Image processing method and device

Abstract

The invention discloses an image processing method and device, and is applied to the technical field of image processing. The method comprises the steps of obtaining an initial stripe image corresponding to an original image based on the original image; performing threshold compensation on the initial stripe image to obtain a target stripe image; wherein the threshold compensation is used for adjusting the stripe amplitude of pixels in the initial stripe image; and performing stripe removal on the original image by using the target stripe image to obtain a target image. The method can ensure that the image still keeps high detail definition and overall quality after stripe noise is removed.

Description

Technical Field

[0001] The present application belongs to the field of image processing technology, and in particular, to an image processing method and device. Background Art

[0002] With the continuous advancement of medical imaging technology, the requirements for imaging systems have also increased. In this context, X-ray imaging systems are inevitably affected by various electromagnetic interferences during operation. These interferences will be superimposed on the X-ray images, forming stripe artifacts, thereby interfering with the display of image details.

[0003] In the existing technology, the variational model, frequency filtering and other methods are mainly used for processing. Although these methods can remove stripe noise to a certain extent, they are also easy to introduce ringing artifacts, which causes the information in the image close to the stripe frequency to be destroyed. This information is crucial to the overall quality of the image and subsequent analysis. Once destroyed, the details of the image will become blurred or distorted, thus affecting the accuracy of the image information. Summary of the invention

[0004] The embodiments of the present application provide an image processing method and device, which can ensure that the image still maintains high detail clarity and overall quality after removing stripe noise.

[0005] In a first aspect, an embodiment of the present application provides an image processing method, comprising: based on an original image, obtaining an initial stripe image corresponding to the original image; performing threshold compensation on the initial stripe image to obtain a target stripe image; wherein the threshold compensation is used to adjust the stripe amplitude of pixels in the initial stripe image; and using the target stripe image to remove stripes from the original image to obtain a target image.

[0006] In a possible implementation of the first aspect, threshold compensation is performed on an initial stripe image to obtain a target stripe image, including: acquiring a target area of ​​the original image, the target area being an area in the original image where a noise value exceeds a preset value; calculating a stripe amplitude threshold of a pixel based on the target area; and threshold compensation is performed on the initial stripe image according to the stripe amplitude threshold to obtain a target stripe image.

[0007] In a possible implementation of the first aspect, a fringe amplitude threshold of a pixel is calculated based on a target area, including: constructing a mean curve of the target area according to the stripe direction in an initial fringe image, the mean curve being constructed by the mean of target quality parameter values ​​in the stripe direction of the target area, the target quality parameter value being a grayscale value or a contrast value; constructing an envelope of the mean curve based on the mean curve; and calculating the fringe amplitude threshold of the pixel according to the envelope.

[0008] In a possible implementation of the first aspect, threshold compensation is performed on an initial fringe image according to a fringe amplitude threshold of a pixel to obtain a target fringe image, including: performing soft threshold shrinkage processing on the initial fringe image using the fringe amplitude threshold of a pixel to obtain a target fringe image.

[0009] In a possible implementation of the first aspect, a soft threshold shrinkage process is performed on an initial stripe image using a stripe amplitude threshold of a pixel to obtain a target stripe image, including: traversing all pixels in the initial stripe image to determine at least one pixel whose stripe amplitude is greater than the stripe amplitude threshold; for each pixel in the at least one pixel, shrinking the stripe amplitude of the pixel based on the stripe amplitude threshold to obtain a target stripe amplitude of the pixel; and updating the corresponding pixel in the initial stripe image based on the target stripe amplitude to obtain the target stripe image.

[0010] In a possible implementation manner of the first aspect, pixels in a column of the initial fringe image in the fringe direction correspond to the same target fringe amplitude.

[0011] In a possible implementation of the first aspect, based on an original image, an initial stripe image corresponding to the original image is obtained, including: obtaining multiple initial component images of the original image; the multiple initial component images include an initial component image in the stripe direction; performing stripe denoising on the initial component image in the stripe direction to obtain a target component image in the stripe direction; performing image reconstruction on the target component image in the stripe direction and other initial component images except the initial component image in the stripe direction to obtain a reconstructed image; and obtaining the initial stripe image using the reconstructed image and the original image.

[0012] In a possible implementation of the first aspect, stripe denoising is performed on an initial component image in the stripe direction to obtain a target component image in the stripe direction, including: acquiring high grayscale areas and low grayscale areas of the original image; using the high grayscale areas and low grayscale areas to calculate filter kernel parameters; and using the filter kernel parameters to filter the initial component image in the stripe direction to obtain a target component image in the stripe direction.

[0013] In a possible implementation of the first aspect, multiple initial component images of an original image are obtained, including: performing multi-scale decomposition on the original image to obtain multiple initial component images; the multiple initial component images include an approximate component image and at least one directional component image, the at least one directional component image includes an initial component image in a stripe direction, and the frequency of the at least one directional component image is higher than the frequency of the approximate component image.

[0014] In a second aspect, an embodiment of the present application provides an image processing device, the device comprising:

[0015] The acquisition module is used to acquire an initial fringe image corresponding to the original image based on the original image.

[0016] The threshold compensation module is used to perform threshold compensation on the initial fringe image to obtain a target fringe image; wherein the threshold compensation is used to adjust the fringe amplitude of pixels in the initial fringe image.

[0017] The stripe removal module is used to remove stripes from the original image using the target stripe image to obtain the target image.

[0018] In a third aspect, an embodiment of the present application provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, an image processing method as described in any one of the first aspects above is implemented.

[0019] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the image processing method as described in any one of the above-mentioned first aspects is implemented.

[0020] In a fifth aspect, an embodiment of the present application provides a computer program product. When the computer program product is executed on a computer device, the computer device executes the image processing method described in any one of the above-mentioned first aspects.

[0021] It can be understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here.

