A method for removing honeycomb patterns from coherent fiber bundles based on spatial pixel translation

By integrating and Hilbert transforming the coherent fiber bundle images, the target reconstruction image is generated, and the problems of image quality and resolution improvement in the prior art are solved, and efficient cellular pattern removal and image detail enhancement are achieved.

CN118570079BActive Publication Date: 2024-12-03HUAZHONG UNIV OF SCI & TECH
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202411047271.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-01
Publication Date
2024-12-03
Estimated Expiration
2044-08-01

AI Technical Summary

Technical Problem

The existing methods of removing coherent fiber bundle cellular patterns cannot improve image quality and resolution simultaneously. The image clarity obtained by the filtering method is low, and the interpolation method has limited impact on resolution enhancement.

Method used

By acquiring the integral image and deriving it in clockwise directions of 0°, 90° and 45°, Hilbert transform integration, combining the initial reconstruction image and pixel values ​​in the integral image, the target reconstruction image is generated, and the image details are enhanced. and the improvement of resolution.

Benefits of technology

Effectively remove cellular patterns in coherent fiber bundles, improve image quality and resolution, and achieve high-quality image restoration without the need for additional hardware and reference optical paths, taking into account calculation speed and image recovery quality.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118570079B_ABST
    Figure CN118570079B_ABST
Patent Text Reader

Abstract

The present application relates to the field of image data processing technology, and discloses a method for removing honeycomb patterns of coherent optical fiber bundles based on spatial pixel translation. The method includes obtaining an integral image, which is obtained by preprocessing a sample image and a reference image, and the sample image contains a honeycomb pattern of a coherent optical fiber bundle; the integral image is derived along three clockwise directions of 0°, 90° and 45°, and integrated along the corresponding directions through Hilbert transform to obtain three initial reconstructed images, and the initial reconstructed image is an image obtained after enhancing the details in the integral image; traversing each pixel point in the three initial reconstructed images and the integral image, for the current pixel point, comparing the pixel values ​​of the current pixel point in the three initial reconstructed images and the integral image, and taking the maximum pixel value as the pixel value of the current pixel point in the target reconstructed image to obtain the target reconstructed image. The present method can effectively remove the honeycomb pattern in the coherent optical fiber bundle while enhancing the image details.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of image data processing, and more specifically, to a method for removing honeycomb patterns of coherent optical fiber bundles based on spatial pixel translation, a computer device, a computer-readable storage medium, and a computer program product. Background Art

[0002] At present, most advanced endoscopes, such as two-photon endoscopes, fluorescence confocal endoscopes, optical coherence tomography endoscopes, and quantitative phase imaging endoscopes, use coherent fiber bundles as their key components. Endoscopes based on coherent fiber bundles have strong in vivo observation capabilities, which are helpful for histological and pathological observations. The cellular level resolution obtained by assembling coherent fiber bundles with micro-objective lenses with magnification exceeds the limitations of traditional endoscopes.

[0003] However, there are inherent limitations in coherent fiber bundles, which consist of tens of thousands of cores. The cladding surrounding each core impedes the propagation of light, causing a pixelated image, often referred to as a "honeycomb pattern," that degrades image quality. Furthermore, system resolution is limited by the spacing between the cores, and ultimately by the core diameter. In addition, mutual coupling between the cores introduces a blurring effect, further reducing image clarity.

[0004] To remove the honeycomb pattern of coherent fiber bundles, the most commonly used methods include filtering and interpolation. Filtering methods include Gaussian smoothing and frequency filtering, which have advantages in computational efficiency. Interpolation methods enhance image clarity by identifying each fiber core and interpolating the cladding area using the core center pixel value.

