A fiber bundle confocal endoscope image noise modeling method, device and medium

Abstract

The present invention relates to a method, device, and medium for modeling noise in fiber bundle confocal endoscope images. The method comprises the following steps: obtaining an original image and a fiber image of the same pixel size as the original image; obtaining a circular cross-section of a single fiber in the fiber image based on a physical motion simulation algorithm; performing two-dimensional particle stacking on the circular cross-section to obtain a noise mask; adjusting the contrast of the original image to that of a white light imaging image; adjusting the grayscale value of the noise mask based on the transmittance of the white light imaging image; superimposing the matrix of the noise mask and the original image to obtain a first noise image; setting a distortion influence range, segmenting the first noise image into sub-images within the distortion influence range, performing nonlinear distortion scaling on the sub-images to obtain a second noise image; and adding salt and pepper noise to obtain a resulting noise image. Compared with the prior art, the present invention can simulate fiber bundle confocal endoscope images under various honeycomb noise conditions.

Description

Technical Field

[0001] The present invention relates to the technical field of endoscopes, and in particular to a method, device and medium for modeling image noise of a fiber bundle confocal endoscope. Background Art

[0002] Confocal imaging technology improves the signal-to-noise ratio and resolution of images by placing spatial pinholes on the light source and detector. Compared to conventional widefield illumination imaging, confocal imaging can effectively filter out information from the sample's out-of-focus plane, reducing the interference of out-of-focus noise on imaging. Based on the high-resolution observation capability of confocal imaging technology, technicians have combined high-resolution confocal technology with in vivo endoscopy technology to achieve real-time high-resolution in vivo exploration.

[0003] Currently, mature confocal microendoscopes are typically based on an in vitro scanning system using an image-transmitting fiber bundle. This scanning method replaces the mechanical scanning components of the probe in a single fiber system by adding two high-speed scanning galvanometers to the optical path, significantly reducing the size of the endoscope probe. In this system, each fiber in the bundle acts as both a point source of illumination and a point detector, achieving the confocal function.

[0004] Although fiber bundle confocal endoscopes have many advantages such as high resolution, slim endoscope probe, and microscopic imaging, their imaging is completed through imaging optical fibers. Each optical fiber in the optical fiber bundle acts as both an illumination point light source and a collection point detector. Therefore, the optical fiber image displayed is divided by the opaque areas between the individual optical fibers in the optical fiber bundle, which appears as honeycomb noise, such as Figure 2 shown.

[0005] The presence of honeycomb noise seriously affects the observation of pathological tissues during diagnosis and treatment by fiber bundle confocal endoscopes, and relevant algorithms are urgently needed to filter and remove it. The development of relevant filtering removal algorithms requires a large number of fiber bundle confocal endoscope images as a data set. However, since the composition and shape of honeycomb noise are closely related to the arrangement of individual optical fibers in the fiber bundle, the honeycomb noise formed by each fiber bundle is different. If only images obtained from a few fiber bundles are used as a data set, the final generated algorithm model will be difficult to generalize to confocal endoscope images of other fiber bundles. The high cost of the fiber bundle itself also makes it difficult to purchase in large quantities, and it is impossible to obtain a large number of fiber bundles for comparing the different effects of honeycomb noise generated by different imaging fiber bundles. Therefore, a fiber bundle confocal endoscope image noise modeling method is needed to simulate fiber bundle confocal endoscope images under various different honeycomb noises. Summary of the Invention

[0006] The purpose of the present invention is to provide a fiber bundle confocal endoscope image noise modeling method in order to overcome the defects of the above-mentioned prior art. This method simulates the arrangement of optical fibers inside the imaging fiber bundle by introducing two-dimensional particle stacking, improves the realism of the noise, and makes it more consistent with the image noise during actual observation by introducing contrast adjustment, adding salt and pepper noise and edge distortion. It solves the problem of how to simulate the fiber bundle confocal endoscope image under various different honeycomb noises, thereby overcoming the problem in the prior art that the high cost of the fiber bundle itself makes it difficult to purchase optical fibers in large quantities and it is impossible to obtain a large number of fiber bundles for comparing the different effects of honeycomb noise generated by different imaging fiber bundles. It can provide a large amount of noise image data of the fiber bundle, provide a data set for the development of related image filtering algorithms, and improve the reliability and versatility of the image filtering algorithms.