[0022] Compared with the prior art, the embodiments of the present application have the following beneficial effects: the present application adjusts the fringe amplitude by threshold compensation on the initial fringe image, so as to adjust the fringe amplitude in the initial fringe image to within a preset amplitude range, thereby achieving the purpose of optimizing the fringe image and making it more suitable for removing fringe from the original image. Through the threshold compensation process, the fringe noise can be effectively reduced while protecting the important information close to the fringe frequency in the original image, so that the target fringe image obtained is closer to the real fringe noise characteristics. So that the target image obtained by removing fringe from the original image using the target fringe image can retain the important information close to the fringe frequency to the maximum extent while removing the fringe noise. So as to ensure that the target image still maintains a high detail clarity and overall quality after removing the fringe noise, it is helpful to improve the accuracy of medical diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0024] Figure 1 is a structural schematic diagram of an image processing system provided by an embodiment of the present application;

[0025] Figure 2 It is a flowchart of an image processing method provided by an embodiment of the present application;

[0026] Figure 3 This is a schematic diagram of an envelope effect provided by another embodiment of the present application;

[0027] Figure 4 This is a schematic diagram comparing the effects of a threshold compensation provided by an embodiment of the present application;

[0028] Figure 5 This is a schematic diagram of a process for obtaining a reconstructed image provided by an embodiment of the present application;

[0029] Figure 6 This is a schematic diagram for comparing the effects of a filter kernel parameter provided by an embodiment of the present application;

[0030] Figure 7 It is a flowchart of an image processing method provided by an embodiment of the present application;

[0031] Figure 8 is a structural schematic diagram of an image processing device provided in an embodiment of the present application;

[0032] Fig. 9 It is a structural diagram of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0033] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0034] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0035] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0036] As used in the specification and appended claims of this application, the term "if" may be interpreted as "when" or "upon" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if [described condition or event] is detected" may be interpreted as meaning "upon determination" or "in response to determining" or "upon detection of [described condition or event]" or "in response to detecting [described condition or event]", depending on the context.

[0037] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0038] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0039] During the scanning process, the X-ray imaging system will inevitably be affected by various electromagnetic interferences, resulting in the presence of horizontal or vertical stripes in the reconstructed image, which is the so-called stripe noise. The presence of stripe noise will greatly reduce the image quality and is not conducive to the subsequent analysis of the image. Therefore, how to effectively remove image stripe noise has become an urgent problem to be solved in the field of medical imaging technology. In the prior art, while removing stripe noise, it is also easy to introduce ringing artifacts, resulting in the destruction of information in the image that is close to the stripe frequency. This information is crucial to the overall quality and subsequent analysis of the image. Once destroyed, the details of the image will become blurred or distorted, thereby affecting the accuracy of the image information.

[0040] Based on this, an embodiment of the present application provides an image processing method and device, the technical principle of which is: first, the original image is decomposed at multiple scales to obtain multiple initial component images. These initial component images include not only approximate component images, but also at least one directional component image, and one of the at least one directional component images is specific to the stripe direction. Next, filtering is performed on the initial component image in the stripe direction. Then, based on the directional component image that has been filtered, the original image is reconstructed to generate a reconstructed image, thereby achieving preliminary removal of stripe noise in the original image. Before filtering, the filter kernel parameters used are determined based on the low grayscale area and high grayscale area in the original image.

[0041] On this basis, the initial fringe image is extracted from the original image based on the reconstructed image, and the initial fringe image is threshold compensated to ensure that its fringe amplitude can be adjusted to a preset range, thereby obtaining the target fringe image. Subsequently, the target fringe image is used to perform fringe removal processing on the original image, so that the obtained target image can effectively remove the fringe noise while retaining important information close to the fringe frequency to the maximum extent. In this way, even after removing the fringe noise, the target image can still maintain a high level of detail clarity and overall quality, thereby helping to improve the accuracy of medical diagnosis.

[0042] To facilitate the use of this embodiment, see Figure 1 The architecture of the image processing system 100 shown in the figure includes a computer device 101 and an imaging device 102 .

[0043] The imaging device 102 can scan the target object (scanning object) in the detection area or the scanning area to obtain the scanned data of the target object. For example, the imaging device 102 can use high-energy rays (such as X-rays, etc.) to scan the target object to collect scanned data related to the target object, such as a three-dimensional image. The target object may include biological or non-biological objects. As an example only, the target object may include a patient, an artificial object (such as an artificial phantom), etc. For another example, the target object may include a patient's part to be examined, such as an organ and / or tissue (such as the head, ear, nose, mouth, neck, chest, abdomen, liver, gallbladder, pancreas, spleen, kidney, spine, heart or tumor tissue, etc.).

[0044] Furthermore, the imaging device 102 can perform image reconstruction based on the raw data obtained by scanning the target object, obtain initial full-angle images and initial limited-angle images corresponding to the multiple continuous time phases, and remove image artifacts from the initial full-angle images and initial limited-angle images corresponding to the multiple continuous time phases to obtain corrected images, so as to transmit the corrected images to the computer device 101.

[0045] In some embodiments, the imaging device 102 may include a single-modality scanning device and / or a multi-modality scanning device. The single-modality scanning device may be a computed tomography (CT) device, a digital radiography (DR) device, a magnetic resonance imaging (MRI) device, a positron emission tomography-computed tomography (PET-CT), or a positron emission tomography-magnetic resonance imaging (PET-MRI). The multi-modality scanner may include any combination of the multi-modalities of the above-mentioned devices. The embodiment of the present application is described by taking a CT device as an example.

[0046] As an example, a CT device may include a computer tomography scanner and a scanning bed coupled to the computer tomography scanner, wherein the computer tomography scanner forms a scanning cavity, the scanning bed is used to support the detection object, and the scanning bed is movable and moves the detection object into the scanning cavity.

[0047] The computer tomography scanner includes an X-ray source module and a detector module. The X-ray source module may include an X-ray tube and associated control circuits. The X-ray tube may generate X-rays during a computer tomography (CT) process. The X-ray source may be of various types, including, for example, a rotating anode X-ray tube or a fixed anode X-ray tube. The detector module may generate absorption information of X-rays in the X, Y or Z directions. The X-ray source module may emit X-ray signals to an object of interest.

[0048] In some embodiments, the function, size, type, geometry, position, amount, and / or amplitude of the X-ray source module and / or the detector module may be determined or changed according to one or more specific conditions. In some embodiments of the present application, the detector module may include a scintillation detector, a semiconductor detector, etc. In some embodiments, the X-ray source module and the detector module may be designed to surround the object to form a tunnel type or an open type.

[0049] In an exemplary solution, the computer device 101 may have a general or special computing device environment or configuration. For example: a personal computer, a server computer, a handheld device or a portable device, a tablet device, a multi-processor device, a distributed computing environment including any of the above devices or devices, etc. The computer device 101 may have different names, such as user equipment (UE), access device, terminal unit, terminal station, mobile station, mobile station, remote station, remote terminal, mobile device, wireless communication device, terminal agent or terminal device, etc.

[0050] Furthermore, the computer device 101 may run an operating system, which may include but is not limited to Android system, IOS system, Linux system, Unix, Windows system, etc., and may also include a user interface (UI) layer, which can provide external display of image processing results through the UI layer.

[0051] In one possible implementation, a communication connection is established between the computer device 101 and the imaging device 102 via a wired or wireless network. Optionally, the wireless network or wired network uses standard communication technology and / or protocols. The network is typically the Internet, but may be any network, including but not limited to a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a mobile, wired or wireless network, a private network or any combination of a virtual private network.

