Striped image denoising method and device, electronic equipment and storage medium

Abstract

The application relates to the technical field of image processing, and discloses a stripe image denoising method and device, electronic equipment and a storage medium. A target stripe image to be denoised is acquired, and a target stripe direction corresponding to the target stripe image is acquired. Morphological opening operation is performed on the target stripe image by using a preset linear structure element to determine a stripe main structure image. A final denoised image is determined according to the stripe main structure image. According to the application, the target stripe image is acquired, and morphological opening operation is performed on the target stripe image by using the linear structure element, so that point noise and block noise can be effectively suppressed in a strong noise background, stripe breakage, blurring or widening can be avoided, and the unification of direction-selective noise suppression and structure preservation is realized. In addition, the target stripe direction is used as an input parameter, so that stripe images with different directions or different rotating postures can be processed, that is, the geometric features of the target stripe image are combined with the morphological filtering method, and the direction adaptive capability is enhanced.

Description

Technical Field

[0001] This application relates to the field of image processing technology, specifically to methods, apparatus, electronic devices, and storage media for stripe image denoising. Background Technology

[0002] Striped images are widely found in systems such as ultrafast optical measurement, stripe camera imaging, line-scan fluorescence microscopy, and structured light illumination imaging. Their prominent feature is that the signal extends continuously along a certain principal direction to form a striped structure. However, in actual imaging processes, they are often subject to various interferences such as scattering background, detector readout noise, dark current noise, and photon statistical noise. Especially under low light or deep imaging conditions, the image quality is severely degraded.

[0003] Denoising methods in related technologies mainly fall into three categories: The first category is spatial domain filtering methods, such as Gaussian filtering, median filtering, and bilateral filtering. These methods suppress noise through weighted averaging or nonlinear statistics within a local neighborhood. However, they are essentially isotropic smoothing processes, which can easily cause blurring of stripe edges and loss of detail while removing noise, leading to breakage or widening of thin stripes. The second category is frequency domain filtering methods, such as Wiener filtering. These methods construct filters based on the differences in the distribution of signal and noise in the frequency domain, but they rely on the assumption of noise stationarity and require knowledge of the power spectrum. Real-world striped images often contain complex factors such as non-uniform backgrounds and spatially correlated noise, making it difficult for frequency domain assumptions to hold, thus limiting the filtering effect and easily leading to structural distortion or residual noise. In addition, statistical modeling-based methods can improve denoising performance to some extent, but due to the lack of explicit modeling of the image's geometric structure, especially the failure to fully utilize the directional continuity features of the striped image, signal and noise often aliasing occurs in the same representation space, thus affecting the recovery of weak structures. Summary of the Invention

[0004] This application provides a stripe image denoising method, apparatus, electronic device, and storage medium to solve the problems in related technologies where removing noise can easily cause blurred stripe edges, loss of details, breakage or widening of thin stripes, limited filtering effect, and easy occurrence of structural distortion or residual noise.

[0005] In a first aspect, this application provides a method for denoising striped images, including:

[0006] Obtain the target stripe image to be denoised, and obtain the target stripe direction corresponding to the target stripe image;

[0007] The target stripe image is subjected to morphological opening operation by a preset linear structuring element to determine the main stripe structure image; the linear structuring element is a structuring element that is multi-pixel in the linear direction and single-pixel in the direction perpendicular to the linear direction, and the linear direction is consistent with the direction of the target stripe.

[0008] The final denoised image is determined based on the stripe principal structure image;

[0009] The step of acquiring the target stripe image to be denoised includes:

[0010] Obtain the original stripe image and obtain the original stripe direction corresponding to the original stripe image;

[0011] Determine the angle between the original stripe direction and the target stripe direction, rotate the original stripe image according to the angle, and determine the intermediate stripe image; the stripe direction corresponding to the intermediate stripe image is the target stripe direction;

[0012] The target stripe image is determined based on the intermediate stripe image.

[0013] The stripe image denoising method provided in this embodiment acquires the target stripe image and its direction, and performs morphological opening operations on the target stripe image using linear structuring elements whose linear direction matches the target stripe direction. This effectively suppresses point and block noise in strong noise backgrounds while maintaining the continuity and integrity of the stripe structure, avoiding stripe breaks, blurring, or widening, thus achieving a balance between direction-selective noise suppression and structure preservation. Furthermore, by using the target stripe direction as an input parameter, stripe images with different directions or rotational postures can be processed, combining the geometric features of the target stripe image with morphological filtering methods to enhance direction adaptability.

[0014] In some optional implementations, obtaining the original stripe direction corresponding to the original stripe image includes:

[0015] The original stripe image is segmented to determine at least one stripe region corresponding to the original stripe image;

[0016] Determine the stripe direction corresponding to each of the stripe regions, and determine the original stripe direction corresponding to the original stripe image based on the stripe direction of each of the stripe regions.

[0017] In some optional implementations, segmenting the original stripe image to determine at least one stripe region corresponding to the original stripe image includes:

[0018] The original stripe image is projected according to a preset direction to determine the projection curve corresponding to the original stripe image; the projection curve is used to represent the pixel projection value corresponding to each pixel position point in the direction perpendicular to the preset direction.

[0019] Identify the local maxima in the projected curve;

[0020] For any given local maximum point, within a preset range corresponding to the local maximum point, determine two local minimum points located on either side of the local maximum point; and,

[0021] The corresponding stripe region is determined based on the positions of the two local minimum points as boundary points;

[0022] And / or,

[0023] The step of determining the stripe direction corresponding to each of the stripe regions, and determining the original stripe direction corresponding to the original stripe image based on the stripe direction of each of the stripe regions, includes:

[0024] For any one of the striped regions, perform a Radon transform to determine the projection value of the striped region under multiple projection angles, and take the projection angle corresponding to the largest projection value as the region stripe angle corresponding to the striped region.

[0025] Stripe regions whose stripe angles meet preset angle conditions are defined as valid stripe regions; and...

[0026] The original stripe direction corresponding to the original stripe image is determined by weighted averaging of the stripe angles corresponding to each of the effective stripe regions.

