Method and device for automatically determining irregular-shaped areas of ultrasonic images

Abstract

The invention discloses a method and device for automatically determining a heterogeneous region of an ultrasonic image. The method comprises: performing a histogram equalization operation on each frame of ultrasonic image determined in an ultrasonic video to obtain all the equalized ultrasonic images, and performing gradient enhancement on each frame of the equalized ultrasonic image to obtain all the enhanced ultrasonic images; performing image morphological expansion on each frame of the enhanced ultrasonic image to obtain all the expanded ultrasonic images, and performing superposition processing on each frame of the expanded ultrasonic image to obtain all the superimposed ultrasonic images; performing a connected domain recognition operation on each frame of the superimposed ultrasonic image to obtain the heterogeneous region of each frame of the superimposed ultrasonic image to be analyzed. In this way, by performing histogram equalization, gradient enhancement, image expansion, superposition processing and connected domain recognition operations on the ultrasonic image, the processing accuracy of the ultrasonic image is improved, and then the heterogeneous region of the ultrasonic image to be analyzed can be accurately obtained, so as to accurately analyze the heterogeneous region.

Description

Technical Field

[0001] The present invention relates to the technical field of ultrasonic image processing, and in particular to a method and a device for automatically determining an abnormal-shaped area of ​​an ultrasonic image. Background Art

[0002] With the rapid development of medical technology, ultrasound imaging has been widely used in the medical field. It has the advantages of simple operation, strong real-time performance and few side effects. However, traditional ultrasound images have defects such as high noise, blurred boundaries, and unclear individual feature differences, which makes it difficult for subsequent technicians to perform accurate image analysis based on ultrasound images, limiting the further development of ultrasound imaging research. It can be seen that it is particularly important to provide a method that can improve the accuracy of ultrasound image processing. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a method and device for automatically determining the irregular region of an ultrasonic image, which can improve the reliability and accuracy of processing the ultrasonic image, thereby facilitating the accurate acquisition of the irregular region of the ultrasonic image to be analyzed, thereby facilitating the accurate analysis operation of the irregular region.

[0004] In order to solve the above technical problems, the first aspect of the present invention discloses a method for automatically determining a shaped area of ​​an ultrasound image, the method comprising:

[0005] Determining a plurality of frames of ultrasound images to be processed from a preset ultrasound video;

[0006] Performing a histogram equalization operation on each frame of the ultrasonic image to obtain all equalized ultrasonic images, and performing a gradient enhancement operation on each frame of the equalized ultrasonic image to obtain all enhanced ultrasonic images;

[0007] Performing an image morphological dilation operation on each frame of the enhanced ultrasonic image to obtain all dilated ultrasonic images, and performing a stacking operation on each frame of the dilated ultrasonic image to obtain all stacked ultrasonic images;

[0008] A connected domain recognition operation is performed on each frame of the superimposed ultrasonic image to obtain a special-shaped region of each frame of the superimposed ultrasonic image to be analyzed.

[0009] As an optional implementation, in the first aspect of the present invention, the performing a histogram equalization operation on each frame of the ultrasound image to obtain all equalized ultrasound images includes:

[0010] For each of the ultrasonic images, determine a pixel value of each first pixel in the ultrasonic image, and determine a pixel parameter corresponding to each grayscale in the ultrasonic image according to the pixel values ​​of all the first pixels;

[0011] For each of the ultrasound images, determining a pixel cumulative probability parameter corresponding to each of the grayscales according to pixel parameters corresponding to all of the grayscales, and calculating a grayscale mapping value corresponding to each of the grayscales according to the pixel cumulative probability parameter corresponding to each of the grayscales and the grayscale value of the target grayscale in the ultrasound image;

[0012] For each of the ultrasonic images, grayscale mapping processing is performed on all the grayscales according to the grayscale mapping values ​​corresponding to all the grayscales to obtain an equalized ultrasonic image corresponding to the ultrasonic image.

[0013] As an optional implementation, in the first aspect of the present invention, performing a gradient enhancement operation on each frame of the equalized ultrasonic image to obtain all enhanced ultrasonic images includes:

[0014] For each of the equalized ultrasonic images, determine a gradient calculation direction parameter of the equalized ultrasonic image according to a preset edge detection requirement parameter of the equalized ultrasonic image, and determine a pixel gradient calculation template matching the equalized ultrasonic image and associated pixels corresponding to each second pixel in the equalized ultrasonic image according to the gradient calculation direction parameter;

[0015] For each second pixel in each of the equalized ultrasound images, a pixel gradient parameter corresponding to the second pixel is calculated according to a pixel value of the second pixel and a pixel value of an associated pixel corresponding to the second pixel, and a pixel convolution operation is performed on the pixel value of the second pixel according to the pixel gradient parameter corresponding to the second pixel and the pixel gradient operation template to obtain an enhanced pixel value corresponding to the second pixel;

[0016] For each of the equalized ultrasonic images, an enhanced ultrasonic image corresponding to the equalized ultrasonic image is determined according to the enhanced pixel values ​​corresponding to all the second pixels in the equalized ultrasonic image.

[0017] As an optional implementation, in the first aspect of the present invention, performing an image morphological dilation operation on each frame of the enhanced ultrasound image to obtain all the dilated ultrasound images includes:

[0018] For each of the enhanced ultrasound images, determining an image quality parameter of the enhanced ultrasound image, and determining an image dilation template of the enhanced ultrasound image according to the image quality parameter;

[0019] For each of the enhanced ultrasonic images, determine a first binary parameter corresponding to each of the third pixels according to the pixel value of each of the third pixels in the enhanced ultrasonic image, and perform a dilation convolution operation on the first binary parameter corresponding to each of the third pixels according to the first binary parameter corresponding to each of the third pixels and the image dilation template to obtain a target binary parameter corresponding to each of the third pixels;

[0020] For each of the enhanced ultrasonic images, a dilated ultrasonic image corresponding to the enhanced ultrasonic image is determined according to the target binary parameters corresponding to all the third pixels.

[0021] As an optional implementation, in the first aspect of the present invention, before determining the image dilation template of the enhanced ultrasound image according to the image quality parameter, the method further includes:

[0022] Determining an image corrosion requirement corresponding to the enhanced ultrasound image according to the image quality parameter, and judging whether the image corrosion requirement is greater than a preset corrosion requirement threshold;

[0023] When the judgment result is no, triggering the execution of the operation of determining the image expansion template of the enhanced ultrasound image according to the image quality parameter;

[0024] When the judgment result is yes, the image erosion template of the enhanced ultrasonic image is determined according to the image quality parameters; the second binary parameter corresponding to each third pixel in the enhanced ultrasonic image is determined according to the pixel value of each third pixel, and the second binary parameter corresponding to each third pixel and the image erosion template are used to perform an erosion convolution operation on the second binary parameter corresponding to each third pixel to obtain the third binary parameter corresponding to each third pixel; the eroded ultrasonic image corresponding to the enhanced ultrasonic image is determined according to the third binary parameters corresponding to all the third pixels, and the reference image quality parameters of the eroded ultrasonic image are determined; the image quality parameters are updated according to the reference image quality parameters, and the operation of determining the image dilation template of the enhanced ultrasonic image according to the image quality parameters is triggered.

[0025] As an optional implementation, in the first aspect of the present invention, performing a superposition processing operation on each frame of the expanded ultrasonic image to obtain all superimposed ultrasonic images includes:

[0026] For each of the expanded ultrasonic images, determining a plurality of frames of associated ultrasonic images associated with the expanded ultrasonic image from all the other expanded ultrasonic images except the expanded ultrasonic image among all the expanded ultrasonic images, and determining a valid image area of ​​the expanded ultrasonic image according to a binary pixel parameter corresponding to the expanded ultrasonic image and binary pixel parameters corresponding to all the associated ultrasonic images;

[0027] For each of the expanded ultrasonic images, according to the effective image area of ​​the expanded ultrasonic image, an ultrasonic image to be superimposed that matches the effective image area is determined from all the associated ultrasonic images, and according to the ultrasonic image to be superimposed, an image superposition operation is performed on the expanded ultrasonic image to obtain a superimposed ultrasonic image corresponding to the expanded ultrasonic image.