[0052] For ease of understanding and explanation, the following Figures 2 to 7 The posture detection method, device and computer equipment provided by the embodiments of the present application are described in detail.

[0053] Figure 2 FIG. 1 is a flow chart of an image processing method according to an embodiment of the present application, which can be executed by an imaging device 102 or a computer device 101. Figure 2 As shown, the method includes:

[0054] 201. Based on the original image, obtain an initial fringe image corresponding to the original image.

[0055] The direction of the stripes in the initial stripe image is related to the direction of the stripe noise in the original image.

[0056] Optionally, the stripe noise appears in the image as noise with certain periodicity, directionality and stripe distribution. The directionality of the stripe noise is usually horizontal or vertical, and it may also appear at any other angle. It is understandable that horizontal and vertical are two relative directions in the image. When the horizontal is determined, the direction perpendicular to it is vertical; or when the vertical is determined, the direction perpendicular to it is horizontal. Any other angle can be recorded as oblique, and the embodiment of the present application does not impose any limitation on this.

[0057] The direction of the stripes in the initial stripe image is consistent with the direction of the stripe noise in the original image. This direction may be horizontal, vertical or of course, if there are stripe noises in multiple directions in the original image, there are multiple corresponding stripe directions.

[0058] Optionally, before obtaining the initial stripe image of the original image in the stripe direction, it also includes: determining the direction of the stripe noise in the original image to determine the stripe direction; the stripe noise can be divided into the following directions: horizontal stripe noise, vertical stripe noise, and oblique stripe noise at any angle.

[0059] Specifically, the stripe noise in the original image may include at least one of horizontal stripe noise, vertical stripe noise, and oblique stripe noise at any angle (such as diagonal stripe noise). Accordingly, the initial stripe image may be an image including at least one of horizontal stripes, vertical stripes, and oblique stripes at any angle (such as diagonal stripes). As an example, if the stripe noise in the original image includes horizontal stripe noise, then the initial stripe image may be a horizontal stripe image; if the stripe noise in the original image includes horizontal stripe noise and vertical stripe noise, then the initial stripe image may be a comprehensive image including horizontal stripes and vertical stripes.

[0060] In actual applications, the operation of electrophysiological equipment will generate electromagnetic radiation, which will interfere with nearby detectors. The directionality of electromagnetic interference usually depends on the layout of the interference source, the distribution of the electromagnetic field, and the sensitivity of the interfered device. Under certain conditions, if there is a certain spatial relationship between the interference source and the interfered device (such as linear arrangement, vertical layout, etc.), it may be observed that the interference signal has a certain directionality. Based on current imaging equipment, in general, the direction of stripe noise is mostly vertical.

[0061] Optionally, the initial stripe image may refer to an image containing only stripe noise.

[0062] In some embodiments, an image with stripe noise, ie, an initial stripe image, can be determined based on an original image and a reconstructed image obtained by performing preliminary noise removal on the original image.

[0063] 202. Perform threshold compensation on the initial fringe image to obtain a target fringe image; wherein the threshold compensation is used to adjust the fringe amplitude of pixels in the initial fringe image.

[0064] In some embodiments, the threshold compensation is used to adjust the fringe amplitude of a pixel in the initial fringe image to within a preset amplitude range corresponding to the pixel.

[0065] Optionally, the fringe amplitude is a characteristic value of a pixel in the initial fringe image, that is, a pixel value.

[0066] It should be noted that the preset amplitude range corresponding to a pixel is limited to the pixel and is not applicable to all pixels of the initial fringe image.

[0067] For each pixel in the initial fringe image, a threshold compensation calculation is performed based on the relationship between the fringe amplitude of the pixel and the preset amplitude range. If the fringe amplitude of the pixel is not within the preset amplitude range, its value is adjusted according to its deviation degree so that it falls within the corresponding preset amplitude range.

[0068] In practical applications, not every pixel needs to be adjusted. If the fringe amplitude of a pixel is within the corresponding preset amplitude range, no adjustment is required.

[0069] 203. Remove stripes from the original image using the target stripe image to obtain a target image.

[0070] Optionally, the target image is obtained by subtracting the original image from the target stripe image; that is, the image obtained by subtracting the pixel data of the original image and the target stripe image at the same pixel position is determined as the target image.

[0071] In some embodiments, the pixel positions in the target stripe image are processed with negative values, and the processed target stripe image is summed with the original image to obtain the target image; that is, the image obtained by adding the pixel data of the processed target stripe image and the original image at the same pixel position is determined as the target image.

[0072] The embodiment of the present application adjusts the fringe amplitude by threshold compensation on the initial fringe image, so as to adjust the fringe amplitude in the initial fringe image to within a preset amplitude range, thereby achieving the purpose of optimizing the fringe image and making it more suitable for removing fringe from the original image. The threshold compensation process can effectively reduce the fringe noise while protecting the important information close to the fringe frequency in the original image, so that the target fringe image obtained is closer to the real fringe noise characteristics. So that the target image obtained by removing fringe from the original image using the target fringe image can retain the important information close to the fringe frequency to the maximum extent while removing the fringe noise. It is ensured that the target image still maintains a high detail clarity and overall quality after removing the fringe noise, which helps to improve the accuracy of medical diagnosis.

[0073] In one embodiment of the present application, threshold compensation is performed on an initial stripe image to obtain a target stripe image, including: acquiring a target area of ​​the original image, the target area being an area in the original image where a noise value exceeds a preset value; calculating a stripe amplitude threshold of a pixel based on the target area; and threshold compensation is performed on the initial stripe image according to the stripe amplitude threshold to obtain a target stripe image.

[0074] Optionally, the fringe amplitude threshold is used to characterize a preset amplitude range corresponding to a pixel.

[0075] In some embodiments, the target area may be one of a low grayscale area, a high grayscale area, a high contrast area, and a low contrast area.

[0076] In practical applications, the specific area of ​​the target area can be determined according to the performance degree of the electromagnetic interference noise in the target area. For example, the preset value includes a first preset value and a second preset value, and the second preset value is greater than the first preset value; when the noise value is greater than or equal to the first preset value and less than the second preset value, the noise is more obvious in the low grayscale area or the low contrast area, and weaker in the high grayscale area or the high contrast area. In order to make the fringe amplitude threshold more accurate, the low grayscale area or the low contrast area can be selected as the target area. When the noise value is greater than or equal to the second preset value, one of the low grayscale area, the high grayscale area, the high contrast area and the low contrast area can be selected as the target area.

