Method and system for image enhancement by ultrasound image speckle noise filtering

Abstract

The application provides a method and system for image enhancement by speckle noise filtering of an ultrasonic image, and the method comprises the following steps: carrying out brightness equalization processing and smoothing filtering on a noisy ultrasonic image; eliminating residual textures in the filtered ultrasonic image to extract components of a complete noise mask; and outputting an image after restoring a real noise-free image. The application equivalently regards an ultrasonic imaging signal as an additive signal of a real ultrasonic imaging signal and a stray noise signal of ultrasonic in a tissue; regards speckle noise and other noises as a whole for processing, and equivalently regards it as a noise mask; calculates the noise mask, and obtains a noise-free real ultrasonic image through simple operation; and the ultrasonic image obtained through the processing method of the application not only removes speckle noise in the ultrasonic image, but also well retains image texture details, and the image contrast is stronger.

Description

Technical Field

[0001] This invention relates to the field of medical ultrasound imaging system technology, and more specifically to a method and system for image enhancement through ultrasound image speckle noise filtering. Background Technology

[0002] Medical ultrasound imaging systems are often subject to noise interference from both the imaging system itself and the external environment during image acquisition, conversion, and transmission. Therefore, actual ultrasound images are always noisy. This noise is primarily random interference, and its main sources are as follows:

[0003] (1) Electronic noise

[0004] The electronic components of medical ultrasound imaging equipment generate noise, including thermal noise, shot noise, power supply ripple noise, common impedance noise, crosstalk noise, and jitter noise, which manifest in a chaotic and disordered manner.

[0005] (2) Environmental noise

[0006] During ultrasound imaging, electromagnetic radiation from electrical equipment in the surrounding environment, as well as power fluctuations in the medical ultrasound imaging equipment itself, can interfere with ultrasound images and create noise.

[0007] (3) Tissue noise

[0008] During ultrasound imaging, as ultrasound waves propagate within the human body, they undergo reflection at different tissue interfaces, generating echoes carrying tissue information. Refraction, diffraction, and scattering also occur. These echoes are typically weak, with most being absorbed and attenuated by the tissue structure. However, as the detection depth increases, the useful signal returning to the probe decreases, while the influence of interfering echoes grows, resulting in greater noise and a lower signal-to-noise ratio. To increase the useful signal, time-controlled gain compensation is used in ultrasound imaging systems, but this also increases noise.

[0009] Among the aforementioned interference noises, the effects are manifested in the image as fine grains, textures, snowflakes, and spots. Among these, spot noise, which is caused by the interaction of scattering and echo, is the most significant noise interference in ultrasound images.

[0010] The speckle noise caused by the superposition of coherent waves during ultrasound imaging severely affects ultrasound images, reducing the target resolution of ultrasound images. As a result, the quality of reconstructed ultrasound images is significantly worse than that of imaging modalities such as CT and MRI, affecting people's observation and understanding of ultrasound images.

[0011] To improve the image quality of ultrasound, speckle noise filtering is usually required. Therefore, speckle noise suppression technology has always been the most important technique for ultrasound image enhancement. However, due to the dependence of speckle noise on human tissue, modeling and filtering speckle noise is very difficult.

[0012] The general requirement for ultrasound image denoising is to effectively suppress speckle noise while preserving image details for diagnosis and analysis. Although numerous studies have reported on speckle noise filtering in ultrasound images, classic filtering methods still lose details such as edges in the image while suppressing noise.

[0013] In existing ultrasound image noise models, speckle noise is considered the main multiplicative noise, and its generation mechanism is generally explained using the discrete scatterer model in random scattering models. In this model, the acoustic medium is assumed to be homogeneous, with a large number of scatterers randomly distributed within it. The size and acoustic impedance of these scatterers are randomly distributed, and their spatial distribution follows a Poisson distribution. The randomness of speckle noise can be described using a random walk model in the complex plane.

[0014] Suppose that the vectors of each scatterer within a certain resolution cell are a1, a2, ... a n These scatterers are superimposed to obtain the composite vector. Amplitude The echo signal intensity within this resolution unit. The joint probability density function can be expressed as:

[0015]

[0016] Among them: A r A i Representing the complex plane respectively The real and imaginary parts of ψ, where ψ represents the total scattering intensity of all random scatterers, are expressed as:

[0017]

[0018] If let Then amplitude The probability density function is:

[0019]

[0020] Equation (0-3) can explain the amplitude The probability density distribution of speckle noise follows a Rayleigh distribution, and the shape of its distribution curve remains constant, meaning it is independent of tissue structure. Speckle noise satisfying this distribution is considered completely random speckle noise. Existing ultrasound image speckle noise models are based on this theory, and these models simultaneously contain multiplicative and additive noise.

