Fiber core point cloud positioning method based on probability distribution and related equipment

Through the core point cloud positioning method based on probability distribution, combined with multi-frame image fusion, local standard deviation image and OSTU method processing, the problem of inaccurate core point cloud positioning in micro-endoscopes is solved, and the image quality is improved Diagnostic accuracy.

CN118941635APending Publication Date: 2024-11-12BIOPSEE (SUZHOU) MEDICAL TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410981128.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-22
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

The electronic components characteristics of the resonant mirror and galvometer galvanometer in microscopes change with temperature, resulting in inaccurate positioning of the core point cloud, affecting image quality.

Method used

The core point cloud positioning method based on probability distribution is adopted, through the integration of multi-frame images, obtaining local standard deviation images, OSTU method binary processing, screening the maximum area contour area, determining the actual end face area and point cloud location of the core, combining two The VGOS model is precisely positioned.

Benefits of technology

Significantly reduce the impact of random noise, enhance image quality, improve the accuracy of core positioning and diagnostic accuracy, especially in early stage cancer and micro-lesion detection.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118941635A_ABST
    Figure CN118941635A_ABST
Patent Text Reader

Abstract

The invention discloses a fiber core point cloud positioning method based on probability distribution and related equipment, and the method comprises the steps: carrying out the fusion operation based on an N-frame original image data set, so as to obtain a fused image; acquiring a local standard deviation image of the fused image; performing binarization processing on the local standard deviation image based on an OSTU method to obtain a binarized image; taking a contour region, meeting a preset screening condition, with the maximum area in the binarized image as a fiber core estimated end face region; acquiring a background area and a fiber core estimated center position based on the fiber core estimated end face area; further screening pixel points within a certain range at the junction of the background region and the fiber core estimated end face region based on a preset threshold value, and determining a fiber core actual end face region according to a screening result and the fiber core estimated end face region; performing fiber core positioning operation in the fiber core actual end face area to obtain a fiber core actual position center; and determining a fiber core point cloud in the fiber core actual end face area based on the fiber core actual position center.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of image processing, and more specifically, to a fiber core point cloud positioning method based on probability distribution and related equipment. Background Art

[0002] A microendoscope is a medical device that can be inserted into the human body through channels such as gastroscopes and colonoscopes to obtain local histological images to achieve accurate diagnosis of tiny lesions, gastrointestinal lesions, and early gastrointestinal cancer. There are two important components in the scanning control module of a microendoscope: a resonant mirror and a galvanometer galvanometer. The function of the resonant mirror is to make the light scan rapidly in the horizontal direction, so it is also called an X-galvanometer. The function of the galvanometer galvanometer is to make the light scan vertically, so it is also called a Y-galvanometer. The two work together to obtain a two-dimensional image.

[0003] The resonant mirror and galvanometer galvanometer mirror contain precise electronic components, and the characteristics of these electronic components will change with the change of the working environment temperature. During the imaging process, each image is based on the predetermined core position and generates a core point cloud. Due to the change of the galvanometer characteristics, the core position changes during the imaging process, which leads to inaccurate core point cloud positioning, resulting in image quality degradation. Therefore, it is necessary to propose a more accurate core point cloud positioning method. Summary of the invention

[0004] A series of simplified concepts are introduced in the Summary of the Invention section, which will be further described in detail in the Detailed Description of the Invention section. The Summary of the Invention section of this application does not mean to attempt to define the key features and essential technical features of the claimed technical solution, nor does it mean to attempt to determine the scope of protection of the claimed technical solution.

[0005] In a first aspect, the present application proposes a fiber core point cloud positioning method based on probability distribution, the method comprising:

[0006] Perform a fusion operation based on N frames of original image data sets to obtain a fused image;

[0007] Obtaining a local standard deviation image of the above fused image;

[0008] The local standard deviation image is binarized based on the OSTU method to obtain a binarized image;

[0009] The largest contour area in the binary image that meets the preset screening conditions is used as the estimated end face area of ​​the fiber core, wherein the preset screening conditions include the lowest area condition and the roundness condition;

[0010] Based on the above-mentioned fiber core estimated end face area, a background area and a fiber core estimated center position are obtained;

[0011] Pixel points within a certain range of the intersection of the background area and the estimated end face area of ​​the fiber core are further screened based on a preset threshold, and the actual end face area of ​​the fiber core is determined based on the screening result and the estimated end face area of ​​the fiber core, wherein the preset threshold is determined based on the estimated center position of the fiber core;

[0012] Performing a core positioning operation in the actual end face region of the core to obtain the actual position center of the core;

[0013] A core point cloud is determined in the actual end face region of the core based on the actual position center of the core.

