Abdominal B-ultrasonic data intelligent analysis system for gynaecology and obstetrics

Abstract

The invention relates to the technical field of image enhancement, in particular to an intelligent abdominal B-ultrasonic data analysis system for gynaecology and obstetrics, which comprises an image acquisition module, an image analysis module, an algorithm optimization module and an image enhancement module, the image analysis module is used for acquiring gradient factors of pixel points according to gray difference among the pixel points in the ultrasonic image, performing region segmentation on the ultrasonic image, acquiring a boundary line in the ultrasonic image, and determining the gradient factors of the pixel points according to gray change conditions of all the pixel points from a starting point to an ending point in the boundary line in the vertical direction; obtaining a boundary area pixel point set; the algorithm optimization module is used for optimizing a gamma transformation algorithm by combining the position information of the pixel points in the pixel point set of the boundary area; the image enhancement module is used for image enhancement. According to the method, the enhancement effect of the gamma transformation algorithm on the abdominal ultrasound image is improved.

Description

Technical Field

[0001] The invention relates to the technical field of image enhancement, and in particular to an intelligent analysis system for abdominal B-ultrasound data used in obstetrics and gynecology. Background Art

[0002] The abdominal B-ultrasound image data of obstetrics and gynecology usually uses enhancement methods to visually magnify the relevant contours and detailed features of the target. In general images, the detailed contours of the target will be lost due to the influence of some noise. For example, the umbilical cord of the fetus in the B-ultrasound image of obstetrics and gynecology may be partially incomplete. The darker part is enhanced by gamma transformation, but the boundary between the fetus and the posterior wall becomes blurred during the enhancement process, affecting the overall enhancement effect. In order to solve this blurring effect, the present invention makes relevant improvements to the gamma transformation in this scenario.

[0003] In the process of overall enhancement of B-ultrasound images, gamma transform is used to enhance them in order to eliminate the loss effect of the fetal umbilical cord, but this will lead to a decrease in the distinction between the fetus and the posterior wall, affecting the overall enhancement effect of the image. Summary of the invention

[0004] The present invention provides an intelligent analysis system for abdominal B-ultrasound data used in obstetrics and gynecology to solve the existing problems.

[0005] The abdominal B-ultrasound data intelligent analysis system for obstetrics and gynecology of the present invention adopts the following technical solutions: An embodiment of the present invention provides an intelligent analysis system for abdominal B-ultrasound data for obstetrics and gynecology, the system comprising the following modules: An image acquisition module is used to perform ultrasound examination on obstetrics and gynecology patients through a medical B-ultrasound instrument to obtain ultrasound images of the patient's abdomen; An image analysis module is used to obtain a gradient factor of a pixel point according to the grayscale difference between the pixel points in the ultrasound image, and obtain a plurality of regions by performing regional segmentation on the ultrasound image, and obtain a boundary line and a starting point and an ending point of the boundary line in the ultrasound image according to the gradient factor, the grayscale level of the region, and the distribution of the regional edges, and obtain a pixel point set in the boundary area according to the grayscale change of all the pixel points between the starting point and the ending point in the boundary line in the vertical direction; An algorithm optimization module is used to optimize the value-taking method of the gamma parameter in the gamma transform algorithm by combining the position information of the pixel points in the pixel point set of the boundary area, so as to obtain an optimized gamma transform algorithm; The image enhancement module is used to enhance the ultrasound image through an optimized gamma transform algorithm.

[0006] Furthermore, after obtaining the ultrasound image of the patient's abdomen, the method also includes obtaining a target area and a non-target area in the ultrasound image through a semantic segmentation algorithm, wherein the target area is a fetal area in the ultrasound image, and the non-target area is an area outside the fetal area in the ultrasound image.