[0005] However, the image clarity obtained by the filtering method is low, and the effect of the interpolation method on resolution enhancement is limited. Therefore, the current methods for removing the honeycomb pattern of the coherent fiber bundle cannot solve the problem of improving the image quality and resolution while removing the honeycomb pattern. Summary of the invention

[0006] In order to solve the problem of how to improve image quality and resolution while removing honeycomb patterns, the present invention provides a method for removing honeycomb patterns in coherent fiber bundles based on spatial pixel translation, a computer device, a computer-readable storage medium and a computer program product, which can effectively remove honeycomb patterns in coherent fiber bundles while enhancing image details, thereby improving image quality and resolution.

[0007] To achieve the above object, according to a first aspect of the present invention, a method for removing honeycomb patterns of coherent fiber bundles based on spatial pixel translation is provided, the method comprising:

[0008] Acquire an integral image, where the integral image is obtained by preprocessing a sample image and a reference image, wherein the sample image includes a honeycomb pattern of a coherent optical fiber bundle, and the reference image is an image with a uniform background and does not include the sample;

[0009] The integral image is derived along the clockwise directions of 0°, 90° and 45°, and integrated along the corresponding directions through Hilbert transform to obtain three initial reconstructed images. The initial reconstructed image is an image obtained by enhancing the details in the integral image.

[0010] Traverse each pixel point in the three initial reconstructed images and the integral image. For the current pixel point, compare the pixel values ​​of the current pixel point in the three initial reconstructed images and the integral image, take the maximum pixel value as the pixel value of the current pixel point in the target reconstructed image, and obtain the target reconstructed image.

[0011] According to a second aspect of the present invention, there is also provided a computer device, which includes a memory, a processor and a computer program stored in the memory, and the processor executes the computer program to implement the steps of any one of the above methods.

[0012] According to a third aspect of the present invention, there is also provided a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps of any of the above methods are implemented.

[0013] According to a fourth aspect of the present invention, there is also provided a computer program product, comprising a computer program, which implements the steps of any of the above methods when executed by a processor.

[0014] In general, the above technical solutions conceived by the present invention can achieve the following beneficial effects compared with the prior art:

[0015] The method for removing honeycomb patterns in coherent optical fiber bundles based on spatial pixel translation provided by the present invention obtains an integral image by preprocessing a sample image and a reference image; three initial reconstructed images are obtained after detail enhancement of the integral image; and finally a target reconstructed image is obtained based on the three initial reconstructed images and the integral image. The target reconstructed image has a high resolution and can restore the sample with high quality, thereby achieving the purpose of effectively removing the honeycomb pattern in the coherent optical fiber bundle while enhancing the image details. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments 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 creative work.

[0017] Figure 1 A schematic diagram of the structure of an endoscopic imaging device provided in an embodiment of the present application;

[0018] Figure 2 A schematic flow chart of a method for removing honeycomb patterns of a coherent optical fiber bundle based on spatial pixel translation provided in an embodiment of the present application;

[0019] Figure 3 A schematic flow chart of a method for removing honeycomb patterns of a coherent optical fiber bundle based on spatial pixel translation provided in another embodiment of the present application;

[0020] Figure 4 A schematic diagram of the image enhancement effect of various methods for removing honeycomb patterns of coherent fiber bundles provided in embodiments of the present application;

[0021] Figure 5 A schematic diagram of the internal structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0023] The terms "first", "second", "third", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0024] like Figure 1 As shown, an endoscopic imaging device is provided for observing a sample 1 (e.g., a USAF 1951 resolution plate), the device comprising a micro-objective lens 2, an illumination optical fiber 3, a coherent optical fiber bundle 4, a microscope objective lens 5, a tube lens 6, linear polarizers 8 and 10, an amplitude-type spatial light modulator (abbreviated as SLM) 9, and a 4 Lens groups 7 and 11, CMOS camera 12, LED light source 13. The micro-objective lens 2, illumination fiber 3 and endoscope probe cladding 15 constitute a cross-sectional view of the endoscope probe. A chessboard grating 16 is loaded on the SLM.

