Enhanced denoising method for clinical ultrasonic image

Abstract

The invention relates to the field of image processing, in particular to an enhanced denoising method for a clinical ultrasonic image, which comprises the following steps: acquiring a knee joint grayscale image; presetting a search window and a sliding window, dividing the search window into upper and lower half-region grayscale images, and layering the upper and lower half-region grayscale images according to the sliding window; obtaining a feature extraction factor of each layer according to the pixel points of each layer; obtaining upper and lower half region gray level image filter weights according to the feature extraction factor of each layer; obtaining the weight of each pixel point in the search window according to the difference between the pixel points in the sliding window in the search window; obtaining a weighted gray value of each pixel point according to the weights of the gray image filters in the upper and lower half regions and the weight and the gray value of each pixel point in the search window; and obtaining a denoised grayscale image according to the weighted grayscale value of each pixel point. According to the method, the pixel point weight is determined through image processing, and the noise interference influence in the ultrasonic image is weakened or even eliminated.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to an enhancement and denoising method for clinical ultrasonic images. Background Art

[0002] In recent years, ultrasound technology has been widely used in preoperative localization of diseases. Compared with CT imaging, ultrasound imaging technology has the advantages of being non-invasive and radiation-free, simple and rapid, convenient and intuitive, and inexpensive. It has a fast imaging speed and has outstanding advantages in the observation of moving organs. Ultrasound imaging can clearly show the size, shape, internal echoes and blood flow distribution of the lesion area.

[0003] However, the ultrasound imaging process is rather special. Due to the random interference of scatterers, when the ultrasonic wave velocity passes through a slightly inhomogeneous medium, speckle multiplicative noise is generated in the image, which reduces the resolution of the ultrasound image, causes a large amount of image detail information to be lost, and the image edges become blurred, which seriously interferes with the normal distinction between diseased tissue and normal human soft tissue in the subsequent clinical diagnosis process. Summary of the invention

[0004] In order to solve the above problems, the present invention provides a method for enhancing and denoising clinical ultrasound images, the method comprising: Processing the collected knee joint ultrasound image to obtain a knee joint grayscale image; A search window and a sliding window are preset, and the search window is divided into an upper half and a lower half, the upper half is recorded as an upper half grayscale image, and the lower half is recorded as a lower half grayscale image; the upper half grayscale image and the lower half grayscale image are layered according to the width of the sliding window to obtain a plurality of layers; a feature extraction factor of each layer is obtained according to the pixel points of each layer; a filter weight of the lower half grayscale image and a filter weight of the upper half grayscale image are obtained according to the feature extraction factor of each layer; The weight of each pixel in the search window is obtained according to the difference between the pixel in the sliding window corresponding to the central pixel in the search window and the pixel in the sliding window corresponding to the remaining pixels in the search window; the weighted gray value of each pixel is obtained according to the filter weight of the gray image in the lower half area, the filter weight of the gray image in the upper half area, the weight of each pixel in the search window and the gray value of each pixel in the search window; The denoised grayscale image is obtained according to the weighted grayscale value of each pixel; Automatic identification of bone hyperplasia areas is performed based on the denoised grayscale image.

[0005] Furthermore, the step of stratifying the upper half grayscale image and the lower half grayscale image to obtain a plurality of layers according to the width of the sliding window includes the following specific steps: After obtaining the upper grayscale image and the lower grayscale image, the upper grayscale image and the lower grayscale image are divided respectively by the sliding window width, and are evenly divided in the vertical direction by the sliding window width L to obtain several layers.

[0006] Furthermore, the specific steps of obtaining the feature extraction factors of each layer are as follows: Any layer in the grayscale image in the lower half is recorded as the target layer, and the formula for the feature extraction factor of the target layer is: In the formula, represents the feature extraction factor of the target layer, Represents the grayscale value of the i-th pixel in the middle row of the target layer, represents the gray value of the zth pixel in the sliding window neighborhood centered on the i-th pixel in the middle row of the target layer, L represents the width of the sliding window, Represents the total number of pixels in the sliding window, Indicates the number of pixels in the target layer; Similarly, the feature extraction factor of each layer is obtained.

[0007] Furthermore, the filter weights of the lower half grayscale image and the filter weights of the upper half grayscale image are specifically obtained in the following steps: The filter weight formula for the lower half grayscale image is: In the formula, Represents the difference between the feature extraction factors of the mth layer and the next adjacent layer in the lower half grayscale image, Represents the mean of the differences in feature extraction factors between all adjacent layers in the lower half grayscale image, represents the number of layers in the lower half grayscale image, represents the hyperbolic tangent function, Represents the filter weight in the grayscale image in the lower half; According to the filter weights in the grayscale image in the lower half, the filter weights in the grayscale image in the upper half are obtained as follows: .

