Sound shadow image detection system and method based on multi-dimensional feature fusion and structure guidance

Abstract

The invention relates to an acoustic shadow image detection system and method based on multi-dimensional feature fusion and structure guidance. The system comprises a first acquisition module for acquiring ultrasonic radio frequency data and a B-mode image obtained by performing logarithmic compression on the ultrasonic radio frequency data; the second acquisition module is used for acquiring a local scale parameter diagram and a local shape parameter diagram based on the ultrasonic radio frequency data, the local scale parameter diagram comprises local scale parameters corresponding to the positions respectively, and the local shape parameter diagram comprises local shape parameters corresponding to the positions respectively; the sound shadow probability graph acquisition module is used for acquiring a sound shadow probability graph based on the local scale parameter graph and the local shape parameter graph; and the correction module takes the B-mode image as a guide image, and performs sound shadow edge correction on the sound shadow probability graph to obtain a sound shadow detection image. According to the method, a low-echo pathological structure and a normal liquid dark region can be effectively distinguished, and high-precision boundary positioning of an acoustic shadow region can be realized based on refined analysis.

Description

Technical Field

[0001] Several embodiments of this specification relate to the field of sound and shadow detection technology, specifically to a sound and shadow image detection system and method based on multi-dimensional feature fusion and structure guidance. Background Technology

[0002] In ultrasound imaging, acoustic shadowing is an important indicator for diagnosing stones, calcifications, or bone defects. Current acoustic shadowing detection techniques suffer from two main limitations:

[0003] Insufficient specificity: When relying solely on B-mode grayscale or a single energy statistical parameter (such as Nakagami Ω), normal hypoechoic fluid structures such as blood vessel cross-sections, cysts, and gallbladders are easily misjudged as acoustic shadows (false positives) because they all appear as "dark areas" in terms of energy.

[0004] Inaccurate boundary localization: Statistical parametric imaging based on radio frequency (RF) data is limited by the sliding window mechanism, resulting in low resolution and blurry, mosaic-like edges in the generated detection results, making it impossible to achieve pixel-level precise segmentation.

[0005] Therefore, there is an urgent need for a detection method that can effectively distinguish between hypoechoic pathological structures and normal liquid dark areas, and achieve high-precision boundary localization of acoustic shadow regions based on refined analysis. Summary of the Invention

[0006] This specification provides an embodiment of a sound shadow image detection system and method based on multi-dimensional feature fusion and structure guidance, which can effectively distinguish between hypoechoic pathological structures and normal liquid dark areas, and can achieve high-precision boundary positioning of sound shadow regions based on refined analysis.

[0007] The technical solution is as follows:

[0008] Firstly, embodiments of this specification provide a sound and shadow image detection system based on multi-dimensional feature fusion and structure guidance, comprising:

[0009] The first acquisition module acquires ultrasonic radio frequency data and a B-mode image obtained after logarithmic compression of the ultrasonic radio frequency data;

[0010] The second acquisition module acquires a local Nakagami scale parameter map and a local Nakagami shape parameter map based on ultrasound radio frequency data. The local Nakagami scale parameter map includes the local Nakagami scale parameters corresponding to each location, and the local Nakagami shape parameter map includes the local Nakagami shape parameters corresponding to each location.

[0011] The sound and shadow probability map acquisition module acquires the sound and shadow probability map based on the local Nakagami scale parameter map and the local Nakagami shape parameter map;

[0012] The correction module uses the B-mode image as a guide image to perform sound and shadow edge correction on the sound and shadow probability map in order to obtain a sound and shadow detection image.

[0013] As a preferred embodiment, the correction module includes an edge correction unit and an adaptive segmentation unit;

[0014] The edge correction unit uses the B-mode image as a guide image to perform sound and shadow edge correction on the sound and shadow probability map to obtain an edge-corrected image;

[0015] The adaptive segmentation unit performs global adaptive thresholding on the edge-corrected image to obtain the sound and shadow detection image.

[0016] As a preferred embodiment, the edge correction unit uses the B-mode image as a guide image and utilizes the high-frequency edge information in the B-mode image to perform sound and shadow edge correction on the sound and shadow probability map to obtain a sound and shadow detection image.