[0007] Furthermore, the method of obtaining the gradient factor of the pixel points according to the grayscale difference between the pixel points in the ultrasound image, and obtaining a plurality of regions by performing regional segmentation on the ultrasound image, and obtaining the boundary line in the ultrasound image and the starting point and the ending point of the boundary line according to the gradient factor, the grayscale level of the region and the distribution of the regional edge, includes the following specific methods: Obtaining a gradient factor of an arbitrary pixel point according to a grayscale difference between the arbitrary pixel point and an adjacent pixel point; Performing regional segmentation on the ultrasound image to obtain a number of regions, and obtaining a determination gradient value based on the grayscale level and size of the region and the grayscale value of the pixel point at the edge of the region and the target region; When the gradient factor of a pixel point is greater than the judgment gradient value, the pixel point is recorded as a marked point, two curves formed by all the marked points are obtained and recorded as the boundary line, the Euclidean distance between the pixel points of the two boundary lines is obtained, and the two pairs of pixel points with the closest Euclidean distance are recorded as the starting point pair and the ending point pair, respectively, the starting point pair includes two starting points, and the ending point pair includes two ending points.

[0008] Furthermore, the ultrasound image is segmented to obtain a plurality of regions, and a determination gradient value is obtained based on the gray level and size of the region and the gray value of the pixel point at the edge of the region and the target region, including the specific method of: The ultrasound image is segmented using the watershed method to obtain several regions, and the average grayscale of all pixels in each region is obtained as the overall grayscale of the corresponding region, and the region with the largest area and the smallest overall grayscale is recorded as the target background, and the overall grayscale of the target background is recorded as the first grayscale; Obtain pixel points at the edges of all regions corresponding to the ultrasound image segmented by the watershed algorithm, record them as suspected edge points, and obtain the average grayscale of all suspected edge points on the edge of the target region, record them as the second grayscale; The difference between the second grayscale and the first grayscale is used as the determination gradient value.

[0009] Furthermore, the method of obtaining a set of pixel points in the boundary area according to the grayscale changes of all pixel points from the starting point to the end point in the boundary line in the vertical direction includes the following specific methods: Get the actual junction point; Connect the two starting points in the starting point pair, and similarly, connect the two ending points in the ending point pair, and obtain a set of all pixel points in the closed area formed by the starting point pair, the ending point pair, and all actual intersection points, which is recorded as the intersection area pixel point set.

[0010] Furthermore, the specific method for obtaining the actual intersection point is: Any pixel point between the starting point and the ending point on the boundary line is recorded as the target point, and a sequence of grayscale values ​​of A pixels in the vertical direction with the target point as the middle pixel is obtained, which is recorded as a grayscale column, and a number of grayscale columns are obtained, and the number of grayscale columns is equal to the number of columns between the starting point and the ending point, where A is a preset parameter; Get the second-order derivative value of each element in any grayscale column, and record the pixel points corresponding to the elements with second-order derivative values ​​greater than 0 as the actual intersection points.

[0011] Furthermore, the method of optimizing the value of the gamma parameter in the gamma transform algorithm by combining the position information of the pixel points in the pixel point set in the boundary area to obtain the optimized gamma transform algorithm includes the following specific methods: Obtain a set formed by the coordinates of all pixel points in the boundary area pixel point set, recorded as the boundary coordinate set; The gamma transform is optimized using the coordinate information in the boundary coordinate set to obtain an optimized gamma transform algorithm.

[0012] Furthermore, the specific expression of the optimized gamma transform algorithm is: in, Indicates the location of The gray value of the pixel after transformation; is the preset hyperparameter, The ultrasound image is located at The gamma parameters of the pixel during gamma transformation; Indicates the location of The gray value of the pixel.

[0013] Furthermore, the gamma parameters during the gamma transformation also include determining the values ​​of the gamma parameters according to the position information of the pixel points.

[0014] Furthermore, the method of determining the value of the gamma parameter according to the position information of the pixel point includes the following specific methods: When the coordinates of the pixel point belong to the boundary coordinate set, the gamma parameter of the gamma transform algorithm is set to 1; when the coordinates of the pixel point do not belong to the boundary coordinate set, the gamma parameter of the gamma transform algorithm is set to a preset value.