[0014] In a feasible implementation manner, the above fusion operation includes a mean fusion operation and a median fusion operation.

[0015] In a feasible implementation manner, the step of obtaining the local standard deviation image of the fused image includes:

[0016] In the above fused image, a square target frame with a size of side length is preset to traverse the image, and all pixels immediately surrounding the target frame are obtained to calculate their standard deviations as the local standard deviation image.

[0017] In a feasible implementation manner, the above-mentioned obtaining the background area and the estimated core center position based on the estimated core end face area includes:

[0018] The pixels surrounding the estimated end face region of the fiber core and marked as 0 in the binary image are defined as the background region;

[0019] A circle fitting is performed based on the edge points of the estimated end face region of the fiber core to determine the estimated center position of the fiber core.

[0020] In a feasible implementation manner, the pixel points within a certain range of the intersection of the background area and the estimated end face area of ​​the core are further screened based on a preset threshold, and the actual end face area of ​​the core is determined according to the screening result and the estimated end face area of ​​the core, including:

[0021] Calculate the standard deviation of all pixel points within a certain radius from the preset center of the fiber core as the preset threshold;

[0022] The pixel points whose standard deviation within a certain range at the junction of the background area and the estimated end face area of ​​the fiber core reaches the first multiple range of the preset threshold are excluded, and the remaining pixel points in the estimated end face area of ​​the fiber core are used as the actual end face area of ​​the fiber core.

[0023] In a feasible implementation manner, the above-mentioned core positioning operation is performed in the actual end face region of the core to obtain the actual position center of the core, including:

[0024] A traversal operation is performed in the actual end face area of ​​the above-mentioned fiber core with a positioning window, and during the traversal process, the pixel point where the local maximum point coincides with the traversal point is determined as the position center of the primary screening fiber core;

[0025] The mean value of pixels in the above background area is taken as the background brightness;

[0026] The difference between the brightness value of the initial screening position center and the background brightness is compared with the second multiple range of the average of the brightness difference between the surrounding N fiber cores and the background for re-screening to determine the actual position center of the fiber core.

[0027] In a feasible implementation manner, the determining of the core point cloud in the actual end face region of the core based on the actual position center of the core includes:

[0028] Marking each of the above-mentioned actual position centers of the fiber cores and all pixel points in their target neighborhood based on the same serial number and establishing a two-dimensional Gaussian model of brightness distribution, wherein the parameters of the above-mentioned two-dimensional Gaussian model are determined based on the coordinate information and brightness information of the above-mentioned actual position center of the fiber core;

[0029] Traversing the unmarked points in the actual end surface area of ​​the core with a specific traversal frame;

[0030] Calculate the pixel model brightness value according to the above two-dimensional Gaussian model of each of the above unmarked points;

[0031] According to the magnitude relationship between the pixel brightness value of the unmarked point and the brightness value of the pixel model, the point cloud belonging probability corresponding to the point is calculated;

[0032] For each pixel point, select the fiber core to which the point cloud has the highest probability of belonging, and mark the pixel point as the serial number of the fiber core;

[0033] The pixel points with the same serial number as each core are defined as the core point cloud.

[0034] In a second aspect, an embodiment of the present application proposes a core point cloud positioning device based on probability distribution, comprising:

[0035] A first acquisition unit, configured to perform a fusion operation based on the N frames of original image data set to acquire a fused image;

[0036] A second acquisition unit, used for acquiring a local standard deviation image of the fused image;

[0037] A third acquisition unit is used to perform binarization processing on the local standard deviation image based on the OSTU method to obtain a binary image;

[0038] A fourth acquisition unit is used to select the contour area with the largest area that meets the preset screening conditions in the binary image as the estimated end face area of ​​the fiber core, wherein the preset screening conditions include the lowest area condition and the roundness condition;

[0039] A fifth acquisition unit, configured to acquire a background area and an estimated center position of the core based on the estimated end face area of ​​the core;

[0040] A first determining unit is used to further screen the pixel points within a certain range of the intersection of the background area and the estimated end face area of ​​the fiber core based on a preset threshold, and determine the actual end face area of ​​the fiber core according to the screening result and the estimated end face area of ​​the fiber core, wherein the preset threshold is determined based on the estimated center position of the fiber core;

[0041] A sixth acquisition unit, used for performing a core positioning operation in the actual end face region of the core to obtain the actual position center of the core;

[0042] The second determining unit is used to determine the core point cloud in the actual end face area of ​​the core based on the actual position center of the core.

[0043] In a third aspect, an electronic device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor is used to implement the steps of the fiber core point cloud positioning method based on probability distribution as described in any one of the first aspects above when executing the computer program stored in the memory.