[0017] As a preferred embodiment, the sound and shadow probability map acquisition module includes a first processing unit, a second processing unit, and a probability map acquisition unit;

[0018] The first processing unit performs logarithmic processing, normalization processing, and numerical inversion processing on the local Nakagami scale parameters corresponding to each position in the local Nakagami scale parameter map in turn to obtain the first sound and shadow filtering weight corresponding to each position, and then obtains the first sound and shadow filtering weight map. The lower the local Nakagami scale parameter in the first sound and shadow filtering weight map, the larger the first sound and shadow filtering weight.

[0019] The second processing unit uses a Gaussian function or a Sigmoid function to process the local Nakagami shape parameters corresponding to each position in the local Nakagami shape parameter map to obtain the second sound and shadow filtering weight corresponding to each position, and then obtains the second sound and shadow filtering weight map. The closer the local Nakagami shape parameter in the second sound and shadow filtering weight map is to the value 1, the smaller the second sound and shadow filtering weight is.

[0020] The probability map acquisition unit acquires the sound and shadow probability map based on the first sound and shadow filtering weight map and the second sound and shadow filtering weight map.

[0021] As a preferred solution, the probability map acquisition unit multiplies the first sound and shadow filtering weight corresponding to each position in the first sound and shadow filtering weight map and the second sound and shadow filtering weight corresponding to each position in the second sound and shadow filtering weight map based on the position correspondence to obtain the sound and shadow probability map.

[0022] Secondly, embodiments of this specification provide a sound and shadow image detection method based on multi-dimensional feature fusion and structure guidance, including:

[0023] Acquire ultrasound radio frequency data and B-mode images obtained after logarithmic compression of the ultrasound radio frequency data;

[0024] Based on ultrasound radio frequency data, local Nakagami scale parameter map and local Nakagami shape parameter map are obtained. The local Nakagami scale parameter map includes the local Nakagami scale parameter corresponding to each location, and the local Nakagami shape parameter map includes the local Nakagami shape parameter corresponding to each location.

[0025] Based on the local Nakagami scale parameter map and the local Nakagami shape parameter map, the sound-shadow probability map is obtained;

[0026] Using the B-mode image as a guide image, sound and shadow edge correction is performed on the sound and shadow probability map to obtain a sound and shadow detection image.

[0027] As a preferred embodiment, the step of using the B-mode image as a guide image to perform sound shadow edge correction on the sound shadow probability map to obtain a sound shadow detection image includes...

[0028] Using the B-mode image as a guide image, sound and shadow edge correction is performed on the sound and shadow probability map to obtain an edge-corrected image;

[0029] Global adaptive thresholding is applied to the edge-corrected image to obtain the sound and shadow detection image.

[0030] As a preferred embodiment, the step of using the B-mode image as a guide image to perform sound and shadow edge correction on the sound and shadow probability map to obtain an edge-corrected image includes:

[0031] The B-mode image is used as the guide image, and the high-frequency edge information in the B-mode image is used to perform sound shadow edge correction on the sound shadow probability map to obtain a sound shadow detection image.

[0032] As a preferred embodiment, the step of obtaining the sound-shadow probability map based on the local Nakagami scale parameter map and the local Nakagami shape parameter map includes:

[0033] Logarithmic processing, normalization processing, and numerical inversion processing are performed on the local Nakagami scale parameters corresponding to each position in the local Nakagami scale parameter map to obtain the first sound shadow screening weight corresponding to each position, thereby obtaining the first sound shadow screening weight map. The lower the local Nakagami scale parameter in the first sound shadow screening weight map, the larger the first sound shadow screening weight.

[0034] The local Nakagami shape parameters corresponding to each position in the local Nakagami shape parameter map are processed using the Gaussian function or the Sigmoid function to obtain the second sound shadow filtering weight corresponding to each position, thereby obtaining the second sound shadow filtering weight map. The closer the local Nakagami shape parameter in the second sound shadow filtering weight map is to the value 1, the smaller the second sound shadow filtering weight is.

[0035] Based on the first sound and shadow filtering weight map and the second sound and shadow filtering weight map, a sound and shadow probability map is obtained.