[0007] The purpose of the present invention can be achieved by the following technical solutions:

[0008] A method for modeling image noise in a fiber bundle confocal endoscope comprises the following steps:

[0009] S1. Obtain an original image and an optical fiber image of the same pixel size as the original image, obtain a circular cross-section of a single optical fiber in the optical fiber image based on a physical motion simulation algorithm, perform two-dimensional particle stacking on the circular cross-section to obtain a two-dimensional stacked image, construct an inscribed regular polygon of the circular cross-section within the two-dimensional stacked image, analyze the stress condition of the circular cross-section, adjust the shape of the regular polygon according to the stress condition, divide the circular cross-section into a translucent portion and an opaque portion based on the adjusted inscribed regular polygon, and blur the edges of the translucent portion and the opaque portion to obtain a noise mask;

[0010] S2. Obtain a white light imaging image, adjust the contrast of the original image to the contrast of the white light imaging image, and adjust the grayscale value of the noise mask based on the transmittance of the white light imaging image;

[0011] S3, converting the noise mask image and the original image into matrices with grayscale values ​​as elements, superimposing the two matrices pixel by pixel, and using the superimposed result as the first noise image;

[0012] S4, setting a distortion influence range, cutting the first noise image into sub-images within the distortion influence range, performing nonlinear distortion scaling on the sub-images, and reassembling the distorted and scaled sub-images to obtain a second noise image;

[0013] S5. Add salt and pepper noise to the second noise image to obtain a result noise image.

[0014] Furthermore, the analysis of the stress condition of the circular cross section and the adjustment of the shape of the regular polygon according to the stress condition are specifically as follows:

[0015] The line connecting the centers of the two contacting circular sections is used as the force direction of the two circular sections, and the regular polygons inscribed in the circular sections are regarded as multiple hinged connecting rods. The shape of each regular polygon is adjusted according to the force direction of the circular sections.

[0016] Furthermore, the criterion for determining whether two circular cross sections are in contact with each other is: when the distance between the center of one circular cross section and the center of another circular cross section is less than the diameter of the circular cross section, the two circular cross sections are determined to be in contact with each other.

[0017] Furthermore, the inscribed regular polygon of the circular cross-section is a regular hexagon.

[0018] Furthermore, the distortion influence range is obtained based on the optical path of the fiber bundle confocal endoscope and the working parameters of the scanning system.

[0019] Furthermore, the distortion influence range is an area near the edge of the first noise image that occupies 5% to 10% of the total length.

[0020] Furthermore, the light-transmitting portion and the light-opaque portion are filled with different colors.

[0021] Furthermore, the salt and pepper noise is noise with uneven black and white ratios and sparse distribution.

[0022] A fiber bundle confocal endoscope image noise modeling device includes a memory and one or more processors. The memory stores executable code. When the one or more processors execute the executable code, they are used to implement the above-mentioned fiber bundle confocal endoscope image noise modeling method.

[0023] A computer-readable storage medium stores a program, which, when executed by a processor, implements the above-mentioned fiber bundle confocal endoscope image noise modeling method.

[0024] Compared with the prior art, the present invention has the following beneficial effects:

[0025] The present invention obtains a circular cross-section of a single optical fiber, performs two-dimensional particle stacking on the circular cross-section, and obtains a two-dimensional stacked image, thereby simulating the optical fiber arrangement morphology inside the imaging optical fiber bundle. In addition, the present invention constructs an inscribed regular polygon of the circular cross-section within the two-dimensional stacked image, considers the force on the optical fiber, and introduces contrast adjustment, adds salt and pepper noise and edge distortion to obtain a noise image through superposition. This method considers multiple optical fiber noise influencing factors such as force between optical fibers, contrast, salt and pepper noise and edge distortion, so that the noise image is more consistent with the image noise during actual observation. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 is a flow chart of the present invention;

[0027] Figure 2 This is an optical fiber image of the honeycomb noise of the present invention;

[0028] Figure 3 Schematic diagram of two-dimensional particle stacking of the present invention;

[0029] Figure 4 This is an original diagram of an embodiment of the present invention;

[0030] Figure 5 is a noise mask diagram according to an embodiment of the present invention;

[0031] Figure 6 This is the noise image resulting from the superimposed noise of the present invention. DETAILED DESCRIPTION

[0032] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0033] This specific embodiment provides a fiber bundle confocal endoscope image noise modeling method, which simulates fiber bundle confocal endoscope images under various honeycomb noise conditions. The flow chart of the method is as follows: Figure 1 shown.

