N;

[0130] S233, determine whether the vertices spacing of the N-1 pole pieces falls within the preset spacing range at the same time; if so, generate a first detection result, which records that the alignment of each pole piece in the wound battery cell is qualified; if not, generate a second detection result, which records that the alignment of some pole pieces in the wound battery cell is unqualified.

[0131] After performing the above step S233, either only the first detection result or the second detection result can be output, or the first detection result or the second detection result can be output together with the final wound cell image, or the first detection result or the second detection result can be output together with the initial wound cell image and the final wound cell image. The present invention does not make any limitation in this regard, so as to provide a certain reference basis for those skilled in the art to judge whether to improve the wound cell.

[0132] In the embodiment of the present invention, when the number of electrode sheets included in the wound cell is large, a clearer image of the wound cell can be obtained through the above image processing method, and finally the alignment degree of the electrode sheets of the wound cell can be detected more accurately.

[0133] Please refer to Figure 13 , Figure 13 which is a schematic diagram of the composition of an image processing system provided by an embodiment of the present invention. The image processing system includes the following:

[0134] The first module 310 is used to obtain an initial wound cell image;

[0135] The second module 320 is used to denoise the initial wound cell image to obtain a first wound cell image;

[0136] The third module 330 is used to perform logarithmic transformation on the first wound cell image to adjust its brightness value to obtain a second wound cell image;

[0137] The fourth module 340 is used to enhance the edges of the second wound cell image to obtain a third wound cell image;

[0138] The fifth module 350 is used to perform dilation and Gaussian filtering on the third wound cell image to obtain a fourth wound cell image;

[0139] The sixth module 360 is used to adjust the contrast of the fourth wound cell image to obtain a final wound cell image.

[0140] The content in the embodiment of the above image processing method is applicable to the embodiment of the image processing system. The functions realized by the image processing system are the same as those in the embodiment of the above image processing method, and the beneficial effects achieved are the same as those in the embodiment of the above image processing method, and will not be elaborated here.

[0141] Please refer to Figure 14 , Figure 14 which is a schematic diagram of the composition of a wound cell alignment degree detection system provided by an embodiment of the present invention. The wound cell alignment degree detection system includes the following:

[0142] The acquisition module 410 is configured to acquire multiple images of wound electric cores through an x-ray detection device.

[0143] The processing module 420 is configured to process the multiple images of wound electric cores by using the above image processing method to obtain a final image of a wound electric core.

[0144] The judgment module 430 is configured to judge the qualification of the alignment degree of each pole piece feature included in the final image of the wound electric core.

[0145] The content in the embodiment of the above method for detecting the alignment degree of wound electric cores is applicable to the embodiment of the system for detecting the alignment degree of wound electric cores. The functions implemented by the system for detecting the alignment degree of wound electric cores are the same as those in the embodiment of the above method for detecting the alignment degree of wound electric cores, and the beneficial effects achieved are the same as those in the embodiment of the above method for detecting the alignment degree of wound electric cores, and will not be elaborated here.

[0146] In addition, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements any one of the above method embodiments. The computer-readable storage medium includes, but is not limited to, any type of disk (including floppy disks, hard disks, optical disks, CD-ROMs, and magneto-optical disks), ROM (Read-Only Memory), RAM (Random Access Memory), EPROM (Erasable Programmable Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), flash memory, magnetic cards, or optical cards. That is, the storage device includes any medium that stores or transmits information in a readable form by a device (such as a computer, mobile phone, etc.), and can be a read-only memory, a magnetic disk, or an optical disk, etc.

[0147] In addition, Figure 15 FIG. is a schematic hardware structure diagram of a computer device provided by an embodiment of the present invention. The computer device includes devices such as a processor 520, a memory 530, an input unit 540, and a display unit 550. Those skilled in the art can understand. Figure 15The illustrated device structure components do not constitute a limitation on all devices, and may include more or fewer components than shown, or combine certain components. The memory 530 can be used to store the computer program 510 and each functional module. The processor 520 runs the computer program 510 stored in the memory 530, thereby performing various functional applications and data processing of the device. The memory can be an internal memory or an external memory, or include both an internal memory and an external memory. The internal memory can include a read-only memory (ROM), a programmable ROM (PROM), an electrically programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), a flash memory, or a random access memory. The external memory can include a hard disk, a floppy disk, a USB flash drive, a magnetic tape, etc. The memory 530 disclosed in the embodiments of the present invention includes, but is not limited to, these types of memories. The memory 530 disclosed in the embodiments of the present invention is only an example and not a limitation.

[0148] The input unit 540 is used to receive the input of signals and receive the keywords input by the user. The input unit 540 can include a touch panel and other input devices. The touch panel can collect the touch operations of the user on or near it (such as the operations of the user using any suitable object or accessory such as a finger, a stylus, etc. on or near the touch panel), and drive the corresponding connection device according to a pre-set program; the other input devices can include, but are not limited to, one or more of a physical keyboard, function keys (such as play control keys, switch keys, etc.), a trackball, a mouse, a joystick, etc. The display unit 550 can be used to display the information input by the user or the information provided to the user and various menus of the terminal device. The display unit 550 can be in the form of a liquid crystal display, an organic light-emitting diode, etc. The processor 520 is the control center of the terminal device, connects various parts of the entire device using various interfaces and lines, and performs various functions and processes data by running or executing the software programs and / or modules stored in the memory 530, and calling the data stored in the memory 530.