[0015] The beneficial effect of the technical solution of the present invention is that by performing grayscale difference analysis on the pixels in the ultrasound image, the gradient factors of each area in the image can be accurately obtained. This analysis helps to accurately identify the boundaries of different tissues in complex abdominal ultrasound images. The image is divided into multiple small areas through regional segmentation technology, and analyzed in combination with features such as gradient factors, grayscale levels, and edge distribution. This fine-grained image segmentation technology can help distinguish different tissues and structures in ultrasound images, such as the boundary between the fetus and the uterine wall, thereby improving the accuracy of image processing, and further dynamically adjust the gamma parameters in the gamma transform algorithm based on the position information of the pixels in the boundary area. This optimization step makes the image enhancement effect more refined and targeted, and effectively improves the enhancement effect of the gamma transform algorithm on abdominal ultrasound images. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 The structure diagram of the intelligent analysis system of abdominal B-ultrasound data used in obstetrics and gynecology of the present invention; Figure 2 Shown is an example of an ultrasound image of a patient's abdomen; Figure 3 Shown are example images of ultrasound images after conventional gamma shifting. DETAILED DESCRIPTION

[0018] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation, structure, features and effects of the intelligent analysis system for abdominal B-ultrasound data for obstetrics and gynecology proposed by the present invention are described in detail below in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0019] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0020] The specific scheme of the intelligent analysis system for abdominal B-ultrasound data for obstetrics and gynecology provided by the present invention is described in detail below with reference to the accompanying drawings.

[0021] See also Figure 1 , which shows a structural block diagram of an intelligent analysis system for abdominal B-ultrasound data for obstetrics and gynecology provided by an embodiment of the present invention, the system includes the following modules: The image acquisition module 101 performs ultrasonic examination on a gynecological patient using a medical B-ultrasound instrument to obtain an ultrasonic image of the patient's abdomen.

[0022] It should be noted that if Figure 2 The figure shows an example of an ultrasound image of a patient's abdomen. Since B-ultrasound images are generated by the reflection of ultrasound waves, and the umbilical cord and other soft tissues (such as amniotic fluid, fetus, etc.) have different acoustic properties from the surrounding tissues, the blood vessels and tissues in the umbilical cord have weaker penetration of ultrasound waves, resulting in weaker detected echoes. The structure of the umbilical cord (such as blood vessels) may not reflect ultrasound waves as clearly as other tissues, resulting in incomplete or blurred image information, which in turn affects the clarity of the imaging. In particular, when the fetus is in a deeper position or the uterine wall is thicker, the penetration of ultrasound waves will be reduced, which may result in incomplete image information in the umbilical cord area. Figure 3 The figure shows an example of an ultrasound image after conventional gamma change. Therefore, in order to improve the completeness of information in the ultrasound image, the embodiment of the present invention processes the ultrasound image by an image enhancement method.

[0023] In order to realize the intelligent analysis system of abdominal B-ultrasound data for obstetrics and gynecology proposed in this embodiment, it is necessary to first collect ultrasound images. The specific process is as follows: First, an ultrasound image of the patient's abdomen is collected using a medical B-ultrasound device.

[0024] Then, the Gaussian filtering algorithm is used to denoise the ultrasound image.

[0025] Finally, a target area and a non-target area in the ultrasound image are obtained by a semantic segmentation algorithm, wherein the target area is the fetal area in the ultrasound image, and the non-target area is the area outside the fetal area in the ultrasound image.

[0026] So far, the ultrasound image is obtained through the above method.

[0027] The image analysis module 102 obtains the gradient factor of the pixel points according to the grayscale difference between the pixel points in the ultrasound image, and obtains a plurality of regions by performing regional segmentation on the ultrasound image, and obtains the boundary line and the starting point and the ending point of the boundary line in the ultrasound image according to the gradient factor, the grayscale level of the region and the distribution of the regional edge, and obtains the pixel point set of the boundary area according to the grayscale change of all the pixel points between the starting point and the ending point in the boundary line in the vertical direction.