[0036] As a preferred embodiment, obtaining the sound-shadow probability map based on the first sound-shadow filtering weight map and the second sound-shadow filtering weight map includes:

[0037] Based on positional correspondence, the first sound and shadow filtering weight corresponding to each position in the first sound and shadow filtering weight map and the second sound and shadow filtering weight corresponding to each position in the second sound and shadow filtering weight map are multiplied together to obtain the sound and shadow probability map.

[0038] Thirdly, embodiments of this specification provide an electronic device, including a processor and a memory; the processor is connected to the memory; the memory is used to store executable program code; the processor reads the executable program code stored in the memory to run a program corresponding to the executable program code, so as to perform the steps described in the second aspect of the above embodiments.

[0039] Fourthly, embodiments of this specification provide a computer storage medium storing a plurality of instructions adapted for loading by a processor and executing the steps described in the second aspect of the above embodiments.

[0040] The beneficial effects of the technical solutions provided in some embodiments of this specification include at least the following:

[0041] By combining the local Nakagami scale parameter map and the local Nakagami shape parameter map, a sound shadow probability map can be obtained, which can effectively distinguish between sound shadows and liquid dark areas.

[0042] Using the B-mode image as a guide image, sound shadow edge correction is performed on the sound shadow probability map to achieve high-precision positioning of the sound shadow boundary. This is achieved by using structure-guided filtering to solve the problem of low resolution in statistical imaging. Attached Figure Description

[0043] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0044] Figure 1 A schematic diagram of the structure of a sound and shadow image detection system based on multidimensional feature fusion and structure guidance, according to some embodiments of the present disclosure, is shown.

[0045] Figure 2 A flowchart illustrating a sound and shadow image detection method based on multidimensional feature fusion and structure guidance, according to some embodiments of this disclosure, is shown.

[0046] Figure 3 A schematic block diagram of an electronic device according to some embodiments of the present disclosure is shown. Detailed Implementation

[0047] The technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings.

[0048] The terms "first," "second," "third," etc., in the description, claims, and accompanying drawings are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such processes, methods, products, or apparatus.

[0049] The following description provides examples and does not limit the scope, applicability, or examples set forth in the claims. Changes may be made to the function and arrangement of the described elements without departing from the scope of this specification. Various processes or components may be appropriately omitted, substituted, or added to the examples. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Furthermore, features described with respect to some examples may be combined into other examples.

[0050] Figure 1 Schematic diagrams of the structure of a sound and shadow image detection system based on multi-dimensional feature fusion and structure guidance, according to some embodiments of this disclosure, are shown. Figure 1 As shown, a sound and shadow image detection system based on multidimensional feature fusion and structure guidance can include at least:

[0051] The first acquisition module acquires ultrasonic radio frequency data and a B-mode image obtained after logarithmic compression of the ultrasonic radio frequency data;

[0052] The second acquisition module acquires a local Nakagami scale parameter map and a local Nakagami shape parameter map based on ultrasound radio frequency data. The local Nakagami scale parameter map includes the local Nakagami scale parameter Ω corresponding to each location, and the local Nakagami shape parameter map includes the local Nakagami shape parameter m corresponding to each location.

[0053] The sound and shadow probability map acquisition module acquires the sound and shadow probability map based on the local Nakagami scale parameter map and the local Nakagami shape parameter map;

[0054] The correction module uses the B-mode image as a guide image to perform sound and shadow edge correction on the sound and shadow probability map in order to obtain a sound and shadow detection image.

[0055] Ultrasound radiofrequency data is used for statistical analysis, and B-mode images are used to provide anatomical information.

[0056] The local Nakagami scale parameter Ω in the local Nakagami scale parameter map is used to characterize the magnitude of echo energy, thereby generating an "energy map", in which the sound shadow and liquid regions are both low values.

[0057] The local Nakagami shape parameter *m* in the local Nakagami shape parameter plot is used to characterize the scatterer distribution. It satisfies the following condition:

[0058] Fluid-filled areas (blood vessels / cysts): The local Nakagami shape parameter m stably approaches 1.0 (Note: i.e., it satisfies the Rayleigh distribution).

[0059] Sound shadow region: The local Nakagami shape parameter m value is usually less than 1 or deviates from 1 due to calculation instability (Note: that is, it does not satisfy the Rayleigh distribution).