[0028] As an optional implementation, in the first aspect of the present invention, the connected domain identification operation is performed on each frame of the superimposed ultrasound image to obtain the heterogeneous region of each frame of the superimposed ultrasound image to be analyzed, including:

[0029] For each of the superimposed ultrasonic images, determining a connectivity mark parameter corresponding to each of the fourth pixels in the superimposed ultrasonic images according to a fourth binary parameter corresponding to each of the fourth pixels;

[0030] For each of the superimposed ultrasound images, multiple connected areas in the superimposed ultrasound image are determined based on the connectivity mark parameters corresponding to all the fourth pixels and the preset connected area identification parameters, and a target connected area is determined from all the connected areas based on the area parameters corresponding to each of the connected areas as the heterogeneous area of ​​the superimposed ultrasound image to be analyzed; the area parameters include at least one of an area area parameter, an area pixel number parameter and an area content parameter.

[0031] The second aspect of the present invention discloses a device for automatically determining an abnormally shaped area of ​​an ultrasound image, the device comprising:

[0032] A determination module, used to determine multiple frames of ultrasound images to be processed from a preset ultrasound video;

[0033] An equalization module, used for performing a histogram equalization operation on each frame of the ultrasonic image to obtain all equalized ultrasonic images;

[0034] A gradient enhancement module, used for performing a gradient enhancement operation on each frame of the equalized ultrasonic image to obtain all enhanced ultrasonic images;

[0035] An image expansion module, used for performing an image morphological expansion operation on each frame of the enhanced ultrasonic image to obtain all expanded ultrasonic images;

[0036] A superposition processing module, used for performing a superposition processing operation on each frame of the expanded ultrasonic image to obtain all superpositioned ultrasonic images;

[0037] The region determination module is used to perform a connected domain recognition operation on each frame of the superimposed ultrasonic image to obtain a special-shaped region of each frame of the superimposed ultrasonic image to be analyzed.

[0038] As an optional implementation, in the second aspect of the present invention, the equalization module performs a histogram equalization operation on each frame of the ultrasound image, and a method of obtaining all equalized ultrasound images specifically includes:

[0039] For each of the ultrasonic images, determine a pixel value of each first pixel in the ultrasonic image, and determine a pixel parameter corresponding to each grayscale in the ultrasonic image according to the pixel values ​​of all the first pixels;

[0040] For each of the ultrasound images, determining a pixel cumulative probability parameter corresponding to each of the grayscales according to pixel parameters corresponding to all of the grayscales, and calculating a grayscale mapping value corresponding to each of the grayscales according to the pixel cumulative probability parameter corresponding to each of the grayscales and the grayscale value of the target grayscale in the ultrasound image;

[0041] For each of the ultrasonic images, grayscale mapping processing is performed on all the grayscales according to the grayscale mapping values ​​corresponding to all the grayscales to obtain an equalized ultrasonic image corresponding to the ultrasonic image.

[0042] As an optional implementation, in the second aspect of the present invention, the gradient enhancement module performs a gradient enhancement operation on each frame of the equalized ultrasonic image to obtain all enhanced ultrasonic images, specifically including:

[0043] For each of the equalized ultrasonic images, determine a gradient calculation direction parameter of the equalized ultrasonic image according to a preset edge detection requirement parameter of the equalized ultrasonic image, and determine a pixel gradient calculation template matching the equalized ultrasonic image and associated pixels corresponding to each second pixel in the equalized ultrasonic image according to the gradient calculation direction parameter;

[0044] For each second pixel in each of the equalized ultrasound images, a pixel gradient parameter corresponding to the second pixel is calculated according to a pixel value of the second pixel and a pixel value of an associated pixel corresponding to the second pixel, and a pixel convolution operation is performed on the pixel value of the second pixel according to the pixel gradient parameter corresponding to the second pixel and the pixel gradient operation template to obtain an enhanced pixel value corresponding to the second pixel;

[0045] For each of the equalized ultrasonic images, an enhanced ultrasonic image corresponding to the equalized ultrasonic image is determined according to the enhanced pixel values ​​corresponding to all the second pixels in the equalized ultrasonic image.

[0046] As an optional implementation, in the second aspect of the present invention, the image expansion module includes:

[0047] a determination submodule, configured to determine, for each of the enhanced ultrasound images, an image quality parameter of the enhanced ultrasound image, and determine, according to the image quality parameter, an image dilation template of the enhanced ultrasound image;

[0048] a calculation submodule, configured to determine, for each of the enhanced ultrasonic images, a first binary parameter corresponding to each of the third pixels according to the pixel value of each of the third pixels in the enhanced ultrasonic image, and perform a dilation convolution operation on the first binary parameter corresponding to each of the third pixels according to the first binary parameter corresponding to each of the third pixels and the image dilation template, so as to obtain a target binary parameter corresponding to each of the third pixels;

[0049] The determination submodule is further configured to determine, for each of the enhanced ultrasonic images, a dilated ultrasonic image corresponding to the enhanced ultrasonic image according to the target binary parameters corresponding to all the third pixels.

[0050] As an optional implementation manner, in the second aspect of the present invention, the determining submodule is further used to:

[0051] Before determining the image dilation template of the enhanced ultrasound image according to the image quality parameter, determining the image corrosion requirement corresponding to the enhanced ultrasound image according to the image quality parameter, and judging whether the image corrosion requirement is greater than a preset corrosion requirement threshold;

[0052] When the judgment result is no, triggering the execution of the operation of determining the image expansion template of the enhanced ultrasound image according to the image quality parameter;

[0053] When the judgment result is yes, the image erosion template of the enhanced ultrasonic image is determined according to the image quality parameters; the second binary parameter corresponding to each third pixel in the enhanced ultrasonic image is determined according to the pixel value of each third pixel, and the second binary parameter corresponding to each third pixel and the image erosion template are used to perform an erosion convolution operation on the second binary parameter corresponding to each third pixel to obtain the third binary parameter corresponding to each third pixel; the eroded ultrasonic image corresponding to the enhanced ultrasonic image is determined according to the third binary parameters corresponding to all the third pixels, and the reference image quality parameters of the eroded ultrasonic image are determined; the image quality parameters are updated according to the reference image quality parameters, and the operation of determining the image dilation template of the enhanced ultrasonic image according to the image quality parameters is triggered.

[0054] As an optional implementation, in the second aspect of the present invention, the superposition processing module performs a superposition processing operation on each frame of the expanded ultrasonic image to obtain all superimposed ultrasonic images, specifically including:

[0055] For each of the expanded ultrasonic images, determining a plurality of frames of associated ultrasonic images associated with the expanded ultrasonic image from all the other expanded ultrasonic images except the expanded ultrasonic image among all the expanded ultrasonic images, and determining a valid image area of ​​the expanded ultrasonic image according to a binary pixel parameter corresponding to the expanded ultrasonic image and binary pixel parameters corresponding to all the associated ultrasonic images;

[0056] For each of the expanded ultrasonic images, according to the effective image area of ​​the expanded ultrasonic image, an ultrasonic image to be superimposed that matches the effective image area is determined from all the associated ultrasonic images, and according to the ultrasonic image to be superimposed, an image superposition operation is performed on the expanded ultrasonic image to obtain a superimposed ultrasonic image corresponding to the expanded ultrasonic image.