[0025] The working principle of the device is as follows: the illumination fiber 3 transmits LED light (e.g., wavelength 780 nanometers, half-maximum full width 30 nanometers) to the tip of the probe to form a ring-shaped illumination. Subsequently, the micro-objective lens 2 collects the backscattered light and transmits the sample image at the end face of the far end of the coherent fiber bundle with a certain magnification (e.g., 2.6 times). The sample image is transmitted to the near end (close to the direction of the CMOS camera) through the coherent fiber bundle, passing through 30,000 3-micron diameter fiber cores, which act as pixels in relative positions. The next microscope part is used for quantitative phase imaging, and the end face of the image transmission bundle is imaged with a microscope objective lens 5.

[0026] Assume that the wavefront on the focal plane of the objective lens (i.e., the end of the coherent fiber bundle 4 close to the microscope objective lens 5) is , It is a sample field and coherent fiber bundle transmission matrix The superposition field of :

[0027] .

[0028] Then, After 4 The system is amplitude modulated by the SLM in the Fourier plane, and the phase shift and lateral shearing functions are realized by loading the chessboard grating. The transmittance expression of the chessboard grating projected on the SLM is:

[0029] ,

[0030] in, is the transverse spatial frequency, and is the lateral displacement introduced by the checkerboard grating, is the artificially added phase shift, which is set to 0 here.

[0031] Grating period p exist x and y The directions are set to be the same and can be expressed as ,in is the wavelength, Serves 4 The magnification of the system, Serves 4 The focal length of the system.

[0032] Finally, the diffraction wavefront introduced by the SLM produces a shearing interference pattern on the CMOS camera, which can be expressed as:

[0033] ,

[0034] in, and Alongx and y The phase difference in the direction is introduced by the checkerboard grating projected on the SLM.

[0035] Interference and wavefront shaping enable phase modulation and extraction of differential information from the captured image, which provides a theoretical basis for the method for removing the honeycomb pattern of the coherent fiber bundle based on the spatial pixel shift algorithm (SPS algorithm for short) provided in this embodiment.

[0036] like Figure 2 As shown, a method for removing honeycomb patterns of coherent fiber bundles based on spatial pixel translation is provided. The method can be executed by a terminal or by a server communicating with the terminal through a network. The terminal can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, etc. The server can be an independent server or a server cluster composed of multiple servers. The method is described by taking the application of the method to a terminal as an example, and includes the following steps:

[0037] Step 201, obtaining an integral image.

[0038] The integral image is obtained by preprocessing a sample image and a reference image. The sample image includes a honeycomb pattern of a coherent optical fiber bundle, and the reference image is an image with a uniform background and does not include the sample.

[0039] Step 202, the integral image is differentiated along three clockwise directions of 0°, 90° and 45° respectively and integrated along the corresponding directions through Hilbert transform to obtain three initial reconstructed images, where the initial reconstructed image is an image obtained by enhancing the details in the integral image.

[0040] Among them, the three directions of 0°, 90° and 45° clockwise correspond to x , y and x + y Three directions. The three initial reconstructed images correspond to the clockwise directions of 0°, 90°, and 45° respectively.

[0041] Step 203, traverse each pixel point in the three initial reconstructed images and the integral image, for the current pixel point, compare the pixel values ​​of the current pixel point in the three initial reconstructed images and the integral image, take the maximum pixel value as the pixel value of the current pixel point in the target reconstructed image, and obtain the target reconstructed image.

[0042] In the above-mentioned method for removing honeycomb patterns in coherent fiber bundles based on spatial pixel translation, the sample image and the reference image are preprocessed to obtain an integral image; the integral image is enhanced in detail to obtain three initial reconstructed images; and finally, the target reconstructed image is obtained based on the three initial reconstructed images and the integral image. The target reconstructed image has a high resolution and can restore the sample with high quality, thereby achieving the purpose of effectively removing the honeycomb pattern in the coherent fiber bundle while enhancing the image details. Compared with the prior art, this method does not require additional hardware and reference optical paths, and can achieve real-time removal of honeycomb patterns, taking into account both computing speed and image restoration quality. The image detail restoration capability is stronger than that of the iterative algorithm, and has higher image quality and resolution.