[0077] In a possible implementation, the signal amplitude of the low grayscale area is small, and the superposition of noise will cause the brightness of these areas to fluctuate greatly, so that the noise is more obvious visually. Therefore, the preset amplitude range of the pixel can be determined by the low grayscale area. In a possible implementation, when the target area is a low grayscale area, the initial stripe image is threshold compensated to obtain the target stripe image, including: obtaining a low grayscale area of ​​the original image, wherein the grayscale value in the low grayscale area is less than a preset grayscale threshold; calculating the stripe amplitude threshold of the pixel based on the low grayscale area, and performing threshold compensation on the initial stripe image according to the stripe amplitude threshold of the pixel to obtain the target stripe image.

[0078] Optionally, apply a threshold to determine which pixels belong to the low grayscale area. The threshold is a grayscale value below which pixels are considered low grayscale pixels. Choosing a suitable threshold is important and should be based on the specific content of the image and the definition of the desired low grayscale area. Use the result of the thresholding process to create a mask image where the pixel value of the low grayscale area is 1 (or some non-zero value) and the pixel value of other areas is 0.

[0079] In some embodiments, the threshold compensation method can adopt any of the three methods: soft threshold shrinkage processing, hard threshold shrinkage processing, or semi-soft threshold shrinkage processing, and the embodiments of the present application do not impose any limitations on this.

[0080] Optionally, the pixels in the columns of the initial stripe image in the stripe direction correspond to the same target stripe amplitude. It is understandable that when threshold compensation is performed on the pixels in the same column in the stripe direction in the initial stripe image, the stripe amplitude threshold corresponding to the column is adopted. It is understandable that the column direction in the initial stripe image is consistent with the stripe direction. For example, if the stripe direction is a vertical direction, the column direction in the initial stripe image is also a vertical direction. If the stripe direction is a horizontal direction, the column direction in the initial stripe image is also a horizontal direction.

[0081] In one embodiment of the present application, a stripe amplitude threshold of a pixel is calculated based on a target area, including: constructing a mean curve of the target area according to the stripe direction in an initial stripe image, the mean curve being constructed by the mean of target quality parameter values ​​in the stripe direction of the target area, the target quality parameter value being a grayscale value or a contrast value; constructing an envelope of the mean curve based on the mean curve; and calculating the stripe amplitude threshold of the pixel according to the envelope.

[0082] Optionally, the envelope of the mean value curve may be determined by using a root mean square method, a Hilbert transform method or an extreme value method.

[0083] As an example, determining the envelope of the mean value curve using the root mean square method includes: dividing the mean value curve into multiple curve segments, calculating the root mean square value of each curve segment, and connecting the root mean square values ​​of each curve segment to form an envelope.

[0084] As another example, using the Hilbert transform method to determine the envelope of the mean curve includes: performing a Hilbert transform on the mean curve to obtain an imaginary part of the mean curve, and calculating an instantaneous amplitude based on the mean curve and the imaginary part, where the instantaneous amplitude is the envelope of the mean curve.

[0085] As an example, using the extreme value method to determine the envelope of the mean curve includes: the envelope includes a first envelope and / or a second envelope, and the first envelope and the second envelope are respectively located on both sides of the mean curve. Further, the first envelope and / or the second envelope are constructed based on the mean curve, the first envelope is constructed by multiple local maxima in the mean curve, and the second envelope is constructed by multiple local minima in the mean curve; the fringe amplitude threshold is calculated according to the first envelope and / or the second envelope.

[0086] Optionally, multiple key points are selected from the first envelope, and an average value of the multiple key points is calculated, and the average value is determined as the fringe amplitude threshold. The calculation of the fringe amplitude threshold according to the second envelope is similar, and will not be repeated here.

[0087] Exemplarily, when the stripe direction is vertical, the mean curve of the target area is constructed according to the vertical direction in the initial stripe image. When the stripe direction is horizontal, the mean curve of the target area is constructed according to the horizontal direction in the initial stripe image.

[0088] In some embodiments, assuming that the target area is a low grayscale area, in order to capture the grayscale change trend of the low grayscale area in the stripe direction, and obtain the stripe amplitude threshold corresponding to each pixel accordingly, a method of constructing a value curve can be used. The mean curve can clearly show the trend of grayscale change, and provide a strong basis for the subsequent calculation of the stripe amplitude threshold. Therefore, the stripe amplitude threshold of the pixel is calculated based on the low grayscale area, including: constructing a mean curve of the low grayscale area according to the stripe direction, the mean curve is constructed by the grayscale mean of the low grayscale area in the stripe direction; constructing a first envelope and a second envelope based on the mean curve, the first envelope is constructed by multiple local maxima in the mean curve, and the first envelope is constructed by multiple local minima in the mean curve; the stripe amplitude threshold of the pixel is calculated according to the first envelope and / or the second envelope. It should be noted that the same is true when the target area is a low contrast area, a high grayscale area or a high contrast area, which will not be repeated here.

[0089] Optionally, in the stripe direction, for the pixels of a column, the grayscale values ​​are sorted from large to small, and the first N grayscale values ​​in the column are taken, and the average of the first N grayscale values ​​is calculated as the grayscale mean of the column. The value N can be determined based on a preset percentage; for example, if the preset percentage is taken as the first 20%, assuming that the number of pixels in a column is 100, the corresponding N is 20.

[0090] Optionally, the mean curve is divided into one or more columns to obtain multiple local curves; the first envelope is a curve formed by sequentially connecting the maximum points (i.e., the maximum points) of the local curves; the second envelope is a curve formed by sequentially connecting the minimum points of the local curves. The line connecting the maximum points is also called the "upper envelope", and the line connecting the minimum points is also called the "lower envelope".

[0091] In some embodiments, the upper envelope or the lower envelope is obtained by determining the maximum and minimum points in multiple local curves and then interpolating the maximum or minimum points of the multiple local curves. The interpolation can be performed using interpolation methods such as linear interpolation, polynomial interpolation, spline interpolation, etc. The choice of interpolation method depends on your specific needs and the characteristics of the data. As an example, refer to Figure 3 , which is a schematic diagram of the effects of an upper envelope, a lower envelope and a local curve.

[0092] Optionally, the fringe amplitude threshold is calculated based on the first envelope and the second envelope of the mean curve. The calculation formula is as follows:

[0093]

[0094] Where nThreshiold represents the fringe amplitude threshold; y up represents the first envelope, y low represents the second envelope; j represents the number of columns of low grayscale areas, j=1,…,N, and N represents the number of image columns.

