An image processing method, apparatus, device and storage medium

Abstract

Embodiments of the present application disclose an image processing method, device and equipment, and a storage medium. The method comprises: acquiring a to-be-processed intravascular ultrasound two-dimensional image of an imaging object and a reference intravascular ultrasound two-dimensional image corresponding to the to-be-processed intravascular ultrasound two-dimensional image; performing affine transformation on the to-be-processed intravascular ultrasound two-dimensional image based on initial affine transformation parameters to obtain a transformed intravascular ultrasound two-dimensional image; determining an image distance between the transformed intravascular ultrasound two-dimensional image and the reference intravascular ultrasound two-dimensional image, and determining target affine transformation parameters based on the image distance; processing the to-be-processed intravascular ultrasound two-dimensional image based on the target affine transformation parameters to obtain a target intravascular ultrasound two-dimensional image corresponding to the to-be-processed intravascular ultrasound two-dimensional image; and performing three-dimensional reconstruction according to the target intravascular ultrasound two-dimensional images corresponding to the intravascular ultrasound two-dimensional images to obtain an intravascular ultrasound three-dimensional image of the imaging object. The imaging effect of the intravascular three-dimensional image is improved.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the field of image processing, and particularly relate to an image processing method, device, equipment and storage medium. BACKGROUND

[0002] In the process of using a catheter to perform ultrasound imaging in a blood vessel, since the catheter is in the blood vessel lumen of a living body, the blood vessel is deformed in various forms due to the beating of the heart, which will cause the displacement of the catheter relative to the blood vessel, and thus the ultrasound images collected in the conventional case will appear periodic motion artifacts between frames, and the three-dimensional body sections generated according to these images will often appear a "sawtooth" pattern, which is quite different from the relatively smooth blood vessel in reality, and the imaging effect is poor. It can be seen that how to improve the effect of the intravascular three-dimensional image is a technical problem to be solved. SUMMARY

[0003] Embodiments of the present application provide an image processing method, device, equipment and storage medium to improve the imaging effect of the intravascular three-dimensional image.

[0004] In a first aspect, embodiments of the present application provide an image processing method, comprising:

[0005] obtaining intravascular ultrasound two-dimensional images of an imaging object, determining a to-be-processed intravascular ultrasound two-dimensional image and a reference intravascular ultrasound two-dimensional image corresponding to the to-be-processed intravascular ultrasound two-dimensional image in the intravascular ultrasound two-dimensional images;

[0006] affine transforming the to-be-processed intravascular ultrasound two-dimensional image based on the initial affine transformation parameter to obtain a transformed intravascular ultrasound two-dimensional image;

[0007] determining an image distance between the transformed intravascular ultrasound two-dimensional image and the reference intravascular ultrasound two-dimensional image, and determining a target affine transformation parameter based on the image distance;

[0008] processing the to-be-processed intravascular ultrasound two-dimensional image based on the target affine transformation parameter to obtain a target intravascular ultrasound two-dimensional image corresponding to the to-be-processed intravascular ultrasound two-dimensional image;

[0009] performing three-dimensional reconstruction according to the target intravascular ultrasound two-dimensional images corresponding to the intravascular ultrasound two-dimensional images to obtain an intravascular ultrasound three-dimensional image of the imaging object.

[0010] Optionally, the determining of the image distance between the transformed intravascular ultrasound two-dimensional image and the reference intravascular ultrasound two-dimensional image comprises:

[0011] performing pixel reduction processing on the transformed intravascular ultrasound two-dimensional image to obtain a distance-transformed intravascular ultrasound two-dimensional image;

[0012] perform pixel reduction processing on the reference intravascular ultrasound two-dimensional image to obtain a distance reference intravascular ultrasound two-dimensional image;

[0013] determine a distance between the distance transformed intravascular ultrasound two-dimensional image and the distance reference intravascular ultrasound two-dimensional image as an image distance.

[0014] Optionally, the determining of the distance between the distance transformed intravascular ultrasound two-dimensional image and the distance reference intravascular ultrasound two-dimensional image as the image distance comprises:

[0015] determining a Wasserstein distance between the distance transformed intravascular ultrasound two-dimensional image and the distance reference intravascular ultrasound two-dimensional image as the image distance.

[0016] Optionally, the determining of the target affine transformation parameter based on the image distance comprises:

[0017] determining a back propagation gradient of the image distance, and adjusting a parameter value of the affine transformation parameter based on the back propagation gradient;

[0018] performing affine transformation on the to-be-processed intravascular ultrasound two-dimensional image based on the adjusted affine transformation parameter to obtain a new transformed intravascular ultrasound two-dimensional image;

[0019] determining a new image distance between the new transformed intravascular ultrasound two-dimensional image and the reference intravascular ultrasound two-dimensional image, and determining a new back propagation gradient corresponding to the new image distance;

[0020] iteratively performing the above operations, and when the new back propagation gradient meets an iteration stop condition, associating the affine transformation parameter corresponding to the new back propagation gradient that meets the iteration stop condition as the target affine transformation parameter.

[0021] Optionally, the affine transformation parameter comprises a first direction transformation parameter, a second direction transformation parameter and an angle transformation parameter, and the determining of the back propagation gradient of the image distance comprises:

[0022] respectively determining a first direction propagation gradient of the image distance to the first direction transformation parameter, a second direction propagation gradient of the image distance to the second direction transformation parameter, and an angle propagation gradient of the image distance to the angle transformation parameter;

[0023] determining the first direction propagation gradient, the second direction propagation gradient and the angle propagation gradient as the back propagation gradient of the image distance.

[0024] Optionally, the meeting of the iteration stop condition comprises that an iteration number reaches a set number threshold, and / or the back propagation gradient is not greater than a set gradient threshold.

[0025] Optionally, the initial affine transformation parameter is used to perform affine transformation on the to-be-processed intravascular ultrasound two-dimensional image to obtain a transformed intravascular ultrasound two-dimensional image, including:

[0026] The initial affine transformation parameter is used to perform affine transformation on the to-be-processed intravascular ultrasound two-dimensional image to obtain an affine intravascular ultrasound two-dimensional image;

[0027] The affine intravascular ultrasound two-dimensional image is subjected to interpolation processing to obtain the transformed intravascular ultrasound two-dimensional image.