[0044] In a fourth aspect, the present application further proposes a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the fiber core point cloud positioning method based on probability distribution of any one of the first aspects is implemented.

[0045] In summary, the method proposed in the embodiment of the present application includes: performing a fusion operation based on N frames of original image data sets to obtain a fused image; obtaining a local standard deviation image of the above fused image; binarizing the above local standard deviation image based on the OSTU method to obtain a binary image; taking the largest area contour area that meets the preset screening conditions in the above binary image as the core estimated end face area, wherein the preset screening conditions include the lowest area condition and the roundness condition; obtaining the background area and the core estimated center position based on the above core estimated end face area; further screening the pixel points within a certain range at the junction of the above background area and the above core estimated end face area based on a preset threshold, and determining the core actual end face area based on the screening results and the above core estimated end face area, wherein the above preset threshold is determined based on the above core estimated center position; performing a core positioning operation in the above core actual end face area to obtain the core actual position center; determining the core point cloud in the above core actual end face area based on the above core actual position center. The method proposed in the embodiment of the present application significantly reduces the influence of random noise by merging multiple frames of images, and enhances the overall quality of the image, making the details in the image clearer, and providing a higher quality basis for more sophisticated image analysis. The local standard deviation image can highlight the texture and boundary details in the image, especially when the tissue structure is complex or the boundary is not obvious, so that the subsequent image segmentation and feature recognition are more accurate, which helps to better identify the lesion area. Through the OSTU algorithm, the system can automatically select the optimal threshold for image binarization, reduce the subjectivity of the operation, and improve the standardization and repeatability of the processing flow. By identifying contours that meet specific shape and size conditions, the end face area of ​​the fiber core is accurately estimated, and the accuracy of the core positioning is further improved. At the same time, accurate core positioning is performed in the actual end face area of ​​the determined core, and the core point cloud position is finally determined. This series of operations makes the imaging results highly consistent with the actual tissue structure. Improving the accuracy of core positioning is directly related to the accuracy of diagnosis, especially in the detection of early canceration and micro-lesions. Through the comprehensive application of these steps, this scheme greatly improves the image quality and diagnostic accuracy of microendoscopic imaging technology.

[0046] The fiber core point cloud positioning method based on probability distribution proposed in this application, and other advantages, objectives and features of this application will be partially reflected through the following description, and will also be partially understood by technical personnel in this field through research and practice of this application. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present specification. Also, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0048] Figure 1 A schematic diagram of the process of a fiber core point cloud positioning method based on probability distribution provided in an embodiment of the present application;

[0049] Figure 2 A schematic diagram of a positioning window provided in an embodiment of the present application;

[0050] Figure 3 A structural schematic diagram of a fiber core point cloud positioning device based on probability distribution provided in an embodiment of the present application;

[0051] Figure 4 A schematic diagram of the structure of a core point cloud positioning electronic device based on probability distribution provided in an embodiment of the present application. DETAILED DESCRIPTION

[0052] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" and "having" and any of their variations 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 necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices. The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments.

[0053] See also Figure 1 , which is a schematic diagram of a process of a fiber core point cloud positioning method based on probability distribution provided in an embodiment of the present application, which may specifically include:

[0054] S110, performing a fusion operation based on the N frames of original image data set to obtain a fused image;

[0055] Exemplarily, the confocal mainframe is adjusted to a position where the focus is clear and the image brightness is moderate, N frames of original images are continuously acquired, and a comprehensive fused image is obtained through fusion processing (such as image superposition or averaging) to improve image stability and signal-to-noise ratio.

[0056] S120, obtaining a local standard deviation image of the fused image;

[0057] Exemplarily, the local standard deviation is calculated from the fused image to generate a local standard deviation image, which can highlight the texture and boundary areas of the image and facilitate subsequent feature extraction.

[0058] S130, performing binarization processing on the local standard deviation image based on the OSTU method to obtain a binarized image;

[0059] Exemplarily, the OSTU method is used to automatically select the optimal threshold and convert the local standard deviation image into a binary image to highlight important image features.

[0060] S140, taking the contour region with the largest area that meets the preset screening conditions in the binary image as the estimated end face region of the fiber core, wherein the preset screening conditions include an area condition and a roundness condition;

[0061] Exemplarily, the largest contour that meets the minimum area and circularity conditions is selected in the binary image as the estimated end face area of ​​the fiber core.

[0062] Specifically, the contour of the binary image is located, and the contour area with the largest area is taken as the core end face area. At the same time, the core end face area must meet the screening conditions, including whether the area meets certain conditions and whether the shape is round enough.

