An image enhancement processing method and device, electronic equipment and storage medium

Abstract

The application discloses an image enhancement processing method and device, electronic equipment and storage medium. The method comprises the following steps: determining a first image, the first image is obtained by superimposing a first number of second images, the second image is an X-ray image collected by a first object; determining a first time-dependent vector field based on the first image, the first time-dependent vector field is a time-dependent vector field guided by the first image, the time-dependent vector field is used for indicating the vector field corresponding to the third image and at least one fourth image respectively, the third image belongs to a noise image, and the fourth image is an intermediate state image in a transformation process based on the third image; determining a fifth image based on the first time-dependent vector field and the third image. According to the scheme, the third image is gradually iteratively transformed into the image after the first image is enhanced by the first time-dependent vector field, and the efficiency and accuracy of the enhancement processing of the first image are effectively improved.

Description

Technical Field

[0001] This invention relates to the field of image processing technology, and in particular to an image enhancement processing method, apparatus, electronic device, and storage medium. Background Technology

[0002] X-ray imaging technology is a core technology for industrial non-destructive testing and quality control. In practical industrial applications, to meet the high throughput and real-time inspection requirements of production lines, extend the service life of core components such as X-ray tubes, reduce equipment radiation protection costs, and minimize damage to radiation-sensitive components, rapid scanning or reduced X-ray exposure dose is typically required. Rapid scanning or low-dose imaging can lead to quantum noise, streak artifacts, and decreased resolution in X-ray images, masking the characteristics of minute defects, thus necessitating X-ray image enhancement. However, current X-ray image enhancement technologies suffer from poor accuracy and low efficiency, impacting the accuracy of industrial non-destructive testing and quality control. Summary of the Invention

[0003] This invention provides an image enhancement processing method, apparatus, electronic device, and storage medium to solve the problems of poor accuracy and low efficiency in enhancing X-ray images.

[0004] According to one aspect of the present invention, an image enhancement processing method is provided, the method comprising: A first image is determined, which is obtained by superimposing a first number of second images, wherein the second image is an X-ray image acquired from the first object; A first time-dependent vector field is determined based on the first image. The first time-dependent vector field is a time-dependent vector field guided by the first image. The time-dependent vector field is used to indicate the vector fields corresponding to the third image and at least one fourth image respectively. The third image is a noisy image. The fourth image is an intermediate image in the transformation process based on the third image. The third image is transformed based on the vector field corresponding to the third image. The fourth image is transformed based on the vector field corresponding to the fourth image. The fifth image is determined based on the first time-dependent vector field and the third image. The fifth image is an image obtained by enhancing the first image and is transformed based on the third image.

[0005] According to another aspect of the present invention, an image enhancement processing apparatus is provided, the apparatus comprising: The first determining module is used to determine the first image, which is obtained by superimposing a first number of second images, and the second image is an X-ray image acquired from the first object; The second determining module is used to determine a first time-dependent vector field based on the first image. The first time-dependent vector field is a time-dependent vector field guided by the first image. The time-dependent vector field is used to indicate the vector fields corresponding to the third image and at least one fourth image respectively. The third image is a noisy image. The fourth image is an intermediate image in the transformation process based on the third image. The third image is transformed based on the vector field corresponding to the third image. The fourth image is transformed based on the vector field corresponding to the fourth image. The third determining module is used to determine the fifth image based on the first time-dependent vector field and the third image. The fifth image is an image obtained by enhancing the first image and is obtained by transforming the third image.

[0006] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising: At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the image enhancement processing method of any embodiment of the present invention.

[0007] According to another aspect of the present invention, a computer-readable storage medium is provided that stores computer instructions for causing a processor to execute and implement the image enhancement processing method of any embodiment of the present invention.