[0028] In a second aspect, an embodiment of the present application further provides an image processing device, including:

[0029] A two-dimensional image acquisition module is configured to acquire intravascular ultrasound two-dimensional images of an imaging object, determine a to-be-processed intravascular ultrasound two-dimensional image and a reference intravascular ultrasound two-dimensional image corresponding to the to-be-processed intravascular ultrasound two-dimensional image;

[0030] A two-dimensional image transformation module is configured to perform affine transformation on the to-be-processed intravascular ultrasound two-dimensional image based on the initial affine transformation parameter to obtain a transformed intravascular ultrasound two-dimensional image;

[0031] A target parameter determination module is configured to determine an image distance between the transformed intravascular ultrasound two-dimensional image and the reference intravascular ultrasound two-dimensional image, and determine a target affine transformation parameter based on the image distance;

[0032] A target two-dimensional image module is configured to perform processing on the to-be-processed intravascular ultrasound two-dimensional image based on the target affine transformation parameter to obtain a target intravascular ultrasound two-dimensional image corresponding to the to-be-processed intravascular ultrasound two-dimensional image;

[0033] An intravascular three-dimensional image module is configured to perform three-dimensional reconstruction based on the target intravascular ultrasound two-dimensional image corresponding to each intravascular ultrasound two-dimensional image to obtain an intravascular ultrasound three-dimensional image of the imaging object.

[0034] In a third aspect, an embodiment of the present application further provides a computer device, including:

[0035] One or more processors;

[0036] A storage device configured to store one or more programs;

[0037] When the one or more programs are executed by the one or more processors, the one or more processors implement the image processing method provided by any embodiment of the present application.

[0038] In a fourth aspect, an embodiment of the present application further provides a computer readable storage medium having a computer program stored thereon, and the program is executed by a processor to implement the image processing method provided by any embodiment of the present application.

[0039] The image processing method provided by the embodiment of the present application comprises the following steps: acquiring intravascular ultrasound two-dimensional images of an imaging object; determining a to-be-processed intravascular ultrasound two-dimensional image and a reference intravascular ultrasound two-dimensional image corresponding to the to-be-processed intravascular ultrasound two-dimensional image; performing affine transformation on the to-be-processed intravascular ultrasound two-dimensional image based on initial affine transformation parameters to obtain a transformed intravascular ultrasound two-dimensional image; determining an image distance between the transformed intravascular ultrasound two-dimensional image and the reference intravascular ultrasound two-dimensional image, and determining target affine transformation parameters based on the image distance; processing the to-be-processed intravascular ultrasound two-dimensional image based on the target affine transformation parameters to obtain a target intravascular ultrasound two-dimensional image corresponding to the to-be-processed intravascular ultrasound two-dimensional image; and performing three-dimensional reconstruction on the target intravascular ultrasound two-dimensional images corresponding to the intravascular ultrasound two-dimensional images to obtain an intravascular ultrasound three-dimensional image of the imaging object. The imaging effect of the intravascular three-dimensional image is improved. Each two-dimensional image is processed, and three-dimensional reconstruction is performed based on the processed two-dimensional images, so that the imaging effect of the three-dimensional image is ensured without controlling the signal acquisition process, and the resolution of the three-dimensional image is further ensured without screening too many two-dimensional images. BRIEF DESCRIPTION OF DRAWINGS

[0040] Figure 1 is a flowchart of an image processing method provided by the embodiment one of the present application;

[0041] Figure 2 is a flowchart of an image processing method provided by the embodiment two of the present application;

[0042] Figure 3 is a structural schematic diagram of an image processing device provided by the embodiment three of the present application;

[0043] Figure 4 is a structural schematic diagram of a computer device provided by the embodiment four of the present application. DETAILED DESCRIPTION

[0044] The present application will be further described below in conjunction with the drawings and embodiments. It can be understood that the specific embodiments described herein are only used to explain the present application, but not to limit the present application. In addition, it should be noted that, for the convenience of description, only the parts related to the present application are shown in the drawings, but not all the structures.

[0045] Embodiment one

[0046] Figure 1is a flowchart of an image processing method provided by Embodiment One of the present application. The present embodiment can be applied to the case of intravascular ultrasound imaging. The method can be executed by an image processing device, which can be implemented in software and / or hardware, for example, the image processing device can be configured in a computer device. As shown in Figure 1 The method comprises the following steps:

[0047] S110, acquiring intravascular ultrasound two-dimensional images of the imaging object, determining a to-be-processed intravascular ultrasound two-dimensional image in the intravascular ultrasound two-dimensional images and a reference intravascular ultrasound two-dimensional image corresponding to the to-be-processed intravascular ultrasound two-dimensional image.

[0048] In the present embodiment, the signal acquisition time of the ultrasonic signal acquisition process does not need to be controlled, and the general ultrasonic signal acquisition method in the prior art can be used to acquire the intravascular ultrasound signal. After the intravascular ultrasound signal of the imaging object is acquired, for each imaging layer, the image is reconstructed to obtain the intravascular ultrasound two-dimensional image of the layer. The method of reconstructing the intravascular ultrasound two-dimensional image can refer to the method of ultrasonic image reconstruction in the prior art, which is not limited herein.

[0049] For the plurality of intravascular ultrasound two-dimensional images of the imaging object, any intravascular ultrasound two-dimensional image can be selected as the to-be-processed intravascular ultrasound two-dimensional image, and the reference intravascular ultrasound two-dimensional image corresponding to the to-be-processed intravascular ultrasound two-dimensional image is determined. Optionally, the intravascular ultrasound two-dimensional image with an image frame number interval less than a set interval from the to-be-processed intravascular ultrasound two-dimensional image can be selected as the reference intravascular ultrasound two-dimensional image corresponding to the to-be-processed intravascular ultrasound two-dimensional image. For example, assuming that the set interval is 10, the intravascular ultrasound two-dimensional image with a frame number interval less than 10 (e.g. 6) from the to-be-processed intravascular ultrasound two-dimensional image can be selected as the reference intravascular ultrasound two-dimensional image corresponding thereto.

[0050] Optionally, one or more intravascular ultrasound two-dimensional images can be selected as the to-be-processed intravascular ultrasound two-dimensional image. When multiple intravascular ultrasound two-dimensional images are selected as the to-be-processed intravascular ultrasound two-dimensional image, the reference intravascular ultrasound two-dimensional image corresponding to each to-be-processed intravascular ultrasound two-dimensional image can be selected, and a corresponding reference intravascular ultrasound two-dimensional image can also be set for the plurality of intravascular ultrasound two-dimensional images. In order to simplify the processing data amount and improve the image processing speed, a corresponding reference intravascular ultrasound two-dimensional image can be set for the plurality of intravascular ultrasound two-dimensional images. In order to unify the image processing standard, the intravascular ultrasound two-dimensional images with a continuous set number of frames can be selected as the to-be-processed intravascular ultrasound images, and an intravascular ultrasound two-dimensional image from the intravascular ultrasound two-dimensional images with a continuous set number of frames or the adjacent intravascular ultrasound two-dimensional images can be selected as the reference intravascular ultrasound image corresponding to the to-be-processed intravascular ultrasound image.