[0063] S150, acquiring a background area and an estimated center position of the core based on the estimated end face area of ​​the core;

[0064] Exemplarily, the center position of the estimated end face region of the core is determined, and the background region around it is distinguished to provide a reference for more accurate subsequent processing.

[0065] S160, further screening the pixel points within a certain range at the junction of the background area and the estimated fiber core end face area based on a preset threshold, and determining the actual fiber core end face area according to the screening result and the estimated fiber core end face area, wherein the preset threshold is determined based on the estimated center position of the fiber core;

[0066] Exemplarily, detailed pixel point screening is performed at the junction of the estimated end face area of ​​the fiber core and the background, and the actual end face area of ​​the fiber core is further determined based on a preset threshold.

[0067] S170, performing a core positioning operation in the actual end face region of the core to obtain the actual position center of the core;

[0068] Exemplarily, precise positioning is performed within the confirmed actual end face region of the fiber core to determine the actual center position of the fiber core.

[0069] S180, determining a core point cloud in the actual end face region of the core based on the actual position center of the core.

[0070] Exemplarily, based on the actual position center of the fiber core, a fiber core point cloud is generated in the actual end face region of the fiber core to complete accurate image reconstruction.

[0071] In summary, the method proposed in the embodiment of the present application significantly reduces the influence of random noise by merging multiple frames of images, and enhances the overall quality of the image, making the details in the image clearer, and providing a higher quality basis for more sophisticated image analysis. The local standard deviation image can highlight the texture and boundary details in the image, especially when the tissue structure is complex or the boundary is not obvious, so that the subsequent image segmentation and feature recognition are more accurate, which helps to better identify the lesion area. Through the OSTU algorithm, the system can automatically select the optimal threshold for image binarization, reduce the subjectivity of the operation, and improve the standardization and repeatability of the processing flow. By identifying contours that meet specific shape and size conditions, the end face area of ​​the fiber core is accurately estimated, and the accuracy of the core positioning is further improved. At the same time, accurate core positioning is performed in the actual end face area of ​​the determined core, and the core point cloud position is finally determined. This series of operations makes the imaging results highly consistent with the actual tissue structure. Improving the accuracy of core positioning is directly related to the accuracy of diagnosis, especially in the detection of early canceration and micro-lesions. Through the comprehensive application of these steps, this scheme greatly improves the image quality and diagnostic accuracy of microendoscopic imaging technology.

[0072] In some examples, the fusion operation includes a mean fusion operation and a median fusion operation.

[0073] Exemplarily, mean fusion is performed by calculating the average intensity value of each pixel of multiple frames of images. Specifically, for each pixel, the intensity values ​​of the pixel in all images are added together and then divided by the total number of images to obtain the average intensity value of the pixel.

[0074] Median fusion involves sorting the pixel values ​​at each pixel location and then selecting the middle value as the final pixel value. For each pixel point, the pixel values ​​of that point are extracted from all images, and these values ​​are sorted and the median is taken as the pixel value of the fused image at that point.

[0075] By combining these two fusion methods, this solution can make full use of their respective advantages and select the most appropriate fusion strategy according to different image content and specific application requirements, thereby optimizing image quality and improving the diagnostic accuracy and reliability of endoscopic images. This has important clinical significance for the early detection and diagnosis of gastrointestinal diseases, especially microlesions and early cancers.

[0076] In some examples, the step of obtaining a local standard deviation image of the fused image includes:

[0077] In the above fused image, a square target frame with a size of side length is preset to traverse the image, and all pixels immediately surrounding the target frame are obtained to calculate their standard deviations as the local standard deviation image.

[0078] For example, when processing a fused image, a square target box is first defined, and the side length of the box is preset. This side length is usually selected based on the resolution of the image and the required level of detail. A smaller side length may capture more subtle changes, while a larger side length may be more suitable for capturing changes in a larger area.

[0079] Slide the square target frame from one end to the other on the fused image to cover every possible position of the image. This sliding is usually done pixel by pixel to ensure that all areas are covered so that no possible details are missed.

[0080] At each location covered by the target box, calculate the standard deviation of all pixels in the box. The local standard deviation is a statistic that measures the degree of discreteness of the pixel value distribution. The calculation formula is:

[0081]

[0082] Where N is the total number of pixels in the target box, x i is the value of the i-th pixel in the target box, and μ is the average value of all pixel values ​​in the target box.

[0083] The calculated standard deviation value of each position is assigned to the corresponding central pixel position, thereby constructing a complete local standard deviation image. In this way, the value of each pixel represents the degree of discreteness of the pixel value in its corresponding area.

[0084] Specifically, the local standard deviation image of the fused image is calculated by traversing the image, taking all pixels in the area around the traversed pixel, and calculating their standard deviations. The size of the neighborhood is a square area with a certain side length centered on P, and the side length value range can be {9,13}.