[0043] In one embodiment, the integral image includes a high-definition image and a blurred edge. In the above step 202, the integral image is derived along three clockwise directions of 0°, 90° and 45° respectively and integrated along the corresponding directions through Hilbert transform to obtain three initial reconstructed images, including:

[0044] ,

[0045] in, is the initial reconstructed image, is the integral image, For high-definition images, are directions, corresponding to x , y and x + y , indicating the three directions of 0°, 90° and 45° clockwise, is the Cauchy principal value.

[0046] In this embodiment, by taking the derivative of the integral image along three clockwise directions of 0°, 90° and 45° respectively, the smoothing effect of the edge of the integral image can be removed and the high-definition image can be retained.

[0047] In one embodiment, the above step 203, using the maximum pixel value as the pixel value of the current pixel point in the target reconstructed image to obtain the target reconstructed image, includes using the maximum pixel value as the pixel value of the current pixel point in the candidate reconstructed image to obtain the candidate reconstructed image; using the intensity adjustment factor The candidate reconstructed image is adjusted so that the grayscale difference between the candidate reconstructed image and the sample image is less than a preset value, thereby obtaining a target reconstructed image.

[0048] in, It is a constant factor between 0 and 1, which is used to adjust the intensity average. The preset value is determined according to the actual test requirements. When the difference between the grayscale of the candidate reconstructed image and the sample image is less than the preset value, the grayscale of the candidate reconstructed image is closer to that of the sample image.

[0049] In one embodiment, the above method for removing honeycomb patterns of coherent fiber bundles based on spatial pixel translation further includes preprocessing the sample image and the reference image to obtain an integral image. The specific steps are as follows:

[0050] Step 1) Obtain sample image and reference image;

[0051] Step 2) spatially shifting the filtered sample image and the reference image to obtain a first sub-image, a second sub-image, and a third sub-image of the sample image and the reference image;

[0052] Step 3) Based on the sample image and the first sub-image, the second sub-image, and the third sub-image of the reference image, calculate the sample image along x Axis and y Difference image of the axis;

[0053] Step 4) Based on the sample image along x Axis and y The differential image of the axis is obtained by spiral integration.

[0054] In one embodiment, the above step 2) performs spatial pixel shifting on the filtered sample image and the reference image to obtain the first sub-image, the second sub-image, and the third sub-image of the sample image and the reference image, respectively, including shifting the filtered sample image and the reference image by one pixel in three clockwise directions of 0°, 90°, and 45° to obtain the first sub-image, the second sub-image, and the third sub-image of the sample image and the reference image, respectively.

[0055] In one embodiment, the above step 3) calculates the sample image along the sample image based on the first sub-image, the second sub-image, and the third sub-image of the reference image. x Axis and y The difference image of the axis includes calculating the sample image along the axis based on the sample image and its first sub-image, the second sub-image and the third sub-image. x Axis and y The initial differential image of the axis; based on the reference image and its first sub-image, the second sub-image and the third sub-image, calculate the reference image along x Axis and y Axis difference image; the sample image is moved along x The initial difference image of the axis minus the reference image along x The difference image of the axis is obtained by taking the sample image along x Axis difference image; the sample image is moved along y The initial difference image of the axis minus the reference image along y The difference image of the axis is obtained by taking the sample image along y Axis difference image.

[0056] In one embodiment, the above step 4) is based on the sample image along x Axis and y The differential image of the axis is obtained by spiral integration, including:

[0057] ,

[0058] ,

[0059] ,

[0060] in, is the integral image, is a sample image along the x The difference image of the axis, is a sample image along the y The difference image of the axis, is the transverse spatial frequency, and is the lateral displacement introduced by the checkerboard grating, is the Wiener filter parameter, represents the Fourier transform, represents the inverse Fourier transform.