[0028] It should be noted that the growth environment of the fetus determines the distribution law of the grayscale value of the pixels in the B-ultrasound image. Observing the edge of the area where the fetus is located (i.e., the target area), the difference between its grayscale value and the grayscale value of the growth environment is obvious, so the gradient value can be used here to find the edge of the target area; similarly, the difference in grayscale value between the boundary line of the non-target area and the pixel points of the growth environment of the target area is also obvious, and the gradient value is used to obtain the edge of the non-target area. The position where the distance between the two edges changes the slowest is the starting position and the end position of the boundary line, because the distance change of the edge of the non-boundary line is very obvious, and it suddenly slows down at the boundary line. The grayscale value difference between the boundary line and the environment can be quantified by the second-order derivative of the grayscale value, because it presents a "concave" feature in the image, then the first-order derivative of the discrete function composed of the grayscale values ​​of the surrounding pixels is 0 and the second-order derivative is greater than 0. In summary, this method is used to determine the spatial position function of the boundary line.

[0029] Specifically, in step S201, a boundary line and a starting point and an ending point of the boundary line in an ultrasonic image are obtained.

[0030] It should be noted that the discrete function composed of the grayscale values ​​of the pixels between the starting position and the ending position is calculated, and the second-order derivative of this discrete function is calculated. The purpose of this step is to obtain two boundary lines. Because the grayscale value of the pixel points on the boundary line is lower than that of the surrounding pixels, its first-order derivative is 0 and the second-order derivative is greater than 0, so that the spatial position function of the boundary line can be obtained.

[0031] As an embodiment, a method for obtaining a boundary line and a starting point and an ending point of the boundary line includes: First, the gradient factor of an arbitrary pixel is obtained according to the grayscale difference between the arbitrary pixel and adjacent pixels.

[0032] As an embodiment, the specific calculation method of the gradient factor of any pixel point is: in, Indicates that it is located The gradient factor of the pixel point, Indicates that it is located The gray value of the pixel, Indicates that it is located The gray value of the pixel, Indicates that it is located The gray value of the pixel, Indicates that it is located The gray value of the pixel, Indicates that it is located The gray value of the pixel.

[0033] Then, the ultrasound image is segmented to obtain a number of regions, and a determination gradient value is obtained based on the grayscale level and size of the region and the grayscale value of the pixel point at the edge of the region and the target region.

[0034] As an embodiment, the specific method for determining the gradient value is: (1) Use the watershed method to segment the ultrasound image into several regions. Obtain the average grayscale of all pixels in each region as the overall grayscale of the corresponding region. The region with the largest area and the smallest overall grayscale is recorded as the target background. The overall grayscale of the target background is recorded as the first grayscale.

[0035] (2) Obtain pixel points at the edges of all regions corresponding to the ultrasound image segmented by the watershed algorithm, record them as suspected edge points, and obtain the average grayscale of all suspected edge points on the edge of the target region, record them as the second grayscale.

[0036] (3) The difference between the second grayscale and the first grayscale is used as the gradient value .

[0037] Finally, when the gradient factor of a pixel point is greater than the judgment gradient value, the pixel point is recorded as a marked point, two curves formed by all the marked points are obtained and recorded as the boundary line, the Euclidean distance between the pixel points of the two boundary lines is obtained, and the two pairs of pixel points with the closest Euclidean distance are recorded as the starting point pair and the ending point pair, respectively. The starting point pair includes two starting points, and the ending point pair includes two ending points.

[0038] It should be noted that the purpose of obtaining the suspected edge points in the embodiment of the present invention is to distinguish the edge of the target area and the edge of the non-target area in the image. In the embodiment of the present invention, a starting point pair corresponds to a starting position, and an ending point pair corresponds to an ending position, such as Figure 2 The figure shows a schematic diagram of the boundary line, wherein the target edge line is the edge line of the target area, and the non-target edge line is the edge line of the non-target area.

[0039] Step S202, obtaining a set of pixel points in the boundary area according to the grayscale changes of all pixel points from the starting point to the end point in the boundary line in the vertical direction.