[0085] By using the method proposed in the embodiment of the present application, the local standard deviation image can highlight the texture and boundary details in the original image. By emphasizing the local changes in the image, this method helps with subsequent image analysis tasks, such as feature recognition, edge detection, and target classification. By adjusting the size of the target box, it can flexibly adapt to different image analysis requirements, so that the method can be widely used in different medical imaging scenarios.

[0086] In some examples, the obtaining of the background area and the estimated center position of the core based on the estimated end face area of ​​the core includes:

[0087] The pixels surrounding the estimated end face region of the fiber core and marked as 0 in the binary image are defined as the background region;

[0088] A circle fitting is performed based on the edge points of the estimated end face region of the fiber core to determine the estimated center position of the fiber core.

[0089] Exemplarily, the pixels surrounding the core end face region and marked as 0 in the binary image are defined as the background region, and the average value of the pixels in the background region is taken as the background brightness. The core center is estimated by performing circle fitting based on the edge points of the core end face region.

[0090] In some examples, the pixel points within a certain range of the intersection of the background area and the estimated end face area of ​​the fiber core are further screened based on a preset threshold, and the actual end face area of ​​the fiber core is determined according to the screening result and the estimated end face area of ​​the fiber core, including:

[0091] Calculate the standard deviation of all pixel points within a certain radius from the preset center of the fiber core as the preset threshold;

[0092] The pixel points whose standard deviation within a certain range at the junction of the background area and the estimated end face area of ​​the fiber core reaches the first multiple range of the preset threshold are excluded, and the remaining pixel points in the estimated end face area of ​​the fiber core are used as the actual end face area of ​​the fiber core.

[0093] Exemplarily, first, in the estimated core end face area, all pixels within a certain radius centered on the estimated core center are selected. The standard deviation of the intensity values ​​of these pixels is calculated, and the standard deviation is used as a preset threshold for subsequent screening operations. This standard deviation reflects the variability of pixel intensity in the core center area, providing a benchmark for subsequent comparisons, thereby helping to identify and exclude abnormal or inconsistent pixels.

[0094] At the junction of the background area and the estimated end face area of ​​the fiber core, check all pixels within a certain range. Compare whether the standard deviation of these pixels reaches the first multiple range of the above preset threshold. If the standard deviation of a pixel exceeds this range, these pixels are excluded. This step can effectively exclude those pixels that may have abnormal standard deviations due to noise, imaging errors or other factors, and retain more stable and consistent pixels as part of the actual end face area of ​​the fiber core.

[0095] After completing the above screening, the remaining pixels are considered to be the actual end face area of ​​the fiber core. These remaining pixels show relatively consistent and stable optical properties, which more accurately represent the actual position of the fiber core.

[0096] Specifically, the standard deviation of all pixel points within a certain radius from the core center is calculated as the preset threshold TH, and the pixel points within a certain range at the junction of the background area and the core end face area are taken to re-judge whether they belong to the core end face. The judgment condition is: the standard deviation reaches a certain multiple of TH, and the first multiple range may be {0.7, 0.9}. Points that do not belong to the core end face are excluded, and the remaining core end face area is defined as the final core end face area.

[0097] This method can reduce the image quality degradation and inaccurate positioning caused by changes in the galvanometer characteristics of the imaging device, environmental interference, etc. The actual end face area of ​​the fiber core finally determined will be closer to the real tissue structure, thereby improving the accuracy and reliability of diagnosis.

[0098] In some examples, the above-mentioned core positioning operation is performed in the actual end face region of the core to obtain the actual position center of the core, including:

[0099] A traversal operation is performed in the actual end face area of ​​the above-mentioned fiber core with a positioning window, and during the traversal process, the pixel point where the local maximum point coincides with the traversal point is determined as the position center of the primary screening fiber core;

[0100] The mean value of pixels in the above background area is taken as the background brightness;

[0101] The difference between the brightness value of the initial screening position center and the background brightness is compared with the second multiple range of the average of the brightness difference between the surrounding N fiber cores and the background for re-screening to determine the actual position center of the fiber core.

[0102] Exemplarily, within the determined actual end face region of the fiber core, a positioning window is used to perform a traversal operation. The positioning window may be a small square or circular region used for local analysis of the image.

[0103] During the traversal process, find the local maximum point, that is, find the pixel with the highest brightness in the positioning window. Whenever the local maximum point in the positioning window coincides with the current traversal point, record this point as the center of the initial screening fiber core position.