[0095] In this embodiment, by using the method of constructing the envelope curve using local maximum and minimum values, the interference of image noise can be effectively reduced, thereby improving the robustness of the algorithm. The design of the envelope curve is intended to highlight the main trends and features of the mean curve, while filtering out unnecessary noise and details. More importantly, using the information contained in the envelope curve, the fringe amplitude threshold of each pixel can be dynamically calculated according to the local features of the image. This adaptive method not only improves the accuracy of fringe detection, but also enhances the algorithm's adaptability to different grayscale changes, making fringe detection more robust and reliable.

[0096] In one embodiment of the present application, in order to improve the adaptability of threshold compensation, the initial fringe image is threshold compensated by soft threshold shrinkage. Therefore, the initial fringe image is threshold compensated according to the fringe amplitude threshold of the pixel to obtain the target fringe image, including: using the fringe amplitude threshold of the pixel to perform soft threshold shrinkage processing on the initial fringe image to obtain the target fringe image.

[0097] In a possible implementation, hard threshold processing is performed on the initial fringe image using a fringe amplitude threshold of a pixel to obtain a target fringe image.

[0098] Optionally, all pixels in the initial stripe image are traversed to determine at least one pixel whose stripe amplitude is greater than a stripe amplitude threshold; the stripe amplitude of at least one pixel in the initial stripe image is updated to a preset value (eg, 0) to obtain a target stripe image.

[0099] In another possible implementation, a pixel fringe amplitude threshold is used to perform semi-soft threshold processing on the initial fringe image to obtain a target fringe image.

[0100] Optionally, for pixels in at least one pixel, based on the pixel's fringe amplitude threshold, semi-soft threshold shrinkage processing is performed on the pixel's fringe amplitude to obtain the pixel's target fringe amplitude; and all target fringe amplitudes corresponding to at least one pixel are used to update the corresponding pixels in the initial fringe image to obtain the target fringe image.

[0101] Optionally, for the fringe amplitude x of the pixel in the i-th row and j-th column of the initial fringe image ij , if the fringe amplitude x ij The absolute value of is greater than the fringe amplitude threshold nThreshiold of the jth column j , then the processed fringe amplitude (i.e. target fringe amplitude) f(x ij ,nThreshiold j ) is calculated as follows:

[0102]

[0103] Among them, sign(x ij ) represents the sign function, when x ij Returns 1 when x is positive. ij If it is negative, it returns -1, and if it is zero, it returns 0. This calculation formula indicates that when the fringe amplitude x in the initial fringe image ij When the absolute value of is greater than the fringe amplitude threshold, the fringe amplitude x ij The absolute value of minus the fringe amplitude threshold nThreshiold jGet the fringe amplitude difference, and then get the fringe amplitude difference and (1+α*nThreshiold j ) ratio to shrink, and the fringe amplitude x ij The sign of is used as the positive or negative value of the target fringe amplitude. ij The absolute value of the fringe amplitude threshold nThreshiold is less than or equal to j , not the fringe amplitude x of the pixel ij No processing: In this way, semi-soft thresholding can shrink the pixels in the initial fringe image that are larger than the fringe amplitude threshold, while retaining the important features of the image. The parameter α controls the degree of shrinkage and can be adjusted according to the specific application scenario. ij ,nThreshiold j ) and then f(x ij ,nThreshiold j ) updates the fringe amplitude of the corresponding pixel in the initial fringe image, and uses the updated initial fringe image as the target fringe image.

[0104] Optionally, a soft threshold shrinkage process is performed on the initial stripe image using a stripe amplitude threshold of a pixel to obtain a target stripe image, including: traversing all pixels in the initial stripe image to determine at least one pixel whose stripe amplitude is greater than the stripe amplitude threshold; for each pixel in the at least one pixel, shrinking the stripe amplitude of the pixel based on the stripe amplitude threshold to obtain a target stripe amplitude of the pixel; and updating the corresponding pixel in the initial stripe image based on the target stripe amplitude to obtain the target stripe image.

[0105] Optionally, first, it is necessary to traverse each pixel in the initial stripe image and compare its amplitude with the stripe amplitude threshold, and determine all pixels whose stripe amplitude is greater than the stripe amplitude threshold, and these pixels will be used as objects for subsequent shrinkage processing.

[0106] Furthermore, the target fringe amplitude f(x ij ,nThreshiold j ):

[0107] f(x ij ,nThreshiold j )=sign(x ij )*max(|x ij |-nThreshiold j ,0)

[0108] Among them, sign(x ij ) represents the sign function, when x ij Returns 1 when x is positive.ij If it is negative, it returns -, and if it is zero, it returns 0. This calculation formula indicates that when the fringe amplitude x in the initial fringe image ij When the absolute value of is greater than the fringe amplitude threshold, the fringe amplitude x ij The absolute value of minus the fringe amplitude threshold nThreshiold j Get the stripe amplitude difference, and then shrink it by comparing it with 0. If it is greater than 0, take the stripe amplitude difference while retaining the stripe amplitude x ij The positive or negative value of the fringe amplitude x ij The difference between the positive and negative fringe amplitudes is used as the target fringe amplitude; if it is less than 0, 0 is taken as the target fringe amplitude, and 0 is used as the new fringe amplitude of the pixel in the i-th row and j-th column in the initial fringe image. ij The absolute value of the fringe amplitude threshold nThreshiold is less than or equal to j , not the fringe amplitude x of the pixel ij No action is taken.

[0109] When f(x ij ,nThreshiold j ) and then f(x ij ,nThreshiold j ) updates the fringe amplitude of the corresponding pixel in the initial fringe image, and uses the updated initial fringe image as the target fringe image.

[0110] It can be understood that after the soft threshold shrinkage process, the original stripe image X stripe All pixels in the image that are larger than the fringe amplitude threshold will be appropriately shrunk, thus obtaining a processed fringe image Y stripe (i.e., the target stripe image). The processed stripe image Y stripe While retaining the main features of the original stripes, unnecessary noise or details are reduced, making subsequent image analysis or processing more accurate and efficient.

[0111] Reference Figure 4 , Figure 4 (a) is the original image. Figure 4 (b) is the target image without threshold compensation. Figure 4 (c) is the target image after threshold compensation. Figure 4 (b) and Figure 4 By comparing with (c), it can be clearly seen that the method of the embodiment of the present application can retain the detailed information of the original image, that is, the guide wire 41 marked in the figure. Figure 4 The clarity of the guide wire 41 in (c) is significantly higher than Figure 4 The clarity of the guide wire 41 in (b).