[0021] y = x × n m +n a (0-4)

[0022] In equation (0-4), x represents the desired noise-free image, and n m Represents multiplicative noise, n a Let represent additive noise, and y represent the observed noisy image. In the field of medical ultrasound imaging, additive noise (such as electronic noise) is generally considered to have a much smaller impact on the image than multiplicative noise. Therefore, in actual denoising filtering, the effect of additive noise is ignored. Thus, after ignoring additive noise, the observed image becomes:

[0023] y = x × n m (0-5)

[0024] Starting from the image model of equation (0-5), various methods for removing speckle noise have been developed, such as anisotropic diffusion filtering, bilateral filtering, and BM3D noise reduction methods. In practical applications, the processed ultrasound images have indeed effectively suppressed speckle noise and enhanced details such as image edges, demonstrating good performance.

[0025] However, existing speckle noise suppression methods all start from ultrasound image noise models, such as those disclosed in Chinese patent CN104156928B, which process hypothetical speckle noise distribution characteristics. They lack an analysis of the speckle noise distribution characteristics from the perspective of the actual overall medical ultrasound imaging process. Furthermore, while existing methods can remove noise and preserve details from a biological vision perspective, they suffer from defects such as uneven image layers and small gray-level gradients in flat areas. (See also...) Figure 7 Currently, mainstream medical ultrasound images employ anisotropic diffusion noise processing in their noise models. Anisotropic diffusion is a heat transfer method; in flat regions, as the number of iterations increases (which equates to longer heat diffusion time in flat regions, eventually reaching thermal equilibrium and making the temperature uniform), the brightness within the flat region becomes uniform, eliminating any brightness variations. Summary of the Invention

[0026] The purpose of this invention is to overcome the shortcomings of the existing technology and provide a method and system for image enhancement by filtering speckle noise in ultrasound images.

[0027] The objective of this invention is achieved through the following technical solution:

[0028] A method for image enhancement by speckle noise filtering of ultrasound images includes the following steps:

[0029] Select the pixel (x,y) in the noisy ultrasound image;

[0030] The noisy ultrasound image is subjected to brightness equalization processing;

[0031] Determine the stray noise brightness B of human tissue during ultrasound propagation. sm ;

[0032] The noisy ultrasound image is smoothed and filtered to obtain a smooth brightness set A(x,y) that approximates speckle noise;

[0033] Remove residual textures from the filtered ultrasound image and extract the components R(x,y) of the entire noise mask;

[0034] The noise mask M(x, y) that meets the conditions is obtained by using the constraints.

[0035] Using the following formula

[0036]

[0037] The noise-free true luminance value B of the pixel (x,y) is obtained by restoration. r (x, y); where B(x, y) is the brightness value of the pixel (x, y) in the noisy ultrasound image;

[0038] The noise-free true brightness values ​​of all pixels are restored sequentially and output to form an enhanced ultrasound image.

[0039] Preferably, the "brightness equalization processing of the noisy ultrasound image" specifically includes,

[0040] The entire frame of the noisy ultrasound image is traversed, and the maximum brightness value is selected as the image illumination L. The image brightness balance coefficient L is obtained by dividing the brightness B(x,y) of the pixel (x,y) by the image illumination L. b , will L b Normalized to [0 1], the constraint is expressed as:

[0041]

[0042] The above formula is used to perform brightness equalization processing on the pixels of the noisy ultrasound image.

[0043] Preferably, the stray noise brightness B of human tissue during ultrasound propagation is determined. sn The range is 0.6-1.0.

[0044] Preferably, in the phrase "smoothing the noisy ultrasound image to obtain a smooth brightness set A(x, y) that approximates speckle noise", the smoothing filtering method includes at least one of Gaussian filtering, median filtering, and mean filtering.

[0045] Preferably, the step of "smoothing the noisy ultrasound image to obtain a smooth brightness set A(x, y) that approximates speckle noise" specifically includes, through the following formula,

[0046] A(x, y) is obtained by guassian(B(x, y)).