[0051] S120, affine transform the to-be-processed intravascular ultrasound two-dimensional image based on the initial affine transform parameter to obtain a transformed intravascular ultrasound two-dimensional image.

[0052] It can be understood that the motion artifact in the image is caused by the motion of the blood vessel. Based on this, the pixel distribution in the to-be-processed intravascular ultrasound two-dimensional image can be adjusted through affine transform to reduce the motion artifact in the to-be-processed intravascular ultrasound two-dimensional image. That is, the to-be-processed intravascular ultrasound two-dimensional image is subjected to affine transform, which is essentially a corresponding adjustment of the pixel value distribution in the to-be-processed intravascular ultrasound two-dimensional image, adjusting the pixel value corresponding to the motion artifact to a reasonable coordinate to reduce the motion artifact in the image.

[0053] Optionally, the to-be-processed intravascular ultrasound two-dimensional image is distributed along a first direction and a second direction perpendicular to each other. Based on this, in order to adjust the pixel distribution of the to-be-processed intravascular ultrasound two-dimensional image, the affine transform parameter can include a first direction transform parameter, a second direction transform parameter and an angle transform parameter. The first direction transform parameter is used to adjust the offset of the pixel point in the first direction, the second direction transform parameter is used to adjust the offset of the pixel point in the second direction, and the angle transform parameter is used to adjust the offset of the pixel value in the angle. Therefore, based on the affine transform parameter, the displacement, rotation and deformation of the pixel value in the to-be-processed intravascular ultrasound two-dimensional image can be realized.

[0054] Optionally, affine transforming the to-be-processed intravascular ultrasound two-dimensional image based on the initial affine transform parameter to obtain a transformed intravascular ultrasound two-dimensional image includes: affine transforming the to-be-processed intravascular ultrasound two-dimensional image according to the initial affine transform parameter to obtain an affine intravascular ultrasound two-dimensional image; and performing interpolation processing on the affine intravascular ultrasound two-dimensional image to obtain the transformed intravascular ultrasound two-dimensional image. Affine transforming the to-be-processed intravascular ultrasound two-dimensional image according to the initial affine transform parameter can be: for each pixel coordinate in the to-be-processed intravascular ultrasound two-dimensional image, obtaining a new pixel coordinate based on the pixel coordinate and the initial affine transform parameter, and matching the pixel value of the pixel coordinate to the new pixel coordinate. After performing the above operation on all pixel coordinates, the affine intravascular ultrasound two-dimensional image can be obtained.

[0055] After performing an affine transformation on the intravascular ultrasound 2D image to be processed based on the affine transformation parameters, the pixel coordinates of each pixel in the image will be projected onto a new coordinate system. However, these new coordinates are often non-integer, meaning they will not overlap with the pixel coordinates of the output image. Therefore, the question arises as to how to define the value of each pixel in the new image. Assume the intravascular ultrasound 2D image to be processed includes pixel coordinates 11, 12, 21, and 22, corresponding to pixel values ​​11, 12, 21, and 22, respectively. After performing the affine transformation on the image, pixel value 11 may be adjusted to pixel coordinate 22, and pixel value 12 may be adjusted to pixel coordinate 21, resulting in missing pixel values ​​for coordinates 11 and 12. Therefore, it is necessary to supplement the pixel values ​​for coordinates 11 and 12 to obtain a complete intravascular ultrasound image. This problem can be solved using bicubic interpolation. The specific method of bicubic interpolation is as follows: For a pixel coordinate, find the original pixel values ​​after the 16 nearest projections around it, that is, 16 points on a nine-square grid centered on the missing pixel. Then, fit a cubic curve to every 4 points that lie on a straight line, resulting in 4 cubic curves. Then, orthogonally cut the 4 cubic curves obtained through the missing pixel coordinates to obtain 4 points, and fit another cubic curve. The value of this curve at the missing pixel coordinates is the output interpolation. This yields a transformed two-dimensional intravascular ultrasound image.

[0056] For example, suppose the pixel coordinates are (x in ,y in The initial affine transformation parameters include the angle transformation parameter θ and the translation parameters (i.e., the first direction transformation parameter and the second direction transformation parameter) v. x ,v y Then we can define the translation-radial transformation matrix A. translation for:

[0057]

[0058] Define the rotational affine transformation matrix A rotation for:

[0059]

[0060] Then, the translational affine transformation matrix and the rotational affine transformation matrix are combined to obtain the affine transformation matrix A. t :

[0061]

[0062] The pixel matrix (x) of the affine intravascular ultrasound two-dimensional image can then be obtained using the following formula. out ,y out ):

[0063]

[0064] S130、determining the image distance between the transformed intravascular ultrasound two-dimensional image and the reference intravascular ultrasound two-dimensional image, determining the target affine transformation parameter based on the image distance.

[0065] After obtaining the transformed intravascular ultrasound two-dimensional image, the image distance between the transformed intravascular ultrasound two-dimensional image and the reference intravascular ultrasound two-dimensional image is calculated. The smaller the image distance between the two images, the closer the two images. The affine transformation parameter at this time is the target affine transformation parameter.

[0066] It can be understood that the resolution of the transformed intravascular ultrasound two-dimensional image obtained after processing the to-be-processed intravascular ultrasound two-dimensional image based on different affine transformation parameters may be different. In order to accurately calculate the image distance between the transformed intravascular ultrasound two-dimensional image and the reference intravascular ultrasound two-dimensional image, the resolution of the image needs to be adjusted to obtain the transformed intravascular ultrasound two-dimensional image and the reference intravascular ultrasound two-dimensional image with the same pixels for distance calculation. Generally, the resolution of the transformed intravascular ultrasound two-dimensional image is higher than that of the reference intravascular ultrasound two-dimensional image. The resolution of the transformed intravascular ultrasound two-dimensional image can be directly reduced to be the same as that of the reference intravascular ultrasound two-dimensional image, and then the image distance between the transformed intravascular ultrasound two-dimensional image and the reference intravascular ultrasound two-dimensional image can be calculated.