[0008] Furthermore, the specific steps for obtaining the weight of each pixel in the search window are as follows: The formula for the weight of the vth pixel in the search window is: In the formula, It is expressed as the mean square error corresponding to the vth pixel in the grayscale image in the lower half, Indicates the number of pixels in the search window. It represents the mean square error corresponding to the j-th pixel in the grayscale image in the lower half, represents the weight of the vth pixel in the grayscale image in the lower half, Represents an exponential function with a natural constant as its base.

[0009] Furthermore, the specific steps for obtaining the mean square error corresponding to the vth pixel in the lower half grayscale image are as follows: Get the central pixel point of the search window in the grayscale image of the lower half, record it as the target central pixel point, get the sliding window of the target central pixel point, record it as the target sliding window window; get the sliding window of the vth pixel point in the search window, record it as the marked sliding window window; take the square sum of the difference between the grayscale values ​​of the pixel point in the target sliding window window and all the pixel points in the marked sliding window window as the mean square error corresponding to the vth pixel point in the grayscale image of the lower half.

[0010] Furthermore, the specific steps of obtaining the weighted grayscale value of each pixel are as follows: The formula for the weighted grayscale value of each pixel is: In the formula, Represents the gray value of the vth pixel in the gray image in the lower half, Represents the gray value of the rth pixel in the gray image of the upper half, represents the weight of the vth pixel in the grayscale image in the lower half, represents the weight of the rth pixel in the grayscale image in the upper half, Indicates the number of pixels in the search window. Represents the weighted gray value of the center pixel of the search window. represents the filter weight in the grayscale image in the lower half, Represents the filter weights in the upper grayscale image.

[0011] Furthermore, the step of dividing the search window into an upper grayscale image and a lower grayscale image comprises the following specific steps: The search window is divided into an upper half and a lower half, the upper half is recorded as an upper half grayscale image, and the lower half is recorded as a lower half grayscale image.

[0012] Furthermore, the step of obtaining a denoised grayscale image according to the weighted grayscale value of each pixel point includes the following specific steps: The weighted gray value of each pixel is taken as the real gray value of each pixel, and the image composed of the real gray value of each pixel in the knee joint gray image is recorded as the denoised gray image.

[0013] The beneficial effect of the technical solution of the present invention is that accurate identification of the situation of the bone hyperplasia area is a more important method in medical image processing, but due to image quality problems, there is noise information in the image, and the visualization degree of the bone spur area is not very high. Therefore, a denoising process must be performed on the image first. In order to better retain the information of the feature area, the present invention sets a search window, divides the search window into upper and lower half areas along the center point, and layers the search window along the y-axis direction, constructs a model to extract the feature information of each layer, fits the feature values ​​output by each layer, and uses the fitting residual as the total weight of the upper and lower half areas in the search window, and then calculates the similarity between the sliding window in the lower half area and the sliding window of the central pixel point, and distributes the weights of each half area instead of local mean filtering weighted smoothing. The present invention makes the filtering weights concentrated on the bone spur area, so that the effective information of the bone spur area can be retained as much as possible during the filtering operation, thereby improving the image quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] 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.

[0015] Figure 1 The present invention is a flowchart of the steps of a method for enhancing and denoising clinical ultrasound images. DETAILED DESCRIPTION

[0016] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation, structure, features and effects of a method for enhancing and denoising clinical ultrasound images proposed by the present invention 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.

[0017] 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.

[0018] The following is a detailed description of a specific scheme of a clinical ultrasound image enhancement and denoising method provided by the present invention in conjunction with the accompanying drawings.

[0019] See also Figure 1 , which shows a flowchart of a method for enhancing and denoising clinical ultrasound images provided by an embodiment of the present invention, the method comprising the following steps: Step S001: Acquire a knee joint ultrasonic image, and pre-process the knee joint ultrasonic image to obtain a knee joint grayscale image.

[0020] It should be noted that knee osteophyte is not an ordinary joint inflammation, but a long-term degenerative disease of cartilage. The kneecap is composed of the lower end of the femur, the upper end of the tibia and the patella, and is covered by a joint capsule with cruciate ligaments and meniscus. It is the most complex and largest weight-bearing joint in the human body. Therefore, the identification of knee osteophyte is of utmost importance and has certain therapeutic significance.