[0047] Preferably, the step of "eliminating residual textures in the filtered ultrasound image and extracting the components R(x, y) of the entire noise mask" specifically includes the following formula:

[0048] R(x,y) = A(x,y) - guassian(|B(x,y) - A(x,y)|) is calculated.

[0049] Preferably, the constraint condition is:

[0050] M(x,y)=max(min(pR(x,y),B(x,y)),0)

[0051] Where p is an introduced constant.

[0052] Preferably, p takes the value [0.90 0.95].

[0053] This invention also discloses a system for image enhancement through speckle noise filtering of ultrasound images, comprising:

[0054] The selection unit is used to select a pixel (x,y) in a noisy ultrasound image;

[0055] A brightness equalization unit is used to perform brightness equalization processing on the noisy ultrasound image.

[0056] The stray noise brightness calculation unit is used to determine the stray noise brightness B of human tissue during ultrasound propagation. sn ;

[0057] A smoothing filter unit is used to smooth the noisy ultrasound image to obtain a smooth brightness set A(x,y) that approximates speckle noise;

[0058] The residual texture removal unit is used to remove residual textures in the filtered ultrasound image and extract the components R(x,y) of the entire noise mask.

[0059] The noise mask calculation unit is used to obtain a noise mask M(x, y) that meets the conditions through constraints.

[0060] The noise-free true luminance value calculation unit is used to calculate the luminance value using the following formula.

[0061]

[0062] The noise-free true luminance value B of the pixel (x,y) is obtained by restoration. r (x, y); where B(x, y) is the brightness value of the pixel (x, y) in the noisy ultrasound image;

[0063] The output unit is used to sequentially restore the noise-free true brightness value of all pixels and output it to form an enhanced ultrasound image.

[0064] Preferably, the smoothing filter unit includes at least one of a Gaussian filter unit, a median filter unit, and a mean filter unit.

[0065] Preferably, the stray noise brightness B of human tissue during ultrasound propagation is determined. sn The range is 0.6-1.0.

[0066] The beneficial effects of this invention are mainly reflected in the following aspects: the ultrasound imaging signal is equivalent to the additive signal of the real ultrasound imaging signal and the stray noise signal of ultrasound in the tissue; the speckle noise and other noises are treated as a whole and are equivalent to a noise mask; after calculating the noise mask, a noise-free real ultrasound image can be obtained through simple calculation. The ultrasound image obtained by the processing method of this invention not only removes speckle noise in the ultrasound image, but also preserves the image texture details well, and the image contrast is stronger. Attached Figure Description

[0067] The technical solution of the present invention will be further described below with reference to the accompanying drawings:

[0068] Figure 1 The original image in the embodiments of this invention;

[0069] Figure 2 Noise masking obtained through a Gaussian filter;

[0070] Figure 3 Noise masking obtained through a median filter;

[0071] Figure 4 Noise masking obtained through a mean filter;

[0072] Figure 5 : The output image after median smoothing filtering;

[0073] Figure 6 The output image after mean smoothing filtering;

[0074] Figure 7 : The output image after anisotropic diffusion processing;

[0075] Figure 8The preferred embodiment of the present invention is the output image after Gaussian smoothing filtering;

[0076] Figure 9 : A schematic diagram of the process of this invention. Detailed Implementation

[0077] The present invention will now be described in detail with reference to the specific embodiments shown in the accompanying drawings. However, these embodiments are not limited to the present invention, and any structural, methodological, or functional modifications made by those skilled in the art based on these embodiments are included within the scope of protection of the present invention.

[0078] During the propagation of ultrasound in human tissues, physical processes such as scattering, diffraction, reflection, and refraction occur. Ultimately, the reflected and scattered ultrasound signals are received, and the amplitude of the received signals is converted into pixels of different brightness, displaying a grayscale image, i.e., B-mode imaging. In this process, the scattered and diffracted ultrasound signals appear as speckle noise and other noise signals in the image, while a true ultrasound tissue tomographic image should be formed by the reflected ultrasound signals.