[0027] In some optional implementations, determining the target stripe image based on the intermediate stripe image includes:

[0028] The intermediate stripe image is filtered to determine the background image corresponding to the intermediate stripe image;

[0029] The target stripe image is determined by subtracting the intermediate stripe image from the background image.

[0030] In some optional implementations, the number of linear structuring elements is multiple, and each linear structuring element corresponds to a different scale; the step of performing morphological opening operations on the target stripe image using preset linear structuring elements to determine the main stripe structure image includes:

[0031] By performing morphological opening operations at corresponding scales on the target stripe image using multiple linear structuring elements, the stripe substructure image corresponding to each linear structuring element is determined.

[0032] The individual stripe substructure images are fused together to determine the main stripe structure image.

[0033] In some optional implementations, determining the final denoised image based on the stripe principal structure image includes:

[0034] Calculate the linear residual between the target stripe image and the stripe principal structure image;

[0035] The linear residual is added to the stripe master structure image to obtain a stripe enhancement image;

[0036] The final denoised image is determined based on the stripe enhancement image.

[0037] In some optional implementations, calculating the linear residual between the target stripe image and the stripe principal structure image includes:

[0038] The difference between the target stripe image and the stripe principal structure image is calculated to obtain the residual image;

[0039] The standard deviation of the residual image is determined based on the absolute deviation of the median, and the standard deviation is multiplied by a preset residual threshold coefficient to determine the residual threshold.

[0040] The residual image is binarized and thresholded based on the residual threshold to determine the binarized residual image;

[0041] Connectivity analysis is performed on the binarized residual image to extract the geometric features of each connected component;

[0042] The connected regions that satisfy the linear condition for the geometric features are taken as linear residual regions, and the linear residuals are determined based on the regions in the residual image that overlap with the linear residual regions.

[0043] The stripe image denoising method provided in this embodiment acquires the original stripe image and automatically estimates its original stripe direction, then rotates it to a preset target direction and performs background flattening processing. This effectively eliminates non-uniform background interference and can process stripe images with different directions or rotation postures. It combines the geometric features of the target stripe image with morphological filtering methods to enhance the direction adaptability. Furthermore, it uses multi-scale linear structuring elements consistent with the target stripe direction to perform morphological opening operations on the target stripe image. By extracting and fusing details and main features through structuring elements of different scales, it can effectively suppress point and block noise in strong noise backgrounds while maintaining the continuity and integrity of the stripe structure, avoiding stripe breakage, blurring, or widening. In addition, by calculating the residual between the target stripe image and the main stripe structure image, and adaptively determining the residual threshold based on the median absolute deviation, and combining connected component geometric features to filter linear residual regions, it selectively adds weak stripe signals that meet the linear condition back to the main structure image. This can restore the details of weak stripes that have been over-suppressed while denoising, avoiding signal loss.

[0044] Secondly, this application provides a stripe image denoising apparatus, comprising:

[0045] The target image module is used to acquire the target stripe image to be denoised and to acquire the target stripe direction corresponding to the target stripe image.

[0046] The stripe master structure module is used to perform morphological opening operations on the target stripe image using preset linear structuring elements to determine the stripe master structure image. A linear structuring element is a multi-pixel element in the linear direction and a single pixel element in the direction perpendicular to the linear direction, with the linear direction consistent with the direction of the target stripes.

[0047] The denoising image module is used to determine the final denoised image based on the stripe master structure image.

[0048] The target image module includes:

[0049] The original image submodule is used to acquire the original stripe image and obtain the original stripe direction corresponding to the original stripe image.

[0050] The angle determination submodule is used to determine the angle between the original stripe direction and the target stripe direction. Based on the angle, the original stripe image is rotated to determine the intermediate stripe image; the stripe direction corresponding to the intermediate stripe image is the target stripe direction.

[0051] The target image submodule is used to determine the target stripe image based on the intermediate stripe image.

[0052] Thirdly, this application provides an electronic device, including: a memory and a processor, which are communicatively connected to each other. The memory stores computer instructions, and the processor executes the computer instructions to perform the stripe image denoising method of the first aspect or any corresponding embodiment described above.

[0053] Fourthly, this application provides a computer-readable storage medium storing computer instructions for causing a computer to perform the stripe image denoising method of the first aspect or any corresponding embodiment described above.

[0054] Fifthly, this application provides a computer program product, including computer instructions for causing a computer to execute the stripe image denoising method of the first aspect or any corresponding embodiment described above. Attached Figure Description

[0055] To more clearly illustrate the technical solutions in the specific embodiments of this application or the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0056] Figure 1 This is a schematic diagram illustrating an application scenario according to an embodiment of this application;

[0057] Figure 2 This is a schematic flowchart of a first method for stripe image denoising according to an embodiment of this application;

[0058] Figure 3 This is a schematic diagram of a second process for a stripe image denoising method according to an embodiment of this application;

[0059] Figure 4 This is a flowchart of a residual analysis algorithm according to an embodiment of this application;

[0060] Figure 5 This is a structural block diagram of a stripe image denoising apparatus according to an embodiment of this application;

[0061] Figure 6 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0062] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0063] It is understood that before using the technical solutions disclosed in the various embodiments of this application, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in this application in an appropriate manner in accordance with relevant laws and regulations, and user authorization should be obtained.

[0064] As one optional application scenario in the embodiments of this application, such as Figure 1 As shown, application 101 is installed in terminal device 110, and user 130 can interact with application 101 through terminal device 110 and / or access device of terminal device 110.

[0065] For example, application 101 can be any application that provides image denoising related services. Figure 1 In the application scenario shown, if application 101 is active, the terminal device 110 can display the interface 102 of application 101. The interface 102 may include various pages that application 101 can provide, such as an image upload page, etc.

[0066] Terminal device 110 may be a mobile terminal, a fixed terminal, or a portable terminal, etc., including but not limited to mobile phones, desktop computers, laptop computers, multimedia tablets, or any combination thereof, including accessories and peripherals of these devices or any combination thereof. In some embodiments, terminal device 110 may also support any type of interface.