[0021] Specifically, an ultrasonic image of a knee joint is collected, and the ultrasonic image of the knee joint is preprocessed into grayscale to obtain a preprocessed grayscale image of the knee joint.

[0022] At this point, a grayscale image of the knee joint is obtained.

[0023] Step S002: Dynamically adjust the search window size according to the characteristics of the knee joint edge, use the width of the sliding window to layer the upper half grayscale image and the lower half grayscale image, obtain the feature extraction factor of each layer according to the pixels of each layer, obtain the filter weight according to the feature extraction factor of each layer, and use the filter weight to obtain the weighted grayscale value of each pixel.

[0024] It should be noted that knee spurs are small bone protrusions formed under the knee cartilage, which usually appear on the joint surface between the femoral condyle and the tibial plateau. The joint edge under the normal cartilage is relatively flat, without protrusions, and the gap between the knee joints is relatively sufficient; while the gap between the knee joints with bone spurs will become smaller, and abnormal protrusions will appear on the bone edge. It is necessary to analyze the detailed information between the knee joints in the blurred image.

[0025] Since bones have a higher grayscale value, they appear brighter in the image. The same is true for bone spurs, which grow on the surface of cartilage, so they appear slightly weaker in the joint space, and their bone density is also slightly weaker.

[0026] Since two windows are required when using non-local mean filtering, a search window width M and a sliding window width L are set here. In this embodiment, the search window width M=49 and the sliding window width L=7 are used as examples for description. This embodiment does not make specific limitations, and M and L can be determined according to the specific implementation situation. The search window and the sliding window are both slid from left to right and from top to bottom with a step size of 1. Because bone spurs can only exist on the upper or lower surface of the joint gap, the search window is layered along the vertical direction according to the direction of the y-axis in the rectangular coordinate system to obtain multiple horizontal layers, and as the different levels in the y-axis direction increase, the distribution of pixel points in the sliding window at different levels is analyzed and counted.

[0027] Specifically, the area corresponding to each sliding of the search window on the grayscale image of the knee joint is recorded as the grayscale image of the search window. First, due to the characteristic of the bone spur being in the middle position, the grayscale image of the search window is first divided into grayscale images of the upper and lower regions, recorded as the upper half grayscale image and the lower half grayscale image; now the lower half grayscale image is analyzed, and the analysis process of the upper half grayscale image and the lower half grayscale image is the same. After obtaining the lower half grayscale image, the lower half grayscale image is divided again by the width of the sliding window, and is evenly divided by a width of L in the vertical direction to obtain multiple horizontal layers; similarly, the division of the upper half grayscale image is the same.

[0028] It should be further explained that, since the specific area of ​​the bone spur is unknown, we only know that the bone spur area is a distribution of pixels with a certain shape, and the grayscale of the pixels in the bone spur area should be similar to that of the pixels in the bone area. The existence of noise points makes the grayscale of some pixels not so obvious; however, the grayscale of the pixels outside the bone spur area is similar, and the grayscale value is relatively low. The average of all pixels in the sliding window with the pixel in the bone spur area as the center pixel will be higher, while the average grayscale of all pixels in the sliding window with the pixel outside the bone spur area as the center pixel is lower. The grayscale value of the center pixel of the sliding window in the bone spur area is higher, and the distribution density of the highlight points in the sliding window is also higher.

[0029] Specifically, the analysis is performed according to the layers divided in the grayscale image in the lower half, and any layer in the grayscale image in the lower half is recorded as the target layer. The feature extraction factor of the target layer in the grayscale image in the lower half is obtained according to all the pixels in the middle row of the target layer. The feature extraction factor of the target layer in the grayscale image in the lower half can be expressed by the following formula: In the formula, represents the feature extraction factor of the target layer, Represents the grayscale value of the i-th pixel in the middle row of the target layer, represents the gray value of the zth pixel in the sliding window neighborhood centered on the i-th pixel in the middle row of the target layer, that is, the gray value of the zth pixel in the window excluding the central pixel. L represents the width of the sliding window. Represents the total number of pixels in the sliding window, Indicates the length of the target layer, that is, the number of pixels in the middle row of the target layer.

[0030] Similarly, the feature extraction factor of each layer is obtained.