[0034] The image noise modeling method for fiber bundle confocal endoscope includes the following steps:

[0035] S1. Obtain the original image and an optical fiber image with the same pixel size as the original image, obtain the circular cross-section of a single optical fiber in the optical fiber image based on a physical motion simulation algorithm, perform two-dimensional particle stacking on the circular cross-section to obtain a two-dimensional stacking image, construct an inscribed regular polygon of the circular cross-section, analyze the force condition of the circular cross-section, adjust the shape of the regular polygon according to the force condition, divide the circular cross-section into a translucent part and an opaque part based on the adjusted inscribed regular polygon, blur the edges of the translucent part and the opaque part to obtain a noise mask.

[0036] The specific steps of S1 include:

[0037] A fiber image is acquired using a fiber bundle confocal endoscope, and a USAF resolution test chart is used as the original image. The fiber image is resized to the same size as the original image. Based on the size ratio of the fiber image and the field of view of the fiber bundle confocal endoscope, the pixel size of a single fiber in the imaging fiber bundle in the fiber image can be determined.

[0038] USAF resolution test chart Figure 4 As shown in FIG, the USAF resolution test chart is a resolution test chart that is widely used to test the resolution capability of optical imaging systems.

[0039] In some embodiments, the height of the pixel size of a single optical fiber in the optical fiber bundle is the height of the optical fiber image under the field of view of the current optical fiber bundle confocal endoscope divided by the average number of single optical fibers in the height direction, and the width of the pixel size of a single optical fiber in the optical fiber bundle is the width of the optical fiber image under the field of view of the current optical fiber bundle confocal endoscope divided by the average number of single optical fibers in the width direction.

[0040] The cross section of a single fiber can be determined based on the pixel size of the single fiber.

[0041] Based on a physical motion simulation algorithm, the cross section of a single optical fiber is treated as a circular cross section with a radius that randomly fluctuates within a certain range. The radius of the circular cross section is determined by the pixel size of the individual optical fibers and the fluctuating radius range. Two-dimensional particle stacking is performed on the circular cross section to obtain the results of the physical motion simulation, which is a two-dimensional stacked image. The two-dimensional stacked image simulates the arrangement of individual optical fibers during the preparation process of the imaging fiber bundle.

[0042] The schematic diagram of two-dimensional stacking is as follows Figure 3 As shown, Figure 3 The middle is the circular cross section of the optical fiber after two-dimensional particle stacking.

[0043] In some embodiments, the radius has a floating range of ±10% to ±20%.

[0044] Output and record the center positions of all circular sections in the two-dimensional stacking image after stacking is completed. With the center position of each circular section as the center, create a regular polygon inscribed in the circular section with an outline size smaller than the circular section. The part inside the regular polygon is used as the light-transmitting part of the optical fiber bundle, and the part outside the regular polygon in the circular section is used as the opaque part.

[0045] like Figure 3 As shown, the circular cross section of a single optical fiber is placed in a plane rectangular coordinate system, the center positions of all circular cross sections are obtained with the plane rectangular coordinate system as a reference, and a regular polygon inscribed in the circular cross section with an outline size smaller than the circular cross section is created with the center position of each circular cross section as the center.

[0046] In some embodiments, the inscribed regular polygon is a regular hexagon, but may also be a regular octagon, a regular decagon, a regular dodecagon, etc. The more sides a regular polygon has, the more accurate the calculation result is, but the computational complexity is greater. Conversely, the fewer sides a regular polygon has, the smaller the computational complexity is, but the calculation result is less accurate than a regular polygon with more sides. Therefore, those skilled in the art can select the number of sides of the inscribed regular polygon as needed, taking into account both accuracy and computational complexity.