[0008] The technical solution of this invention involves determining a first image, which is obtained by superimposing a first number of second images. The second images are X-ray images acquired from a first object. The resolution of the first image is less than a preset resolution threshold, or the first image contains noise or artifacts. Based on the first image, a first time-dependent vector field is determined. This first time-dependent vector field is a time-dependent vector field guided by the first image. The time-dependent vector field is used to indicate the vector fields corresponding to a third image and at least one fourth image, respectively. The third image is a noisy image, and the fourth image is an intermediate state image in the transformation process based on the third image. This achieves the description of the instantaneous change direction and instantaneous change rate of each point in the image through the vector field corresponding to the image, thereby enabling the first... The third image can be advanced along the vector field corresponding to the third image by one time step to obtain the fourth image. The fourth image can be advanced along the vector field corresponding to the fourth image by one time step to obtain a new fourth image. By iteratively advancing in this way, images that meet the expected conditions can be obtained. The fifth image is determined based on the first time-dependent vector field and the third image. The fifth image is the image after enhancing the first image. The fifth image is obtained by transforming the third image. This realizes the step-by-step transformation of the third image into the image after enhancing the first image by controlling the first time-dependent vector field, which improves the resolution of the first image and removes the noise and artifacts present in the first image, thereby effectively improving the efficiency and accuracy of enhancing the first image.

[0009] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0010] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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.

[0011] Figure 1 A flowchart of an image enhancement processing method provided in an embodiment of the present invention; Figure 2 A flowchart of another image enhancement processing method provided in an embodiment of the present invention; Figure 3 This is a schematic diagram illustrating the training of a second neural network according to an embodiment of the present invention; Figure 4 A schematic diagram of a fifth image determination process provided in an embodiment of the present invention; Figure 5 This is a schematic diagram of the structure of an image enhancement processing device provided in an embodiment of the present invention; Figure 6 This is a schematic diagram of the structure of an electronic device that implements an image enhancement processing method according to an embodiment of the present invention. Detailed Implementation

[0012] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0013] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0014] Figure 1 This is a flowchart illustrating an image enhancement processing method provided in an embodiment of the present invention. This embodiment is applicable to enhancing X-ray images. The method can be executed by an image enhancement processing device, which can be implemented in hardware and / or software and configured in an electronic device implementing the image enhancement processing method. Figure 1 As shown, the image enhancement processing method includes: S101. Determine the first image, which is obtained by superimposing a first number of second images. The second image is an X-ray image acquired from the first object.

[0015] Specifically, an X-ray image can refer to an image formed based on the intensity of X-rays after they penetrate an object. A first number of second images can be obtained by acquiring multiple images of the first object using a device for acquiring X-ray images. Furthermore, a first image is obtained by superimposing the first number of second images, and the resolution of the first image is less than a preset resolution threshold, or the first image contains noise or artifacts. For example, the first number can be 4 or 8.

[0016] S102. Determine a first time-dependent vector field based on the first image. The first time-dependent vector field is a time-dependent vector field guided by the first image. The time-dependent vector field is used to indicate the vector fields corresponding to the third image and at least one fourth image respectively. The third image is a noisy image. The fourth image is an intermediate image in the transformation process based on the third image. The third image is transformed based on the vector field corresponding to the third image. The fourth image is transformed based on the vector field corresponding to the fourth image.

[0017] Here, the time-dependent vector field refers to a vector field that changes with the transition time and state of the third image. The transition state of the third image can be reflected by the fourth image. The vector field corresponding to the third image can be used to describe the instantaneous change direction and instantaneous change rate of each point in the third image. The vector field corresponding to the fourth image can be used to describe the instantaneous change direction and instantaneous change rate of each point in the fourth image. Advancing the third image along the vector field corresponding to the third image by one time step yields the fourth image. Advancing the fourth image along the vector field corresponding to the fourth image by one time step yields a new fourth image, and this process is iteratively advanced until an image that meets the expected conditions is obtained. Specifically, by determining the time-dependent vector field based on the first image, using the first image as a guiding condition, the first time-dependent vector field can be obtained.