[0067] It should be noted that, Figure 1 This is merely an example of an application scenario and does not limit the scope of protection of this application.

[0068] The embodiments of this application will be described below with reference to the accompanying drawings. It should be understood that the pages shown in the drawings are merely examples, and various page designs are possible in practice. The various graphic elements on the page may have different arrangements and different visual representations, one or more elements may be omitted or replaced, and one or more other elements may also be present; no limitations are made in the embodiments of this application. Furthermore, the embodiments are mainly described below with reference to terminal device 110. It should be understood that the actions described relative to terminal device 110 can be performed by application 101 on terminal device 110.

[0069] This application provides a stripe image denoising method. By acquiring the target stripe image and the target stripe direction, and performing morphological opening operations on the target stripe image using linear structuring elements whose linear direction is consistent with the target stripe direction, it can effectively suppress point and block noise in a strong noise background, while maintaining the continuity and integrity of the stripe structure, avoiding stripe breakage, blurring or widening, and achieving the unity of direction-selective noise suppression and structure preservation.

[0070] According to an embodiment of this application, a method for denoising striped images 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. 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.

[0071] This embodiment provides a method for denoising striped images, which can be used in the aforementioned terminal devices, such as desktop computers and laptop computers. Figure 2 This is a flowchart of a stripe image denoising method according to an embodiment of this application, as follows: Figure 2 As shown, the process includes the following steps:

[0072] Step S201: Obtain the target stripe image to be denoised, and obtain the target stripe direction corresponding to the target stripe image.

[0073] In the process of denoising stripe images, the first step is to acquire a stripe image to be processed, i.e., the target stripe image. The target stripe image can be any digital image with a stripe structure, such as a stripe camera image, a line scan micrograph, or a fluorescence stripe image. Target stripe images are usually affected by noise, resulting in low visibility of the stripe structure, thus requiring denoising processing. Simultaneously with acquiring the target stripe image, the target stripe direction can also be obtained. The target stripe direction describes the direction of the stripes in the target stripe image and can be represented by an angle value (e.g., the angle relative to the horizontal or vertical direction).

[0074] Step S202: Perform morphological opening operations on the target stripe image using preset linear structuring elements to determine the main stripe structure image. A linear structuring element is a structuring element that has multiple pixels in the linear direction and a single pixel in the direction perpendicular to the linear direction, with the linear direction consistent with the direction of the target stripes.

[0075] After determining the target stripe image, a morphological opening operation can be performed on the target stripe image using a preset linear structuring element. The linear structuring element is a specially shaped structuring element that can cover multiple pixels (i.e., have a certain length) in the linear direction, while covering only a single pixel (i.e., a width of 1 pixel) in the direction perpendicular to the linear direction. In this embodiment, the linear direction of the linear structuring element is consistent with the target stripe direction, enabling denoising operations to perform directional constraint denoising based on the stripe direction of the image, avoiding structural damage caused by denoising operations in related technologies. Furthermore, using the target stripe direction as an input parameter allows processing stripe images with different directions or rotational postures, expanding the applicability. The linear direction refers to the direction corresponding to the axis covering multiple pixels in the linear structuring element. For example, if the target stripe direction is vertical, the linear direction of the linear structuring element can also be vertical, meaning the linear structuring element covers multiple pixels in the vertical direction and a single pixel in the horizontal direction.

[0076] Morphological opening involves first eroding the target stripe image, then dilating the erosion result. For example, a linear structuring element is first slid across the target stripe image pixel by pixel. At each pixel, the minimum value of all pixels within the structuring element's coverage area is taken and assigned to the corresponding position in the output image, thus obtaining the erosion result. Using this erosion result as input, the same linear structuring element is slid across the image pixel by pixel again, and the maximum value within the coverage area is taken as the output, thus performing the dilution operation. After completing the morphological opening operation, point and block noise in the target stripe image can be suppressed while preserving the stripe structure, resulting in the main stripe structure image.

[0077] Step S203: Determine the final denoised image based on the stripe principal structure image.

[0078] After acquiring the main stripe structure image, the final denoised image can be determined based on the main stripe structure image. In this embodiment, the main stripe structure image can be directly used as the final denoised image, or post-processing operations can be performed on the main stripe structure image to enhance the visual effect. For example, post-processing operations can be grayscale range adjustment, Gaussian smoothing, or contrast adjustment. After the post-processing operation is completed, the post-processing result is used as the final denoised image.

[0079] The stripe image denoising method provided in this embodiment acquires the target stripe image and its direction, and performs morphological opening operations on the target stripe image using linear structuring elements whose linear direction matches the target stripe direction. This effectively suppresses point and block noise in strong noise backgrounds while maintaining the continuity and integrity of the stripe structure, avoiding stripe breaks, blurring, or widening, thus achieving a balance between direction-selective noise suppression and structure preservation. Furthermore, by using the target stripe direction as an input parameter, stripe images with different directions or rotational postures can be processed, combining the geometric features of the target stripe image with morphological filtering methods to enhance direction adaptability.

[0080] This embodiment provides a method for denoising striped images, which can be used in the aforementioned terminal devices, such as desktop computers and laptop computers. Figure 3 This is a flowchart of a stripe image denoising method according to an embodiment of this application, as follows: Figure 3 As shown, the process includes the following steps:

[0081] Step S301: Obtain the target stripe image to be denoised, and obtain the target stripe direction corresponding to the target stripe image.

[0082] Specifically, step S301, “acquiring the target stripe image to be denoised,” includes steps S3011 to S3013.

[0083] Step S3011: Obtain the original stripe image and obtain the original stripe direction corresponding to the original stripe image.

[0084] When acquiring the target stripe image, the original stripe image can be obtained. The original stripe image can be a noisy stripe image uploaded by the user. Since the stripes in the original stripe image may exist in any direction, such as horizontal, vertical, or at a certain tilt angle, after acquiring the original stripe image, the original stripe direction corresponding to the original stripe image can also be obtained, that is, the actual direction of the stripes in the original stripe image, to simplify subsequent processing. In this embodiment, the original stripe direction can be manually marked by the user based on the image content, or it can be automatically calculated using direction estimation algorithms or other methods.