[0031] in, It represents the distribution density of highlighted pixels in the sliding window with the i-th pixel as the center pixel in the middle row of the target layer. Because there is noise in the sliding window, there is no way to accurately capture the bright and dark pixels. Therefore, the number of highlighted pixels cannot be used to represent the distribution density. The average grayscale in the window can be used to indirectly represent the distribution density of highlighted pixels. The higher the average grayscale of the pixels in the sliding window, the more highlighted pixels there are in the window, that is, the greater the distribution density of highlighted pixels. Represents the Euclidean norm of the grayscale value of the central pixel and the distribution density of the highlight pixels in the sliding window with the central pixel as the pixel; there are M sliding windows in each layer.

[0032] According to the feature extraction factor of the target layer, the feature extraction factor of each layer in the grayscale image in the lower half is obtained, and the feature extraction factor of each layer in the grayscale image in the lower half is obtained from top to bottom to obtain a set of sequences A, which are recorded as . Where n represents the total number of layers in the grayscale image in the lower half, represents the feature extraction factor of the i-th layer in the lower half grayscale image. Similarly, the feature extraction factors of all layers in the upper half grayscale image can be obtained.

[0033] It should be noted that the degree of fit of the fitting line is then analyzed to obtain the difference between the actual degree of fit and the degree of fit under ideal conditions. This difference is used to allocate the filter weight, and the feature information of the bone spur area is retained as much as possible during the filtering process. In actual scenarios, the bone spur presents a "pyramid shape" from top to bottom, that is, the feature extraction of each layer from top to bottom is incremental, and all will be analyzed by the difference in feature extraction factors between adjacent layers. If there is a bone spur, it will present a "pyramid shape", that is, the difference in feature extraction factors between adjacent layers is not much.

[0034] Specifically, the difference of the feature extraction factors between all adjacent layers in the grayscale image of the lower half is obtained, represented by sequence B, which is recorded as ,in, Represents the difference between the feature extraction factors of the mth layer and the m+1th layer in the grayscale image of the lower half, and n-1 represents the grayscale image of the lower half. According to the characteristics of bone spurs, if there are bone spurs in the image, the variance in sequence B should be very small, that is, each value in sequence B is considered equal. Therefore, when sequence B meets the characteristics of bone spurs, a straight line can be obtained by linear fitting of all values ​​in sequence A. Therefore, when there are bone spurs, the closer the curve obtained by fitting the feature extraction factors of each layer in sequence A is to a straight line, the smaller the variance of all data in the corresponding sequence B, that is, when the variance is closer to 0, the more likely it is that bone spurs exist in the search window.

[0035] Then the formula for the filter weight in the lower half grayscale image can be expressed as: In the formula, Represents the difference between the feature extraction factors of the mth layer and the next adjacent layer in the lower half grayscale image, Represents the mean of the differences in feature extraction factors between all adjacent layers in the lower half grayscale image, represents the number of layers in the grayscale image in the lower half, represents the hyperbolic tangent function, Represents the filter weights in the lower half of the grayscale image.

[0036] Among them, To represent the output value of each layer in the lower half Ideally, the change of each layer from the "spur tip" to the bone area should be a linearly increasing fitting effect. The value of will be close to 0. Use the th function to Normalized to between 0 and 1; The smaller the value of , the more the output value of each layer increases linearly, and it is believed that the central pixel of the sliding window is more likely to be the pixel of the bone spur area. This is also the effect of true fitting, which means that the filter weights are changed from To 0 is a fitting effect that does not belong to the bone spur area. The value is used as the weight of the pixel point in the non-spur area.

[0037] According to the weight of the filter in the grayscale image in the lower half, the weight of the filter in the grayscale image in the upper half is obtained: .

[0038] The above operation is based on the description of the lower bone spur with the horizontal line of the center point of the search window as the dividing line. The calculation of the upper bone spur can be obtained in the same way. Compare the fitting results of the upper and lower half grayscale images and take The smallest side is the suspected bone spur area.

[0039] It should be noted that the above calculation It is based on the difference under the ideal situation, that is, the distribution of the real pixel points in the bone spur area. The overall weight of the area below the horizontal dividing line of the search window is ; The area above the dividing line is considered to be an area without bone spurs, and the weight of this area is The area below the dividing line not only contains the bone spur area, but also the noise area outside the bone spur is in the area below the dividing line of the search window. However, the distribution of the bone spur area is like a "pyramid" distribution. The closer to the cartilage surface, the more pixels are distributed in the bone spur area, the more sliding windows there are, and the similarity between the sliding windows is higher, 1- The weight should be more allocated to the pixels in the bone spur area, so as to ensure that the feature information of the bone spur area will not be smoothed too much during the filtering process.