[0061] In one embodiment, Figure 3 As shown, the method for removing honeycomb patterns of coherent fiber bundles based on spatial pixel translation includes five steps.

[0062] The first step is to take two input images, one is the endoscopic image with the sample (referred to as the sample image) and the other is an image without the sample with a uniform background used as a reference (referred to as the reference image). I and reference image Filtering is performed using a spatial Gaussian filter to smooth the honeycomb pattern.

[0063] The second step is to filter the sample image and reference image Shift one pixel in each of the three directions of 0°, 90° and 45°, plus the original Figure 1 A total of four images are obtained (i.e., filtered sample images / 、 The first sub-image of the sample image 、 The second sub-image of the sample image and the third sub-image of the sample image , the filtered reference image / , the first sub-image of the reference image , the second sub-image of the reference image and the third sub-image of the reference image ).

[0064] Since the filtered image can be regarded as a slowly varying signal, we assume that its amplitude remains constant across adjacent pixel values, i.e. .

[0065] Since the checkerboard pattern on the SLM x and y The samples have the same period in the direction. x and y Phase shift in direction Roughly the same, no drastic changes. However, right x Changes in direction show greater sensitivity, while y The change in direction is relatively negligible. Same here.

[0066] Therefore, when pixel displacement occurs along different axes, and The additional phase shift Therefore, in x Direction The pixel displacement changes but remain unchanged. On the contrary, y Moving one pixel in the direction will modify But it will not affect . These changes can be expressed as:

[0067] ,

[0068] ,

[0069] ,

[0070] .

[0071] In order to simplify the calculation of the sub-image, it is assumed that the amplitude on the image plane remains unchanged on adjacent pixel values, which can be expressed as:

[0072] .

[0073] Therefore, taking the filtered sample image as an example, the filtered sample image 、 The first sub-image of the sample image 、 The second sub-image of the sample image and the third sub-image of the sample image It can be expressed as:

[0074] ,

[0075] ,

[0076] ,

[0077] .

[0078] In the same way, the filtered reference image is calculated , the first sub-image of the reference image , the second sub-image of the reference image and the third sub-image of the reference image .

[0079] The third step is to calculate the x Axis and y To simplify the calculation, assume that the added phase shift value .

[0080] To calculate the sample image along x Initial difference image of the axis , the sample image is along y Initial difference image of the axis For example, we get:

[0081] ,

[0082] .

[0083] In the same way, the reference image is calculated along x Difference image of the axis and the reference image along y Difference image of the axis .

[0084] Finally, the sample image is calculated along x Difference image of the axis , the sample image is along y Difference image of the axis ,get:

[0085] ,

[0086] .

[0087] The fourth step is to reconstruct the image by using spiral integral (i.e., the integral image in the above embodiment):

[0088] ,

[0089] ,

[0090] ,

[0091] in, is the integral image, is a sample image along the x The difference image of the axis, is a sample image along the y The difference image of the axis, is the transverse spatial frequency, and is the lateral displacement introduced by the checkerboard grating, is the Wiener filter parameter, represents the Fourier transform, represents the inverse Fourier transform.

[0092] The fifth step is to integrate the image Perform detail enhancement to obtain the final reconstructed image (i.e., the target reconstructed image). There is inevitably a blurring effect at the edge of , which can be considered as a smooth function, affecting the presentation of low-frequency details. Assume that the smooth part is , then the high-definition image after removing the smoothing effect is , which can be expressed as: .

[0093] In order to enhance this part of details, the integral image is firstly derived along the three directions of 0°, 90° and 45° to remove the smooth response of the edge, and then the Hilbert transform is integrated along the corresponding directions to obtain three images with enhanced details (i.e., the initial reconstructed images in the above embodiment). The three initial reconstructed images can be uniformly expressed as:

[0094] ,

[0095] in, is the initial reconstructed image, is the integral image, For high-definition images, are directions, corresponding to x , y and x + y , indicating the three directions of 0°, 90° and 45° clockwise, is the Cauchy principal value.