[0047] The force applied to a single optical fiber is determined based on the contact relationship between each circular cross-section. When the distance between the center of one circular cross-section and the center of another is less than the cross-section diameter, the two circular cross-sections are considered to be in contact. For two circular cross-sections in contact, the line connecting their centers is the force direction for the two circular cross-sections. The regular polygons inscribed within the circular cross-sections are treated as multiple hinged connecting rods. The shape of each regular polygon is adjusted according to the force direction of the circular cross-sections to simulate the force deformation of each optical fiber in the fiber bundle.

[0048] The transparent part is filled with white, the opaque part is filled with black, and the edges of the transparent and opaque parts are blurred using a convolution algorithm that can recognize image contours to obtain a noise mask. Figure 5 shown.

[0049] In some embodiments, the filling of the light-transmitting portion and the light-opaque portion may be other colors or other forms, as long as the convolution algorithm can identify the edges of the light-transmitting portion and the light-opaque portion.

[0050] S2. Obtain a white light imaging image, adjust the contrast of the original image to the contrast of the white light imaging image, and adjust the grayscale value of the noise mask based on the transmittance of the white light imaging image.

[0051] The specific steps of S2 include:

[0052] The fiber bundle endoscope probe is placed close to a white light source for image acquisition. This produces an image of the fiber bundle confocal endoscope imaging the white light source. Based on the image contrast and transmittance of the imaging fiber bundle when imaging the white light source, the contrast of the original image is adjusted to match the contrast of the actual imaging. The grayscale value of the noise mask is also adjusted based on the transmittance of the white light image. This is because the noise grid in actual imaging is not completely opaque. Therefore, the grayscale value of the noise mask is adjusted to reflect the actual imaging effect, so that the noise mask more closely matches the image obtained through the fiber bundle confocal endoscope.

[0053] S3. Convert the noise mask image and the original image into matrices with grayscale values ​​as elements, superimpose the two matrices pixel by pixel, and use the superimposed result as the first noise image.

[0054] In this method, the single-pixel grayscale values ​​of the noise mask and the original image are set to be in the range of 0 to 255. The transparent part of the noise mask is filled with white, and the opaque part of the noise mask is filled with black. The closer the pixel in the noise mask is to white, the greater the transmittance, and the greater the proportion of the grayscale value of the original image when the two matrices are superimposed pixel by pixel. Conversely, the closer the pixel in the noise mask is to black, the smaller the transmittance, and the smaller the proportion of the grayscale value of the original image when the two matrices are superimposed pixel by pixel.

[0055] Assuming that the single pixel value of the superimposed matrix is ​​P_outi,j, the single pixel value of the noise mask matrix is ​​P_maski,j, and the single pixel value of the original image matrix is ​​P_imgi,j, the calculation expression of the single pixel value of the superimposed matrix is:

[0056]

[0057] The matrix single pixel values ​​constitute the gray value matrix of the superimposed first noise image.

[0058] The superimposed result is used as the first noise image.

[0059] S4. Set the distortion influence range, cut the first noise image into sub-images within the distortion influence range, perform nonlinear distortion scaling on the sub-images, and reassemble the distorted and scaled sub-images to obtain a second noise image.

[0060] The image edges of fiber bundle confocal endoscopes will produce a contraction-like distortion, which is related to the imaging principle of fiber bundle confocal endoscopes. Fiber bundle confocal endoscopes use two-dimensional scanning performed by galvanometers in the X and Y directions. The periodic oscillation of the galvanometers will cause the physical distance between pixels at the edge of the image to be greater than the physical distance between pixels in the center of the image, which will cause the image edges to produce a contraction-like distortion.

[0061] The size and range of edge distortion of the image are usually related to the optical path and scanning system, but not to the imaging fiber bundle itself. Therefore, the distortion influence range of the image can be set according to the optical path and scanning system working parameters of the fiber bundle confocal endoscope, such as the galvanometer scanning angle, the distance between the galvanometer and the end face of the imaging fiber bundle, and other parameters.