[0057] As an optional implementation, in the second aspect of the present invention, the region determination module performs a connected domain identification operation on each frame of the superimposed ultrasound image to obtain the heterogeneous region of each frame of the superimposed ultrasound image to be analyzed, specifically comprising:

[0058] For each of the superimposed ultrasonic images, determining a connectivity mark parameter corresponding to each of the fourth pixels in the superimposed ultrasonic images according to a fourth binary parameter corresponding to each of the fourth pixels;

[0059] For each of the superimposed ultrasound images, multiple connected areas in the superimposed ultrasound image are determined based on the connectivity mark parameters corresponding to all the fourth pixels and the preset connected area identification parameters, and a target connected area is determined from all the connected areas based on the area parameters corresponding to each of the connected areas as the heterogeneous area of ​​the superimposed ultrasound image to be analyzed; the area parameters include at least one of an area area parameter, an area pixel number parameter and an area content parameter.

[0060] A third aspect of the present invention discloses another device for automatically determining an abnormally shaped area of ​​an ultrasonic image, the device comprising:

[0061] A memory storing executable program code;

[0062] a processor coupled to the memory;

[0063] The processor calls the executable program code stored in the memory to execute the method for automatically determining the irregular-shaped area of ​​an ultrasound image disclosed in the first aspect of the present invention.

[0064] The fourth aspect of the present invention discloses a computer storage medium, wherein the computer storage medium stores computer instructions, and when the computer instructions are called, they are used to execute the method for automatically determining the abnormal-shaped area of ​​the ultrasound image disclosed in the first aspect of the present invention.

[0065] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0066] In the embodiment of the present invention, a histogram equalization operation is performed on each frame of ultrasonic image to obtain all equalized ultrasonic images, and gradient enhancement is performed on each frame of equalized ultrasonic image to obtain all enhanced ultrasonic images; image dilation is performed on each frame of enhanced ultrasonic image to obtain all dilated ultrasonic images, and superposition processing is performed on each frame of dilated ultrasonic image to obtain all superimposed ultrasonic images; a connected domain recognition operation is performed on each frame of superimposed ultrasonic image to obtain a heterogeneous region of each frame of superimposed ultrasonic image to be analyzed. In this way, by performing histogram equalization, gradient enhancement, image dilation, superposition processing and connected domain recognition operations on ultrasonic images, the processing reliability and accuracy of ultrasonic images can be improved, which is conducive to accurately obtaining the heterogeneous region of the ultrasonic image to be analyzed, thereby facilitating the accurate analysis operation of the heterogeneous region. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments 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 creative work.

[0068] Figure 1 It is a flow chart of a method for automatically determining a shaped area of ​​an ultrasound image disclosed in an embodiment of the present invention;

[0069] Figure 2 It is a flow chart of another method for automatically determining a special-shaped area of ​​an ultrasound image disclosed in an embodiment of the present invention;

[0070] Figure 3 It is a schematic diagram of the structure of a device for automatically determining an abnormally shaped area of ​​an ultrasonic image disclosed in an embodiment of the present invention;

[0071] Figure 4 It is a schematic structural diagram of another device for automatically determining a special-shaped area of ​​an ultrasonic image disclosed in an embodiment of the present invention;

[0072] Figure 5 It is a structural schematic diagram of another device for automatically determining irregular-shaped areas of an ultrasonic image disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0073] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0074] The terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish different objects rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, device, product or end including a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units inherent to these processes, methods, products or ends.

[0075] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0076] The present invention discloses a method and device for automatically determining the irregular region of an ultrasonic image, which can improve the reliability and accuracy of processing the ultrasonic image, thereby being conducive to accurately obtaining the irregular region of the ultrasonic image to be analyzed, thereby being conducive to realizing the accurate analysis operation of the irregular region.

[0077] Embodiment 1

[0078] See also Figure 1 , Figure 1 1 is a flow chart of a method for automatically determining an abnormally shaped area of ​​an ultrasound image disclosed in an embodiment of the present invention. Figure 1 The described method for automatically determining the irregular regions of ultrasound images can be applied to extract irregular regions of various types of ultrasound images, such as cardiac ultrasound images, thyroid ultrasound images, abdominal organ ultrasound images, etc., which is not limited in the embodiments of the present invention. Optionally, the method can be implemented by an ultrasound image processing device, which can be an ultrasound image detection device, or a local server or cloud server for monitoring the ultrasound image processing process, etc., which is not limited in the embodiments of the present invention. Figure 1 As shown, the method for automatically determining the irregular-shaped area of ​​the ultrasound image may include the following operations:

[0079] 101. Determine multiple frames of ultrasound images to be processed from a preset ultrasound video.

[0080] In the embodiment of the present invention, the ultrasound video is composed of multiple frames of original ultrasound images. Optionally, it can be based on the image quality parameters corresponding to all the original ultrasound images in the preset ultrasound video, and multiple frames of ultrasound images to be processed whose image quality parameters meet the preset processing conditions are determined from all the original ultrasound images; it can also be based on determining the image change parameters corresponding to the original ultrasound images in the preset ultrasound video, and according to the image change parameters corresponding to the original ultrasound images in pairs, all ultrasound images to be processed whose image change parameters are greater than or equal to the preset change parameter threshold are determined from all the original ultrasound images.

[0081] 102. Perform a histogram equalization operation on each frame of the ultrasonic image to obtain all equalized ultrasonic images, and perform a gradient enhancement operation on each frame of the equalized ultrasonic image to obtain all enhanced ultrasonic images.

[0082] In an embodiment of the present invention, further, performing a histogram equalization operation on each frame of ultrasound image to obtain all equalized ultrasound images may include: determining object parameters of the object to be analyzed corresponding to all ultrasound images, and determining analysis requirement parameters corresponding to all ultrasound images based on the object parameters of the object to be analyzed; and performing a histogram equalization operation on each frame of ultrasound image based on the analysis requirement parameters to obtain all equalized ultrasound images.

[0083] In this optional embodiment, the object parameters of the object to be analyzed may include position parameters, operation parameters, size parameters, etc. of the object to be analyzed, such as position parameters of the apical four chambers, operation parameters of the parasternal short axis apex, size parameters of the parasternal short axis center chamber, etc. Further optionally, the analysis requirement parameters corresponding to all ultrasound images may include annotation requirement parameters for annotation points of the object to be analyzed, analysis clarity requirement parameters, etc.

[0084] 103. Perform an image morphological dilation operation on each frame of the enhanced ultrasonic image to obtain all dilated ultrasonic images, and perform a stacking operation on each frame of the dilated ultrasonic image to obtain all stacked ultrasonic images.

[0085] In the embodiment of the present invention, further, the noise parameters corresponding to each frame of the enhanced ultrasound image may be determined first, and then, according to the noise parameters corresponding to all the enhanced ultrasound images, the image morphological dilation operation may be performed on each frame of the enhanced ultrasound image to obtain all the dilated ultrasound images. Optionally, the noise parameters may include a noise size parameter, a noise type parameter, and the like.

[0086] 104. Perform a connected domain recognition operation on each frame of the superimposed ultrasound image to obtain a special-shaped region of each frame of the superimposed ultrasound image to be analyzed.

[0087] In the embodiment of the present invention, the obtained irregular region of the superimposed ultrasound image can be understood as the most important image region in the superimposed ultrasound image that needs to be analyzed. Optionally, the irregular region can be a sector region, a rectangular region, or other types of regions, which is not limited in the embodiment of the present invention.

[0088] It can be seen that the implementation of the embodiments of the present invention can improve the processing reliability and accuracy of ultrasonic images by performing histogram equalization, gradient enhancement, image dilation, overlay processing and connected domain identification operations on ultrasonic images, which is beneficial to accurately obtain the irregular region of the ultrasonic image to be analyzed, thereby facilitating the accurate analysis operation of the irregular region.