[0067] When the resolution is high, a large amount of calculation is required to calculate the image distance between images. In order to reduce the amount of calculation of the processor, the pixels of the transformed intravascular ultrasound two-dimensional image and the reference intravascular ultrasound two-dimensional image can be simultaneously transformed for distance calculation. Optionally, the image distance between the transformed intravascular ultrasound two-dimensional image and the reference intravascular ultrasound two-dimensional image is determined, including: performing pixel reduction processing on the transformed intravascular ultrasound two-dimensional image to obtain a distance-transformed intravascular ultrasound two-dimensional image; performing pixel reduction processing on the reference intravascular ultrasound two-dimensional image to obtain a distance-reference intravascular ultrasound two-dimensional image; and determining the distance between the distance-transformed intravascular ultrasound two-dimensional image and the distance-reference intravascular ultrasound two-dimensional image as the image distance. In an embodiment, the average pooling method can be used to reduce the pixels of the transformed intravascular ultrasound two-dimensional image and the reference intravascular ultrasound two-dimensional image to obtain the corresponding distance-transformed intravascular ultrasound two-dimensional image and distance-reference intravascular ultrasound two-dimensional image, and the distance between the distance-transformed intravascular ultrasound two-dimensional image and the distance-reference intravascular ultrasound two-dimensional image is taken as the image distance. Specifically, for the transformed intravascular ultrasound two-dimensional image, the pixels in the pooling region can be replaced by their average value through average pooling, and then two-dimensional convolution is performed on the entire picture with the step length as the pooling edge length, and then a low-dimensional picture is obtained as the distance-transformed intravascular ultrasound two-dimensional image. For the reference intravascular ultrasound two-dimensional image, the pixels in the pooling region can be replaced by their average value through average pooling, and then two-dimensional convolution is performed on the entire picture with the step length as the pooling edge length, and then a low-dimensional picture is obtained as the distance-reference intravascular ultrasound two-dimensional image.

[0068] On the basis of the above scheme, the distance between the distance-transformed intravascular ultrasound two-dimensional image and the distance-reference intravascular ultrasound two-dimensional image is determined as the image distance, including: taking the Wasserstein distance between the distance-transformed intravascular ultrasound two-dimensional image and the distance-reference intravascular ultrasound two-dimensional image as the image distance. Optionally, the Wasserstein distance between the distance-transformed intravascular ultrasound two-dimensional image and the distance-reference intravascular ultrasound two-dimensional image can be directly calculated as the image distance. Considering the calculation efficiency caused by the amount of calculation, the amount of calculation consumed in the processing of the Wasserstein distance is large. In order to improve the image processing speed, the Sinkhorn distance between the distance-transformed intravascular ultrasound two-dimensional image and the distance-reference intravascular ultrasound two-dimensional image can be calculated as an approximation of the Wasserstein distance as the image distance. Using the Sinkhorn distance to approximate the Wasserstein distance to calculate the image distance between images can not only retain the differentiable characteristics of the Wasserstein distance, but also improve the calculation speed.

[0069] In an embodiment of the present application, the method for determining the target affine transformation parameter based on the image distance comprises: determining the back propagation gradient of the image distance, adjusting the parameter value of the affine transformation parameter based on the back propagation gradient; performing affine transformation on the to-be-processed IVUS two-dimensional image based on the adjusted affine transformation parameter to obtain a new transformed IVUS two-dimensional image; determining the image distance between the new transformed IVUS two-dimensional image and the reference IVUS two-dimensional image as a new image distance, and determining the new back propagation gradient corresponding to the new image distance; iteratively performing the above operations, and when the new back propagation gradient meets the iteration stop condition, associating the affine transformation parameter corresponding to the new back propagation gradient meeting the iteration stop condition as the target affine transformation parameter. Overall, the distance between the distance-transformed IVUS two-dimensional image and the distance reference IVUS two-dimensional image is changed by continuously adjusting the affine transformation parameter. When the distance between the distance-transformed IVUS two-dimensional image and the distance reference IVUS two-dimensional image is small enough, it can be determined that the affine transformation parameter at this time is the target affine transformation parameter. Specifically, the back propagation gradient of the image distance can be calculated, the affine transformation parameter can be adjusted based on the back propagation gradient, then the new transformed IVUS two-dimensional image can be obtained based on the adjusted affine transformation parameter, the image distance between the new transformed IVUS two-dimensional image and the reference IVUS two-dimensional image can be determined as a new image distance, the back propagation gradient of the new image distance can be calculated, and the iteration is performed. After the back propagation gradient is calculated, it is judged whether the iteration stop condition is met, and the affine transformation parameter when the iteration stop condition is met is associated as the target affine transformation parameter. The adjustment of the affine transformation parameter can be: after the back propagation gradient is determined, the affine transformation parameter is updated along the negative gradient.

[0070] In the above process, the iteration stop condition is met, including that the iteration number reaches a set number threshold, and / or the back propagation gradient is not greater than a set gradient threshold. Optionally, the set number threshold and / or the set gradient threshold can be set in advance. When any condition is met, it is judged that the iteration stop condition is met, and the iteration operation is stopped.

[0071] Optionally, the affine transformation parameters comprise a first direction transformation parameter, a second direction transformation parameter and an angle transformation parameter, and determining the back propagation gradient of the image distance comprises: determining a first direction propagation gradient of the image distance on the first direction transformation parameter, a second direction propagation gradient of the image distance on the second direction transformation parameter, and an angle propagation gradient of the image distance on the angle transformation parameter, respectively; and taking the first direction propagation gradient, the second direction propagation gradient and the angle propagation gradient as the back propagation gradient of the image distance. When the affine transformation parameters are specifically the first direction transformation parameter, the second direction transformation parameter and the angle transformation parameter, the corresponding back propagation gradient can be specifically the first direction propagation gradient of the image distance on the first direction transformation parameter, the second direction propagation gradient of the image distance on the second direction transformation parameter, and the angle propagation gradient of the image distance on the angle transformation parameter. For example, assuming that the image distance is W R,C , the affine transformation parameters comprise a first direction transformation parameter v x , a second direction transformation parameter v y and an angle transformation parameter θ, the back propagation gradient of the scalar W R，C can be obtained by an automatic differentiation program, and the first direction propagation gradient , the second direction propagation gradient and the angle propagation gradient are obtained as the back propagation gradient of the image distance, respectively. It can be understood that when the back propagation gradient comprises the first direction propagation gradient, the second direction propagation gradient and the angle propagation gradient, the first gradient threshold corresponding to the first direction propagation gradient, the second gradient threshold corresponding to the second direction propagation gradient and the angle gradient threshold corresponding to the angle propagation gradient can be set, respectively. The first gradient threshold and the second gradient threshold can be the same value or different values.