[0062] In some embodiments, an area near the edge of the image that occupies 5% to 10% of the total length is set as the distortion affected area.

[0063] In some embodiments, an area near the edge of the image that occupies 10% to 15% of the total length is set as the distortion affected area.

[0064] The distortion influence range is obtained based on the optical path and working parameters of the scanning system of the fiber bundle confocal endoscope, and has nothing to do with the imaging fiber bundle itself. Different optical paths and scanning systems correspond to different distortion influence ranges. Different distortion influence ranges should be set according to the working parameters of different optical paths and scanning systems, and are not limited to the distortion influence ranges in the above embodiments.

[0065] The first noise image, with the noise mask superimposed on it, is segmented into a sub-image within the distortion range, obtaining an image not within the distortion range and a sub-image within the distortion range. Nonlinear distortion scaling is performed on the sub-image within the distortion range. The scaled sub-image and the image not within the distortion range are reassembled to obtain a second noise image.

[0066] S5. Add salt and pepper noise to the second noise image to obtain a result noise image.

[0067] Since the generation of fiber bundle confocal endoscope images is completed by two-dimensional scanning, the two-dimensional scanning process will cause random pixel noise due to the timing error of the photoelectric sensor response time and the sampling circuit, which manifests as a salt and pepper noise with uneven black and white ratio and sparse distribution on the image.

[0068] Salt and pepper noise is added to the second noise image that has been superimposed with noise and edge distortion to simulate random noise points in the scanning imaging process.

[0069] The resulting noise image after superposition is as follows Figure 6 shown.

[0070] The present invention obtains the circular cross-section of a single optical fiber, performs two-dimensional particle stacking on the circular cross-section, and obtains a two-dimensional stacked image, thereby simulating the arrangement of optical fibers inside the imaging optical fiber bundle, and constructing an inscribed regular polygon of the circular cross-section in the two-dimensional stacked image, taking into account the stress conditions of the optical fiber, and by introducing contrast adjustment, adding salt and pepper noise and edge distortion, superimposing a noise image to achieve modeling of the noise of the optical fiber bundle confocal endoscope image, so that the noise image is more consistent with the image noise during actual observation. The optical fiber bundle confocal endoscope image noise modeling method of the present invention overcomes the problem in the prior art that the high cost of the optical fiber bundle itself makes it difficult to purchase optical fibers in large quantities, and it is impossible to obtain a large number of optical fiber bundles for comparing the different effects of cellular noise generated by different imaging optical fiber bundles. It can provide a large amount of noise image data of the optical fiber bundle, thereby providing a data set for the development of related image filtering algorithms, thereby improving the reliability and versatility of the image filtering algorithms.

[0071] The present invention also provides a computer-readable storage medium, which stores a computer program, which can be used to execute the above Figure 1 A method for modeling image noise in fiber bundle confocal endoscopes is provided.

[0072] The present invention also provides a fiber bundle confocal endoscope image noise modeling device. The device includes a memory and one or more processors, and the memory stores executable code. When the one or more processors execute the executable code, they are used to implement the above-mentioned fiber bundle confocal endoscope image noise modeling method. At the hardware level, the fiber bundle confocal endoscope image noise modeling device includes a processor, an internal bus, a network interface, a memory and a non-volatile memory, and of course may also include hardware required for other services. The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it to implement the above-mentioned data acquisition method. Of course, in addition to software implementation, the present invention does not exclude other implementation methods, such as logic devices or a combination of software and hardware, etc., that is, the execution subject of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.

[0073] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, apparatus, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0074] The present invention is described with reference to flowcharts of methods, apparatuses, and computer program products according to embodiments of the present invention. It should be understood that each process in the flowcharts, as well as combinations of processes in the flowcharts, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts. Figure 1 A device that specifies functions in a process or multiple processes.

[0075] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A function specified in a process or multiple processes.

[0076] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1The steps of the functions specified in a process or multiple process boxes.

[0077] Computer-readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory media such as modulated data signals and carrier waves.

[0078] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0079] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0080] The present invention may be described in the general context of computer-executable instructions, such as program modules, executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, and the like that perform specific tasks or implement specific abstract data types. The present invention may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media, including storage devices.