[0089] Embodiment 2

[0090] See also Figure 2 , Figure 2FIG. 1 is a flow chart of another method for automatically determining a shaped region of an ultrasound image disclosed in an embodiment of the present invention. Figure 2 The described method for automatically determining the irregular regions of ultrasound images can be applied to extract irregular regions of various types of ultrasound images, such as cardiac ultrasound images, thyroid ultrasound images, abdominal organ ultrasound images, etc., which is not limited in the embodiments of the present invention. Optionally, the method can be implemented by an ultrasound image processing device, which can be an ultrasound image detection device, or a local server or cloud server for monitoring the ultrasound image processing process, etc., which is not limited in the embodiments of the present invention. Figure 2 As shown, the method for automatically determining the irregular-shaped area of ​​the ultrasound image may include the following operations:

[0091] 201. Determine multiple frames of ultrasound images to be processed from a preset ultrasound video.

[0092] 202. For each ultrasound image, determine a pixel value of each first pixel in the ultrasound image, and determine a pixel parameter corresponding to each grayscale in the ultrasound image according to the pixel values ​​of all the first pixels.

[0093] In an embodiment of the present invention, optionally, the pixel parameter corresponding to each grayscale may include a pixel quantity parameter corresponding to the grayscale, and may also include a pixel contribution parameter corresponding to the grayscale, wherein the pixel quantity parameter is a non-negative integer (that is, there is no decimal type parameter), and the pixel contribution parameter is a parameter greater than or equal to 0 (that is, there may be a decimal type parameter).

[0094] 203. For each ultrasound image, determine the pixel cumulative probability parameter corresponding to each grayscale according to the pixel parameters corresponding to all grayscales, and calculate the grayscale mapping value corresponding to each grayscale according to the pixel cumulative probability parameter corresponding to each grayscale and the grayscale value of the target grayscale in the ultrasound image.

[0095] In the embodiment of the present invention, further, calculating the grayscale mapping value corresponding to each grayscale according to the pixel cumulative probability parameter corresponding to each grayscale and the grayscale value of the target grayscale in the ultrasound image may include:

[0096] Calculate the product parameter between the pixel cumulative probability parameter corresponding to each grayscale and the grayscale value of the target grayscale in the ultrasound image; determine the grayscale mapping adjustment curve for all grayscales according to the preset analysis requirement parameters, and calculate the grayscale mapping value corresponding to each grayscale according to the product parameter between the pixel cumulative probability parameter corresponding to all grayscales and the grayscale value of the target grayscale in the ultrasound image and the grayscale mapping adjustment curve.

[0097] 204. For each ultrasonic image, grayscale mapping processing is performed on all grayscales according to grayscale mapping values ​​corresponding to all grayscales to obtain an equalized ultrasonic image corresponding to the ultrasonic image.

[0098] In the embodiment of the present invention, the grayscale value corresponding to each grayscale is replaced by the grayscale mapping value corresponding to the grayscale, so as to obtain the equalized ultrasonic image corresponding to the ultrasonic image.

[0099] 205. Perform a gradient enhancement operation on each frame of the equalized ultrasonic image to obtain all enhanced ultrasonic images.

[0100] 206. Perform an image morphological dilation operation on each frame of the enhanced ultrasonic image to obtain all dilated ultrasonic images, and perform a stacking operation on each frame of the dilated ultrasonic image to obtain all stacked ultrasonic images.

[0101] 207. Perform a connected domain recognition operation on each frame of the superimposed ultrasonic image to obtain a special-shaped region of each frame of the superimposed ultrasonic image to be analyzed.

[0102] In the embodiment of the present invention, for other descriptions of step 201 and step 205 to step 207, please refer to the detailed description of step 101 to step 104 in the first embodiment, and the embodiment of the present invention will not be repeated here.

[0103] It can be seen that the implementation of the embodiment of the present invention can determine the pixel parameters corresponding to each grayscale in the ultrasonic image according to the pixel values ​​of all the first pixels in the ultrasonic image, and then determine the pixel cumulative probability parameters corresponding to each grayscale, and calculate the grayscale mapping value corresponding to each grayscale to realize the grayscale mapping processing operation, which is conducive to improving the reliability and accuracy of the grayscale mapping processing operation in the ultrasonic image, and then conducive to improving the image contrast effect of the obtained equalized ultrasonic image, so as to improve the reliability and accuracy of the subsequent gradient enhancement operation of the equalized ultrasonic image.

[0104] In an optional embodiment, the step 205 performs a gradient enhancement operation on each frame of the equalized ultrasound image to obtain all enhanced ultrasound images, including:

[0105] For each equalized ultrasonic image, according to the preset edge detection requirement parameters of the equalized ultrasonic image, a gradient calculation direction parameter of the equalized ultrasonic image is determined, and according to the gradient calculation direction parameter, a pixel gradient calculation template matching the equalized ultrasonic image and an associated pixel corresponding to each second pixel in the equalized ultrasonic image are determined;

[0106] For each second pixel in each equalized ultrasound image, a pixel gradient parameter corresponding to the second pixel is calculated according to the pixel value of the second pixel and the pixel value of the associated pixel corresponding to the second pixel, and a pixel convolution operation is performed on the pixel value of the second pixel according to the pixel gradient parameter corresponding to the second pixel and a pixel gradient operation template to obtain an enhanced pixel value corresponding to the second pixel;

[0107] For each equalized ultrasonic image, an enhanced ultrasonic image corresponding to the equalized ultrasonic image is determined according to enhanced pixel values ​​corresponding to all second pixels in the equalized ultrasonic image.

[0108] In this optional embodiment, optionally, the gradient calculation direction parameter may be 0°, 45°, 90°, 135°, etc. Further optionally, the pixel gradient calculation template may be a pixel gradient calculation template of the sobel operator type, or a pixel gradient calculation template of other operator types, which may be further determined according to the sensitivity parameters of the corresponding operator type to noise and the noise processing requirement parameters.

[0109] It can be seen that this optional embodiment can determine the gradient calculation direction parameters of the equalized ultrasonic image according to the edge detection requirement parameters of the equalized ultrasonic image, and then determine the pixel gradient operation template of the equalized ultrasonic image and the associated pixels corresponding to the second pixel, and perform a pixel convolution operation on the second pixel according to the pixel gradient operation template and the associated pixels to achieve the gradient enhancement operation on the equalized ultrasonic image. In this way, not only can the gradient enhancement of the equalized ultrasonic image be flexibly performed based on the edge detection requirement parameters, but the reliability and accuracy of the enhanced ultrasonic image can also be improved, thereby improving the effectiveness of the gradient enhancement operation on the equalized ultrasonic image to enhance the degree of image change of the equalized ultrasonic image.

[0110] In another optional embodiment, in the above step 206, the image morphological dilation operation is performed on each frame of the enhanced ultrasound image to obtain all the dilated ultrasound images, including:

[0111] For each enhanced ultrasound image, determining an image quality parameter of the enhanced ultrasound image, and determining an image dilation template of the enhanced ultrasound image according to the image quality parameter;

[0112] For each enhanced ultrasonic image, according to the pixel value of each third pixel in the enhanced ultrasonic image, determine the first binary parameter corresponding to each third pixel, and according to the first binary parameter corresponding to each third pixel and the image dilation template, perform a dilation convolution operation on the first binary parameter corresponding to each third pixel to obtain a target binary parameter corresponding to each third pixel;

[0113] For each enhanced ultrasonic image, a dilated ultrasonic image corresponding to the enhanced ultrasonic image is determined according to the target binary parameters corresponding to all third pixels.

[0114] In this optional embodiment, the image quality parameter may optionally include image noise, image target region continuity, etc. Further optionally, the dilation convolution operation may be performed once or multiple times.

[0115] It can be seen that this optional embodiment can perform a corresponding dilation convolution operation on the third pixel in the enhanced ultrasonic image according to the image quality parameter of the enhanced ultrasonic image, so as to obtain a dilated ultrasonic image corresponding to the enhanced ultrasonic image. In this way, the image morphological dilation operation can be flexibly performed on the enhanced ultrasonic image, thereby improving the reliability and accuracy of the obtained dilated ultrasonic image, reducing the image noise in the enhanced ultrasonic image and ensuring the continuity of the image, which is conducive to the subsequent accurate superposition processing of the dilated ultrasonic image.