[0072] S140, processing the to-be-processed intravascular ultrasound two-dimensional image based on the target affine transformation parameter to obtain a target intravascular ultrasound two-dimensional image corresponding to the to-be-processed intravascular ultrasound two-dimensional image.

[0073] After the target affine transformation parameter is determined, the to-be-processed intravascular ultrasound two-dimensional image is processed based on the target affine transformation parameter, such as affine conversion, interpolation and dimension reduction, to obtain a target intravascular ultrasound two-dimensional image corresponding to the to-be-processed intravascular ultrasound two-dimensional image.

[0074] Based on the above method, the processing of all intravascular ultrasound two-dimensional images of the imaging object can be completed, and a target intravascular ultrasound two-dimensional image corresponding to each intravascular ultrasound two-dimensional image can be obtained.

[0075] S150, performing three-dimensional reconstruction according to the target intravascular ultrasound two-dimensional images corresponding to the intravascular ultrasound two-dimensional images to obtain an intravascular ultrasound three-dimensional image of the imaging object.

[0076] The intravascular ultrasound three-dimensional image of the imaging object can be obtained by three-dimensional reconstruction of the target intravascular ultrasound two-dimensional image corresponding to each intravascular ultrasound two-dimensional image of the imaging object. The three-dimensional reconstruction manner can refer to the three-dimensional reconstruction manner in the prior art, and is not limited herein.

[0077] The image processing method provided in the embodiment of the present application can obtain the intravascular ultrasound two-dimensional images of the imaging object, determine the to-be-processed intravascular ultrasound two-dimensional image in the intravascular ultrasound two-dimensional images and the reference intravascular ultrasound two-dimensional image corresponding to the to-be-processed intravascular ultrasound two-dimensional image, perform affine transformation on the to-be-processed intravascular ultrasound two-dimensional image based on the initial affine transformation parameter to obtain a transformed intravascular ultrasound two-dimensional image, determine the image distance between the transformed intravascular ultrasound two-dimensional image and the reference intravascular ultrasound two-dimensional image, determine the target affine transformation parameter based on the image distance, process the to-be-processed intravascular ultrasound two-dimensional image based on the target affine transformation parameter to obtain the target intravascular ultrasound two-dimensional image corresponding to the to-be-processed intravascular ultrasound two-dimensional image, and perform three-dimensional reconstruction on the target intravascular ultrasound two-dimensional image corresponding to each intravascular ultrasound two-dimensional image to obtain the intravascular ultrasound three-dimensional image of the imaging object. The imaging effect of the intravascular three-dimensional image is improved. By processing each two-dimensional image and performing three-dimensional reconstruction based on the processed two-dimensional image, the imaging effect of the three-dimensional image is ensured without controlling the signal acquisition process, and the resolution of the three-dimensional image is further ensured without screening too many two-dimensional images.

[0078] Embodiment two

[0079] Figure 2 is a flowchart of an image processing method provided in the embodiment two of the present application. Based on the above scheme, the present embodiment provides a preferred embodiment.

[0080] Overall, the method provided in the present embodiment mainly relies on gradient descent to update the affine parameter, minimizes the Wasserstein distance between the two images, and makes the overlap of the features of the two images after affine conversion close to global maximization on the two-dimensional plane to adjust the motion artifacts in the image.

[0081] Specifically, any intravascular ultrasound two-dimensional image obtained by signal acquisition and reconstruction of the imaging object can be taken as a single input image (i.e., the to-be-processed intravascular ultrasound two-dimensional image) R, and the anchor image (i.e., the reference intravascular ultrasound two-dimensional image) C corresponding to the image R is determined. As shown in the following formula (1), the image adjustment can be performed through the following steps: Figure 2

[0082] (1) For the anchor image C, as the positioning standard, no transformation processing is performed, and the image resolution is reduced to obtain the matrix C1.

[0083] ​Specifically, the anchor image C can be an intravascular ultrasound two-dimensional image with the frame number of the interval image of the input image R within a set frame number threshold. The anchor image C can be reduced in image pixels by average pooling to obtain C1.

[0084] (2) Initialize the input image transformation parameter matrix (i.e., initial affine transformation parameters) A.

[0085] The variables θ, v x , v y can be initialized as initial affine transformation parameters, where v x , v y are the first direction transformation parameter and the second direction transformation parameter, respectively, and θ is the angle transformation parameter. Optionally, the initial affine transformation parameters can be set by an automatic differentiation program or can be set according to experience. After determining the initial variables θ, v x , v y as the initial affine transformation parameters, linear operations performed thereon will be recorded for gradient calculation.

[0086] (3) Perform affine transformation on R to obtain matrix R1 through the matrix A.

[0087] Define the variables according to the pre-defined affine transformation method A t , perform coordinate projection on a single input image R to obtain new coordinates, and obtain matrix R1 (i.e., an affine intravascular ultrasound two-dimensional image).

[0088] (4) Perform bicubic interpolation to fill pixels after matrix R1 to obtain R2 (i.e., a transformed intravascular ultrasound two-dimensional image).

[0089] Specifically, after transformation, the converted coordinates need to be projected into new coordinates, and then the value of each output pixel is calculated by bicubic interpolation. The operation is completed by combining the linear operation interface provided by the automatic differentiation program to obtain R2.

[0090] (5) Reduce the dimension of R2 by mean pooling, and then obtain a matrix R3 with the same size as C1.

[0091] Optionally, the image pixels can be reduced by average pooling, and the operation can be completed by combining the linear operation interface provided by the automatic differentiation program to obtain R3.

[0092] (6) Calculate the Wasserstein distance between C1 and R3 to obtain W R,C .

[0093] Optionally, the Wasserstein distance between C1 and R3 is calculated by using the signal angle iteration. The signal angle distance between the reduced dimension anchor image C1 and R3 is calculated by using the signal angle iteration, which can be completed by combining the linear operation interface provided by the automatic differentiation program to obtain the scalar W R,C.

[0094] (7) The gradient W R,C is back-propagated to the transformation parameters A by an automatic differentiation program to obtain the gradient The parameters A are updated.

[0095] The scalar W R,C is back-propagated by an automatic differentiation program to obtain the gradient and The affine transformation parameters v x ,v y and theta are then updated along the negative gradient at a learning rate of 1e-3.

[0096] (8) The steps 4-7 are iterated until the update gradient is lower than a predetermined threshold to determine the corresponding affine transformation parameters v x ,v y and theta.