[0081] The various embodiments of the present invention are described in a progressive manner. Similar portions between the various embodiments can be referred to in conjunction with each other. Each embodiment focuses on the differences between the other embodiments. In particular, the system embodiment is generally similar to the method embodiment, so its description is relatively simple. For relevant portions, refer to the description of the method embodiment.

[0082] The above describes in detail the preferred embodiments of the present invention. It should be understood that those skilled in the art can make numerous modifications and variations based on the concepts of the present invention without inventive effort. Therefore, any technical solutions that can be derived by those skilled in the art through logical analysis, reasoning, or limited experimentation based on the concepts of the present invention and the prior art should be within the scope of protection defined by the claims.

Claims

1. A method for modeling image noise in fiber bundle confocal endoscope, characterized in that: The method comprises the following steps: S1. Obtain an original image and an optical fiber image with the same pixel size as the original image, perform two-dimensional particle stacking on the circular cross-section of a single optical fiber in the optical fiber image based on a physical motion simulation algorithm to obtain a result of the physical motion simulation, which is a two-dimensional stacked image, construct an inscribed regular polygon of the circular cross-section in the two-dimensional stacked image, analyze the stress condition of the circular cross-section, adjust the shape of the regular polygon according to the stress condition, divide the circular cross-section into a translucent part and an opaque part based on the adjusted inscribed regular polygon, blur the edges of the translucent part and the opaque part to obtain a noise mask; S2. Obtain a white light imaging image, adjust the contrast of the original image to the contrast of the white light imaging image, and adjust the grayscale value of the noise mask based on the transmittance of the white light imaging image; S3, converting the noise mask image and the original image into matrices with grayscale values ​​as elements, superimposing the two matrices pixel by pixel, and using the superimposed result as the first noise image; S4, setting a distortion influence range, cutting the first noise image into sub-images within the distortion influence range, performing nonlinear distortion scaling on the sub-images, and reassembling the distorted and scaled sub-images to obtain a second noise image; S5. Add salt and pepper noise to the second noise image to obtain a result noise image.

2. The method for modeling image noise of a fiber bundle confocal endoscope according to claim 1, characterized in that: The analysis of the stress condition of the circular cross section and the adjustment of the shape of the regular polygon according to the stress condition are specifically as follows: The line connecting the centers of the two contacting circular sections is used as the force direction of the two circular sections, and the regular polygons inscribed in the circular sections are regarded as multiple hinged connecting rods. The shape of each regular polygon is adjusted according to the force direction of the circular sections.

3. The method for modeling image noise of a fiber bundle confocal endoscope according to claim 2, characterized in that: The criterion for determining whether two circular cross sections are in contact with each other is: when the distance between the center of one circular cross section and the center of another circular cross section is less than the diameter of the circular cross section, the two circular cross sections are determined to be in contact with each other.

4. The method for modeling image noise of a fiber bundle confocal endoscope according to claim 1, characterized in that: The inscribed regular polygon of the circular cross section is a regular hexagon.

5. The method for modeling image noise of a fiber bundle confocal endoscope according to claim 1, characterized in that: The distortion influence range is obtained based on the optical path of the fiber bundle confocal endoscope and the working parameters of the scanning system.

6. The method for modeling image noise of a fiber bundle confocal endoscope according to claim 5, characterized in that: The distortion influence range is an area near the edge of the first noise image that occupies 5% to 10% of the total length.

7. The method for modeling image noise of a fiber bundle confocal endoscope according to claim 1, characterized in that: The light-transmitting portion and the light-impermeable portion are filled with different colors.

8. The method for modeling image noise of a fiber bundle confocal endoscope according to claim 1, characterized in that: The salt and pepper noise is noise with uneven black and white ratios and sparse distribution.

9. A fiber bundle confocal endoscope image noise modeling device, characterized in that: The invention comprises a memory and one or more processors, wherein the memory stores executable code, and when the one or more processors execute the executable code, they are used to implement the fiber bundle confocal endoscope image noise modeling method according to any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that A program is stored thereon, and when the program is executed by a processor, the fiber bundle confocal endoscope image noise modeling method according to any one of claims 1 to 8 is implemented.

Images