[0060] In this invention, the local Nakagami shape parameter m is estimated using the inverse normalized variance method, and its calculation formula is: m = E[I] 2 / Var(I), where I is the square of the radio frequency signal envelope (i.e., intensity), and E[⋅] and Var(⋅) represent the mean and variance within the local sliding window, respectively.

[0061] According to the statistical theory of ultrasound scattering, when an ultrasound beam passes through fluid-filled tissues such as blood vessels, cysts, or gallbladders, the echo signal envelope follows a Rayleigh distribution because this region contains a large number of randomly distributed tiny scatterers (such as red blood cells) and there is no highly coherent specular reflection. The local Nakagami shape parameter m is a general expression of the Rayleigh distribution; when the local Nakagami shape parameter m = 1, it indicates that the Rayleigh distribution is satisfied.

[0062] Furthermore, although sound shadows are also dark areas, they are caused by signal attenuation or loss. Although pure thermal noise also follows a Rayleigh distribution, due to computational instability at extremely low signal-to-noise ratios, the local Nakagami shape parameter m usually deviates from the value of 1 or fluctuates significantly.

[0063] Therefore, in ultrasound images, the local Nakagami shape parameter m value of fluid-filled dark areas (such as blood vessels) stably approaches the value of 1. In contrast, due to severe signal attenuation and electronic noise interference, or due to the non-uniformity of tissue boundaries, the local Nakagami shape parameter m value of acoustic shadowing areas usually deviates from the value of 1. This invention utilizes this physical characteristic to effectively distinguish between "fluid-filled dark areas" and "acoustic shadowing dark areas" by detecting whether the local Nakagami shape parameter m value is close to the value of 1, thereby eliminating false positive interference such as blood vessels.

[0064] Therefore, it can be understood that the location of a sound shadow or fluid region can be initially determined by the local Nakagami scale parameter map. Further, the location can be re-evaluated by the local Nakagami shape parameter map, thus eliminating fluid regions and ultimately identifying the sound shadow region. In other words, by combining the local Nakagami scale parameter map and the local Nakagami shape parameter map to obtain a sound shadow probability map, the sound shadow and fluid dark region can be effectively distinguished. This avoids the problem of misjudging normal hypoechoic fluid structures such as blood vessel cross-sections, cysts, and gallbladders as sound shadows when relying solely on B-mode grayscale or a single energy statistical parameter (such as Nakagami Ω).

[0065] Furthermore, in the embodiments of this specification, the B-mode image is used as a guide image to perform sound shadow edge correction on the sound shadow probability map in order to achieve high-precision positioning of the sound shadow boundary, that is, to use structure-guided filtering to solve the problem of low statistical imaging resolution.

[0066] It should also be noted that in the embodiments of this specification, the sound shadow probability map is obtained first, and then the B-mode image is used as a guide image for structure-guided filtering based on the sound shadow probability map, which can further improve the positioning accuracy of the sound shadow boundary. Because the sound shadow probability map has significantly enhanced the contrast between the sound shadow area and the non-sound shadow area, the filter can more focused on correcting the sound shadow boundary, avoiding the interference of "false positive" dark areas such as blood vessels and cysts in the original image on the filtering process, and preventing the sound shadow boundary from shifting.

[0067] Using the above method, a sound and shadow detection image with accurate sound and shadow region judgment and precise sound and shadow boundary delineation can be obtained.

[0068] In some embodiments of this specification, the correction module includes an edge correction unit and an adaptive segmentation unit;

[0069] The edge correction unit uses the B-mode image as a guide image to perform sound and shadow edge correction on the sound and shadow probability map to obtain an edge-corrected image;

[0070] The adaptive segmentation unit performs global adaptive thresholding on the edge-corrected image (e.g., using the Otsu method to generate a high-precision binarized sound and shadow mask) to obtain the sound and shadow detection image.

[0071] The edge correction unit uses the B-mode image as a guide image and utilizes high-frequency edge information (such as tissue interfaces) in the B-mode image to perform sound shadow edge correction on the sound shadow probability map to obtain a sound shadow detection image.

[0072] In the embodiments of this specification, local Nakagami shape parameters are introduced as "texture filters" to automatically identify and exclude interference items such as blood vessels and cysts, thus solving the misjudgment problem of "detecting every black element" in traditional methods.