[0085] In some optional implementations, step S3011, "obtaining the original stripe direction corresponding to the original stripe image", includes steps a1 and a2.

[0086] Step a1: Segment the original stripe image to determine at least one stripe region corresponding to the original stripe image.

[0087] Step a2: Determine the stripe direction of each stripe region, and determine the original stripe direction of the original stripe image based on the stripe direction of each region.

[0088] In determining the original stripe direction, the original stripe image can be segmented to divide it into one or more stripe regions. Each stripe region corresponds to a stripe structure in the original stripe image. For example, the user can manually segment the original stripe image. The specific method for segmenting the original image is not limited in this embodiment. After determining the stripe regions, the direction of the stripe structure in each region can be calculated to obtain the stripe direction corresponding to each region. The stripe directions of each region are then fused, for example, by calculating the median, mode, and mean of the stripe directions in each region, to determine the original stripe direction corresponding to the original stripe image.

[0089] In some optional implementations, step a1, “segmenting the original stripe image and determining at least one stripe region corresponding to the original stripe image,” includes steps a11 to a14.

[0090] Step a11: Project the original stripe image according to a preset direction to determine the projection curve corresponding to the original stripe image. The projection curve is used to represent the pixel projection value corresponding to each pixel position point in the vertical direction of the preset direction.

[0091] In determining each stripe region, the original stripe image can be projected in a preset direction. This preset direction can be a row direction (horizontal), a column direction (vertical), or the direction of the target stripe. The projection process involves accumulating the pixel values ​​of the image along this preset direction, thus obtaining the corresponding projection curve. For example, the y-axis of this projection curve is a coordinate axis parallel to the preset direction, and the x-axis is a coordinate axis perpendicular to the preset direction.

[0092] For example, the above-mentioned "projecting the original stripe image according to a preset direction to determine the projection curve corresponding to the original stripe image" may specifically include: for each pixel position point arranged along the first direction in the original stripe image, determining the pixel projection value corresponding to the pixel position point according to the pixel value of the pixel position point in the second direction; the second direction is a direction parallel to the preset direction, and the first direction is perpendicular to the second direction; determining the projection curve corresponding to the original stripe image according to the pixel projection value corresponding to each pixel position point.

[0093] In this embodiment, the original striped image can first be normalized to obtain a normalized image. ,in Where I is the original stripe image, , These represent the maximum and minimum gray values ​​in the original striped image. Then, the image obtained after normalization is... Determine the direction of the stripes.

[0094] For example, when the preset direction is vertical, calculate the vertical projection curve. The methods can be:

[0095] ;

[0096] in, Let be the vertical projection value (i.e., pixel projection value) of the j-th column (i.e., the j-th pixel position), where i is the row coordinate of the original stripe image, j is the column coordinate of the original stripe image, and N is the height of the original stripe image (i.e., the number of pixel rows). It can be understood that, at this point, the preset direction is the vertical direction, the direction perpendicular to the preset direction is the horizontal direction, and the pixel positions j are arranged horizontally, resulting in a projection curve... It can represent the pixel projection value corresponding to any pixel location.

[0097] After obtaining the vertical projection curve, Gaussian smoothing can be used to suppress projection curve noise:

[0098] ;

[0099] in, It is a one-dimensional Gaussian smoothing kernel. This is the smoothing parameter. The smoothed projection curve is the projection curve corresponding to the original striped image.

[0100] Step a12: Determine the local maxima in the projected curve.

[0101] After determining the projection curve corresponding to the original stripe image, local maxima can be found on the projection curve. A local maximum is a point on the projection curve where the projected value is greater than the projected value of its adjacent position. Because the pixel values ​​within the stripe region are relatively high, a peak is formed at that position after projection, so local maxima are usually located near the center of the stripe region. In this embodiment, a peak detection algorithm can be used to extract all local maxima points that are greater than the peak detection threshold using a preset peak detection threshold.

[0102] Step a13: For any local maximum point, within the preset range corresponding to the local maximum point, determine two local minimum points located on both sides of the local maximum point.

[0103] For each local maximum point, search for local minimum points within a preset range on both sides of the local maximum point (e.g., within a certain width area centered on the local maximum point). For example, the left and right search intervals can be determined according to the preset search radius, and the local minimum points of the projected curves in the left and right search intervals are calculated respectively, which are the positions where the projected values ​​are the smallest.

[0104] Step a14: Determine the corresponding striped region based on the positions of the two local minima as boundary points.

[0105] After determining two local minima, these two local minima can be used as the left and right boundaries of the corresponding fringe regions, achieving adaptive segmentation of the fringe regions. Since the minima of the projection curve usually correspond to the background gaps between fringe regions, positioning the boundaries at the minima can maximize the inclusion of the main fringe signal within the region while excluding irrelevant background on both sides.

[0106] In some optional implementations, step a2, "determining the stripe direction corresponding to each stripe region and determining the original stripe direction corresponding to the original stripe image based on the stripe direction of each region", includes steps a21 to a23.

[0107] Step a21: Perform a Radon transform on any striped region to determine the projection value of the striped region under multiple projection angles, and take the projection angle corresponding to the largest projection value as the region stripe angle corresponding to the striped region.

[0108] Step a22: Select the stripe regions whose stripe angles meet the preset angle conditions as valid stripe regions.

[0109] Step a23: Perform a weighted average of the fringe angles corresponding to each effective fringe region to determine the original fringe direction corresponding to the original fringe image.

[0110] For any striped region, a Radon Transform can be performed, which involves performing a line integral along a straight line at different angles to obtain the projection value at each projection angle. In other words, for any striped region, a Radon Transform is performed at multiple projection angles (e.g., angles selected at equal intervals from 0° to 180°) to obtain the projection value corresponding to each projection angle. The projection angle with the largest projection value is the region stripe angle corresponding to that striped region, which is used to represent the region stripe direction of the striped region.