[0104] Calculate the average brightness of all pixels outside the actual end face area of ​​the fiber core, that is, in the background area. This value is defined as the background brightness. The background brightness is used to help distinguish the fiber core and non-fiber core areas.

[0105] Only when the brightness of the initial screening position center is significantly higher than the second multiple range of the background brightness, can this point be confirmed as the actual position center of the fiber core, ensuring that the screened fiber core position center is true and reliable.

[0106] Specifically, the points in the core end face area are traversed, and the location of the local maximum value is located with a certain window. The window value is determined according to the situation, such as Figure 2 As shown, for example, a 13-neighborhood centered at the point is used as a positioning window.

[0107] If the traversal point coincides with the local maximum point, it means that the point is a possible core position and it is marked. After all possible core positions are marked, possible core positions with lower core brightness are excluded. The judgment standard for lower core brightness can be: the difference between its brightness and the background brightness does not reach a certain multiple of the average of the brightness differences between the surrounding N cores and the background. N can be {20,100}, and the second multiple range can be {0.5,0.75}. The remaining core positions are marked as the last core positions.

[0108] The method proposed in this embodiment greatly improves the accuracy of core center positioning and ensures the correctness of the core position by screening local maximum points and comparing with the background brightness. It ensures that the actual position of the core is determined only when the core brightness is significantly higher than the background brightness, thereby reducing misdiagnosis due to image noise or other non-core structures. The entire process can be performed through automated software, reducing manual intervention by doctors or technicians in imaging analysis and improving the efficiency and repeatability of the diagnostic process.

[0109] In some examples, determining the core point cloud in the actual end face region of the core based on the actual position center of the core includes:

[0110] Marking each of the above-mentioned actual position centers of the fiber cores and all pixel points in their target neighborhood based on the same serial number and establishing a two-dimensional Gaussian model of brightness distribution, wherein the parameters of the above-mentioned two-dimensional Gaussian model are determined based on the coordinate information and brightness information of the above-mentioned actual position center of the fiber core;

[0111] Traversing the unmarked points in the actual end surface area of ​​the core with a specific traversal frame;

[0112] Calculate the pixel model brightness value according to the above two-dimensional Gaussian model of each of the above unmarked points;

[0113] According to the magnitude relationship between the pixel brightness value of the unmarked point and the brightness value of the pixel model, the point cloud belonging probability corresponding to the point is calculated;

[0114] For each pixel point, select the fiber core to which the point cloud has the highest probability of belonging, and mark the pixel point as the serial number of the fiber core;

[0115] The pixel points with the same serial number as each core are defined as the core point cloud.

[0116] Exemplarily, for all the pixels in the determined center of the actual position of the core and its neighborhood, a two-dimensional Gaussian model is established based on the coordinate information and brightness information of these points. The two-dimensional Gaussian model is a common method used to describe the degree to which data points are clustered around the center, where parameters (such as mean and covariance) describe the shape and direction of this distribution. The mean is usually set to the coordinate of the center of the actual position of the core, and the covariance structure is determined based on the variability of the brightness distribution. This model helps simulate the brightness values ​​of each pixel, assuming that they are distributed around the center according to a Gaussian distribution.

[0117] Using a specific traversal box, all unmarked points are traversed in the actual end face area of ​​the fiber core. This step is to determine the point cloud belonging probability of these unmarked points.

[0118] For each unmarked point traversed, the theoretical pixel brightness value is calculated using the two-dimensional Gaussian model established previously. This value represents the brightness that the pixel at that position should have if there are no other influencing factors.

[0119] The actual pixel brightness value of each unlabeled point is compared with the model brightness value calculated from the two-dimensional Gaussian model. Based on the difference between the two, the probability that this point belongs to the core point cloud is calculated. If the actual brightness value is close to the model brightness value, the probability of belonging is high; if the difference is large, the probability of belonging is low.

[0120] For each pixel point, select the core with the highest probability of belonging, and mark the pixel point as the serial number of the core. Then, all the pixels corresponding to the same core serial number are defined as the point cloud of the core.

[0121] Specifically, the core is traversed, and the pixel where the core is located and its eight neighboring pixels (upper left, upper, upper right, left, right, lower left, lower and lower right) are marked with the same sequence number. A two-dimensional Gaussian model is established for the brightness distribution of the core and its eight neighbors. The model has four parameters A, μ x , μ y and σ , where A is the brightness of the center point of the Gaussian model, μx and μ y is the coordinate of the center point of the Gaussian model, σ The standard deviation of the Gaussian distribution traverses the unmarked pixel point P in the core end face area, and its brightness value is L1. Take all the cores in its local neighborhood. The size of the neighborhood is a square area with a certain side length centered on P, and the side length range is {7,11}. Calculate the probability that the pixel P belongs to each core. The probability is calculated as follows: According to the two-dimensional Gaussian model of the brightness distribution of each core, calculate the model brightness value L2 of the pixel P. If L1>L2, the probability is L2 / L1, if L1≤L2, the probability is L1 / L2; take the core corresponding to the maximum probability value, and mark the pixel point P as the serial number of the core. Define the pixel points with the same serial number as each core as the core point cloud.