[0073] In the embodiments described in this specification, guided filtering is performed in conjunction with B-mode images to eliminate block artifacts in the statistical parameter map, resulting in smooth and accurate sound and shadow contours with pixel-level precision.

[0074] In the embodiments described in this specification, the fusion algorithm uses soft fusion instead of hard cutoff, combined with Otsu adaptive segmentation, which can adapt to images with different gains and depths without manual parameter adjustment.

[0075] In some embodiments of this specification, the sound and shadow probability map acquisition module includes a first processing unit, a second processing unit, and a probability map acquisition unit;

[0076] The first processing unit performs logarithmic processing, normalization processing, and numerical inversion processing on the local Nakagami scale parameters corresponding to each position in the local Nakagami scale parameter map in turn to obtain the first sound and shadow filtering weight corresponding to each position, and then obtains the first sound and shadow filtering weight map. The lower the local Nakagami scale parameter in the first sound and shadow filtering weight map, the larger the first sound and shadow filtering weight.

[0077] The second processing unit uses a Gaussian function or a Sigmoid function to process the local Nakagami shape parameters corresponding to each position in the local Nakagami shape parameter map to obtain the second sound and shadow filtering weight corresponding to each position, and then obtains the second sound and shadow filtering weight map. The closer the local Nakagami shape parameter in the second sound and shadow filtering weight map is to the value 1, the smaller the second sound and shadow filtering weight is.

[0078] The probability map acquisition unit acquires the sound and shadow probability map based on the first sound and shadow filtering weight map and the second sound and shadow filtering weight map.

[0079] The probability map acquisition unit multiplies the first sound and shadow filtering weight corresponding to each position in the first sound and shadow filtering weight map and the second sound and shadow filtering weight corresponding to each position in the second sound and shadow filtering weight map based on the position correspondence to obtain the sound and shadow probability map.

[0080] The following is a specific example of the calculation process in obtaining the sound and shadow probability map:

[0081] 1. Set the window size (e.g., 15 pixels), and calculate the local Nakagami scale parameter Ω(x, y) and local Nakagami shape parameter m(x, y) based on the set window size and ultrasonic radio frequency data (RF envelope signal), where (x, y) are the position coordinates.

[0082] 2. Perform logarithmic, normalization, and numerical inversion processing on the local Nakagami scale parameter Ω(x,y) in sequence to obtain the first sound and shadow screening weight S(x,y) corresponding to Ω(x,y). The specific calculation formula used for the processing can be, but is not limited to: S(x,y)=1-Normalize[logΩ(x,y)], where S(x,y) represents the first sound and shadow screening weight corresponding to the position coordinate (x,y), and Normalize[⋅] represents normalization.

[0083] 3. Use a Gaussian or Sigmoid function to process the local Nakagami shape parameter m(x,y) to suppress the weights of local Nakagami shape parameters m(x,y) that are close to the value 1. The calculation formula used for processing can be, but is not limited to: W(x,y) = 1 - exp[-[m(x,y) - 1] 2 / σ 2 ], σ represents the sensitivity coefficient, which can be set to 0.1, but is not limited to 0.1. W(x,y) represents the second sound and shadow filtering weight corresponding to the position coordinates (x,y).

[0084] 4. The formulas for calculating the sound and shadow probabilities at each location in the sound and shadow probability diagram can be, but are not limited to, the following:

[0085] G(x,y)=S(x,y)*W(x,y);

[0086] Where G(x, y) represents the probability of sound and shadow corresponding to the position coordinates (x, y) in the sound and shadow probability diagram. The higher the probability of sound and shadow, the more likely it is to be sound and shadow.

[0087] Figure 2 The following are schematic flowcharts illustrating some embodiments of the sound and shadow image detection method based on multidimensional feature fusion and structure guidance. The various embodiments in this specification are described in a progressive manner, with reference to each other for similar or identical parts. Each embodiment focuses on describing the differences from other embodiments. In particular, the sound and shadow image detection method embodiments are basically similar to the sound and shadow image detection system embodiments, so the description is relatively simple; relevant details can be found in the descriptions of the sound and shadow image detection system embodiments.