[0111] In this embodiment, the method for calculating the projection value using Radon transform can be as follows:

[0112] ;

[0113] in, For the projection angle, For the projection distance, For the Dirac function, Let x and y represent any striped region. The x and y coordinates.

[0114] After determining the projection values ​​of the fringe region at various projection angles, it can be... As the regional fringe angle of this fringe area, where... .

[0115] After obtaining the fringe angles of each region, it is possible to calculate whether the fringe angles meet preset angle conditions. Fringe regions that meet these conditions are considered valid fringe regions, thus removing unreliable regions that may provide incorrect direction estimates due to noise, occlusion, or fringe breaks. In this embodiment, the lower quartile and upper quartile of the fringe angles of each region can be calculated. Fringe regions whose fringe angles lie between the lower and upper quartiles, and whose projection peak value (i.e., the maximum projection value) is higher than a preset proportion (e.g., a preset proportion of 0.8) of the global projection peak value in each fringe region, are considered valid fringe regions.

[0116] After determining the effective fringe regions, the fringe angles corresponding to each effective fringe region can be weighted and averaged to obtain the weighted average angle, which represents the original fringe direction of the original fringe image. In this embodiment, the weight of the corresponding fringe angle can be determined based on the projection peak value corresponding to each effective fringe region, or the area of ​​the effective fringe region can be used as the weight; wherein, the larger the projection peak value or area, the greater its corresponding weight.

[0117] Step S3012: Determine the angle between the original fringe direction and the target fringe direction. Rotate the original fringe image according to the angle to determine the intermediate fringe image. The fringe direction corresponding to the intermediate fringe image is the target fringe direction.

[0118] After obtaining the original fringe direction, the angle between the original fringe direction and the target fringe direction can be calculated. The target fringe direction can be a pre-defined standard direction, such as vertical or horizontal. After obtaining the angle between the original and target fringe directions, the original fringe image can be rotated based on this angle, so that the fringe direction of the rotated original fringe image is consistent with the target fringe direction. After rotation, the rotated original fringe image becomes the intermediate fringe image. Rotating the original fringe image provides a fixed direction for subsequent morphological opening operations, reducing algorithm complexity and improving computational efficiency.

[0119] Step S3013: Determine the target stripe image based on the intermediate stripe image.

[0120] After obtaining the intermediate stripe image, it can be used directly as the target stripe image, or the intermediate stripe image can be processed by background flattening and other processes. The processed intermediate stripe image is the target stripe image.

[0121] In some alternative implementations, step S3013, “determining the target stripe image based on the intermediate stripe image,” includes steps b1 and b2.

[0122] Step b1: Filter the middle stripe image to determine the background image corresponding to the middle stripe image.

[0123] Step b2: Subtract the background image from the intermediate stripe image to determine the target stripe image.

[0124] Since the intermediate stripe image may contain slowly changing background noise, the intermediate stripe image can be flattened. For example, a large-scale Gaussian filter can be used to perform low-frequency filtering on the intermediate stripe image to determine the low-frequency background image corresponding to the intermediate stripe image.

[0125] In this embodiment, the low-frequency background image The methods to obtain it can be:

[0126] ;

[0127] in, The image shows the middle stripes. It is a large-scale Gaussian filter kernel.

[0128] After determining the low-frequency background image, image flattening can be achieved through subtraction, i.e.:

[0129] ;

[0130] in, This refers to the target stripe image after image flattening.

[0131] Step S302: Perform morphological opening operations on the target stripe image using preset linear structuring elements to determine the main stripe structure image. A linear structuring element is a structuring element that has multiple pixels in the linear direction and a single pixel in the direction perpendicular to the linear direction, with the linear direction consistent with the direction of the target stripes.

[0132] Please see details Figure 2 Step S202 of the illustrated embodiment will not be described again here.

[0133] In some optional implementations, there are multiple linear structuring elements, and each linear structuring element corresponds to a different scale. Step S302, "performing a morphological opening operation on the target stripe image using preset linear structuring elements to determine the main structure image of the stripes," includes steps c1 and c2.

[0134] Step c1 involves performing morphological opening operations at corresponding scales on the target stripe image using multiple linear structuring elements to determine the stripe substructure image corresponding to each linear structuring element.

[0135] Step c2: Fuse the images of each stripe substructure to determine the main stripe structure image.

[0136] As mentioned earlier, morphological opening operations can be performed on the target stripe image using preset linear structuring elements to determine the main structure image of the stripes. In this step, there can be multiple linear structuring elements, and each linear structuring element corresponds to a different scale (i.e., the length of the linear structuring element).

[0137] For example, the number of linear structuring elements can be 4, with corresponding scales of 7, 11, 17, and 25. If the target stripe direction is vertical, the size of the linear structuring element with scale 7 can be 7×1 (a linear structuring element is a 7-row, 1-column structuring element). Similarly, the sizes of the other three linear structuring elements are 11×1, 17×1, and 25×1, respectively. Using structuring elements of multiple scales helps capture stripe features at different scales. Short-scale structuring elements are used to preserve detailed features, while long-scale structuring elements are used to extract and enhance large structures and main directional features in the image. For each linear structuring element, a morphological opening operation is performed on the target stripe image using that linear structuring element. The specific method of the morphological opening operation can be found in step S202 and will not be repeated here. After each linear structuring element completes the morphological opening operation on the target stripe image, the stripe substructure image corresponding to the linear structuring element can be determined. The maximum or average value of each pixel in the stripe substructure images at all scales is taken to fuse the multiple stripe substructure images, thus obtaining the stripe main structure image.

[0138] Step S303: Determine the final denoised image based on the stripe principal structure image.

[0139] Please see details Figure 2 Step S203 of the illustrated embodiment will not be described again here.

[0140] In some alternative implementations, step S303, “determining the final denoised image based on the stripe master structure image,” includes steps d1 to d3.

[0141] Step d1: Calculate the linear residual between the target stripe image and the stripe principal structure image.