[0079] Speckle noise is essentially a wave interference phenomenon. Speckle interference is a type of particle interference inherent in medical ultrasound, radar, and optical coherence. In fact, the interfaces between different tissues on the surface of human skin and within the body are extremely rough on a wavelength scale. In images obtained by ultrasound imaging systems from these structures, the sound waves from numerous scatterers will inevitably interfere with each other, resulting in speckle patterns. Existing models model the reflectivity function as a set of scatterers, revealing the cause of this phenomenon and the influence of interference. In ultrasound imaging, due to the widespread presence of speckle noise, the interface between fat and stroma in subcutaneous fat forms numerous strong scatterers, and the distribution of these scatterers is extremely irregular, no longer conforming to the assumed Poisson distribution. These scatterers then form a large amount of speckle noise, which is the fundamental reason why the background image always appears "hazy." Figure 1 As shown.

[0080] In this invention, the ultrasound propagation process is understood as an additive signal of the real ultrasound imaging signal and the stray signal of ultrasound in the tissue. The speckle noise and other noises are treated as a whole, equivalent to a noise mask.

[0081] The ultrasound image signal model is shown in equation (1):

[0082] B(x, y) = B r (x, y)e -αd(x，y) +B sn (1-e -αd(x，y) (1) Wherein:

[0083] B(x, y) -- The brightness value of pixel (x, y) in a noisy ultrasound image;

[0084] Br(x, y) -- The noise-free real brightness value of pixel (x, y);

[0085] d(x, y) -- the distance of the corresponding tissue with inherent luminance B0(x, y);

[0086] B sn --Spurious noise brightness in human tissues during ultrasound propagation;

[0087] α -- Absorption coefficient of human tissue.

[0088] The noise mask for pixel (x,y) is given by equation (2):

[0089] M(x, y) = B sn (1-e-α d(x，y) (2)

[0090] Equation (1) can be rewritten from equation (2) into equation (3):

[0091]

[0092] The noise-free true luminance value Br(x,y) of pixel (x,y) can be expressed as Equation (4):

[0093]

[0094] Therefore, it can be seen from equation (4) that as long as the stray noise brightness B is calculated, sn With a noise mask M(x,y), the true brightness of the pixel (x,y) can be restored.

[0095] Based on this, such as Figure 9 As shown, the present invention provides a method for image enhancement by speckle noise filtering of ultrasound images, comprising the following steps:

[0096] Select the pixel (x,y) in the noisy ultrasound image;

[0097] The noisy ultrasound image is subjected to brightness equalization processing;

[0098] Determine the stray noise brightness B of human tissue during ultrasound propagation. sn ;

[0099] The noisy ultrasound image is smoothed and filtered to obtain a smooth brightness set A(x,y) that approximates speckle noise;

[0100] Remove residual textures from the filtered ultrasound image and extract the components R(x,y) of the entire noise mask;

[0101] The noise mask M(x, y) that meets the conditions is obtained by using the constraints.

[0102] Through the following formula

[0103]

[0104] The noise-free true luminance value BT of the pixel (x,y) is obtained by restoration. ( (x, y); where B(x, y) is the brightness value of the pixel (x, y) in the noisy ultrasound image;

[0105] The noise-free true brightness values ​​of all pixels are restored sequentially and output to form an enhanced ultrasound image.

[0106] The following is a detailed description of the above steps:

[0107] In the initial step, scene illumination is estimated for the entire frame of the noisy ultrasound image to eliminate brightness attenuation caused by distance changes during ultrasound signal propagation. First, the entire frame of the noisy ultrasound image is traversed, and its maximum brightness value is selected as the image illumination L, i.e., standard white. Then, the image brightness balance coefficient L is obtained by dividing the brightness value B(x,y) of pixel (x,y) in the noisy image by the image illumination L. b , will L b Normalized to [0 1], the constraint is expressed as:

[0108]

[0109] The pixels of the noisy ultrasound image are subjected to brightness equalization processing using Equation (5).

[0110] This invention employs brightness equalization processing, using the brightest point in the image as the equivalent light source. The brightness of other pixels in the image is adjusted according to different illumination levels. This differs from conventional ultrasound applications, which use different depths of ultrasound propagation for brightness compensation, i.e., time gain compensation (TGC).

[0111] B sn The parameter is used for brightness equalization of the entire image, ensuring that the brightness of noise is consistent between deep and shallow tissue areas. This value is a brightness compensation parameter and can be adjusted arbitrarily according to your needs, with a preferred adjustment range of 0.6-1.0. Thus, by calculating M(x, y), the noise-free image B can be reconstructed. r(x, y).