[0040] First, any pixel point between the starting point and the end point on the boundary line is recorded as the target point, and the vertical direction is obtained with the target point as the middle pixel point. The sequence formed by the grayscale values ​​of pixels is recorded as a grayscale column, and a number of grayscale columns are obtained. The number of grayscale columns is equal to the number of columns from the starting point to the end point. are preset parameters.

[0041] It should be noted that the parameters are preset based on experience. The value is 30, which can be adjusted according to actual conditions and is not specifically limited in the embodiment of the present invention.

[0042] Then, the second-order derivative value of each element in any grayscale column is obtained, and the pixel points corresponding to the elements whose second-order derivative values ​​are greater than 0 are recorded as the actual intersection points.

[0043] It should be noted that in the ultrasound image, since there is a long and narrow area belonging to the target background between the target area representing the fetal area and the non-target area, there are corresponding boundaries on both sides of the area, which means that there is a grayscale change at the corresponding boundaries, that is, each grayscale column element contains two maximum points, that is, the actual intersection points, and several actual intersection points are obtained.

[0044] Finally, the two starting points in the starting point pair are connected, and similarly, the two ending points in the ending point pair are also connected to obtain a set of all pixel points in the closed area formed by the starting point pair, the ending point pair and all actual intersection points, which is recorded as the intersection area pixel point set.

[0045] So far, the pixel point set in the boundary area is obtained through the above method.

[0046] The algorithm optimization module 103 optimizes the value-taking method of the gamma parameter in the gamma transformation algorithm in combination with the position information of the pixel points in the boundary area pixel point set to obtain an optimized gamma transformation algorithm.

[0047] It should be noted that if the grayscale information is added to the gamma transform only for the grayscale distribution information of the boundary line, so that the gamma transform retains or specifies the change value of this grayscale range, similar grayscale values ​​in other regions will also be retained or transformed, resulting in the loss of features of some regions, affecting the enhancement effect. Therefore, the embodiment of the present invention adds position information to perform gamma transform on the grayscale values ​​of pixels in a certain area.

[0048] Specifically, first, a set formed by the coordinates of all pixel points in the boundary area pixel point set is obtained, and recorded as the boundary coordinate set.

[0049] It should be noted that the formula for gamma transform is: ,in is the gray value of the pixel after gamma transformation, and is a positive constant, is the gamma parameter of the gamma transform; It is usually a preset value, which is preset to 1 based on experience and can be adjusted according to actual conditions. It can be seen from the formula expression of the gamma transform that only the grayscale information of the pixel point is used in the gamma transform function.

[0050] Then, the gamma transform is optimized using the coordinate information in the boundary coordinate set to obtain an optimized gamma transform algorithm.

[0051] As an embodiment, the specific expression of the optimized gamma transform algorithm is: in, Indicates the location of The gray value of the pixel after transformation; is the preset hyperparameter, The ultrasound image is located at The gamma parameters of the pixel during gamma transformation; Indicates the location of The gray value of the pixel.

[0052] Among them, the value of the gamma parameter is determined according to the position information of the pixel point. When the coordinates of the pixel point belong to the boundary coordinate set, the gamma parameter of the gamma transform algorithm is set to 1; when the coordinates of the pixel point do not belong to the boundary coordinate set, the gamma parameter of the gamma transform algorithm is set to a preset value.

[0053] As an embodiment, the specific method of obtaining the value of the gamma parameter is: in, is the preset gamma parameter; is the set of boundary coordinates.

[0054] It should be noted that the embodiment of the present invention optimizes the gamma transform algorithm so that when the optimized gamma transform algorithm is subsequently used to enhance the ultrasound image, the gamma transform processing method for the ultrasound image can be adjusted according to the spatial position of the pixel point in the image, that is, when the spatial position of the pixel point in the ultrasound image satisfies the boundary function, the gamma transform outputs the original gray value. , when it is not satisfied, the grayscale value after the gamma transformation is performed is output. Usually, the gamma parameters of the gamma transformation algorithm can be preset according to the actual needs of image enhancement, so the specific numerical value is not limited in the embodiment of the present invention.