[0096] According to the Hilbert transform properties, the initial reconstructed image is a high-definition image The reciprocal of the square of the sum direction convolution.

[0097] In order to prevent the convolution kernel from weakening the details, traverse the three initial reconstructed images With integral image Each pixel value of the image is retained, and only the value with the largest pixel value in all images is retained, that is: traverse the three initial reconstructed images , , and integral image For each pixel in the image, for the current pixel, compare the three initial reconstructed images , , and integral image The maximum pixel value is used as the pixel value of the current pixel in the candidate reconstructed image to obtain a candidate reconstructed image.

[0098] Finally, the intensity adjustment factor Adjust the candidate reconstructed image so that the grayscale of the candidate reconstructed image is close to that of the sample image, and finally reconstruct the target reconstructed image .

[0099] refer to Figure 4 , the USAF 1951 resolution plate was used as the test sample to evaluate the image enhancement capabilities of various honeycomb pattern methods that remove coherent fiber bundles. Figure 4 As shown, the Gaussian filtered image ( Figure 4 (b) Compared with the original image ( Figure 4 (a) has a smooth background but exhibits speckle effects. In contrast, the interpolated image ( Figure 4 (c) has higher clarity, the background is more uniform and pure, and the element lines show higher contrast than the Gaussian filtered image. Image processed by the alternating direction method of multiple multiplications (ADMM) ( Figure 4 (d) shows comparable quality to the interpolated image, but there is still some noise between the element lines. However, the processing results of our proposed method of removing the honeycomb pattern of the coherent fiber bundle based on spatial pixel translation show the most significant quality improvement, as shown in Figure 4 (e) shows that the background and element lines in this image are very clear and clean, and there is no noise between the element lines.

[0100] In a preliminary comparison, Figure 4 (ce) It is easier to distinguish the details of the lines of each element.

[0101] In addition, if Figure 4As shown in (f), by recording the intensity values ​​corresponding to the white line parts in the black solid frame, the processing quality of each algorithm is also compared. The contour curve of the method based on spatial pixel translation to remove the honeycomb pattern of the coherent fiber bundle has the largest range of variation, is smoother than the contour curve of the original image, and has the highest signal-to-noise ratio. Compared with other methods, the line edges are more recognizable, showing that this method has excellent performance in improving image quality and clarity.

[0102] The present application also provides a computer, the internal structure diagram of the computer device can be as follows Figure 5 As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a method for removing a coherent fiber bundle honeycomb pattern based on spatial pixel translation is implemented. The display unit of the computer device is used to form a visually visible image, and can be a display screen, a projection device or a virtual reality imaging device. The input device of the computer device can be a touch layer covered on the display screen, or a button, trackball or touchpad set on the computer device casing, or an external keyboard, touchpad or mouse, etc.

[0103] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0104] like Figure 5 As shown, the present application also provides a computer device, which includes a memory, a processor, and a computer program stored in the memory, and the processor executes the computer program to implement the steps in the above-mentioned method embodiments.

[0105] The present application also provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented. The computer-readable storage medium may include, but is not limited to, any type of disk, including a floppy disk, an optical disk, a DVD, a CD-ROM, a micro drive, and a magneto-optical disk, a ROM, a RAM, an EPROM, an EEPROM, a DRAM, a VRAM, a flash memory device, a magnetic card or an optical card, a nanosystem (including a molecular memory IC), or any type of medium or device suitable for storing instructions and / or data.

[0106] The present application also provides a computer program product, including a computer program, which implements the steps in the above-mentioned method embodiments when executed by a processor.

[0107] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0108] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0109] The above is only an exemplary embodiment of the present disclosure, and the scope of the present disclosure cannot be limited thereto. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the specification and practicing the disclosure here, those skilled in the art will easily think of the implementation scheme of the present disclosure. This application is intended to cover any modification, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the technical field not recorded in the present disclosure. The description and examples are only regarded as exemplary, and the scope and spirit of the present disclosure are defined by the claims.