[0097] After the affine transformation parameters are updated, the above steps are repeated to perform affine conversion, interpolation, dimension reduction, calculation of Wasserstein distance, calculation of gradient, and then update on the input image until the update gradient is lower than the set threshold to obtain the latest affine transformation parameters.

[0098] (9) The input image R is subjected to affine conversion by the target affine transformation parameters v x ,v y and theta and is interpolated to obtain an adjusted image as the target intravascular ultrasound two-dimensional image output.

[0099] The above operations are performed on all two-dimensional images to obtain the processed target intravascular ultrasound two-dimensional image output.

[0100] (10) The processed target intravascular ultrasound two-dimensional image output is subjected to three-dimensional reconstruction to obtain an intravascular ultrasound three-dimensional image.

[0101] It should be noted that the step with a large amount of calculation in image processing is the calculation of the distance in the signal-angle domain. Since the distance calculation of each image does not have a dependency relationship, a plurality of images can be combined into a multi-dimensional tensor for batch calculation in the signal-angle iteration. In combination with the current open-source GPU interface for automatic differentiation, the time of the entire calculation process can be greatly reduced.

[0102] The method provided in the embodiments of the application does not require auxiliary instruments and only needs conventional signal acquisition. The catheter does not cause any influence whether it is manually or automatically pulled back during signal acquisition, and no additional signal acquisition time is added. After the two-dimensional image is reconstructed from the signal acquisition, any image sample can be processed, so that all samples are retained and the resolution of the three-dimensional body in the later stage is ensured. ​

[0103] Embodiment Three

[0104] Figure 3 is a structural schematic diagram of an image processing device provided by Embodiment Three of the present application. The image processing device can be implemented in software and / or hardware, for example, the image processing device can be configured in a computer device. As shown in the figure, the device comprises a two-dimensional image acquisition module 310, a two-dimensional image transformation module 320, a target parameter determination module 330, a target two-dimensional image module 340 and an intravascular three-dimensional image module 350, wherein: Figure 3

[0105] The two-dimensional image acquisition module 310 is configured to acquire intravascular ultrasound two-dimensional images of an imaging object, determine a to-be-processed intravascular ultrasound two-dimensional image in the intravascular ultrasound two-dimensional images and a reference intravascular ultrasound two-dimensional image corresponding to the to-be-processed intravascular ultrasound two-dimensional image;

[0106] The two-dimensional image transformation module 320 is configured to perform affine transformation on the to-be-processed intravascular ultrasound two-dimensional image based on initial affine transformation parameters, to obtain a transformed intravascular ultrasound two-dimensional image;

[0107] The target parameter determination module 330 is configured to determine an image distance between the transformed intravascular ultrasound two-dimensional image and the reference intravascular ultrasound two-dimensional image, and determine target affine transformation parameters based on the image distance;

[0108] The target two-dimensional image module 340 is configured to process the to-be-processed intravascular ultrasound two-dimensional image based on the target affine transformation parameters, to obtain a target intravascular ultrasound two-dimensional image corresponding to the to-be-processed intravascular ultrasound two-dimensional image;

[0109] The intravascular three-dimensional image module 350 is configured to perform three-dimensional reconstruction according to the target intravascular ultrasound two-dimensional image corresponding to each intravascular ultrasound two-dimensional image, to obtain an intravascular ultrasound three-dimensional image of the imaging object.

[0110] ​The embodiment of the application acquires intravascular ultrasound two-dimensional images of an imaging object, determines a to-be-processed intravascular ultrasound two-dimensional image in the intravascular ultrasound two-dimensional images and a reference intravascular ultrasound two-dimensional image corresponding to the to-be-processed two-dimensional intravascular ultrasound image, performs affine transformation on the to-be-processed intravascular ultrasound two-dimensional image based on initial affine transformation parameters to obtain a transformed intravascular ultrasound two-dimensional image, determines an image distance between the transformed intravascular ultrasound two-dimensional image and the reference intravascular ultrasound two-dimensional image, determines target affine transformation parameters based on the image distance, processes the to-be-processed intravascular ultrasound two-dimensional image based on the target affine transformation parameters to obtain a target intravascular ultrasound two-dimensional image corresponding to the to-be-processed intravascular ultrasound two-dimensional image, and performs three-dimensional reconstruction on the target intravascular ultrasound two-dimensional images corresponding to the intravascular ultrasound two-dimensional images to obtain an intravascular ultrasound three-dimensional image of the imaging object. The imaging effect of the intravascular three-dimensional image is improved. By processing each two-dimensional image and performing three-dimensional reconstruction based on the processed two-dimensional images, the imaging effect of the three-dimensional image is ensured without controlling the signal acquisition process, and the resolution of the three-dimensional image is further ensured without screening too many two-dimensional images.

[0111] Optionally, on the basis of the above scheme, the target parameter determination module 330 is specifically configured to:

[0112] perform pixel reduction processing on the transformed intravascular ultrasound two-dimensional image to obtain a distance-transformed intravascular ultrasound two-dimensional image;

[0113] perform pixel reduction processing on the reference intravascular ultrasound two-dimensional image to obtain a distance-reference intravascular ultrasound two-dimensional image;

[0114] determine the distance between the distance-transformed intravascular ultrasound two-dimensional image and the distance-reference intravascular ultrasound two-dimensional image as the image distance.

[0115] Optionally, on the basis of the above scheme, the target parameter determination module 330 is specifically configured to:

[0116] determine the Wasserstein distance between the distance-transformed intravascular ultrasound two-dimensional image and the distance-reference intravascular ultrasound two-dimensional image as the image distance.

[0117] Optionally, on the basis of the above scheme, the target parameter determination module 330 is specifically configured to:

[0118] determine the backpropagation gradient of the image distance, and adjust the parameter value of the affine transformation parameter based on the backpropagation gradient;

[0119] perform affine transformation on the to-be-processed intravascular ultrasound two-dimensional image based on the adjusted affine transformation parameter to obtain a new transformed intravascular ultrasound two-dimensional image;

[0120] determine an image distance between the new transformed intravascular ultrasound two-dimensional image and the reference intravascular ultrasound two-dimensional image as a new image distance, and determine a new back propagation gradient corresponding to the new image distance;

[0121] iteratively perform the above operations, and when the new back propagation gradient meets an iteration stop condition, associate an affine transformation parameter corresponding to the new back propagation gradient meeting the iteration stop condition as the target affine transformation parameter.