[0018] For example, a time-dependent vector field can be adopted This is represented. Where, when... hour, This represents the third image. Output the vector field corresponding to the third image; when hour, This represents the fourth image. Output the vector field corresponding to the fourth image. The first-time dependent vector field can be obtained using... To represent, where The first image represents the first time-dependent vector field, which can determine the vector field corresponding to the third image based on the third image, time, and the first image, and determine the vector field corresponding to the fourth image based on the fourth image, time, and the first image.

[0019] As an optional embodiment of the present invention, the pixel values ​​in the third image follow a Gaussian distribution with a mean of 0 and a standard deviation of 1. The third image... This allows the pixel values ​​in the third image to have a mean of 0 and a standard deviation of 1, thus facilitating a stable and continuous transition, improving convergence speed, and making it applicable to different X-ray images.

[0020] S103. Determine the fifth image based on the first time-dependent vector field and the third image. The fifth image is the image after enhancing the first image. The fifth image is obtained by transforming the third image.

[0021] Specifically, enhancing the first image can refer to increasing its resolution, removing noise, or removing artifacts. By controlling the third image through a first time-dependent vector field to iteratively transform it, a fifth image can be obtained, effectively improving the efficiency of enhancing the first image.

[0022] As an optional implementation of the present invention, determining a fifth image based on a first time-dependent vector field and a third image includes: establishing a first ordinary differential equation based on the first time-dependent vector field, wherein the first derivative in the first ordinary differential equation is equal to the first time-dependent vector field, and the first derivative is the first derivative of the transition state of the third image with respect to time; and determining a fifth image based on the third image and the first ordinary differential equation.

[0023] Specifically, the first time-dependent vector field represents a vector field that changes with the transition time and transition state of the third image under the guidance of the first image. Therefore, based on the first derivative of the transition state of the third image with respect to time and the first time-dependent vector field, a first ordinary differential equation can be constructed. Furthermore, solving for the transition state of the third image based on the third image and the first ordinary differential equation yields the fifth image.

[0024] For example, the first ordinary differential equation can be represented as follows: ; ; in, Indicates time; Representing the transition state of the third image, by solving... time This will give you the fifth image.

[0025] As an optional implementation of the present invention, determining the fifth image based on the third image and the first ordinary differential equation includes: solving the first ordinary differential equation based on the third image using the first-order Euler method or the second-order improved Euler method to determine the fifth image.

[0026] Specifically, the fifth image is determined by solving the first ordinary differential equation using the first-order Euler method based on the third image, including the following steps A1-A5: Step A1: Determine the first time step of the second quantity. The sum of the first time steps of the second quantity is 1.

[0027] Step A2: Use the third image as the reference image.

[0028] Step A3: Determine the vector field corresponding to the reference image based on the first image and the reference image through the first time-dependent vector field.

[0029] Step A4: The product of the vector field corresponding to the reference image and the first time step is taken as the change amount corresponding to the reference image, and the sum of the changes of the reference image and the reference image is taken as the target image.

[0030] Step A5: Use the target image as the new reference image, and re-execute the process of determining the vector field corresponding to the reference image based on the first image and the reference image through the first time-dependent vector field until the number of iterations equals the second number, and then use the target image as the fifth image.

[0031] For example, the solution to the first ordinary differential equation using the first-order Euler method can be represented as follows: ; in, Indicates the first time step; Indicates a reference image; This represents the target image.

[0032] The fifth image is determined by solving the first ordinary differential equation using the second-order improved Euler method based on the third image, including the following steps B1-B8: Step B1: Determine the first time step of the second quantity. The sum of the first time steps of the second quantity is 1.

[0033] Step B2: Use the third image as the reference image.

[0034] Step B3: Determine the vector field corresponding to the reference image based on the first image and the reference image through the first time-dependent vector field.

[0035] Step B4: The product of the vector field corresponding to the reference image and the first time step is taken as the change amount corresponding to the reference image, and the sum of the changes of the reference image and the reference image is taken as the candidate image.