[0055] It should be noted that the boundary between the fetus and the posterior wall is quantified to obtain a boundary function, and the boundary function is added to the gamma transform to obtain an improved gamma transform. This method effectively solves the problem of blurred boundary after gamma transform and optimizes the enhancement effect of B-ultrasound fetal images.

[0056] It should be noted that the gamma transform enhances or weakens the contrast of certain areas by adjusting the brightness distribution of the image. Since there is usually a small brightness difference at the junction of the fetus and the posterior wall, such an area may be blurred due to excessive contrast adjustment after the gamma transform, making the boundary between the fetus and the posterior wall unclear. Therefore, in order to solve the problem of blurred boundary between the fetus and the posterior wall in the B-ultrasound image after gamma transform, the embodiment of the present invention makes relevant improvements to the corresponding part of the gamma transform and adds restrictions. The improved method is to quantify the boundary and obtain the corresponding grayscale information and spatial information. The grayscale information is to find out the grayscale range and grayscale distribution characteristics of the boundary; the spatial information is to find out the spatial position of the boundary, because if only the corresponding grayscale range is transformed, the grayscale values ​​of pixels with similar ranges in other non-border areas will also be changed, thereby losing the original features. The two pieces of information obtained are combined to construct a boundary function, and the function is brought into the gamma transform to obtain an improved gamma transform and add restrictions, thereby solving the problem of blurred boundary.

[0057] At this point, the optimized gamma transform algorithm is obtained through the above method.

[0058] The image enhancement module 104 enhances the ultrasound image by using an optimized gamma transform algorithm.

[0059] Specifically, the ultrasound image is enhanced by using an optimized gamma transform algorithm to obtain an enhanced ultrasound image, in which the umbilical cord is clear and the boundary between the fetus and the posterior wall is clear and complete.

[0060] At this point, this embodiment is completed.

[0061] 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, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An intelligent analysis system for abdominal B-ultrasound data in obstetrics and gynecology, characterized in that: The system includes the following modules: An image acquisition module is used to perform ultrasound examination on obstetrics and gynecology patients through a medical B-ultrasound instrument to obtain ultrasound images of the patient's abdomen; An image analysis module is used to obtain a gradient factor of a pixel point according to the grayscale difference between the pixel points in the ultrasound image, and obtain a plurality of regions by performing regional segmentation on the ultrasound image, and obtain a boundary line and a starting point and an ending point of the boundary line in the ultrasound image according to the gradient factor, the grayscale level of the region, and the distribution of the regional edges, and obtain a pixel point set in the boundary area according to the grayscale change of all the pixel points between the starting point and the ending point in the boundary line in the vertical direction; An algorithm optimization module is used to optimize the value-taking method of the gamma parameter in the gamma transform algorithm by combining the position information of the pixel points in the pixel point set of the boundary area, so as to obtain an optimized gamma transform algorithm; An image enhancement module, used for enhancing ultrasound images through an optimized gamma transform algorithm; The method of optimizing the value of the gamma parameter in the gamma transform algorithm by combining the position information of the pixel points in the pixel point set in the boundary area to obtain the optimized gamma transform algorithm includes the following specific methods: Obtain a set formed by the coordinates of all pixel points in the boundary area pixel point set, recorded as the boundary coordinate set; The gamma transform is optimized by using the coordinate information in the boundary coordinate set to obtain an optimized gamma transform algorithm; The specific expression of the optimized gamma transform algorithm is: in, Indicates the location of The gray value of the pixel after transformation; is the preset hyperparameter, The ultrasound image is located at The gamma parameters of the pixel during gamma transformation; Indicates the location of The gray value of the pixel.

2. The intelligent analysis system for abdominal B-ultrasound data for obstetrics and gynecology according to claim 1, characterized in that: After obtaining the ultrasound image of the patient's abdomen, the method also includes obtaining the target area and non-target area in the ultrasound image through a semantic segmentation algorithm, wherein the target area is the fetal area in the ultrasound image, and the non-target area is the area outside the fetal area in the ultrasound image.