[0122] Optionally, based on the scheme, the affine transformation parameter includes a first direction transformation parameter, a second direction transformation parameter and an angle transformation parameter, and the target parameter determination module 330 is specifically configured to:

[0123] determine a first direction propagation gradient of the image distance to the first direction transformation parameter, a second direction propagation gradient of the image distance to the second direction transformation parameter, and an angle propagation gradient of the image distance to the angle transformation parameter;

[0124] take the first direction propagation gradient, the second direction propagation gradient and the angle propagation gradient as the back propagation gradient of the image distance.

[0125] Optionally, based on the scheme, the iteration stop condition is met including that the iteration number reaches a set number threshold, and / or the back propagation gradient is not greater than a set gradient threshold.

[0126] Optionally, based on the scheme, the two-dimensional image transformation module 320 is specifically configured to:

[0127] perform affine transformation on the to-be-processed intravascular ultrasound two-dimensional image according to the initial affine transformation parameter to obtain an affine intravascular ultrasound two-dimensional image;

[0128] perform interpolation processing on the affine intravascular ultrasound two-dimensional image to obtain the transformed intravascular ultrasound two-dimensional image.

[0129] The image processing device provided in the embodiments of the present application can execute the image processing method provided in any of the embodiments of the present application, and has the corresponding function modules and beneficial effects of the execution method.

[0130] Embodiment Four

[0131] Figure 4 is a structural schematic diagram of a computer device provided in Embodiment Four of the present application. Figure 4 A block diagram of an example computer device 412 suitable for implementing an embodiment of the present application is shown. Figure 4 The computer device 412 shown is merely an example, and should not bring any limitation to the functions and use range of the embodiments of the present application.

[0132] As Figure 4As shown, the computer device 412 is in the form of a general- purpose computer device. The components of the computer device 412 can include, but are not limited to, one or more processors 416, system memory 428, and a bus 418 that connects the various system components, including the system memory 428 and the processor 416.

[0133] The bus 418 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, a graphics accelerator bus, a processor or local bus using any of a variety of bus architectures including an industry standard architecture (ISA), microchannel architecture (MAC), enhanced ISA (EISA), Video Electronics Standards Association (VESA) local bus, and a peripheral component interconnect (PCI) bus.

[0134] The computer device 412 typically includes a variety of computer system readable media. Such media can be any available media that is located either in or out of the computer device 412, such as volatile and non-volatile media, removable and non-removable media.

[0135] The system memory 428 can include computer system readable media in the form of volatile memory, such as random access memory (RAM) 430 and / or cache memory 432. The computer device 412 can further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, a storage device 434 can be used for reading from and writing to non-removable, non-volatile magnetic media (e.g., a "hard drive"). Figure 4 Not shown, a computer device 412 typically includes a variety of computer system readable media. Such media can be any available media that is located either in or out of the computer device 412, such as volatile and non-volatile media, removable and non-removable media. Figure 4 Not shown, a computer device 412 typically includes a variety of computer system readable media. Such media can be any available media that is located either in or out of the computer device 412, such as volatile and non-volatile media, removable and non-removable media.

[0136] Program / utility 440 having a set (at least one) of program modules 442 can be stored in, for example, memory 428 by way of example, such program modules 442 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data, each of which or a combination can include implementation of a network environment. Program modules 442 generally carry out the functions and / or methodologies of embodiments of the present application as described herein.

[0137] Computer device 412 can also communicate with one or more external devices 414 such as a keyboard, a pointing device, a display 424, etc.; one or more devices that enable a user to interact with computer device 412; and / or any devices (e.g., network card, modem, etc.) that enable computer device 412 to communicate with one or more other computing devices. Such communication can occur via Input / Output (I / O) interface(s) 422. Still yet, computer device 412 can communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or the Internet) through network adapter 420. As an example, network adapter 420 can include a modem, a network card (wireless or wired), or other well-known interface devices. As depicted, network adapter 420 communicates with the other components of computer device 412 via bus 418. It should be appreciated that although not shown, other hardware and / or software modules could be used in conjunction with computer device 412. Examples, include, but are not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data archival storage systems, etc.

[0138] Processor 416 executes various function applications and data processing by running programs stored in system memory 428, such as implementing the image processing method provided by any embodiments of the present application, which comprises:

[0139] Obtaining intravascular ultrasound two-dimensional images of an imaging object, determining a to-be-processed intravascular ultrasound two-dimensional image and a reference intravascular ultrasound two-dimensional image corresponding to the to-be-processed intravascular ultrasound two-dimensional image in the intravascular ultrasound two-dimensional images;

[0140] Performing affine transformation on the to-be-processed intravascular ultrasound two-dimensional image based on the initial affine transformation parameter, to obtain a transformed intravascular ultrasound two-dimensional image;

[0141] Determining an image distance between the transformed intravascular ultrasound two-dimensional image and the reference intravascular ultrasound two-dimensional image, and determining a target affine transformation parameter based on the image distance;

[0142] Processing the to-be-processed intravascular ultrasound two-dimensional image based on the target affine transformation parameter, to obtain a target intravascular ultrasound two-dimensional image corresponding to the to-be-processed intravascular ultrasound two-dimensional image;

[0143] Performing three-dimensional reconstruction according to the target intravascular ultrasound two-dimensional images corresponding to the intravascular ultrasound two-dimensional images, to obtain an intravascular ultrasound three-dimensional image of the imaging object.

[0144] Of course, those skilled in the art can understand that the processor can also implement the technical solutions of the image processing method provided by any embodiments of the present application.

[0145] Embodiment five

[0146] The fifth embodiment of the present application also provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the image processing method provided by the embodiments of the present application, and the method comprises the following steps:

[0147] An intravascular ultrasound two-dimensional image of an imaging object is acquired, a to-be-processed intravascular ultrasound two-dimensional image in the intravascular ultrasound two-dimensional image and a reference intravascular ultrasound two-dimensional image corresponding to the to-be-processed intravascular ultrasound two-dimensional image are determined;

[0148] An affine transformation is performed on the to-be-processed intravascular ultrasound two-dimensional image based on the initial affine transformation parameter, and a transformed intravascular ultrasound two-dimensional image is obtained;

[0149] An image distance between the transformed intravascular ultrasound two-dimensional image and the reference intravascular ultrasound two-dimensional image is determined, and a target affine transformation parameter is determined based on the image distance;

[0150] The to-be-processed intravascular ultrasound two-dimensional image is processed based on the target affine transformation parameter, and a target intravascular ultrasound two-dimensional image corresponding to the to-be-processed intravascular ultrasound two-dimensional image is obtained;

[0151] Three-dimensional reconstruction is performed on the target intravascular ultrasound two-dimensional images corresponding to the intravascular ultrasound two-dimensional images, and an intravascular ultrasound three-dimensional image of the imaging object is obtained.