[0112] Because the noise mask M(x,y) is approximately equal at similar depths in human tissue, B(x,y) is approximately smooth at similar depths in human tissue.

[0113] The noisy ultrasound image is then smoothed to obtain a smooth brightness set A(x, y) that approximates speckle noise. The smoothing filtering method includes at least one or more of Gaussian filtering, median filtering, and mean filtering. Preferably, the method of this invention uses a Gaussian filter to smooth B(x, y) to obtain a smooth brightness set A(x, y) that approximates speckle noise.

[0114] in:

[0115] A(x,y)=guassian(B(x,y)) (6)

[0116] Because some texture features still remain in the ultrasound image after filtering, and these texture features belong to human tissue and are definitely not part of the noise masking, it is necessary to remove the texture and ensure that only components belonging to the noise masking are retained. Therefore, the calculation formula is used:

[0117] R(x,y)=A(x,y)-guassian(|B(x,y)-A(x,y)|) (7)

[0118] The residual texture in the filtered ultrasound image is eliminated, and the components R(x,y) of the entire noise mask are extracted. Because the noisy ultrasound image contains texture and flat areas, the texture is a line with fine edges and a large difference in brightness from the flat area. After smoothing, the texture is also smoothed, and the difference in brightness between it and the flat area becomes smaller. In the above formula (7), B(x,y)-A(x,y) is used to subtract the texture.

[0119] However, all components R(x, y) of the ultrasound image noise mask still need to be constrained by two conditions:

[0120] M(x,y)=max(min(pR(x,y),B(x,y)),0) (8)

[0121] In equation (8), p is an introduced constant with a value of [0.90 0.95]. In reality, tissue echoes inevitably contain some impurity molecules. The role of the constant p is to avoid completely removing the noise mask. Completely removing the noise mask would make the ultrasound image look unrealistic and would lose the sense of depth. Therefore, by introducing the constant p, a portion of the noise mask covering the deep tissue is selectively retained.

[0122] It should be noted that the process of obtaining the ultrasound mask is a screening process. After two layers of screening, an ultrasound mask with a darker brightness is obtained.

[0123] Condition 1: min(pR(x, y), B(x, y))=X

[0124] R(x, y) is a mask used as an intermediate variable. This mask cannot be brighter than the corresponding pixel in the Gaussian-smoothed image B(x, y), so the smaller of the two values ​​should be taken.

[0125] Condition 2: max(X, k);

[0126] M(x, y) is the final noise mask. How dark the mask is depends on the value of k being compared. In this invention, k = 0.

[0127] Finally, the determined ultrasonic noise mask M(x,y) is calculated, and the noise-free true brightness value B of the restored pixel (x,y) is estimated according to Equation (4). r (x, y), for ease of explanation, B in this embodiment sn Taking the value 1, we get:

[0128]

[0129] The noise-free true brightness values ​​of all pixels are restored sequentially, and the results are output to form an enhanced ultrasound image, such as... Figure 8 As shown.

[0130] Specific implementation process:

[0131] The same ultrasound image was processed for speckle noise reduction using an anisotropic diffusion method commonly used by many ultrasound diagnostic equipment manufacturers. The parameters of the anisotropic diffusion method are as follows:

[0132] The iteration count is iter = 7, K = 9, and lambda = 0.23.

[0133] The processed image is as follows Figure 7 As shown.

[0134] When extracting noise masks, Gaussian filters, median filters, and mean filters are selected respectively, such as... Figures 2 to 6 As shown, a comparative verification was conducted. Based on the actual filtering effect, the size of the median filter was selected as 7, and the size of the mean filter was selected as 11. From the extracted noise mask, the Gaussian filter is smoother than the median and mean filters, avoiding the loss of image details and texture.

[0135] Therefore, in summary, the ultrasound image obtained by the processing method of the present invention not only removes speckle noise in the ultrasound image, but also preserves the image texture details well, while the image contrast is stronger.