[0036] Step B5: Determine the vector field corresponding to the candidate image based on the first image and the candidate image through the first time-dependent vector field.

[0037] Step B6: Take the average value of the vector field corresponding to the reference image and the vector field corresponding to the candidate image as the average vector field.

[0038] Step B7: The product of the average vector field and the first time step is taken as the average change, and the sum of the reference image and the average change is taken as the target image.

[0039] Step B8: Use the target image as the new reference image, and re-execute the process of determining the vector field corresponding to the reference image based on the first image and the reference image through the first time-dependent vector field until the number of iterations equals the second number, and then use the target image as the fifth image.

[0040] For example, the solution to the first ordinary differential equation using the second-order improved Euler method can be represented as follows: ; ; in, This represents a candidate image.

[0041] The technical solution of this invention involves determining a first image, which is obtained by superimposing a first number of second images. The second images are X-ray images acquired from a first object. The resolution of the first image is less than a preset resolution threshold, or the first image contains noise or artifacts. Based on the first image, a first time-dependent vector field is determined. This first time-dependent vector field is a time-dependent vector field guided by the first image. The time-dependent vector field is used to indicate the vector fields corresponding to a third image and at least one fourth image, respectively. The third image is a noisy image, and the fourth image is an intermediate state image in the transformation process based on the third image. This achieves the description of the instantaneous change direction and instantaneous change rate of each point in the image through the vector field corresponding to the image, thereby enabling the first... The third image can be advanced along the vector field corresponding to the third image by one time step to obtain the fourth image. The fourth image can be advanced along the vector field corresponding to the fourth image by one time step to obtain a new fourth image. By iteratively advancing in this way, images that meet the expected conditions can be obtained. The fifth image is determined based on the first time-dependent vector field and the third image. The fifth image is the image after enhancing the first image. The fifth image is obtained by transforming the third image. This realizes the step-by-step transformation of the third image into the image after enhancing the first image by controlling the first time-dependent vector field, which improves the resolution of the first image and removes the noise and artifacts present in the first image, thereby effectively improving the efficiency and accuracy of enhancing the first image.

[0042] Figure 2This is a flowchart of another image enhancement processing method provided by an embodiment of the present invention. The technical solution of this embodiment further optimizes the process of determining a first time-dependent vector field based on a first image in the aforementioned embodiments, based on the technical solutions of the above embodiments. Solutions not described in detail in this embodiment can be found in the above embodiments. This embodiment can be combined with various optional solutions in one or more of the above embodiments. Figure 2 As shown, the image enhancement processing method includes: S201. Determine the first image, which is obtained by superimposing a first number of second images. The second image is an X-ray image acquired from the first object.

[0043] S202. Determine the first prediction model. The first prediction model is a conditional flow matching generation model using a first neural network. The first neural network is pre-trained based on X-ray images and a time-dependent vector field guided by X-ray images.

[0044] Specifically, the Conditional Flow Matching (CFM) generative model can fit a deterministic vector field to a neural network, thus transforming the generation task into a regression learning process of the neural network parameterized vector field. A time-dependent vector field guided by X-ray images can be used to control the transformation of noisy images into enhanced X-ray images. Based on the X-ray images and the time-dependent vector field guided by the X-ray images, fitting and training an initial neural network generates a first neural network. Then, building a conditional flow matching generative model based on this first neural network yields a first prediction model.

[0045] As an optional implementation of this invention, the training process of the first neural network includes: determining a first training dataset and a first path, wherein the first training dataset includes a sixth image and a seventh image, the sixth image is obtained by superimposing a first number of eighth images, the eighth image is an X-ray image acquired from a second object, the seventh image is an image after enhancing the sixth image, and the first path is a conditional probability path that transforms a noisy image into the seventh image; and training the second neural network based on the first training dataset and the first path to obtain the first neural network.