3. The intelligent analysis system for abdominal B-ultrasound data for obstetrics and gynecology according to claim 2, characterized in that: The method of obtaining the gradient factor of the pixel points according to the grayscale difference between the pixel points in the ultrasound image, and obtaining a plurality of regions by performing regional segmentation on the ultrasound image, and obtaining the boundary line and the starting point and the ending point of the boundary line in the ultrasound image according to the gradient factor, the grayscale level of the region and the distribution of the regional edge, includes the following specific methods: Obtaining a gradient factor of an arbitrary pixel point according to a grayscale difference between the arbitrary pixel point and an adjacent pixel point; Performing regional segmentation on the ultrasound image to obtain a number of regions, and obtaining a determination gradient value based on the grayscale level and size of the region and the grayscale value of the pixel point at the edge of the region and the target region; When the gradient factor of a pixel point is greater than the judgment gradient value, the pixel point is recorded as a marked point, two curves formed by all the marked points are obtained and recorded as the boundary line, the Euclidean distance between the pixel points of the two boundary lines is obtained, and the two pairs of pixel points with the closest Euclidean distance are recorded as the starting point pair and the ending point pair, respectively, the starting point pair includes two starting points, and the ending point pair includes two ending points.

4. The intelligent analysis system for abdominal B-ultrasound data for obstetrics and gynecology according to claim 3, characterized in that: The method of performing region segmentation on the ultrasound image to obtain a plurality of regions and obtaining a determination gradient value based on the grayscale level and size of the regions and the grayscale value of the pixel points at the edge of the regions and the target region includes: The ultrasound image is segmented using the watershed method to obtain several regions, and the average grayscale of all pixels in each region is obtained as the overall grayscale of the corresponding region, and the region with the largest area and the smallest overall grayscale is recorded as the target background, and the overall grayscale of the target background is recorded as the first grayscale; Obtain pixel points at the edges of all regions corresponding to the ultrasound image segmented by the watershed algorithm, record them as suspected edge points, and obtain the average grayscale of all suspected edge points on the edge of the target region, record them as the second grayscale; The difference between the second grayscale and the first grayscale is used as the determination gradient value.

5. The intelligent analysis system for abdominal B-ultrasound data for obstetrics and gynecology according to claim 1, characterized in that: The specific method of obtaining the pixel point set in the junction area according to the grayscale change of all pixel points from the starting point to the end point in the junction line in the vertical direction includes: Get the actual junction point; Connect the two starting points in the starting point pair, and similarly, connect the two ending points in the ending point pair, and obtain a set of all pixel points in the closed area formed by the starting point pair, the ending point pair, and all actual intersection points, which is recorded as the intersection area pixel point set.

6. The intelligent analysis system for abdominal B-ultrasound data for obstetrics and gynecology according to claim 5, characterized in that: The specific method for obtaining the actual intersection point is: Any pixel point between the starting point and the ending point on the boundary line is recorded as the target point, and a sequence of grayscale values ​​of A pixels in the vertical direction with the target point as the middle pixel is obtained, which is recorded as a grayscale column, and a number of grayscale columns are obtained, and the number of grayscale columns is equal to the number of columns between the starting point and the ending point, where A is a preset parameter; Get the second-order derivative value of each element in any grayscale column, and record the pixel points corresponding to the elements with second-order derivative values ​​greater than 0 as the actual intersection points.

7. The intelligent analysis system for abdominal B-ultrasound data for obstetrics and gynecology according to claim 1, characterized in that: The gamma parameters during the gamma transformation also include determining the value of the gamma parameters according to the position information of the pixel points.

8. The intelligent analysis system for abdominal B-ultrasound data for obstetrics and gynecology according to claim 7, characterized in that: The specific method of determining the value of the gamma parameter according to the position information of the pixel point includes: When the coordinates of the pixel point belong to the boundary coordinate set, the gamma parameter of the gamma transform algorithm is set to 1; when the coordinates of the pixel point do not belong to the boundary coordinate set, the gamma parameter of the gamma transform algorithm is set to a preset value.

Images