[0122] By using a Gaussian model based on actual brightness information, this solution can more accurately simulate and predict the brightness distribution of the fiber core, thereby improving the accuracy of fiber core point cloud positioning, reducing dependence on manual intervention, and improving the consistency and repeatability of operations through automated data analysis. It is suitable for fiber core identification and point cloud construction of various complexities and can adapt to different imaging conditions and requirements.

[0123] See also Figure 3 , a structural schematic diagram of a fiber core point cloud positioning device based on probability distribution provided in an embodiment of the present application may include:

[0124] A first acquisition unit 21 is used to perform a fusion operation based on N frames of original image data sets to obtain a fused image;

[0125] A second acquisition unit 22, used to acquire a local standard deviation image of the fused image;

[0126] A third acquisition unit 23 is used to perform binarization processing on the local standard deviation image based on the OSTU method to obtain a binary image;

[0127] A fourth acquisition unit 24 is used to select the contour area with the largest area that meets the preset screening conditions in the above binary image as the estimated end face area of ​​the fiber core, wherein the preset screening conditions include the lowest area condition and the roundness condition;

[0128] A fifth acquisition unit 25, configured to acquire a background area and an estimated core center position based on the estimated core end face area;

[0129] A first determining unit 26 is used to further screen the pixel points within a certain range of the intersection of the background area and the estimated fiber core end face area based on a preset threshold, and determine the actual fiber core end face area according to the screening result and the estimated fiber core end face area, wherein the preset threshold is determined based on the estimated center position of the fiber core;

[0130] A sixth acquisition unit 27, configured to perform a core positioning operation in the actual end face region of the core to obtain the actual position center of the core;

[0131] The second determining unit 28 is used to determine the core point cloud in the actual end face region of the core based on the actual position center of the core.

[0132] like Figure 4 As shown, an embodiment of the present application also provides an electronic device 300, including a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and executable on the processor, and when the processor 320 executes the computer program 311, the steps of any of the above-mentioned methods for core point cloud positioning based on probability distribution are implemented.

[0133] Since the electronic device introduced in this embodiment is a device used to implement a fiber core point cloud positioning device based on probability distribution in the embodiment of the present application, based on the method introduced in the embodiment of the present application, the technical personnel in this field can understand the specific implementation mode of the electronic device of this embodiment and its various variations. Therefore, how the electronic device implements the method in the embodiment of the present application is not introduced in detail here. As long as the equipment adopted by the technical personnel in this field to implement the method in the embodiment of the present application, it belongs to the scope of protection of this application.

[0134] In the specific implementation process, when the computer program 311 is executed by the processor, it can achieve Figure 1 Any implementation manner in the corresponding embodiments.

[0135] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0136] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0137] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, 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 computer, or other programmable data processing device to generate 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 flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0138] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate 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 process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0139] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0140] The embodiment of the present application also provides a computer program product, which includes computer software instructions. When the computer software instructions are executed on a processing device, the processing device executes the process of core point cloud positioning based on probability distribution in the corresponding embodiment.

[0141] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on the computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website site, a computer, a server or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium can be any available medium that a computer can store or a data storage device such as a server or a data center that includes one or more available media integration. The available medium can be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid state drive (SSD)), etc.

[0142] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0143] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0144] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0145] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0146] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), disk or optical disk and other media that can store program codes.

[0147] The above embodiments 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, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some of the technical features therein by equivalents. However, these 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.

Claims

1. A fiber core point cloud positioning method based on probability distribution, characterized in that: include: Perform a fusion operation based on N frames of original image data sets to obtain a fused image; Acquire a local standard deviation image of the fused image; Binarizing the local standard deviation image based on the OSTU method to obtain a binary image; The largest contour area in the binary image that meets the preset screening conditions is used as the estimated end face area of ​​the fiber core, wherein the preset screening conditions include an area condition and a roundness condition; Acquire the background area and the estimated center position of the core based on the estimated end face area of ​​the core; Pixel points within a certain range at the junction of the background area and the estimated core end face area are further screened based on a preset threshold, and the actual core end face area is determined based on the screening result and the estimated core end face area, wherein the preset threshold is determined based on the estimated center position of the core; Performing a core positioning operation in the actual end face region of the core to obtain the actual position center of the core; A core point cloud is determined in the actual end face region of the core based on the actual position center of the core.