[0088] Furthermore, it should be understood that the numbers in the flowchart of the method do not indicate the order in which these steps are executed. Some or all of these steps can be executed in parallel, or their execution order can be interchanged, and this disclosure does not impose any restrictions in this regard. Figure 2 The methods described may also include additional steps not shown and / or the steps shown may be omitted, and the scope of this disclosure is not limited in this respect.

[0089] like Figure 2 As shown, the sound and shadow image detection method based on multidimensional feature fusion and structure guidance can include at least:

[0090] Step 202: Acquire ultrasound radiofrequency data and the B-mode image obtained after logarithmic compression of the ultrasound radiofrequency data;

[0091] Step 204: Based on the ultrasound radio frequency data, obtain the local Nakagami scale parameter map and the local Nakagami shape parameter map. The local Nakagami scale parameter map includes the local Nakagami scale parameters corresponding to each location, and the local Nakagami shape parameter map includes the local Nakagami shape parameters corresponding to each location.

[0092] Step 206: Obtain the sound-shadow probability map based on the local Nakagami scale parameter map and the local Nakagami shape parameter map;

[0093] Step 208: Using the B-mode image as a guide image, perform sound and shadow edge correction on the sound and shadow probability map to obtain a sound and shadow detection image.

[0094] In some embodiments of this specification, the step of using the B-mode image as a guide image to perform sound and shadow edge correction on the sound and shadow probability map to obtain a sound and shadow detection image includes...

[0095] Step 2082: Using the B-mode image as the guide image, perform sound and shadow edge correction on the sound and shadow probability map to obtain an edge-corrected image;

[0096] Step 2084: Perform global adaptive thresholding on the edge correction image to obtain the sound and shadow detection image.

[0097] In some embodiments of this specification, the step of using the B-mode image as a guide image to perform sound and shadow edge correction on the sound and shadow probability map to obtain an edge-corrected image includes:

[0098] The B-mode image is used as the guide image, and the high-frequency edge information in the B-mode image is used to perform sound shadow edge correction on the sound shadow probability map to obtain a sound shadow detection image.

[0099] In some embodiments of this specification, obtaining the sound-shadow probability map based on the local Nakagami scale parameter map and the local Nakagami shape parameter map includes:

[0100] Step 2062: Perform logarithmic processing, normalization processing, and numerical inversion processing on the local Nakagami scale parameters corresponding to each position in the local Nakagami scale parameter map to obtain the first sound shadow screening weight corresponding to each position, and then obtain the first sound shadow screening weight map. The lower the local Nakagami scale parameter in the first sound shadow screening weight map, the larger the first sound shadow screening weight.

[0101] Step 2064: Use the Gaussian function or the Sigmoid function to process the local Nakagami shape parameters corresponding to each position in the local Nakagami shape parameter map to obtain the second sound shadow filtering weight corresponding to each position, and then obtain the second sound shadow filtering weight map. The closer the local Nakagami shape parameter in the second sound shadow filtering weight map is to the value 1, the smaller the second sound shadow filtering weight is.

[0102] Step 2066: Based on the first sound and shadow filtering weight map and the second sound and shadow filtering weight map, obtain the sound and shadow probability map.

[0103] In some embodiments of this specification, obtaining the sound-shadow probability map based on the first sound-shadow filtering weight map and the second sound-shadow filtering weight map includes:

[0104] Based on positional correspondence, the first sound and shadow filtering weight corresponding to each position in the first sound and shadow filtering weight map and the second sound and shadow filtering weight corresponding to each position in the second sound and shadow filtering weight map are multiplied together to obtain the sound and shadow probability map.

[0105] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this specification are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in or transmitted through a computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available media may be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., Digital Versatile Discs (DVDs)), or semiconductor media (e.g., Solid State Disks (SSDs)).

[0106] Figure 3 A block diagram of an electronic device 300 that can implement various embodiments of the present disclosure is shown. For example... Figure 3As shown, the electronic device 300 includes a processor 310, a disk drive 320, an input / output interface 330, a network interface 340, and a memory 350. The processor 310, disk drive 320, input / output interface 330, network interface 340, and memory 350 can communicate with each other via a communication bus 360.

[0107] The processor 310 can be implemented using a general-purpose CPU, microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits to execute relevant programs in order to implement the technical solution provided in this application.