[0149] As an embodiment, the computer device includes a processor 520, a memory 530, and a computer program 510. The computer program 510 is stored in the memory 530 and is configured to be executed by the processor 520. The computer program 510 is configured to execute any one of the above method embodiments.

[0150] Although the description of the present application has been quite detailed and has particularly described several of the described embodiments, it is not intended to be limited to any of these details or embodiments or any particular embodiment, but rather should be regarded as effectively covering the intended scope of the present application by reference to the appended claims, considering the prior art to provide a broad interpretation of these claims. In addition, the present application has been described above with embodiments foreseeable by the inventors for the purpose of providing a useful description, and those non-substantive changes to the present application that are not currently foreseeable may still represent equivalent changes to the present application.

Claims

1. An image processing method, characterized in that The method includes: Obtaining an initial wound cell image; Denosing the initial wound cell image to obtain a first wound cell image; Performing logarithmic transformation on the first wound cell image to adjust its brightness value to obtain a second wound cell image; Performing edge enhancement on the second wound cell image to obtain a third wound cell image; Performing dilation and Gaussian filtering on the third wound cell image to obtain a fourth wound cell image; Adjusting the contrast of the fourth wound cell image to obtain a final wound cell image.

2. The image processing method according to claim 1, wherein The obtaining of the initial wound cell image includes: Obtaining a plurality of wound cell images collected by an x-ray detection device for the wound cell, and then determining the average gray value of all pixel points in each wound cell image; Selecting, from the plurality of wound cell images, the wound cell image whose average gray value exceeds a preset gray threshold as the initial wound cell image.

3. The image processing method according to claim 1, wherein The denosing of the initial wound cell image to obtain a first wound cell image includes: Converting the initial wound cell image into a grayscale image; Performing convolution operation on the grayscale image using a mean filter convolution kernel of size 3×3 to obtain the denoised first wound cell image.

4. The image processing method according to claim 1, wherein The performing of edge enhancement on the second wound cell image to obtain a third wound cell image includes: Performing convolution operation on the second wound cell image in the y-axis direction using a preset enhancement matrix of size 3×3 to obtain the edge-enhanced third wound cell image.

5. The image processing method according to claim 4, wherein The preset enhancement matrix is: kernel = (-9, -5, -9; 0, 1, 0; 9, 15, 9).

6. The image processing method according to claim 1, wherein The performing of dilation and Gaussian filtering on the third wound cell image to obtain a fourth wound cell image includes: Performing dilation operation on the third wound cell image in the vertical direction to obtain a fourth initial wound cell image; Performing Gaussian filtering operation on the fourth initial wound cell image in the horizontal direction to obtain a preliminary filtered image; Converting the preliminary filtered image into a 32-bit single-channel image and performing Gaussian filtering operation on it to obtain a secondary filtered image; Dividing the preliminary filtered image by the secondary filtered image to obtain a ratio image; Converting the ratio image into an 8-bit single-channel image and using it as the fourth wound cell image.

7. The image processing method according to claim 1, wherein The adjusting of the contrast of the fourth wound cell image to obtain a final wound cell image includes: Performing power-law transformation on the gray value of each pixel point in the fourth wound cell image to obtain the final wound cell image with increased overall contrast.

8. A method for detecting the alignment degree of wound electric cores, characterized in that, The method includes: Collecting images of the wound cell by an x-ray detection device to obtain a plurality of wound cell images; Processing the plurality of wound cell images using the image processing method according to any one of claims 1 to 7 to obtain a final wound cell image; Judging the qualification of the alignment degree of each pole piece feature included in the final wound cell image.

9. An image processing system, characterized in that, The system includes: A first module for obtaining an initial wound cell image; A second module for denosing the initial wound cell image to obtain a first wound cell image; The third module is used to perform logarithmic transformation on the first wound battery cell image to adjust its brightness value, obtaining a second wound battery cell image; The fourth module is used to perform edge enhancement on the second wound battery cell image, obtaining a third wound battery cell image; The fifth module is used to perform dilation and Gaussian filtering on the third wound battery cell image, obtaining a fourth wound battery cell image; The sixth module is used to perform contrast adjustment on the fourth wound battery cell image, obtaining a final wound battery cell image.

10. A winding battery cell alignment detection system, characterized in that, The system includes: An acquisition module, which is used to acquire multiple wound battery cell images of a wound battery cell through an x-ray detection device; A processing module, which is used to process the multiple wound battery cell images by using the image processing method described in any one of claims 1 to 7, obtaining a final wound battery cell image; A judgment module, which is used to judge the qualification of the alignment degree of each pole piece feature included in the final wound battery cell image.

Images