[0116] In yet another optional embodiment, before determining the image dilation template of the enhanced ultrasound image according to the image quality parameter in the above step, the method further includes:

[0117] Determining the image corrosion requirement corresponding to the enhanced ultrasound image according to the image quality parameter, and judging whether the image corrosion requirement is greater than a preset corrosion requirement threshold;

[0118] When the judgment result is no, triggering the execution of the operation of determining the image expansion template of the enhanced ultrasound image according to the image quality parameter in the above step;

[0119] When the judgment result is yes, determine the image erosion template of the enhanced ultrasonic image according to the image quality parameters; determine the second binary parameter corresponding to each third pixel according to the pixel value of each third pixel in the enhanced ultrasonic image, and perform an erosion convolution operation on the second binary parameter corresponding to each third pixel according to the second binary parameter corresponding to each third pixel and the image erosion template to obtain the third binary parameter corresponding to each third pixel; determine the eroded ultrasonic image corresponding to the enhanced ultrasonic image according to the third binary parameters corresponding to all third pixels, and determine the reference image quality parameters of the eroded ultrasonic image; update the image quality parameters according to the reference image quality parameters, and trigger the execution of the operation of determining the image dilation template of the enhanced ultrasonic image according to the image quality parameters in the above steps.

[0120] In this optional embodiment, based on the image quality parameters, it is determined whether it is necessary to first perform an image erosion operation on the enhanced ultrasonic image. If so, the enhanced ultrasonic image is firstly subjected to image erosion, and then, based on the obtained image quality parameters of the eroded ultrasonic image, an image morphological dilation operation is performed on it; if not, the enhanced image is directly subjected to image dilation processing. It should be noted that by performing erosion on the enhanced ultrasonic image, image noise can be removed, but the image will be compressed. Therefore, performing dilation processing on the eroded ultrasonic image can remove image noise while maintaining the image content of the original enhanced ultrasonic image.

[0121] It can be seen that this optional embodiment can perform a corrosion demand determination operation on the enhanced ultrasonic image according to the image quality parameters of the enhanced ultrasonic image, and then perform an image corrosion operation on the enhanced ultrasonic image when there is an image corrosion demand, and then perform an image dilation operation on the corroded ultrasonic image. In this way, the image content of the original enhanced ultrasonic image can be maintained while removing image noise, which is beneficial to improving the processing reliability and accuracy of the enhanced ultrasonic image, thereby facilitating better realization of subsequent precise overlay processing of the expanded ultrasonic image.

[0122] In yet another optional embodiment, the step 206 performs a superposition processing operation on each frame of the expanded ultrasound image to obtain all superimposed ultrasound images, including:

[0123] For each expanded ultrasonic image, determining a plurality of frames of associated ultrasonic images associated with the expanded ultrasonic image from all other expanded ultrasonic images except the expanded ultrasonic image among all the expanded ultrasonic images, and determining a valid image area of ​​the expanded ultrasonic image according to a binary pixel parameter corresponding to the expanded ultrasonic image and binary pixel parameters corresponding to all the associated ultrasonic images;

[0124] For each expanded ultrasonic image, according to the effective image area of ​​the expanded ultrasonic image, an ultrasonic image to be superimposed that matches the effective image area is determined from all associated ultrasonic images, and according to the ultrasonic image to be superimposed, an image superposition operation is performed on the expanded ultrasonic image to obtain a superimposed ultrasonic image corresponding to the expanded ultrasonic image.

[0125] In this optional embodiment, the image superposition operation can be understood as an exclusive-OR superposition operation between the expanded ultrasonic image and its corresponding ultrasonic image to be superimposed. Optionally, the multiple frames of associated ultrasonic images associated with the expanded ultrasonic image can be multiple frames of adjacent ultrasonic images adjacent to the expanded ultrasonic image, or multiple frames of similar ultrasonic images whose image similarity to the expanded ultrasonic image is greater than or equal to a preset similarity threshold. Further, the ultrasonic image to be superimposed that matches the effective image area can be understood as an ultrasonic image to be superimposed that can make up for the incomplete defect area in the effective image area.

[0126] It can be seen that this optional embodiment can determine the effective image area of ​​the expanded ultrasonic image based on the binary parameters of the pixels corresponding to the expanded ultrasonic image and the binary parameters of the pixels corresponding to all the associated ultrasonic images, and then determine the ultrasonic image to be superimposed that matches the effective image area from all the associated ultrasonic images to realize the superimposition processing operation of the expanded ultrasonic image. In this way, the pixels of the effective image area in the expanded ultrasonic image can be accurately overlapped, and then the incomplete defective area in the effective image area can be effectively compensated, so as to improve the reliability and accuracy of the superimposition processing of the expanded ultrasonic image, so as to realize the subsequent accurate connected domain recognition operation of the superimposed ultrasonic image.

[0127] In yet another optional embodiment, the connected domain identification operation is performed on each frame of the superimposed ultrasound image in step 207 to obtain the heterogeneous region of each frame of the superimposed ultrasound image to be analyzed, including:

[0128] For each superimposed ultrasonic image, determining a connectivity mark parameter corresponding to each fourth pixel according to a fourth binary parameter corresponding to each fourth pixel in the superimposed ultrasonic image;

[0129] For each superimposed ultrasound image, multiple connected areas in the superimposed ultrasound image are determined according to the connectivity marking parameters corresponding to all fourth pixels and the preset connected area identification parameters, and a target connected area is determined from all the connected areas according to the area parameters corresponding to each connected area as the heterogeneous area of ​​the superimposed ultrasound image to be analyzed.

[0130] In this optional embodiment, the connected domain identification operation can be understood as selecting the area in the superimposed ultrasound image that is a block or contains important information to obtain the irregular area to be analyzed, so that the subsequent technicians can perform ultrasound image analysis operations on the irregular area. Optionally, the connected region identification parameter can be a 4-neighborhood identification parameter, an 8-neighborhood identification parameter, a D-neighborhood identification parameter, etc. Further optionally, the region parameter includes at least one of a region area parameter, a region pixel number parameter, and a region content parameter.

[0131] It can be seen that this optional embodiment can determine multiple connected areas in the superimposed ultrasonic image according to the connectivity mark parameters corresponding to each fourth pixel in the superimposed ultrasonic image and the preset connected area identification parameters, and determine the irregular area of ​​the superimposed ultrasonic image to be analyzed from all the connected areas. In this way, the reliability and accuracy of determining the irregular area to be analyzed can be improved, and then the precise analysis of the ultrasonic image areas that are blocked or contain important information can be improved, which is conducive to the smooth progress of ultrasonic imaging research.

[0132] Embodiment 3

[0133] See also Figure 3 , Figure 3 Schematic diagram of the structure of an automatic determination device for an ultrasonic image irregular region disclosed in an embodiment of the present invention. Figure 3 As shown, the automatic determination device of the ultrasonic image irregular area may include:

[0134] A determination module 301 is used to determine a plurality of frames of ultrasound images to be processed from a preset ultrasound video;

[0135] An equalization module 302 is used to perform a histogram equalization operation on each frame of the ultrasound image to obtain all equalized ultrasound images;

[0136] A gradient enhancement module 303 is used to perform a gradient enhancement operation on each frame of the equalized ultrasonic image to obtain all enhanced ultrasonic images;

[0137] An image expansion module 304 is used to perform an image morphological expansion operation on each frame of the enhanced ultrasound image to obtain all expanded ultrasound images;

[0138] A superposition processing module 305 is used to perform a superposition processing operation on each frame of the expanded ultrasonic image to obtain all superimposed ultrasonic images;

[0139] The region determination module 306 is used to perform a connected domain recognition operation on each frame of the superimposed ultrasound image to obtain a special-shaped region of each frame of the superimposed ultrasound image to be analyzed.