[0040] Specifically, the central pixel point of the search window in the grayscale image of the lower half is obtained, recorded as the target central pixel point, and the sliding window of the target central pixel point is obtained, recorded as the target sliding window window; then the sliding window windows of other pixel points in the search window are obtained, recorded as the marked sliding window window. The result of the square sum of the difference between the grayscale values ​​of the pixel point in the target sliding window window and all the pixel points in the marked sliding window window is taken as the grayscale mean square error between the target sliding window window and the marked sliding window window, which is used to represent the mean square error corresponding to the central pixel point in the marked sliding window window. Since there is no difference between itself and itself, the mean square error of the target central pixel point is 0. After obtaining the mean square error corresponding to each pixel point, the weight of each pixel point in the grayscale image of the lower half can be represented by the proportion of the mean square error corresponding to each pixel point in the sum of the mean square errors corresponding to all the pixel points in the grayscale image of the lower half.

[0041] Then the weight of the vth pixel in the grayscale image in the lower half can be expressed as: In the formula, It is expressed as the mean square error corresponding to the vth pixel in the grayscale image in the lower half, Indicates the number of pixels in the search window. It represents the mean square error corresponding to the j-th pixel in the grayscale image in the lower half, represents the weight of the vth pixel in the grayscale image in the lower half, Represents an exponential function with a natural constant as its base.

[0042] in, It represents the sum of the mean square errors corresponding to all pixels in the grayscale image in the lower half. The weight of a single pixel in the lower half grayscale image is represented by the proportion of the mean square error corresponding to a single pixel in the lower half grayscale image to the sum of the mean square errors corresponding to all pixels in the lower half grayscale image.

[0043] Under normal circumstances, the weight of the search window is divided into two, with the upper and lower areas each accounting for 0.5. After the filter weights determined above, the weight of the lower area is (The upper area is for the upper bone spur, and the calculation method is the same as the lower area), and then according to The value of will be the overall weight of the lower half of the area Then distribute it.

[0044] Then the gray value corresponding to each pixel after weighting is: In the formula, Represents the gray value of the vth pixel in the gray image in the lower half, Represents the gray value of the rth pixel in the gray image of the upper half, represents the weight of the vth pixel in the grayscale image in the lower half, represents the weight of the rth pixel in the grayscale image in the upper half, Indicates the number of pixels in the search window. Represents the weighted gray value of the center pixel of the search window. represents the filter weight in the grayscale image in the lower half, Represents the filter weights in the upper grayscale image.

[0045] Similarly, the weighted grayscale value of each pixel in the grayscale image of the knee joint is obtained, and the weighted grayscale value of each pixel is used as the real grayscale value of each pixel.

[0046] in, It represents the weighted weight of each pixel point in the lower half that may belong to the bone spur area; Indicates the weighted weight of each pixel in the upper half of the area; the weighted average of the pixels in the upper and lower areas can get the true gray value of the central pixel. This makes the weight of the pixels that are most likely to belong to the bone spur area larger, and in the process of non-local mean filtering, it can better ensure that these pixels will not be over-smoothed and better retain effective information.

[0047] At this point, the true grayscale value of each pixel in the knee joint grayscale image is obtained.

[0048] Step S003: Obtain a denoised grayscale image according to the weighted grayscale value of each pixel.

[0049] Through the above analysis of the search window, the real gray value of the pixel in the center of the search window is obtained, and then the real gray value corresponding to each pixel in the knee joint gray image is obtained by analyzing the search window corresponding to each pixel, and the denoised gray image is obtained according to the real gray value of each pixel. At this point, the noise removal of the knee joint ultrasound image is completed.

[0050] At this point, this embodiment is completed.