[0108] The memory 350 can be implemented in the form of ROM (Read Only Memory), RAM (Read Access Memory), static memory, dynamic storage devices, etc. The memory 350 can store the operating system 351 used to control the operation of the electronic device 300, and the basic input / output system (BIOS) 352 used to control the low-level operations of the electronic device 300. Additionally, it can store a web browser 353, a data storage management system 354, etc. In summary, when the technical solution provided in this application is implemented through software or firmware, the relevant program code is stored in the memory 350 and is called and executed by the processor 310.

[0109] Input / output interface 330 is used to connect input / output modules to realize information input and output. Input / output modules can be configured as components in the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touch screens, microphones, various sensors, etc., and output devices may include displays, speakers, vibrators, indicator lights, etc.

[0110] Network interface 340 is used to connect a communication module (not shown in the figure) to enable communication and interaction between the device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0111] Bus 360 includes a pathway for transmitting information between various components of the device, such as processor 310, disk drive 320, input / input interface 330, network interface 340, and memory 350.

[0112] It should be noted that although the above-described device only shows the processor 310, disk drive 320, input / output interface 330, network interface 340, memory 350, bus 360, etc., in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the method of this application, and does not necessarily include all the components shown in the figures.

[0113] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0114] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing. Furthermore, although operations are depicted in a specific order, this should be understood as requiring that such operations be performed in the specific order shown or in sequential order, or requiring that all illustrated operations be performed to achieve the desired result. In certain environments, multitasking and parallel processing may be advantageous. Similarly, while several specific implementation details are included in the foregoing discussion, these should not be construed as limiting the scope of this disclosure. Certain features described in the context of individual embodiments may also be implemented in combination in a single implementation. Conversely, various features described in the context of a single implementation may also be implemented individually or in any suitable sub-combination in multiple implementations.

[0115] Although the subject matter has been described using language specific to structural features and / or methodological logic, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are merely illustrative examples of implementing the claims.

Claims

1. A sound and shadow image detection system based on multi-dimensional feature fusion and structure guidance, characterized in that, include: The first acquisition module acquires ultrasonic radio frequency data and a B-mode image obtained after logarithmic compression of the ultrasonic radio frequency data; The second acquisition module acquires a local Nakagami scale parameter map and a local Nakagami shape parameter map based on ultrasound radio frequency data. The local Nakagami scale parameter map includes the local Nakagami scale parameters corresponding to each location, and the local Nakagami shape parameter map includes the local Nakagami shape parameters corresponding to each location. The sound and shadow probability map acquisition module acquires the sound and shadow probability map based on the local Nakagami scale parameter map and the local Nakagami shape parameter map; The correction module uses the B-mode image as a guide image to perform sound and shadow edge correction on the sound and shadow probability map in order to obtain a sound and shadow detection image.

2. The acoustic image detection system based on multi-dimensional feature fusion and structure guidance according to claim 1, characterized in that, The correction module includes an edge correction unit and an adaptive segmentation unit; The edge correction unit uses the B-mode image as a guide image to perform sound and shadow edge correction on the sound and shadow probability map to obtain an edge-corrected image; The adaptive segmentation unit performs global adaptive thresholding on the edge-corrected image to obtain the sound and shadow detection image.

3. The acoustic image detection system based on multi-dimensional feature fusion and structure guidance according to claim 2, characterized in that, The edge correction unit uses the B-mode image as a guide image and utilizes the high-frequency edge information in the B-mode image to perform sound and shadow edge correction on the sound and shadow probability map to obtain a sound and shadow detection image.

4. The acoustic image detection system based on multi-dimensional feature fusion and structure guidance according to claim 1, characterized in that, The sound and shadow probability map acquisition module includes a first processing unit, a second processing unit, and a probability map acquisition unit; The first processing unit performs logarithmic processing, normalization processing, and numerical inversion processing on the local Nakagami scale parameters corresponding to each position in the local Nakagami scale parameter map in turn to obtain the first sound and shadow filtering weight corresponding to each position, and then obtains the first sound and shadow filtering weight map. The lower the local Nakagami scale parameter in the first sound and shadow filtering weight map, the larger the first sound and shadow filtering weight. The second processing unit uses a Gaussian function or a Sigmoid function to process the local Nakagami shape parameters corresponding to each position in the local Nakagami shape parameter map to obtain the second sound and shadow filtering weight corresponding to each position, and then obtains the second sound and shadow filtering weight map. The closer the local Nakagami shape parameter in the second sound and shadow filtering weight map is to the value 1, the smaller the second sound and shadow filtering weight is. The probability map acquisition unit acquires the sound and shadow probability map based on the first sound and shadow filtering weight map and the second sound and shadow filtering weight map.