[0140] It can be seen that implementation Figure 3 The described automatic determination device for the irregular region of an ultrasonic image can improve the processing reliability and accuracy of the ultrasonic image by performing histogram equalization, gradient enhancement, image expansion, superposition processing and connected domain identification operations on the ultrasonic image, which is conducive to accurately obtaining the irregular region of the ultrasonic image to be analyzed, thereby facilitating the accurate analysis operation of the irregular region.

[0141] In an optional embodiment, the equalization module 302 performs a histogram equalization operation on each frame of the ultrasound image to obtain all equalized ultrasound images in the following manner:

[0142] For each ultrasound image, determine the pixel value of each first pixel in the ultrasound image, and determine the pixel parameter corresponding to each grayscale in the ultrasound image according to the pixel values ​​of all the first pixels;

[0143] For each ultrasound image, according to the pixel parameters corresponding to all grayscales, the pixel cumulative probability parameter corresponding to each grayscale is determined, and according to the pixel cumulative probability parameter corresponding to each grayscale and the grayscale value of the target grayscale in the ultrasound image, the grayscale mapping value corresponding to each grayscale is calculated;

[0144] For each ultrasonic image, grayscale mapping processing is performed on all grayscales according to grayscale mapping values ​​corresponding to all grayscales to obtain an equalized ultrasonic image corresponding to the ultrasonic image.

[0145] It can be seen that implementation Figure 4 The described automatic determination device for the irregular-shaped area of ​​an ultrasonic image can determine the pixel parameters corresponding to each grayscale in the ultrasonic image based on the pixel values ​​of all the first pixels in the ultrasonic image, and then determine the pixel cumulative probability parameters corresponding to each grayscale, and calculate the grayscale mapping value corresponding to each grayscale to realize the grayscale mapping processing operation. This is conducive to improving the reliability and accuracy of the grayscale mapping processing operation in the ultrasonic image, and further conducive to improving the image contrast effect of the obtained equalized ultrasonic image, thereby facilitating improving the reliability and accuracy of the subsequent gradient enhancement operation of the equalized ultrasonic image.

[0146] In another optional embodiment, the gradient enhancement module 303 performs a gradient enhancement operation on each frame of the equalized ultrasound image to obtain all enhanced ultrasound images in the following manner:

[0147] For each equalized ultrasonic image, according to the preset edge detection requirement parameters of the equalized ultrasonic image, a gradient calculation direction parameter of the equalized ultrasonic image is determined, and according to the gradient calculation direction parameter, a pixel gradient calculation template matching the equalized ultrasonic image and an associated pixel corresponding to each second pixel in the equalized ultrasonic image are determined;

[0148] For each second pixel in each equalized ultrasound image, a pixel gradient parameter corresponding to the second pixel is calculated according to the pixel value of the second pixel and the pixel value of the associated pixel corresponding to the second pixel, and a pixel convolution operation is performed on the pixel value of the second pixel according to the pixel gradient parameter corresponding to the second pixel and a pixel gradient operation template to obtain an enhanced pixel value corresponding to the second pixel;

[0149] For each equalized ultrasonic image, an enhanced ultrasonic image corresponding to the equalized ultrasonic image is determined according to enhanced pixel values ​​corresponding to all second pixels in the equalized ultrasonic image.

[0150] It can be seen that implementation Figure 4 The described automatic determination device for the irregular-shaped area of ​​the ultrasonic image can determine the gradient calculation direction parameters of the equalized ultrasonic image according to the edge detection requirement parameters of the equalized ultrasonic image, and then determine the pixel gradient operation template of the equalized ultrasonic image and the associated pixels corresponding to the second pixel, and perform a pixel convolution operation on the second pixel according to the pixel gradient operation template and the associated pixels to achieve a gradient enhancement operation on the equalized ultrasonic image. In this way, not only can the gradient enhancement of the equalized ultrasonic image be flexibly performed based on the edge detection requirement parameters, but the reliability and accuracy of the enhanced ultrasonic image can also be improved, thereby improving the effectiveness of the gradient enhancement operation on the equalized ultrasonic image to enhance the degree of image change of the equalized ultrasonic image.

[0151] In yet another optional embodiment, the image expansion module 304 includes:

[0152] A determination submodule 3041 is used to determine, for each enhanced ultrasound image, an image quality parameter of the enhanced ultrasound image, and determine an image dilation template of the enhanced ultrasound image according to the image quality parameter;

[0153] The calculation submodule 3042 is used to determine, for each enhanced ultrasonic image, a first binary parameter corresponding to each third pixel according to the pixel value of each third pixel in the enhanced ultrasonic image, and perform a dilation convolution operation on the first binary parameter corresponding to each third pixel according to the first binary parameter corresponding to each third pixel and the image dilation template to obtain a target binary parameter corresponding to each third pixel;

[0154] The determination submodule 3041 is further configured to determine, for each enhanced ultrasonic image, the dilated ultrasonic image corresponding to the enhanced ultrasonic image according to the target binary parameters corresponding to all third pixels.

[0155] It can be seen that implementation Figure 4 The described automatic determination device for the irregular-shaped area of ​​an ultrasonic image can perform a corresponding dilation convolution operation on the third pixel in the enhanced ultrasonic image according to the image quality parameter of the enhanced ultrasonic image, thereby obtaining a dilated ultrasonic image corresponding to the enhanced ultrasonic image. In this way, the image morphological dilation operation can be flexibly performed on the enhanced ultrasonic image, thereby improving the reliability and accuracy of the obtained dilated ultrasonic image, reducing the image noise in the enhanced ultrasonic image and ensuring the continuity of the image, which is conducive to the subsequent accurate superposition processing of the dilated ultrasonic image.

[0156] In yet another optional embodiment, the determination submodule 3041 is further configured to:

[0157] Before determining the image dilation template of the enhanced ultrasound image according to the image quality parameter, determining the image corrosion requirement corresponding to the enhanced ultrasound image according to the image quality parameter, and judging whether the image corrosion requirement is greater than a preset corrosion requirement threshold;

[0158] When the judgment result is no, triggering the execution of an operation of determining an image dilation template of the enhanced ultrasound image according to the image quality parameter;

[0159] When the judgment result is yes, determine the image erosion template of the enhanced ultrasonic image according to the image quality parameters; determine the second binary parameter corresponding to each third pixel according to the pixel value of each third pixel in the enhanced ultrasonic image, and perform an erosion convolution operation on the second binary parameter corresponding to each third pixel according to the second binary parameter corresponding to each third pixel and the image erosion template to obtain the third binary parameter corresponding to each third pixel; determine the eroded ultrasonic image corresponding to the enhanced ultrasonic image according to the third binary parameters corresponding to all third pixels, and determine the reference image quality parameters of the eroded ultrasonic image; update the image quality parameters according to the reference image quality parameters, and trigger the execution of the operation of determining the image dilation template of the enhanced ultrasonic image according to the image quality parameters.

[0160] It can be seen that implementation Figure 4 The described automatic determination device for the irregular-shaped area of ​​an ultrasonic image can perform a corrosion demand determination operation on the enhanced ultrasonic image according to the image quality parameters of the enhanced ultrasonic image, and then perform an image corrosion operation on the enhanced ultrasonic image when there is an image corrosion demand, and then perform an image dilation processing on the corroded ultrasonic image. In this way, the image content of the original enhanced ultrasonic image can be maintained while removing image noise, which is beneficial to improving the processing reliability and accuracy of the enhanced ultrasonic image, thereby facilitating better realization of subsequent precise overlay processing of the expanded ultrasonic image.