[0046] The first number of eighth images can be obtained by acquiring multiple images of the second object using an X-ray image acquisition device. The resolution of the sixth image is less than a preset resolution threshold, or noise or artifacts exist in the first image. The seventh image can be obtained by enhancing the sixth image, or by superimposing the third number of eighth images. The resolution of the seventh image is not less than a preset resolution threshold, and there is no noise or artifacts in the seventh image. The first path can be a linear path, for example, the first path can be represented as: ,in, Represents a noisy image. , This represents the seventh image. Based on the first path and time, the intermediate state image in the process of transforming the noisy image into the seventh image can be determined. The second neural network can refer to the initial neural network that has not been trained.

[0047] For details, please refer to Figure 3 The first neural network can be trained using the first path, time, and the sixth image as inputs, and the difference between the seventh image and the noisy image as the training objective. For example, the second neural network can employ a U-Net architecture, where the intermediate image determined by the first path is concatenated with the sixth image and then input into the second neural network.

[0048] As an optional implementation of this invention, the first neural network is obtained by training the second neural network based on the first loss function. The first loss function is determined based on the output result of the second neural network during the training process and the second vector field. The second vector field is determined based on the seventh image and the noisy image.

[0049] Among them, reference Figure 3 The second vector field can be the difference between the seventh image and the noisy image. The second neural network is trained based on the first loss function so that the deviation between the output of the second neural network and the second vector field does not exceed a preset deviation threshold.

[0050] For example, the first loss function can be represented as follows: ; in, Represents the first loss function; This represents the output of the second neural network based on the first path, time, and the sixth image; This represents the sixth image; This represents the second vector field.

[0051] S203. Determine the first time-dependent vector field based on the first image using the first prediction model.

[0052] Specifically, by inputting the first image into the first prediction model, the first neural network in the first prediction model can determine the time-dependent vector field guided by the first image, thereby obtaining the first time-dependent vector field and improving the accuracy and efficiency of determining the first time-dependent vector field.

[0053] S204. Determine the fifth image based on the first time-dependent vector field and the third image. The fifth image is the image after enhancing the first image. The fifth image is obtained by transforming the third image.

[0054] refer to Figure 4 The first neural network uses a linear path during training, allowing the fifth image to be solved using a one-step, first-order Euler method based on the third image and the first time-dependent vector field. .

[0055] The technical solution of this invention involves determining a first image, which is obtained by superimposing a first number of second images, where the second images are X-ray images acquired from a first object; determining a first prediction model, which is a conditional flow matching generation model using a first neural network, which is pre-trained based on the X-ray image and a time-dependent vector field guided by the X-ray image; determining a first time-dependent vector field based on the first image and the first prediction model; the time-dependent vector field guided by the X-ray image can be used to control the transformation of a noisy image into an enhanced image of the X-ray image, thereby training and generating the first neural network based on the X-ray image and the time-dependent vector field guided by the X-ray image, which can effectively improve the accuracy and efficiency of determining the first time-dependent vector field; and determining a fifth image based on the first time-dependent vector field and a third image, where the fifth image is an enhanced image of the first image, obtained by transforming the third image, thus realizing the step-by-step iterative transformation of the third image into an enhanced image of the first image by controlling the third image through the first time-dependent vector field.

[0056] Figure 5 This is a schematic diagram of an image enhancement processing apparatus provided in an embodiment of the present invention. The embodiments of the present invention are applicable to the enhancement processing of X-ray images, and the image enhancement processing apparatus can be implemented in hardware and / or software. Figure 5 As shown, the image enhancement processing apparatus includes: The first determining module 301 is used to determine a first image, which is obtained by superimposing a first number of second images, and the second image is an X-ray image acquired from the first object; The second determining module 302 is used to determine a first time-dependent vector field based on the first image. The first time-dependent vector field is a time-dependent vector field guided by the first image. The time-dependent vector field is used to indicate the vector fields corresponding to the third image and at least one fourth image respectively. The third image is a noisy image. The fourth image is an intermediate image in the transformation process based on the third image. The third image is transformed based on the vector field corresponding to the third image. The fourth image is transformed based on the vector field corresponding to the fourth image. The third determining module 303 is used to determine the fifth image based on the first time-dependent vector field and the third image. The fifth image is an image obtained by enhancing the first image and is obtained by transforming the third image.