[0142] Step d2: Add the linear residual to the stripe master structure image to obtain the stripe enhancement image.

[0143] Step d3: Determine the final denoised image based on the stripe enhancement image.

[0144] In determining the final denoised image based on the main structure image of the stripes, residual calculation can be considered. First, the linear residual between the target stripe image and the main structure image of the stripes is calculated. This linear residual represents linear regions that are not displayed in the main structure image but exist in the target stripe image, such as weak stripe signals that were not fully reconstructed.

[0145] After obtaining the linear residual, it can be added back to the stripe principal structure image to compensate for the loss of signals that were not fully reconstructed and to ensure that the gray values ​​of the reconstructed image are non-negative. At this point, the reconstruction of the stripe image is completed, and the stripe enhanced image is obtained.

[0146] After determining the stripe enhancement image, it can be used as the final denoised image, or a small-scale Gaussian smoothing process can be performed on the stripe enhancement image to remove local spikes and residual high-frequency noise in the image, further optimizing the image quality and obtaining the final denoised image.

[0147] In some optional implementations, step d1, “calculating the linear residual between the target stripe image and the stripe principal structure image,” includes steps d11 to d15.

[0148] Step d11: Calculate the difference between the target stripe image and the stripe principal structure image to obtain the residual image.

[0149] Step d12: Determine the standard deviation of the residual image based on the absolute deviation of the median, and multiply the standard deviation by the preset residual threshold coefficient to determine the residual threshold.

[0150] Step d13: Perform binarization thresholding on the residual image based on the residual threshold to determine the binarized residual image.

[0151] Step d14: Perform connected component analysis on the binarized residual image and extract the geometric features of each connected component.

[0152] Step d15: Connected regions whose geometric features satisfy the linear condition are taken as linear residual regions. Based on the regions in the residual image that coincide with the linear residual regions, the linear residuals are determined.

[0153] When determining the linear residual, it is necessary to calculate the difference between the target fringe image and the main fringe structure image, i.e., the residual image. This residual image represents details and noise that were not captured in the main fringe structure image. For example, continuous, linearly distributed positive residuals are likely weak fringe signals that were not fully reconstructed, while isolated, discrete positive and negative residuals usually correspond to shot noise, salt-and-pepper noise, or reconstruction artifacts. In this embodiment, the residual image... It can be: ,in For the target stripe image, This is the main structure image of the stripes.

[0154] Since the residuals may contain real signals, directly using the classical standard deviation is easily affected by outliers. Therefore, a robust standard deviation method based on the absolute deviation of the median can be used to estimate the standard deviation of the residuals.

[0155] ;

[0156] in, Standard deviation, This indicates taking the median.

[0157] In obtaining the standard deviation Then, the residual threshold can be determined by multiplying the preset residual threshold coefficient k (e.g., k can be 2, 3, etc.) with the standard deviation. .For example, .

[0158] Using this residual threshold, the residual image can be binarized and thresholded. This involves marking all pixels exceeding the residual threshold (e.g., marking them as 1, and the rest as 0), resulting in a binarized residual image. Subsequently, connected component analysis is performed on the binarized residual image to extract the geometric features of each connected component. These geometric features may include area, major axis length, eccentricity, and orientation angle. The geometric features of each connected component are analyzed, identifying those that satisfy the linear condition as linear residual regions, and discarding those that do not, treating them as point-like or block-like noise residuals.

[0159] The linear conditions may include: 1. Area condition: The area of ​​the connected region should be greater than a preset area threshold to ensure that the size of the connected region is sufficient to represent the effective signal; 2. Major axis length condition: The major axis length of the connected region should be greater than a preset major axis threshold to ensure that the connected region has linear characteristics; 3. Eccentricity: The eccentricity of the connected region should be close to 1 (the closer to 1, the more linear), to ensure that the shape of the connected region conforms to the linear characteristics; 4. Direction angle: The direction angle of the connected region should be within a preset angle threshold to ensure that the direction of the connected region is consistent with the direction of the target stripe.

[0160] Finally, the linear residual is determined based on the region in the residual image that overlaps with the linear residual region. In this embodiment, the linear residual region can be multiplied by the residual image to determine the linear residual:

[0161] ;

[0162] in, The residuals are linear. The linear residual region in the binary residual image can be the binary residual image determined after removing connected regions in the binary residual image that do not meet the linear condition.

[0163] Figure 4 This is a flowchart of a residual analysis algorithm according to an embodiment of this application, as follows: Figure 4 As shown, residual calculation is performed on the flattened image (i.e., the target stripe image) and the stripe principal structure image, and binarization segmentation is performed through the residual threshold. Then, connected component analysis is performed, and connected components whose geometric features satisfy the linear condition are taken as linear residual regions. Connected components that do not satisfy the linear condition are removed, and finally, linear residuals are obtained.

[0164] The stripe image denoising method provided in this embodiment acquires the original stripe image and automatically estimates its original stripe direction, then rotates it to a preset target direction and performs background flattening processing. This effectively eliminates non-uniform background interference and can process stripe images with different directions or rotation postures. It combines the geometric features of the target stripe image with morphological filtering methods to enhance the direction adaptability. Furthermore, it uses multi-scale linear structuring elements consistent with the target stripe direction to perform morphological opening operations on the target stripe image. By extracting and fusing details and main features through structuring elements of different scales, it can effectively suppress point and block noise in strong noise backgrounds while maintaining the continuity and integrity of the stripe structure, avoiding stripe breakage, blurring, or widening. In addition, by calculating the residual between the target stripe image and the main stripe structure image, and adaptively determining the residual threshold based on the median absolute deviation, and combining connected component geometric features to filter linear residual regions, it selectively adds weak stripe signals that meet the linear condition back to the main structure image. This can restore the details of weak stripes that have been over-suppressed while denoising, avoiding signal loss.

[0165] This embodiment also provides a stripe image denoising device, which is used to implement the above embodiments and preferred embodiments, and will not be repeated as described herein. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0166] This embodiment provides a stripe image denoising device, such as Figure 5 As shown, it includes:

[0167] The target image module 501 is used to acquire the target stripe image to be denoised and to acquire the target stripe direction corresponding to the target stripe image.