[0161] In another optional embodiment, the superposition processing module 305 performs a superposition processing operation on each frame of the expanded ultrasonic image to obtain all superimposed ultrasonic images, specifically including:

[0162] For each expanded ultrasonic image, determining a plurality of frames of associated ultrasonic images associated with the expanded ultrasonic image from all other expanded ultrasonic images except the expanded ultrasonic image among all the expanded ultrasonic images, and determining a valid image area of ​​the expanded ultrasonic image according to a binary pixel parameter corresponding to the expanded ultrasonic image and binary pixel parameters corresponding to all the associated ultrasonic images;

[0163] For each expanded ultrasonic image, according to the effective image area of ​​the expanded ultrasonic image, an ultrasonic image to be superimposed that matches the effective image area is determined from all associated ultrasonic images, and according to the ultrasonic image to be superimposed, an image superposition operation is performed on the expanded ultrasonic image to obtain a superimposed ultrasonic image corresponding to the expanded ultrasonic image.

[0164] It can be seen that implementation Figure 4 The described automatic determination device for the irregular-shaped area of ​​an ultrasonic image can determine the effective image area of ​​the expanded ultrasonic image based on the binary pixel parameters corresponding to the expanded ultrasonic image and the binary pixel parameters corresponding to all associated ultrasonic images, and then determine the ultrasonic image to be superimposed that matches the effective image area from all associated ultrasonic images to realize the superimposition processing operation of the expanded ultrasonic image. In this way, the pixels of the effective image area in the expanded ultrasonic image can be accurately overlapped, and the incomplete defective area in the effective image area can be effectively compensated, thereby improving the reliability and accuracy of the superimposition processing of the expanded ultrasonic image, so as to realize the subsequent accurate connected domain recognition operation of the superimposed ultrasonic image.

[0165] In another optional embodiment, the region determination module 306 performs a connected domain identification operation on each frame of the superimposed ultrasound image to obtain the irregular region of each frame of the superimposed ultrasound image to be analyzed, specifically including:

[0166] For each superimposed ultrasonic image, determining a connectivity mark parameter corresponding to each fourth pixel according to a fourth binary parameter corresponding to each fourth pixel in the superimposed ultrasonic image;

[0167] For each superimposed ultrasound image, multiple connected areas in the superimposed ultrasound image are determined according to the connectivity marking parameters corresponding to all fourth pixels and the preset connected area identification parameters, and a target connected area is determined from all the connected areas according to the area parameters corresponding to each connected area as the heterogeneous area of ​​the superimposed ultrasound image to be analyzed.

[0168] In this optional embodiment, the region parameter includes at least one of a region area parameter, a region pixel quantity parameter, and a region content parameter.

[0169] It can be seen that implementation Figure 4The described automatic determination device for the irregular region of the ultrasonic image can determine multiple connected regions in the superimposed ultrasonic image according to the connected marking parameters corresponding to each fourth pixel in the superimposed ultrasonic image and the preset connected region identification parameters, and determine the irregular region of the superimposed ultrasonic image to be analyzed from all the connected regions. In this way, the reliability and accuracy of the determination of the irregular region to be analyzed can be improved, and then the precise analysis of the ultrasonic image regions that are blocked or contain important information can be improved, which is conducive to the smooth progress of ultrasonic imaging research.

[0170] Embodiment 4

[0171] See also Figure 5 , Figure 5 FIG. 1 is a schematic diagram of the structure of another device for automatically determining an abnormally shaped area of ​​an ultrasonic image disclosed in an embodiment of the present invention. Figure 5 As shown, the automatic determination device of the ultrasonic image irregular area may include:

[0172] A memory 401 storing executable program codes;

[0173] a processor 402 coupled to the memory 401;

[0174] The processor 402 calls the executable program code stored in the memory 401 to execute the steps of the method for automatically determining the irregular-shaped area of ​​the ultrasound image described in the first embodiment of the present invention or the second embodiment of the present invention.

[0175] Embodiment 5

[0176] An embodiment of the present invention discloses a computer storage medium storing computer instructions. When the computer instructions are called, they are used to execute the steps of the method for automatically determining the abnormal-shaped area of ​​an ultrasound image described in the first embodiment of the present invention or the second embodiment of the present invention.

[0177] Embodiment 6

[0178] An embodiment of the present invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to enable a computer to execute the steps of the method for automatically determining an abnormally shaped area in an ultrasound image described in Embodiment 1 or Embodiment 2.

[0179] The device embodiments described above are only illustrative, wherein the modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules, i.e., they may be located in one place, or they may be distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Those of ordinary skill in the art may understand and implement it without creative work.

[0180] Through the specific description of the above embodiments, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the above technical solution can be essentially or partly contributed to the prior art in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, and the storage medium includes a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically erasable rewritable read-only memory (EEPROM), a compact disc (CD-ROM) or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0181] Finally, it should be noted that the method and device for automatically determining the irregular area of ​​an ultrasonic image disclosed in the embodiment of the present invention disclose only the preferred embodiments of the present invention, which are only used to illustrate the technical solution of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments can still be modified, or some of the technical features therein can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for automatically determining irregular-shaped areas of an ultrasonic image, characterized in that: The method comprises: Determining a plurality of frames of ultrasound images to be processed from a preset ultrasound video; Performing a histogram equalization operation on each frame of the ultrasonic image to obtain all equalized ultrasonic images, and performing a gradient enhancement operation on each frame of the equalized ultrasonic image to obtain all enhanced ultrasonic images; For each of the enhanced ultrasonic images, determine the image quality parameters of the enhanced ultrasonic image, and determine the image dilation template of the enhanced ultrasonic image according to the image quality parameters; for each of the enhanced ultrasonic images, determine the first binary parameter corresponding to each third pixel according to the pixel value of each third pixel in the enhanced ultrasonic image, and perform a dilation convolution operation on the first binary parameter corresponding to each third pixel according to the first binary parameter corresponding to each third pixel and the image dilation template to obtain a target binary parameter corresponding to each third pixel; for each of the enhanced ultrasonic images, determine the dilated ultrasonic image corresponding to the enhanced ultrasonic image according to the target binary parameters corresponding to all the third pixels; For each of the expanded ultrasonic images, a plurality of frames of associated ultrasonic images associated with the expanded ultrasonic image are determined from all other expanded ultrasonic images except the expanded ultrasonic image among all the expanded ultrasonic images, and an effective image area of ​​the expanded ultrasonic image is determined according to a binary pixel parameter corresponding to the expanded ultrasonic image and binary pixel parameters corresponding to all the associated ultrasonic images; for each of the expanded ultrasonic images, an ultrasonic image to be superimposed that matches the effective image area is determined from all the associated ultrasonic images according to the effective image area of ​​the expanded ultrasonic image, and an image superposition operation is performed on the expanded ultrasonic image according to the ultrasonic image to be superimposed to obtain a superimposed ultrasonic image corresponding to the expanded ultrasonic image; Performing a connected domain recognition operation on each frame of the superimposed ultrasonic image to obtain a special-shaped region of each frame of the superimposed ultrasonic image to be analyzed; The step of performing a gradient enhancement operation on each frame of the equalized ultrasonic image to obtain all enhanced ultrasonic images includes: For each of the equalized ultrasonic images, determine a gradient calculation direction parameter of the equalized ultrasonic image according to a preset edge detection requirement parameter of the equalized ultrasonic image, and determine a pixel gradient calculation template matching the equalized ultrasonic image and associated pixels corresponding to each second pixel in the equalized ultrasonic image according to the gradient calculation direction parameter; For each second pixel in each of the equalized ultrasound images, a pixel gradient parameter corresponding to the second pixel is calculated according to a pixel value of the second pixel and a pixel value of an associated pixel corresponding to the second pixel, and a pixel convolution operation is performed on the pixel value of the second pixel according to the pixel gradient parameter corresponding to the second pixel and the pixel gradient operation template to obtain an enhanced pixel value corresponding to the second pixel; For each of the equalized ultrasonic images, determining an enhanced ultrasonic image corresponding to the equalized ultrasonic image according to the enhanced pixel values ​​corresponding to all the second pixels in the equalized ultrasonic image; The step of performing a connected domain recognition operation on each frame of the superimposed ultrasound image to obtain a heterogeneous region of each frame of the superimposed ultrasound image to be analyzed includes: For each of the superimposed ultrasonic images, determining a connectivity mark parameter corresponding to each of the fourth pixels in the superimposed ultrasonic images according to a fourth binary parameter corresponding to each of the fourth pixels; For each of the superimposed ultrasound images, multiple connected areas in the superimposed ultrasound image are determined based on the connectivity mark parameters corresponding to all the fourth pixels and the preset connected area identification parameters, and a target connected area is determined from all the connected areas based on the area parameters corresponding to each of the connected areas as the heterogeneous area of ​​the superimposed ultrasound image to be analyzed; the area parameters include at least one of an area area parameter, an area pixel number parameter and an area content parameter.