5. The acoustic image detection system based on multi-dimensional feature fusion and structure guidance according to claim 4, characterized in that, The probability map acquisition unit multiplies the first sound and shadow filtering weight corresponding to each position in the first sound and shadow filtering weight map and the second sound and shadow filtering weight corresponding to each position in the second sound and shadow filtering weight map based on the position correspondence to obtain the sound and shadow probability map.

6. A method for sound and shadow image detection based on multidimensional feature fusion and structure guidance, based on the sound and shadow image detection system based on multidimensional feature fusion and structure guidance as described in any one of claims 1 to 5, characterized in that, include: Acquire ultrasound radio frequency data and B-mode images obtained after logarithmic compression of the ultrasound radio frequency data; Based on ultrasound radio frequency data, local Nakagami scale parameter map and local Nakagami shape parameter map are obtained. The local Nakagami scale parameter map includes the local Nakagami scale parameter corresponding to each location, and the local Nakagami shape parameter map includes the local Nakagami shape parameter corresponding to each location. Based on the local Nakagami scale parameter map and the local Nakagami shape parameter map, the sound-shadow probability map is obtained; Using the B-mode image as a guide image, sound and shadow edge correction is performed on the sound and shadow probability map to obtain a sound and shadow detection image.

7. The acoustic-shadow image detection method based on multidimensional feature fusion and structure guidance according to claim 6, characterized in that, The process involves using the B-mode image as a guide image to perform sound and shadow edge correction on the sound and shadow probability map to obtain a sound and shadow detection image, including... Using the B-mode image as a guide image, sound and shadow edge correction is performed on the sound and shadow probability map to obtain an edge-corrected image; Global adaptive thresholding is applied to the edge-corrected image to obtain the sound and shadow detection image.

8. The acoustic-shadow image detection method based on multi-dimensional feature fusion and structure guidance according to claim 6, characterized in that, The step of using the B-mode image as a guide image to perform sound and shadow edge correction on the sound and shadow probability map to obtain an edge-corrected image includes: The B-mode image is used as the guide image, and the high-frequency edge information in the B-mode image is used to perform sound shadow edge correction on the sound shadow probability map to obtain a sound shadow detection image.

9. The acoustic-shadow image detection method based on multi-dimensional feature fusion and structure guidance according to claim 1, characterized in that, The process of obtaining the sound-shadow probability map based on the local Nakagami scale parameter map and the local Nakagami shape parameter map includes: Logarithmic processing, normalization processing, and numerical inversion processing are performed on the local Nakagami scale parameters corresponding to each position in the local Nakagami scale parameter map to obtain the first sound shadow screening weight corresponding to each position, thereby obtaining the first sound shadow screening weight map. The lower the local Nakagami scale parameter in the first sound shadow screening weight map, the larger the first sound shadow screening weight. The local Nakagami shape parameters corresponding to each position in the local Nakagami shape parameter map are processed using the Gaussian function or the Sigmoid function to obtain the second sound shadow filtering weight corresponding to each position, thereby obtaining the second sound shadow filtering weight map. The closer the local Nakagami shape parameter in the second sound shadow filtering weight map is to the value 1, the smaller the second sound shadow filtering weight is. Based on the first sound and shadow filtering weight map and the second sound and shadow filtering weight map, a sound and shadow probability map is obtained.

10. The acoustic-shadow image detection method based on multi-dimensional feature fusion and structure guidance according to claim 9, characterized in that, The process of obtaining the sound and shadow probability map based on the first sound and shadow filtering weight map and the second sound and shadow filtering weight map includes: Based on positional correspondence, the first sound and shadow filtering weight corresponding to each position in the first sound and shadow filtering weight map and the second sound and shadow filtering weight corresponding to each position in the second sound and shadow filtering weight map are multiplied together to obtain the sound and shadow probability map.

Images