[0168] The stripe main structure module 502 is used to perform morphological opening operations on the target stripe image using preset linear structuring elements to determine the stripe main structure image. The linear structuring element is a structuring element that has multiple pixels in the linear direction and a single pixel in the direction perpendicular to the linear direction, with the linear direction consistent with the direction of the target stripes.

[0169] The denoising image module 503 is used to determine the final denoised image based on the stripe principal structure image.

[0170] In some alternative implementations, the target image module 501 includes:

[0171] The original image submodule is used to acquire the original stripe image and obtain the original stripe direction corresponding to the original stripe image.

[0172] The angle determination submodule is used to determine the angle between the original fringe direction and the target fringe direction. Based on the angle, the original fringe image is rotated to determine the intermediate fringe image. The fringe direction corresponding to the intermediate fringe image is the target fringe direction.

[0173] The target image submodule is used to determine the target stripe image based on the intermediate stripe image.

[0174] In some alternative implementations, the original image submodule includes:

[0175] A segmentation unit is used to segment the original stripe image and determine at least one stripe region corresponding to the original stripe image.

[0176] The orientation determination unit is used to determine the orientation of the stripes in each stripe region and, based on the orientation of the stripes in each region, to determine the orientation of the original stripe image.

[0177] In some alternative implementations, the segmentation unit includes:

[0178] The projection subunit is used to project the original stripe image according to a preset direction and determine the projection curve corresponding to the original stripe image. The projection curve is used to represent the pixel projection value corresponding to each pixel position point in the vertical direction of the preset direction.

[0179] Local maximum sub-units are used to determine local maxima points in the projected curve.

[0180] The minimum value sub-unit is used to determine two local minimum values ​​located on both sides of any local maximum value within a preset range corresponding to the local maximum value.

[0181] The striped region sub-unit is used to determine the corresponding striped region based on the positions of two local minima as boundary points.

[0182] In some optional implementations, the direction determination unit includes:

[0183] The transformation sub-unit is used to perform Radon transformation on any stripe region to determine the projection value of the stripe region under multiple projection angles, and take the projection angle corresponding to the largest projection value as the region stripe angle corresponding to the stripe region.

[0184] The effective region sub-unit is used to identify stripe regions whose stripe angles meet preset angle conditions as effective stripe regions.

[0185] The direction determination subunit is used to perform a weighted average of the fringe angles corresponding to each effective fringe region to determine the original fringe direction corresponding to the original fringe image.

[0186] In some alternative implementations, the target image submodule includes:

[0187] The filtering unit is used to filter the image of the middle stripes and determine the background image corresponding to the image of the middle stripes.

[0188] The target image unit is used to subtract the background image from the intermediate stripe image to determine the target stripe image.

[0189] In some optional implementations, the number of linear structural elements is multiple, and each linear structural element corresponds to a different scale. The striped main structure module 502 includes:

[0190] The morphological opening operation submodule is used to perform morphological opening operations at corresponding scales on the target stripe image using multiple linear structuring elements to determine the stripe substructure image corresponding to each linear structuring element.

[0191] The fusion submodule is used to fuse the images of the various stripe substructures to determine the main stripe structure image.

[0192] In some alternative implementations, the denoising image module 503 includes:

[0193] The linear residual determination submodule is used to calculate the linear residual between the target stripe image and the stripe principal structure image.

[0194] The enhanced image submodule is used to add linear residuals to the stripe master structure image to obtain the stripe enhanced image.

[0195] The denoising image submodule is used to determine the final denoised image based on the stripe enhancement image.

[0196] In some optional implementations, the linear residual determination submodule includes:

[0197] The residual image unit is used to calculate the difference between the target stripe image and the stripe principal structure image to obtain the residual image.

[0198] The residual threshold unit is used to determine the standard deviation of the residual image based on the absolute deviation of the median, and multiplies the standard deviation by a preset residual threshold coefficient to determine the residual threshold.

[0199] The binarization unit is used to perform binarization thresholding on the residual image based on the residual threshold to determine the binarized residual image.

[0200] The connected component analysis unit is used to perform connected component analysis on the binarized residual image and extract the geometric features of each connected component.

[0201] Linear residual units are used to define connected regions whose geometric features satisfy linear conditions as linear residual regions. Linear residuals are determined based on the regions in the residual image that overlap with the linear residual regions.

[0202] The stripe image denoising apparatus provided in this application can execute the stripe image denoising method provided in any embodiment of this application, and has the corresponding functional modules and beneficial effects for executing the method. Further functional descriptions of the above modules and units are the same as those in the corresponding embodiments described above, and will not be repeated here.

[0203] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0204] The following is a detailed reference. Figure 6 This diagram illustrates a suitable structural schematic for implementing the electronic device described in the embodiments of this application. The electronic device may include a processor (e.g., a central processing unit, graphics processor, etc.) 601, which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 602 or a program loaded from memory 608 into random access memory (RAM) 603. The RAM 603 also stores various programs and data required for the operation of the electronic device. The processor 601, ROM 602, and RAM 603 are interconnected via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0205] Typically, the following devices can be connected to I / O interface 605: input devices 606 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 607 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; memory devices 608 including, for example, magnetic tapes, hard disks, etc.; and communication devices 609. Communication device 609 allows electronic devices to communicate wirelessly or wiredly with other devices to exchange data. Although... Figure 6 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown, and more or fewer devices may be implemented or have instead.

[0206] Specifically, according to embodiments of this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this application include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 609, or installed from a memory 608, or installed from a ROM 602. When the computer program is executed by the processor 601, it performs the functions defined in the stripe image denoising method of embodiments of this application.

[0207] Figure 6 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0208] This application also provides a computer-readable storage medium. The methods described above according to this application can be implemented in hardware or firmware, or implemented as recordable on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and subsequently stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc. Further, the storage medium may also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the stripe image denoising method shown in the above embodiments is implemented.