[0112] In this embodiment, the fringe amplitude threshold can be dynamically calculated according to the characteristics of the original image, so as to smoothly shrink the fringe amplitude greater than the fringe amplitude threshold in the initial fringe image, thereby avoiding the generation of sharp edges or artifacts. This makes the method more flexible and effective when processing different data. By shrinking the fringe amplitude whose absolute value is greater than the fringe amplitude threshold toward zero, the soft threshold shrinkage can effectively weaken the noise in the fringe image, avoiding excessive fringe noise in the fringe image, resulting in poor subsequent fringe removal effect.

[0113] In one embodiment of the present application, in order to extract and separate the initial stripe noise in the original image and provide support for subsequent denoising, the original image can be decomposed to obtain an image corresponding to the direction of the stripe noise to be removed from the original image, and then a preliminary stripe removal and extraction process is performed based on the image to obtain an initial stripe image. Therefore, based on the original image, an initial stripe image corresponding to the original image is obtained, including: obtaining multiple initial component images of the original image; the multiple initial component images include an initial component image in the stripe direction; performing stripe denoising on the initial component image in the stripe direction to obtain a target component image in the stripe direction; reconstructing the target component image in the stripe direction and other initial component images except the initial component image in the stripe direction to obtain a reconstructed image; and obtaining the initial stripe image using the reconstructed image and the original image.

[0114] Optionally, a difference is made between the original image and the reconstructed image, that is, an image obtained by subtracting pixel data of the original image and the reconstructed image at the same pixel position is determined as the target image.

[0115] Optionally, multi-scale fusion is performed on the target component image in the stripe direction and other initial component images except the initial component image in the stripe direction to obtain a reconstructed image.

[0116] In some embodiments, obtaining multiple initial component images of an original image includes: performing multi-scale decomposition on the original image to obtain multiple initial component images; the multiple initial component images include an approximate component image and at least one directional component image, the at least one directional component image includes an initial component image in the direction of stripes, and the frequency of the at least one directional component image is higher than the frequency of the approximate component image.

[0117] Optionally, the multi-scale decomposition method may be an algorithm capable of performing directional decomposition on an image, for example, any one of wavelet transform, Curvelet transform or Gabor transform.

[0118] Optionally, when decomposing, a multi-scale decomposition method may be used to continuously decompose multiple times to obtain multiple initial component images. Specifically, the original image may be decomposed at multiple levels and scales to obtain multiple component images corresponding to the layers, the multiple component images including an approximate component image and at least one directional component image, and the multiple initial component images include the multiple component images corresponding to the layers.

[0119] Furthermore, the initial component image in the stripe direction of the layer is subjected to stripe denoising to obtain the target component image in the stripe direction of the layer. Then, a multi-level and multi-scale fusion method is used to reconstruct the target component image in the stripe direction and other initial component images except the initial component image in the stripe direction to obtain a reconstructed image.

[0120] As an example, taking wavelet transform as an example, suppose that the original image is subjected to five wavelet transforms to obtain multiple component images corresponding to the five layers. Figure 5 When the original image is subjected to the first wavelet transform, multiple component images of the first layer are obtained, which are approximate component image A11, horizontal component image H11, vertical component image V11 and diagonal component image D11; when the second wavelet transform is performed, the wavelet transform is performed on the approximate component image A11; multiple component images of the second layer are obtained, which are approximate component image A12, horizontal component image H12, vertical component image V12 and diagonal component image D12; when the third wavelet transform is performed, the wavelet transform is performed on the approximate component image A12; multiple component images of the third layer are obtained, which are approximate component image A13, horizontal component image H13, vertical component image V13 and diagonal component image D13; and so on, multiple component images of the fourth layer are obtained, which are approximate component image A14, horizontal component image H14, vertical component image V14 and diagonal component image D14; and multiple component images of the fifth layer are obtained, which are approximate component image A15, horizontal component image H15, vertical component image V15 and diagonal component image D15.

[0121] Based on the above, assuming that the vertical direction is the direction of the stripe noise in the original image, the vertical component image of the corresponding layer is subjected to stripe denoising to obtain the target component image of the layer. And using the inverse wavelet transform, multi-level and multi-scale fusion is performed to obtain the target image. Assuming that the horizontal direction is the direction of the stripe noise in the original image, the horizontal component image of the corresponding layer is subjected to stripe denoising to obtain the target component image of the layer. And using the inverse wavelet transform, multi-level and multi-scale fusion is performed to obtain the target image.

[0122] Specifically, refer to Figure 5, remove the stripe noise from the vertical component image V11 to obtain the target component image V21; remove the stripe noise from the vertical component image V12 to obtain the target component image V22; remove the stripe noise from the vertical component image V13 to obtain the target component image V23; remove the stripe noise from the vertical component image V14 to obtain the target component image V24; remove the stripe noise from the vertical component image V15 to obtain the target component image V25. Then, firstly, perform an inverse wavelet transform on the image of the fifth layer, that is, perform an inverse wavelet transform on the target component image V25, the approximate component image A15, the horizontal component image H15, and the diagonal component image D15 to obtain the approximate component image A24 of the fourth layer; then perform an inverse wavelet transform on the image of the fourth layer, that is, perform an inverse wavelet transform on the target component image V24, the approximate component image A14, the horizontal component image H14, and the diagonal component image D14 to obtain the approximate component image A23 of the third layer; then perform an inverse wavelet transform on the image of the third layer, that is, perform an inverse wavelet transform on the target component image V23, the approximate component image A 13. Perform inverse wavelet transform on the horizontal component image H13 and the diagonal component image D13 to obtain the approximate component image A22 of the second layer; then perform inverse wavelet transform on the image of the second layer, that is, perform inverse wavelet transform on the target component image V22, the approximate component image A12, the horizontal component image H12, and the diagonal component image D12 to obtain the approximate component image A21 of the first layer; then perform inverse wavelet transform on the image of the first layer, that is, perform inverse wavelet transform on the target component image V21, the approximate component image A11, the horizontal component image H11, and the diagonal component image D11 to obtain the reconstructed image.

[0123] It should be noted that the reconstructed image based on the curvelet transform or Gabor transform can refer to the above wavelet transform embodiment, which will not be repeated here. It can be understood that the curvelet transform and Gabor transform both have corresponding inverse transforms, namely the inverse curvelet transform and the inverse Gabor transform.

[0124] In one embodiment of the present application, considering that the dynamic range of X-ray images is large and the images contain abundant high-gradient features such as guidewire catheters, a filter width that is too narrow will cause obvious ringing, and a filter that is too wide will lose image information and cause image blur. Therefore, the filter kernel parameters of the filter are determined by using the high-grayscale area and the low-grayscale area of ​​the original image, so that the filter can filter the initial component image in the stripe direction at an appropriate width. Specifically, the initial component image in the stripe direction is subjected to stripe denoising to obtain a target component image in the stripe direction, including: obtaining the high-grayscale area and the low-grayscale area of ​​the original image; using the high-grayscale area and the low-grayscale area to calculate the filter kernel parameters; using the filter kernel parameters to filter the initial component image in the stripe direction to obtain a target component image in the stripe direction.