[0057] Based on any of the above optional technical solutions, optionally, the second determining module 302 includes: a fourth determining unit and a fifth determining unit. The fourth determining unit is used to determine a first prediction model, which is a conditional flow matching generation model using a first neural network, the first neural network being pre-trained based on an X-ray image and a time-dependent vector field guided by the X-ray image; the fifth determining unit is used to determine a first time-dependent vector field based on the first image and the first prediction model.

[0058] Based on any of the above optional technical solutions, optionally, the training process of the first neural network includes: determining a first training dataset and a first path, wherein the first training dataset includes a sixth image and a seventh image, the sixth image is obtained by superimposing a first number of eighth images, the eighth image is an X-ray image acquired from a second object, the seventh image is an image after enhancing the sixth image, and the first path is a conditional probability path that transforms a noisy image into the seventh image; and training the second neural network based on the first training dataset and the first path to obtain the first neural network.

[0059] Based on any of the above optional technical solutions, optionally, the first neural network is obtained by training the second neural network based on the first loss function, the first loss function is determined based on the output result of the second neural network during the training process and the second vector field, and the second vector field is determined based on the seventh image and the noisy image.

[0060] Based on any of the above optional technical solutions, optionally, the third determining module 303 includes: a first establishing unit and a sixth determining unit. The first establishing unit is used to establish a first ordinary differential equation based on a first time-dependent vector field, where the first derivative of the first ordinary differential equation is equal to the first time-dependent vector field, and the first derivative is the first derivative of the transition state of the third image with respect to time; the sixth determining unit is used to determine a fifth image based on the third image and the first ordinary differential equation.

[0061] Based on any of the above optional technical solutions, optionally, the sixth determining unit is specifically used to solve the first ordinary differential equation based on the third image using the first-order Euler method or the second-order improved Euler method to determine the fifth image.

[0062] Based on any of the above optional technical solutions, optionally, the pixel values ​​in the third image follow a Gaussian distribution with a mean of 0 and a standard deviation of 1.

[0063] The technical solution of this invention involves a first determining module 301 determining a first image, which is obtained by superimposing a first number of second images. The second images are X-ray images acquired from a first object. The resolution of the first image is less than a preset resolution threshold, or there is noise or artifacts in the first image. A second determining module 302 determines a first time-dependent vector field based on the first image. The first time-dependent vector field is a time-dependent vector field guided by the first image. This time-dependent vector field indicates the vector fields corresponding to a third image and at least one fourth image. The third image is a noisy image, and the fourth image is an intermediate state image during the transformation process based on the third image. This achieves the description of the instantaneous change direction and instantaneous change rate of each point in the image through the vector field corresponding to the image. The process involves iteratively advancing the third image along the vector field corresponding to the third image to obtain the fourth image, and then iteratively advancing the fourth image along the vector field corresponding to the fourth image to obtain a new fourth image. This iterative advancement yields images that meet the expected conditions. The third determining module 303 determines the fifth image based on the first time-dependent vector field and the third image. The fifth image is an enhanced version of the first image, obtained by transforming the third image. This process controls the third image to iteratively transform into an enhanced version of the first image through the first time-dependent vector field, improving the resolution of the first image and removing noise and artifacts. This effectively improves the efficiency and accuracy of enhancing the first image.

[0064] The image enhancement processing apparatus provided in the embodiments of the present invention can execute the image enhancement processing method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of executing the method.

[0065] Figure 6 This is a schematic diagram of an electronic device implementing an image enhancement processing method according to an embodiment of the present invention. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0066] like Figure 6 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded into the RAM 13 from storage unit 18. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0067] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0068] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as image enhancement processing methods.