[0209] A portion of this application can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to this application through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0210] Although embodiments of this application have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of this application, and all such modifications and variations fall within the scope defined by the appended claims.

Claims

1. A method for denoising striped images, characterized in that, The method includes: Obtain the target stripe image to be denoised, and obtain the target stripe direction corresponding to the target stripe image; The target stripe image is subjected to morphological opening operation by a preset linear structuring element to determine the main stripe structure image; the linear structuring element is a structuring element that is multi-pixel in the linear direction and single-pixel in the direction perpendicular to the linear direction, and the linear direction is consistent with the direction of the target stripe. The final denoised image is determined based on the stripe principal structure image; The step of acquiring the target stripe image to be denoised includes: Obtain the original stripe image and obtain the original stripe direction corresponding to the original stripe image; Determine the angle between the original stripe direction and the target stripe direction, rotate the original stripe image according to the angle, and determine the intermediate stripe image; the stripe direction corresponding to the intermediate stripe image is the target stripe direction; The target stripe image is determined based on the intermediate stripe image; The linear structuring elements are multiple, and each linear structuring element corresponds to a different scale. The step of performing morphological opening operations on the target stripe image using preset linear structuring elements to determine the main stripe structure image includes: By performing morphological opening operations at corresponding scales on the target stripe image using multiple linear structuring elements, the stripe substructure image corresponding to each linear structuring element is determined. The stripe substructure images are fused together to determine the main stripe structure image; Determining the final denoised image based on the stripe principal structure image includes: Calculate the linear residual between the target stripe image and the stripe principal structure image; The linear residual is added to the stripe master structure image to obtain a stripe enhancement image; The final denoised image is determined based on the stripe enhancement image.

2. The method according to claim 1, characterized in that, The step of obtaining the original stripe direction corresponding to the original stripe image includes: The original stripe image is segmented to determine at least one stripe region corresponding to the original stripe image; Determine the stripe direction corresponding to each of the stripe regions, and determine the original stripe direction corresponding to the original stripe image based on the stripe direction of each of the stripe regions.

3. The method according to claim 2, characterized in that, The step of segmenting the original stripe image to determine at least one stripe region corresponding to the original stripe image includes: The original stripe image is projected according to a preset direction to determine the projection curve corresponding to the original stripe image; the projection curve is used to represent the pixel projection value corresponding to each pixel position point in the direction perpendicular to the preset direction. Identify the local maxima in the projected curve; For any given local maximum point, within a preset range corresponding to the local maximum point, determine two local minimum points located on either side of the local maximum point; and, The corresponding stripe region is determined based on the positions of the two local minimum points as boundary points; And / or, The step of determining the stripe direction corresponding to each of the stripe regions, and determining the original stripe direction corresponding to the original stripe image based on the stripe direction of each of the stripe regions, includes: For any one of the striped regions, perform a Radon transform to determine the projection value of the striped region under multiple projection angles, and take the projection angle corresponding to the largest projection value as the region stripe angle corresponding to the striped region. Stripe regions whose stripe angles meet preset angle conditions are defined as valid stripe regions; and... The original stripe direction corresponding to the original stripe image is determined by weighted averaging of the stripe angles corresponding to each of the effective stripe regions.

4. The method according to claim 1, characterized in that, Determining the target stripe image based on the intermediate stripe image includes: The intermediate stripe image is filtered to determine the background image corresponding to the intermediate stripe image; The target stripe image is determined by subtracting the intermediate stripe image from the background image.

5. The method according to claim 1, characterized in that, The calculation of the linear residual between the target stripe image and the stripe principal structure image includes: The difference between the target stripe image and the stripe principal structure image is calculated to obtain the residual image; The standard deviation of the residual image is determined based on the absolute deviation of the median, and the standard deviation is multiplied by a preset residual threshold coefficient to determine the residual threshold. The residual image is binarized and thresholded based on the residual threshold to determine the binarized residual image; Connectivity analysis is performed on the binarized residual image to extract the geometric features of each connected component; The connected regions that satisfy the linear condition for the geometric features are taken as linear residual regions, and the linear residuals are determined based on the regions in the residual image that overlap with the linear residual regions.

6. A stripe image denoising device, characterized in that, The device includes: The target image module is used to acquire the target stripe image to be denoised, and to acquire the target stripe direction corresponding to the target stripe image; The stripe main structure module is used to perform morphological opening operations on the target stripe image using preset linear structural elements to determine the stripe main structure image; the linear structural element is a structural element that is multi-pixel in the linear direction and single-pixel in the direction perpendicular to the linear direction, and the linear direction is consistent with the direction of the target stripe; A denoising image module is used to determine the final denoised image based on the stripe master structure image; The target image module includes: The original image submodule is used to acquire the original stripe image and obtain the original stripe direction corresponding to the original stripe image; Angle determination submodule is used to determine the angle between the original stripe direction and the target stripe direction, rotate the original stripe image according to the angle, and determine the intermediate stripe image; the stripe direction corresponding to the intermediate stripe image is the target stripe direction; A target image submodule is used to determine the target stripe image based on the intermediate stripe image; The striped main structure module includes: The morphological opening operation submodule is used to perform morphological opening operations at corresponding scales on the target stripe image using multiple linear structuring elements to determine the stripe substructure image corresponding to each linear structuring element; the number of linear structuring elements is multiple, and each linear structuring element corresponds to a different scale; The fusion submodule is used to fuse the various stripe substructure images to determine the stripe main structure image; The denoising image module includes: The linear residual determination submodule is used to calculate the linear residual between the target stripe image and the stripe master structure image; An enhanced image submodule is used to add the linear residual to the stripe master structure image to obtain a stripe enhanced image; A denoising image submodule is used to determine the final denoised image based on the stripe enhancement image.

7. An electronic device, characterized in that, include: A memory and a processor are communicatively connected, the memory stores computer instructions, and the processor executes the computer instructions to perform the stripe image denoising method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing a computer to perform the stripe image denoising method according to any one of claims 1 to 5.

Images