[0125] Optionally, the information of the original image is used to adaptively calculate the parameters of the Gaussian filter kernel. By calculating the grayscale threshold nThres, the image is divided into a high grayscale area ψ2 and a low grayscale area ψ1:

[0126] ψ1={z ij |z ij <nThres}

[0127] ψ2={z ij |z ij >nThres}

[0128] The filter kernel parameter σ is obtained by the following formula:

[0129]

[0130] Among them, "std" means standard deviation, z ij Represents the pixel value of the i-th row and j-th column in the original image.

[0131] Furthermore, for the initial component image V in the stripe direction i The filtering process includes: performing one-dimensional Fourier transform on the initial component image in the stripe direction, filtering with a Gaussian filter, and performing inverse Fourier transform to obtain a target component image V with stripe noise removed. i ′ .

[0132] The Gaussian filter is as follows:

[0133]

[0134] Among them, the filter kernel parameter σ determines the range of influence of the Gaussian filter in the frequency domain.

[0135] Optionally, refer to Figure 6 , Figure 6 (a) is the original image. Figure 6 (b) is the filtering effect of fixed filter kernel parameters. Figure 6 (c) is the filtering effect of the adaptive filter kernel parameters. Figure 6 (b) and Figure 6 By comparing with (c), it can be clearly seen that Figure 6 (c) is the filtering effect of adaptive filter kernel parameters is better than Figure 6 (b) is the filtering effect of the adaptive filter kernel parameters.

[0136] In this embodiment, the filter kernel parameters of the filter are determined by the high grayscale area and the low grayscale area of ​​the original image, and the filter parameters can be dynamically and adaptively adjusted according to the image features, thereby more effectively removing the noise of the initial component image in the stripe direction while retaining the details and edge features of the image. This method can not only remove noise in smooth areas, but also retain important structural information in edge areas.

[0137] For better understanding, refer to Figure 7 , the method provided in the embodiment of the present application is exemplarily described, assuming that a 5-order wavelet transform is used to perform multi-scale decomposition on the original image.

[0138] Specifically, the original image is first subjected to five wavelet transforms to obtain the approximate component image, horizontal component image, vertical component image and diagonal component image of the five layers; then, the vertical component image of the layer is subjected to a filtering process including Fourier transform, filtering and inverse Fourier transform to obtain the target component image. Then, the target component image and other initial component images except the initial component image in the stripe direction are subjected to inverse wavelet transform to obtain the reconstructed image; then, the original image and the reconstructed image are subjected to difference processing to obtain the initial stripe image, and the initial stripe image is subjected to threshold compensation to obtain the target stripe image; finally, the target stripe image is subjected to difference processing with the original image to obtain the target image.

[0139] Understandably, Figure 7 The above steps are only briefly described in Figure 7 The corresponding text description.

[0140] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0141] Corresponding to the image processing method described in the above embodiment, Figure 8 A structural block diagram of an image processing device provided in an embodiment of the present application is shown. For ease of explanation, only the parts related to the embodiment of the present application are shown.

[0142] Reference Figure 8 The data processing device includes: an acquisition module 801, a threshold compensation module 802 and a stripe removal module 803, wherein:

[0143] The acquisition module 801 is used to acquire an initial fringe image corresponding to the original image based on the original image.

[0144] The threshold compensation module 802 is used to perform threshold compensation on the initial fringe image to obtain a target fringe image; wherein the threshold compensation is used to adjust the fringe amplitude of the pixels in the initial fringe image.

[0145] The stripe removal module 803 is used to remove stripes from the original image using the target stripe image to obtain the target image.

[0146] In one possible implementation, the threshold compensation module 802 is specifically used to obtain a target area of ​​the original image, where the target area is an area in the original image where the noise value exceeds a preset value; a fringe amplitude threshold of a pixel is calculated based on the target area; and threshold compensation is performed on the initial fringe image according to the fringe amplitude threshold to obtain a target fringe image.

[0147] In one possible implementation, the threshold compensation module 802 is specifically used to construct a mean curve of the target area according to the stripe direction in the initial stripe image, where the mean curve is constructed by the mean of the target quality parameter values ​​in the target area in the stripe direction, where the target quality parameter value is a grayscale value or a contrast value; construct an envelope of the mean curve based on the mean curve; and calculate the stripe amplitude threshold of the pixel according to the envelope.

[0148] In a possible implementation, the threshold compensation module 802 is specifically configured to perform soft threshold shrinkage processing on the initial fringe image using the fringe amplitude threshold of the pixel to obtain the target fringe image.

[0149] In one possible implementation, the threshold compensation module 802 is specifically used to traverse all pixels in the initial stripe image, determine at least one pixel whose stripe amplitude is greater than the stripe amplitude threshold; for each pixel in the at least one pixel, based on the stripe amplitude threshold, shrink the stripe amplitude of the pixel to obtain a target stripe amplitude of the pixel; and update the corresponding pixel in the initial stripe image based on the target stripe amplitude to obtain a target stripe image.

[0150] In a possible implementation, pixels in a row of the initial fringe image in the fringe direction correspond to the same target fringe amplitude.

[0151] In a possible implementation, the acquisition module 801 is specifically used to acquire multiple initial component images of the original image; the multiple initial component images include an initial component image in the stripe direction; the initial component image in the stripe direction is subjected to stripe denoising to obtain a target component image in the stripe direction; the target component image in the stripe direction and other initial component images except the initial component image in the stripe direction are subjected to image reconstruction to obtain a reconstructed image; and the initial stripe image is obtained using the reconstructed image and the original image.

[0152] In one possible implementation, the acquisition module 801 is specifically used to acquire high grayscale areas and low grayscale areas of the original image; use the high grayscale areas and low grayscale areas to calculate the filter kernel parameters; use the filter kernel parameters to filter the initial component image in the stripe direction to obtain the target component image in the stripe direction.

[0153] In one possible implementation, the acquisition module 801 is specifically used to perform multi-scale decomposition on the original image to obtain multiple initial component images; the multiple initial component images include an approximate component image and at least one directional component image, the at least one directional component image includes an initial component image in the direction of the stripes, and the frequency of the at least one directional component image is higher than the frequency of the approximate component image.