[0136] It should be understood that the series of detailed descriptions listed above are merely specific descriptions of feasible implementations of the present invention, and are not intended to limit the scope of protection of the present invention. All equivalent implementations or modifications made without departing from the spirit of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method of image enhancement by speckle noise filtering of ultrasound images, characterized in that: The method comprises the following steps: selecting a pixel point (x, y) in a noisy ultrasonic image; performing brightness equalization processing on the noisy ultrasonic image; Determining the stray noise intensity B of human tissue during ultrasound propagation sn ; performing smoothing filtering on the noisy ultrasonic image to obtain a smoothing brightness set A(x, y) approximating to speckle noise; eliminating residual textures in the filtered ultrasonic image to extract a component R(x, y) of the entire noise mask; obtaining a noise mask M(x, y) meeting a condition through a constraint condition; through the following formula, reduction of the noisy real brightness value B(x,y) of the pixel point (x,y) to obtain the non-noisy real brightness value B(x,y) of the pixel point (x,y) r (x,y); wherein B(x,y) is the brightness value of the pixel point (x,y) in the noisy ultrasound image; sequentially restoring noise-free real brightness values of all pixel points, outputting, and forming an enhanced ultrasonic image; the "performing smoothing filtering on the noisy ultrasonic image to obtain a smoothing brightness set A(x, y) approximating to speckle noise" specifically comprises, through the following formula, the constraint condition is M(x, y) = max(min(pR(x, y), B(x, y)), 0) wherein, p is a constant introduced; the "performing brightness equalization processing on the noisy ultrasonic image" specifically comprises, Traverse the whole frame of the noisy ultrasound image, select the maximum value of its brightness as the image illumination L, and obtain the image brightness balance coefficient L by dividing the brightness B(x, y) of the pixel point (x, y) by the image illumination L b , L b is normalized to [01], and the constraint condition is represented as: performing brightness equalization processing on the pixel of the noisy ultrasonic image through the above formula.

2. The method of claim 1, wherein: Determining the spurious noise intensity B of human tissue during ultrasound propagation sn ranging from 0.6 to 1.

0.

3. The method according to claim 1, wherein the "performing smoothing filtering on the noisy ultrasonic image to obtain a smoothing brightness set A(x, y) approximating to speckle noise" comprises at least one of Gaussian filtering, median filtering and mean filtering.

4. The method of claim 1, wherein: the "eliminating residual textures in the filtered ultrasonic image to extract a component R(x, y) of the entire noise mask" specifically comprises, through the following formula, R(x, y) = A(x, y) - guassian(|B(x, y) - A(x, y)|) is calculated.

5. The method of claim 1, wherein: p is [0.9 0.95].

6. A system for image enhancement by speckle noise filtering of ultrasound images, characterized in that: comprise a selecting unit, configured to select a pixel point (x, y) in a noisy ultrasonic image; a brightness equalization unit, configured to perform brightness equalization processing on the noisy ultrasonic image; a stray noise brightness calculation unit for determining the stray noise brightness B of the human tissue during the ultrasound propagation sn ; a smoothing filtering unit, configured to perform smoothing filtering on the noisy ultrasonic image to obtain a smoothing brightness set A(x, y) approximating to speckle noise; a residual texture elimination unit, configured to eliminate residual textures in the filtered ultrasonic image to extract a component R(x, y) of the entire noise mask; a noise mask calculation unit, configured to obtain a noise mask M(x, y) meeting a condition through a constraint condition; a noise-free real brightness value calculation unit, configured to calculate through the following formula reduction of the noisy real brightness value B(x,y) of the pixel point (x,y) to obtain the non-noisy real brightness value B(x,y) of the pixel point (x,y) r (x,y); wherein B(x,y) is the brightness value of the pixel point (x,y) in the noisy ultrasound image; an output unit, configured to sequentially restore noise-free real brightness values of all pixel points, output, and form an enhanced ultrasonic image; the "performing smoothing filtering on the noisy ultrasonic image to obtain a smoothing brightness set" specifically comprises, through the following formula, A(x, y) = guassian(B(x, y)) is calculated; the constraint condition is M(x, y) = max(min(pR(x, y), B(x, y)), 0) wherein, p is a constant introduced; the "performing brightness equalization processing on the noisy ultrasonic image" specifically comprises, Traverse the whole frame of noisy ultrasound image, select the maximum value of its brightness as image illumination L, obtain image brightness balance coefficient L by dividing the brightness B(x, y) of the pixel point (x, y) by the image illumination L b , L b is normalized to [0 1], and the constraint condition is represented as: performing brightness equalization processing on the pixel of the noisy ultrasonic image through the above formula.

7. The system of claim 6, wherein: Determining the spurious noise intensity B of human tissue during ultrasound propagation sn ranges from 0.6 to 1.0.

Images