2. The method for automatically determining the irregular-shaped area of ​​an ultrasonic image according to claim 1, characterized in that: The step of performing a histogram equalization operation on each frame of the ultrasonic image to obtain all the equalized ultrasonic images includes: For each of the ultrasonic images, determine a pixel value of each first pixel in the ultrasonic image, and determine a pixel parameter corresponding to each grayscale in the ultrasonic image according to the pixel values ​​of all the first pixels; For each of the ultrasound images, determining a pixel cumulative probability parameter corresponding to each of the grayscales according to pixel parameters corresponding to all of the grayscales, and calculating a grayscale mapping value corresponding to each of the grayscales according to the pixel cumulative probability parameter corresponding to each of the grayscales and the grayscale value of the target grayscale in the ultrasound image; For each of the ultrasonic images, grayscale mapping processing is performed on all the grayscales according to the grayscale mapping values ​​corresponding to all the grayscales to obtain an equalized ultrasonic image corresponding to the ultrasonic image.

3. The method for automatically determining the irregular-shaped area of ​​an ultrasonic image according to claim 1, characterized in that: Before determining the image dilation template of the enhanced ultrasound image according to the image quality parameter, the method further includes: Determining an image corrosion requirement corresponding to the enhanced ultrasound image according to the image quality parameter, and judging whether the image corrosion requirement is greater than a preset corrosion requirement threshold; When the judgment result is no, triggering the execution of the operation of determining the image expansion template of the enhanced ultrasound image according to the image quality parameter; When the judgment result is yes, the image erosion template of the enhanced ultrasonic image is determined according to the image quality parameters; the second binary parameter corresponding to each third pixel in the enhanced ultrasonic image is determined according to the pixel value of each third pixel, and the second binary parameter corresponding to each third pixel and the image erosion template are used to perform an erosion convolution operation on the second binary parameter corresponding to each third pixel to obtain the third binary parameter corresponding to each third pixel; the eroded ultrasonic image corresponding to the enhanced ultrasonic image is determined according to the third binary parameters corresponding to all the third pixels, and the reference image quality parameters of the eroded ultrasonic image are determined; the image quality parameters are updated according to the reference image quality parameters, and the operation of determining the image dilation template of the enhanced ultrasonic image according to the image quality parameters is triggered.

4. An automatic determination device for an abnormally shaped area of ​​an ultrasonic image, characterized in that: The device comprises: A determination module, used to determine multiple frames of ultrasound images to be processed from a preset ultrasound video; An equalization module, used for performing a histogram equalization operation on each frame of the ultrasonic image to obtain all equalized ultrasonic images; A gradient enhancement module, used for performing a gradient enhancement operation on each frame of the equalized ultrasonic image to obtain all enhanced ultrasonic images; An image dilation module is used to determine, for each of the enhanced ultrasonic images, an image quality parameter of the enhanced ultrasonic image, and determine, according to the image quality parameter, an image dilation template of the enhanced ultrasonic image; for each of the enhanced ultrasonic images, determine, according to the pixel value of each third pixel in the enhanced ultrasonic image, a first binary parameter corresponding to each third pixel, and, according to the first binary parameter corresponding to each third pixel and the image dilation template, perform a dilation convolution operation on the first binary parameter corresponding to each third pixel to obtain a target binary parameter corresponding to each third pixel; for each of the enhanced ultrasonic images, determine, according to the target binary parameters corresponding to all the third pixels, a dilated ultrasonic image corresponding to the enhanced ultrasonic image; a superposition processing module, configured to determine, for each of the expanded ultrasonic images, from all the other expanded ultrasonic images except the expanded ultrasonic image among all the expanded ultrasonic images, a plurality of frames of associated ultrasonic images associated with the expanded ultrasonic image, and determine, according to the binary pixel parameters corresponding to the expanded ultrasonic image and the binary pixel parameters corresponding to all the associated ultrasonic images, a valid image region of the expanded ultrasonic image; for each of the expanded ultrasonic images, according to the valid image region of the expanded ultrasonic image, determine, from all the associated ultrasonic images, an ultrasonic image to be superimposed that matches the valid image region, and, according to the ultrasonic image to be superimposed, perform an image superposition operation on the expanded ultrasonic image to obtain a superimposed ultrasonic image corresponding to the expanded ultrasonic image; A region determination module, used for performing a connected domain recognition operation on each frame of the superimposed ultrasonic image to obtain a special-shaped region of each frame of the superimposed ultrasonic image to be analyzed; The gradient enhancement module performs a gradient enhancement operation on each frame of the equalized ultrasonic image to obtain all enhanced ultrasonic images in the following manner: For each of the equalized ultrasonic images, determine a gradient calculation direction parameter of the equalized ultrasonic image according to a preset edge detection requirement parameter of the equalized ultrasonic image, and determine a pixel gradient calculation template matching the equalized ultrasonic image and associated pixels corresponding to each second pixel in the equalized ultrasonic image according to the gradient calculation direction parameter; For each second pixel in each of the equalized ultrasound images, a pixel gradient parameter corresponding to the second pixel is calculated according to a pixel value of the second pixel and a pixel value of an associated pixel corresponding to the second pixel, and a pixel convolution operation is performed on the pixel value of the second pixel according to the pixel gradient parameter corresponding to the second pixel and the pixel gradient operation template to obtain an enhanced pixel value corresponding to the second pixel; For each of the equalized ultrasonic images, determining an enhanced ultrasonic image corresponding to the equalized ultrasonic image according to the enhanced pixel values ​​corresponding to all the second pixels in the equalized ultrasonic image; The region determination module performs a connected domain recognition operation on each frame of the superimposed ultrasound image to obtain the irregular region of each frame of the superimposed ultrasound image to be analyzed, specifically including: For each of the superimposed ultrasonic images, determining a connectivity mark parameter corresponding to each of the fourth pixels in the superimposed ultrasonic images according to a fourth binary parameter corresponding to each of the fourth pixels; For each of the superimposed ultrasound images, multiple connected areas in the superimposed ultrasound image are determined based on the connectivity mark parameters corresponding to all the fourth pixels and the preset connected area identification parameters, and a target connected area is determined from all the connected areas based on the area parameters corresponding to each of the connected areas as the heterogeneous area of ​​the superimposed ultrasound image to be analyzed; the area parameters include at least one of an area area parameter, an area pixel number parameter and an area content parameter.

5. An automatic determination device for an abnormally shaped area of ​​an ultrasonic image, characterized in that: The device comprises: A memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the method for automatically determining the irregularly shaped area of ​​an ultrasound image as described in any one of claims 1-3.

6. A computer storage medium, characterized in that: The computer storage medium stores computer instructions, and when the computer instructions are called, they are used to execute the method for automatically determining the abnormal-shaped area of ​​an ultrasound image as described in any one of claims 1-3.

Images