[0154] It should be noted that the information interaction, execution process and other contents between the above-mentioned devices are based on the same concept as the method embodiment of the present application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0155] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0156] Reference below Fig. 9 , Fig. 9 A schematic diagram of the structure of a computer device suitable for implementing the embodiments of the present application is shown. Fig. 9 As shown, the computer device 900 includes a central processing unit (CPU) 901, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 902 or the program loaded from the storage part 908 into the random access memory (RAM) 903. In the RAM 903, various programs and data required for the operation instructions of the system are also stored. The CPU 901, the ROM 902 and the RAM 903 are connected to each other through a bus 904. An input / output (I / O) interface 905 is also connected to the bus 904.

[0157] The following components are connected to the I / O interface 905: an input section 906 including a keyboard, a mouse, etc.; an output section 907 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 908 including a hard disk, etc.; and a communication section 909 including a network interface card such as a LAN card, a modem, etc. The communication section 909 performs communication processing via a network such as the Internet. A drive 910 is also connected to the I / O interface 905 as needed. A removable medium 911, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 910 as needed so that a computer program read therefrom is installed into the storage section 908 as needed.

[0158] In particular, according to an embodiment of the present application, the above reference flow chart Figure 2 The described process can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flow chart. In such an embodiment, the computer program includes a program code for executing the method shown in the flow chart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 909, and / or installed from the removable medium 911. When the computer program is executed by the central processing unit (CPU) 901, the above-mentioned functions defined in the system of the present application are executed.

[0159] It should be noted that the computer-readable medium shown in the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium such as a computer-readable storage medium that can send, propagate or transmit a program for use by or in conjunction with an instruction execution system, device or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0160] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, functions and operating instructions of the system, method and computer program product according to various embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the aforementioned module, program segment or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the box can also occur in a sequence different from that marked in the accompanying drawings. For example, the boxes represented by two connections can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operating instruction, or can be implemented with a combination of dedicated hardware and computer instructions.

[0161] The units or modules involved in the embodiments described in the present application may be implemented by software or hardware. The units or modules described may also be provided in a processor, for example, may be described as: a processor including an illegal person detection unit, a multimodal detection unit, and a recognition unit. The names of these units or modules do not, in some cases, constitute limitations on the units or modules themselves.

[0162] As another aspect, the present application further provides a computer-readable storage medium, which may be included in the computer device described in the above embodiment, or may exist independently without being installed in the computer device. The above computer-readable storage medium stores one or more programs, and when the above programs are used by one or more processors to execute the image processing method described in the present application. For example, it can be executed Figure 2 The individual steps of the image processing method shown.

[0163] The present application embodiment provides a computer program product, which includes instructions. When the instructions are executed, the method described in the present application embodiment is executed. For example, Figure 2 The individual steps of the image processing method shown.

[0164] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the aforementioned disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in this application (but not limited to) by each other to form a technical solution.

[0165] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. An image processing method, characterized in that: include: Based on the original image, obtaining an initial fringe image corresponding to the original image; Performing threshold compensation on the initial fringe image to obtain a target fringe image; wherein the threshold compensation is used to adjust the fringe amplitude of pixels in the initial fringe image; The target stripe image is used to remove stripes from the original image to obtain a target image.

2. The image processing method according to claim 1, characterized in that: The step of performing threshold compensation on the initial fringe image to obtain a target fringe image comprises: Acquire a target area of ​​the original image, where the target area is an area in the original image where the noise value exceeds a preset value; Calculating a fringe amplitude threshold of the pixel based on the target area; The initial fringe image is threshold-compensated according to the fringe amplitude threshold of the pixel to obtain the target fringe image.

3. The image processing method according to claim 2, characterized in that: The step of calculating the fringe amplitude threshold of the pixel based on the target area includes: Constructing a mean value curve of the target area according to the stripe direction in the initial stripe image, wherein the mean value curve is constructed by the mean of the target quality parameter values ​​in the stripe direction of the target area, wherein the target quality parameter value is a gray value or a contrast value; constructing an envelope of the mean curve based on the mean curve; The fringe amplitude threshold of the pixel is calculated according to the envelope.

4. The image processing method according to claim 2 or 3, characterized in that: The step of performing threshold compensation on the initial fringe image according to the fringe amplitude threshold of the pixel to obtain the target fringe image comprises: The initial fringe image is subjected to soft threshold shrinkage processing using the fringe amplitude threshold of the pixel to obtain the target fringe image.

5. The image processing method according to claim 4, characterized in that: The step of performing soft threshold shrinkage processing on the initial fringe image by using the fringe amplitude threshold of the pixel to obtain the target fringe image comprises: Traversing all pixels in the initial fringe image, and determining at least one pixel whose fringe amplitude is greater than the fringe amplitude threshold; For each pixel in the at least one pixel, based on the fringe amplitude threshold, shrinking the fringe amplitude of the pixel to obtain a target fringe amplitude of the pixel; The corresponding pixels in the initial fringe image are updated based on the target fringe amplitude to obtain the target fringe image.

6. The image processing method according to claim 2 or 3, characterized in that: Pixels in a column of the initial fringe image in the fringe direction correspond to the same target fringe amplitude.

7. The image processing method according to any one of claims 1 to 3, characterized in that: The step of acquiring an initial fringe image corresponding to the original image based on the original image includes: Acquire a plurality of initial component images of the original image; the plurality of initial component images include the initial component image in the stripe direction; Performing stripe denoising processing on the initial component image in the stripe direction to obtain a target component image in the stripe direction; Reconstructing the target component image in the stripe direction and other initial component images except the initial component image in the stripe direction to obtain a reconstructed image; The initial fringe image is obtained by using the reconstructed image and the original image.

8. The image processing method according to claim 7, characterized in that: The performing stripe denoising on the initial component image in the stripe direction to obtain the target component image in the stripe direction includes: Acquire a high grayscale area and a low grayscale area of ​​the original image; Calculating filter kernel parameters using the high grayscale area and the low grayscale area; The initial component image in the stripe direction is filtered using the filter kernel parameters to obtain a target component image in the stripe direction.

9. The image processing method according to claim 7, characterized in that: The step of acquiring a plurality of initial component images of the original image comprises: The original image is subjected to multi-scale decomposition to obtain the multiple initial component images; the multiple initial component images include an approximate component image and at least one directional component image, the at least one directional component image includes an initial component image of the stripe direction, and the frequency of the at least one directional component image is higher than the frequency of the approximate component image.

10. An image processing device, characterized in that: include: An acquisition module, used for acquiring an initial fringe image corresponding to the original image based on the original image; A threshold compensation module, used to perform threshold compensation on the initial fringe image to obtain a target fringe image; wherein the threshold compensation is used to adjust the fringe amplitude of pixels in the initial fringe image; The stripe removal module is used to remove stripes from the original image using the target stripe image to obtain a target image.

Images