[0110] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0111] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for removing honeycomb patterns from coherent fiber bundles based on spatial pixel translation. It is characterized in that include: Acquire a sample image and a reference image, wherein the sample image includes a honeycomb pattern of a coherent optical fiber bundle, and the reference image is an image that does not include the sample and has a uniform background; Performing spatial pixel shift on the filtered sample image and the reference image to obtain a first sub-image, a second sub-image, and a third sub-image of the sample image and the reference image, respectively, wherein the spatial pixel shift includes shifting one pixel in three clockwise directions of 0°, 90°, and 45°; Based on the sample image and the first sub-image, the second sub-image, and the third sub-image of the reference image, the sample image is calculated along x Axis and y Difference image of the axis; Based on the sample image along x Axis and y The differential image of the axis is obtained by spiral integration to obtain an integral image, wherein the integral image includes a high-definition image and a blurred edge; The integral image is derived along three directions of 0°, 90° and 45° clockwise respectively and integrated along the corresponding directions through Hilbert transform to obtain three initial reconstructed images, wherein the initial reconstructed image is an image obtained by enhancing the high-definition image, and the initial reconstructed image is the convolution of the high-definition image and the reciprocal of the square of the direction; Traverse each pixel point in the three initial reconstructed images and the integral image, and for the current pixel point, compare the pixel values ​​of the three initial reconstructed images and the current pixel point in the integral image, take the maximum pixel value as the pixel value of the current pixel point in the target reconstructed image, and obtain the target reconstructed image.

2. The method according to claim 1, It is characterized in that The integral image is derived along three directions of 0°, 90° and 45° clockwise respectively and integrated along the corresponding directions through Hilbert transform to obtain three initial reconstructed images, including: , in, is the initial reconstructed image, is the integral image, For high-definition images, are directions, corresponding to x , y and x + y , indicating the three directions of 0°, 90° and 45° clockwise, is the Cauchy principal value.

3. The method according to claim 1, It is characterized in that The step of taking the maximum pixel value as the pixel value of the current pixel point in the target reconstructed image to obtain the target reconstructed image includes: The maximum pixel value is used as the pixel value of the current pixel in the candidate reconstructed image to obtain the candidate reconstructed image; The candidate reconstructed image is adjusted by using an intensity adjustment factor so that the grayscale difference between the candidate reconstructed image and the sample image is less than a preset value, thereby obtaining a target reconstructed image.

4. The method according to claim 1, It is characterized in that The method comprises calculating the sample image along the first sub-image, the second sub-image, and the third sub-image based on the sample image and the reference image. x Axis and y Difference images of the axes, including: Based on the sample image and its first sub-image, second sub-image and third sub-image, the sample image is calculated along x Axis and y Initial difference image of the axis; Based on the reference image and its first sub-image, second sub-image and third sub-image, the reference image is calculated along x Axis and y Difference image of the axis; Place the sample image along x The initial difference image of the axis is minus the reference image along x The difference image of the axis is obtained by taking the sample image along x Difference image of the axis; Place the sample image along y The initial difference image of the axis is minus the reference image along y The difference image of the axis is obtained by taking the sample image along y Axis difference image.

5. The method according to claim 1, It is characterized in that The step of obtaining an integral image by spiral integration based on the differential image of the sample image along the x-axis and the y-axis comprises: , , , in, is the integral image, is a sample image along the x The difference image of the axis, is a sample image along the y The difference image of the axis, is the transverse spatial frequency, and is the lateral displacement introduced by the checkerboard grating, is the Wiener filter parameter, represents the Fourier transform, represents the inverse Fourier transform.

6. A computer device, It is characterized in that The method comprises a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method according to any one of claims 1 to 5.

7. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

8. A computer program product comprising a computer program, It is characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Citation Information

Patent Citations

  • Image reconstruction method and system based on Katsevich algorithm

    CN103714560A

  • Image processing method and device, equipment and storage medium

    CN116342434A