[0051] 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 spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for enhancing and denoising clinical ultrasound images, characterized in that: The method comprises the following steps: Processing the collected knee joint ultrasound image to obtain a knee joint grayscale image; A search window and a sliding window are preset in the grayscale image of the knee joint, and the search window is divided into an upper half and a lower half, the upper half is recorded as an upper half grayscale image, and the lower half is recorded as a lower half grayscale image; the upper half grayscale image and the lower half grayscale image are layered according to the width of the sliding window to obtain a plurality of layers; a feature extraction factor of each layer is obtained according to the pixel points of each layer; and a filter weight of the lower half grayscale image and a filter weight of the upper half grayscale image are obtained according to the change of the feature extraction factor of each layer; The weight of each pixel in the search window is obtained according to the difference between the pixel in the sliding window corresponding to the central pixel in the search window and the pixel in the sliding window corresponding to the remaining pixels in the search window; the weighted gray value of each pixel is obtained according to the filter weight of the gray image in the lower half area, the filter weight of the gray image in the upper half area, the weight of each pixel in the search window and the gray value of each pixel in the search window; The denoised grayscale image is obtained according to the weighted grayscale value of each pixel; The specific steps for obtaining the feature extraction factors of each layer are as follows: Any layer in the grayscale image in the lower half is recorded as the target layer, and the formula for the feature extraction factor of the target layer is: In the formula, represents the feature extraction factor of the target layer, Represents the grayscale value of the i-th pixel in the middle row of the target layer, represents the gray value of the zth pixel in the sliding window neighborhood centered on the i-th pixel in the middle row of the target layer, L represents the width of the sliding window, Represents the total number of pixels in the sliding window, Indicates the number of pixels in the target layer; Similarly, the feature extraction factor of each layer is obtained.

2. The method for enhancing and denoising clinical ultrasound images according to claim 1, characterized in that: The step of layering the upper half grayscale image and the lower half grayscale image according to the width of the sliding window to obtain a plurality of layers includes the following specific steps: After obtaining the upper grayscale image and the lower grayscale image, the upper grayscale image and the lower grayscale image are divided respectively by the sliding window width, and are evenly divided in the vertical direction by the sliding window width L to obtain several layers.

3. The method for enhancing and denoising clinical ultrasound images according to claim 1, characterized in that: The specific steps of obtaining the filter weights of the lower half grayscale image and the filter weights of the upper half grayscale image are as follows: The filter weight formula for the lower half grayscale image is: In the formula, Represents the difference between the feature extraction factors of the mth layer and the next adjacent layer in the lower half grayscale image, Represents the mean of the differences in feature extraction factors between all adjacent layers in the lower half grayscale image, represents the number of layers in the grayscale image in the lower half, represents the hyperbolic tangent function, Represents the filter weight in the grayscale image in the lower half; According to the filter weights in the grayscale image in the lower half, the filter weights in the grayscale image in the upper half are obtained as follows: .

4. The method for enhancing and denoising clinical ultrasound images according to claim 1, characterized in that: The specific steps for obtaining the weight of each pixel in the search window are as follows: The formula for the weight of the vth pixel in the search window is: In the formula, It is expressed as the mean square error corresponding to the vth pixel in the grayscale image in the lower half, Indicates the number of pixels in the search window. It represents the mean square error corresponding to the j-th pixel in the grayscale image in the lower half, represents the weight of the vth pixel in the grayscale image in the lower half, Represents an exponential function with a natural constant as its base.

5. The method for enhancing and denoising clinical ultrasound images according to claim 4, characterized in that: The specific steps for obtaining the mean square error corresponding to the vth pixel in the grayscale image of the lower half are as follows: Get the central pixel point of the search window in the grayscale image of the lower half, record it as the target central pixel point, get the sliding window of the target central pixel point, record it as the target sliding window window; get the sliding window of the vth pixel point in the search window, record it as the marked sliding window window; take the square sum of the difference between the grayscale values ​​of the pixel point in the target sliding window window and all the pixel points in the marked sliding window window as the mean square error corresponding to the vth pixel point in the grayscale image of the lower half.

6. The method for enhancing and denoising clinical ultrasound images according to claim 1, characterized in that: The specific steps for obtaining the weighted grayscale value of each pixel are as follows: The formula for the weighted grayscale value of each pixel is: In the formula, Represents the gray value of the vth pixel in the gray image in the lower half, Represents the gray value of the rth pixel in the gray image of the upper half, represents the weight of the vth pixel in the grayscale image in the lower half, represents the weight of the rth pixel in the grayscale image in the upper half, Indicates the number of pixels in the search window. Represents the weighted gray value of the center pixel of the search window. represents the filter weight in the grayscale image in the lower half, Represents the filter weights in the upper grayscale image.

7. The method for enhancing and denoising clinical ultrasound images according to claim 1, characterized in that: The specific steps of obtaining a denoised grayscale image according to the weighted grayscale value of each pixel are as follows: The weighted gray value of each pixel is taken as the real gray value of each pixel, and the image composed of the real gray value of each pixel in the knee joint gray image is recorded as the denoised gray image.

Images