[0069] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication unit 19, or installed from storage unit 18, or installed from ROM 12. When the computer program is executed by processor 11, it performs the functions defined in the methods of the embodiments of the present invention.

[0070] In some embodiments, the image enhancement processing method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the image enhancement processing method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the image enhancement processing method by any other suitable means (e.g., by means of firmware).

[0071] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transferring data and instructions to the storage system, the at least one input device, and the at least one output device.

[0072] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0073] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. 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 fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0074] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0075] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0076] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0077] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0078] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. An image enhancement processing method, characterized in that, The method includes: A first image is determined, which is obtained by superimposing a first number of second images, wherein the second image is an X-ray image acquired from a first object; A first time-dependent vector field is determined based on the first image. The first time-dependent vector field is a time-dependent vector field guided by the first image. The time-dependent vector field is used to indicate the vector fields corresponding to the third image and at least one fourth image respectively. The third image is a noisy image. The fourth image is an intermediate image in the transformation process based on the third image. The third image is transformed based on the vector field corresponding to the third image. The fourth image is transformed based on the vector field corresponding to the fourth image. A fifth image is determined based on the first time-dependent vector field and the third image. The fifth image is an image obtained by enhancing the first image and is derived from the transformation of the third image.

2. The method according to claim 1, characterized in that, Determining a first time-dependent vector field based on the first image includes: A first prediction model is determined, which is a conditional flow matching generation model using a first neural network. The first neural network is pre-trained based on X-ray images and a time-dependent vector field guided by X-ray images. The first time-dependent vector field is determined based on the first image using the first prediction model.

3. The method according to claim 2, characterized in that, The training process of the first neural network includes: A first training dataset and a first path are determined. The first training dataset includes a sixth image and a seventh image. The sixth image is obtained by superimposing a first number of eighth images. The eighth image is an X-ray image acquired from a second object. The seventh image is an image after enhancing the sixth image. The first path is a conditional probability path that transforms a noisy image into the seventh image. The first neural network is obtained by training the second neural network based on the first training dataset and the first path.

4. The method according to claim 3, characterized in that, The first neural network is obtained by training the second neural network based on the first loss function. The first loss function is determined based on the output result of the second neural network during the training process and the second vector field. The second vector field is determined based on the seventh image and the noisy image.

5. The method according to claim 1, characterized in that, Determining the fifth image based on the first time-dependent vector field and the third image includes: A first ordinary differential equation is established based on the first time-dependent vector field. In the first ordinary differential equation, the first derivative is equal to the first time-dependent vector field. The first derivative is the first derivative of the transition state of the third image with respect to time. The fifth image is determined based on the third image and the first ordinary differential equation.

6. The method according to claim 5, characterized in that, Determining the fifth image based on the third image and the first ordinary differential equation includes: The fifth image is determined by solving the first ordinary differential equation based on the third image using either the first-order Euler method or the second-order improved Euler method.

7. The method according to claim 1, characterized in that, The pixel values ​​in the third image follow a Gaussian distribution with a mean of 0 and a standard deviation of 1.

8. An image enhancement processing apparatus, characterized in that, The device includes: A first determining module is used to determine a first image, which is obtained by superimposing a first number of second images, wherein the second image is an X-ray image acquired from a first object; The second determining module is used to determine a first time-dependent vector field based on the first image. The first time-dependent vector field is a time-dependent vector field guided by the first image. The time-dependent vector field is used to indicate the vector fields corresponding to the third image and at least one fourth image, respectively. The third image is a noisy image. The fourth image is an intermediate image in the transformation process based on the third image. The third image is transformed based on the vector field corresponding to the third image, and the fourth image is transformed based on the vector field corresponding to the fourth image. The third determining module is used to determine a fifth image based on the first time-dependent vector field and the third image, wherein the fifth image is an image after enhancing the first image, and the fifth image is obtained by transforming the third image.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the image enhancement processing method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the image enhancement processing method according to any one of claims 1-7.

Images