2. The method for core point cloud positioning based on probability distribution according to claim 1, characterized in that: The fusion operation includes a mean fusion operation and / or a median fusion operation.

3. The method for core point cloud positioning based on probability distribution according to claim 1, characterized in that: The step of acquiring a local standard deviation image of the fused image comprises: A square target frame with a size of side length is preset in the fused image to traverse the image, and all pixels immediately surrounding the target frame are obtained to calculate their standard deviations as the local standard deviation image.

4. The method for core point cloud positioning based on probability distribution according to claim 1, characterized in that: The obtaining of the background area and the estimated center position of the core based on the estimated end face area of ​​the core includes: The pixels surrounding the estimated end face region of the fiber core and marked as 0 in the binary image are defined as the background region; A circle fitting is performed according to edge points of the estimated end face region of the fiber core to determine the estimated center position of the fiber core.

5. The method for core point cloud positioning based on probability distribution according to claim 1, characterized in that: The pixel points within a certain range of the intersection of the background area and the estimated end face area of ​​the fiber core are further screened based on a preset threshold, and the actual end face area of ​​the fiber core is determined according to the screening result and the estimated end face area of ​​the fiber core, including: Calculate the standard deviation of all pixel points within a certain radius from the preset center of the fiber core as the preset threshold; The pixel points whose standard deviation within a certain range at the junction of the background area and the estimated core end face area reaches the first multiple range of the preset threshold are excluded, and the remaining pixel points of the estimated core end face area are used as the actual core end face area.

6. The method for core point cloud positioning based on probability distribution according to claim 1, characterized in that: The performing of a core positioning operation in the actual end face region of the core to obtain the actual position center of the core includes: Performing a traversal operation with a positioning window in the actual end face area of ​​the fiber core, and determining the pixel point where the local maximum point coincides with the traversal point as the center of the primary screening fiber core position during the traversal process; Taking the mean value of pixels in the background area as background brightness; The difference between the brightness value of the center of the initial screening position and the background brightness is compared with the second multiple range of the average of the brightness difference values ​​of the surrounding N fiber cores and the background for re-screening to determine the actual position center of the fiber core.

7. The method for core point cloud positioning based on probability distribution according to claim 1, characterized in that: Determining the core point cloud in the actual end face region of the core based on the actual position center of the core includes: Marking the actual position center of each fiber core and all pixel points in its target neighborhood based on the same serial number and establishing a two-dimensional Gaussian model of brightness distribution, wherein the parameters of the two-dimensional Gaussian model are determined based on the coordinate information and brightness information of the actual position center of the fiber core; traversing the unmarked points in the actual end surface area of ​​the core with a specific traversal frame; Calculate the pixel model brightness value according to the two-dimensional Gaussian model of each unmarked point; Calculate the point cloud belonging probability corresponding to the point according to the magnitude relationship between the pixel brightness value of the unmarked point and the brightness value of the pixel model; For each pixel point, select the fiber core to which the point cloud has the highest probability of belonging, and mark the pixel point as the serial number of the fiber core; The pixel points having the same sequence number as each fiber core are defined as the fiber core point cloud.

8. A fiber core point cloud positioning device based on probability distribution, characterized in that: include: A first acquisition unit, configured to perform a fusion operation based on the N frames of original image data set to acquire a fused image; A second acquisition unit, used for acquiring a local standard deviation image of the fused image; A third acquisition unit is used to perform binarization processing on the local standard deviation image based on the OSTU method to obtain a binary image; A fourth acquisition unit, configured to use the contour region with the largest area that meets a preset screening condition in the binary image as the estimated end face region of the fiber core, wherein the preset screening condition includes a minimum area condition and a roundness condition; A fifth acquisition unit, configured to acquire a background area and an estimated core center position based on the estimated core end face area; A first determining unit is used to further screen the pixel points within a certain range of the intersection of the background area and the estimated fiber core end face area based on a preset threshold, and determine the actual fiber core end face area according to the screening result and the estimated fiber core end face area, wherein the preset threshold is determined based on the estimated center position of the fiber core; A sixth acquisition unit, configured to perform a core positioning operation in the actual end face region of the core to obtain the actual position center of the core; The second determining unit is configured to determine a core point cloud in an actual end face region of the core based on the actual position center of the core.

9. An electronic device, comprising: A memory and a processor, characterized in that the processor is used to implement the steps of the fiber core point cloud positioning method based on probability distribution as described in any one of claims 1 to 7 when executing the computer program stored in the memory.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the fiber core point cloud positioning method based on probability distribution are implemented as described in any one of claims 1 to 7.