[0152] Of course, the computer readable storage medium provided by the embodiments of the present application and the computer program stored thereon are not limited to the method operations described above, and can also perform related operations of the test system construction method provided by any embodiment of the present application.

[0153] The computer storage medium of the embodiments of the present application can adopt any combination of one or more computer readable media. The computer readable medium can be a computer readable signal medium or a computer readable storage medium. The computer readable storage medium may, for example, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or component, or any combination thereof. More specific examples (non-exhaustive list) of the computer readable storage medium include: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this document, the computer readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device or component.

[0154] A computer readable signal medium can include a propagated data signal with computer executable code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated signal can take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium can be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate or transport programming code.

[0155] Program code embodied on a computer readable medium can be transmitted using any appropriate medium, including but not limited to wireless, wire line, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0156] Computer program code for carrying out operations for aspects of the present application can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0157] Note that the foregoing examples have been provided merely for the purpose of explanation and are in no way to be construed as limiting of the present application. While the application has been described with reference to preferred embodiments and specific mountings, it will be apparent to one skilled in the art that various changes and modifications can be made therein without departing from the spirit and scope of the application. Accordingly, the scope of the application is to be limited only by the appended claims and their equivalents.

Claims

1. An image processing method, characterized by, The method comprises the following steps: acquiring intravascular ultrasound two-dimensional images of an imaging object, determining a to-be-processed intravascular ultrasound two-dimensional image in the intravascular ultrasound two-dimensional images and a reference intravascular ultrasound two-dimensional image corresponding to the to-be-processed intravascular ultrasound two-dimensional image; affine transforming the to-be-processed intravascular ultrasound two-dimensional image based on initial affine transformation parameters to obtain a transformed intravascular ultrasound two-dimensional image; determining an image distance between the transformed intravascular ultrasound two-dimensional image and the reference intravascular ultrasound two-dimensional image, and determining target affine transformation parameters based on the image distance; processing the to-be-processed intravascular ultrasound two-dimensional image based on the target affine transformation parameters to obtain a target intravascular ultrasound two-dimensional image corresponding to the to-be-processed intravascular ultrasound two-dimensional image; performing three-dimensional reconstruction on the target intravascular ultrasound two-dimensional images corresponding to the intravascular ultrasound two-dimensional images to obtain an intravascular ultrasound three-dimensional image of the imaging object; wherein the determining of the target affine transformation parameters based on the image distance comprises: determining a backpropagation gradient of the image distance, and adjusting a parameter value of an affine transformation parameter based on the backpropagation gradient, wherein the adjustment of the affine transformation parameter is: after the backpropagation gradient is determined, the affine transformation parameter is updated along a negative gradient; affine transforming the to-be-processed intravascular ultrasound two-dimensional image based on the adjusted affine transformation parameter to obtain a new transformed intravascular ultrasound two-dimensional image; taking an image distance between the new transformed intravascular ultrasound two-dimensional image and the reference intravascular ultrasound two-dimensional image as a new image distance, and determining a new backpropagation gradient corresponding to the new image distance; iteratively performing the above operations, and when a new backpropagation gradient that meets an iteration stop condition is obtained, regarding the affine transformation parameter associated with the new backpropagation gradient that meets the iteration stop condition as the target affine transformation parameter; wherein the affine transformation parameters comprise a first direction transformation parameter, a second direction transformation parameter, and an angle transformation parameter, the first direction transformation parameter is used to adjust the offset of a pixel point in a first direction, the second direction transformation parameter is used to adjust the offset of a pixel point in a second direction, and the angle transformation parameter is used to adjust the offset of a pixel value in an angle, and the first direction and the second direction are perpendicular to each other; the determining of the backpropagation gradient of the image distance comprises: respectively determining a first direction propagation gradient of the image distance on the first direction transformation parameter, a second direction propagation gradient of the image distance on the second direction transformation parameter, and an angle propagation gradient of the image distance on the angle transformation parameter; taking the first direction propagation gradient, the second direction propagation gradient, and the angle propagation gradient as the backpropagation gradient of the image distance.

2. The method of claim 1, wherein, the determining of the image distance between the transformed intravascular ultrasound two-dimensional image and the reference intravascular ultrasound two-dimensional image comprises: performing pixel reduction processing on the transformed intravascular ultrasound two-dimensional image to obtain a distance transformed intravascular ultrasound two-dimensional image; performing pixel reduction processing on the reference intravascular ultrasound two-dimensional image to obtain a distance reference intravascular ultrasound two-dimensional image; determine a distance between the distance-transformed intravascular ultrasound two-dimensional image and the distance reference intravascular ultrasound two-dimensional image as the image distance.

3. The method of claim 2, wherein, The determining the distance between the distance-transformed intravascular ultrasound two-dimensional image and the distance reference intravascular ultrasound two-dimensional image as the image distance comprises: determining a Wasserstein distance between the distance-transformed intravascular ultrasound two-dimensional image and the distance reference intravascular ultrasound two-dimensional image as the image distance.

4. The method of claim 1, wherein, The satisfying the iteration stop condition comprises that the number of iterations reaches a set number threshold, and / or the back propagation gradient is not greater than a set gradient threshold.

5. The method of claim 1, wherein, The affine transformation of the to-be-processed intravascular ultrasound two-dimensional image based on the initial affine transformation parameter to obtain a transformed intravascular ultrasound two-dimensional image comprises: affine transformation of the to-be-processed intravascular ultrasound two-dimensional image according to the initial affine transformation parameter to obtain an affine intravascular ultrasound two-dimensional image; interpolation processing of the affine intravascular ultrasound two-dimensional image to obtain the transformed intravascular ultrasound two-dimensional image.

6. An image processing apparatus characterized by comprising: comprise: a two-dimensional image acquisition module configured to acquire intravascular ultrasound two-dimensional images of an imaging object, determine a to-be-processed intravascular ultrasound two-dimensional image in the intravascular ultrasound two-dimensional images and a reference intravascular ultrasound two-dimensional image corresponding to the to-be-processed intravascular ultrasound two-dimensional image; a two-dimensional image transformation module configured to perform affine transformation of the to-be-processed intravascular ultrasound two-dimensional image based on an initial affine transformation parameter to obtain a transformed intravascular ultrasound two-dimensional image; a target parameter determination module configured to determine an image distance between the transformed intravascular ultrasound two-dimensional image and the reference intravascular ultrasound two-dimensional image, and determine a target affine transformation parameter based on the image distance; a target two-dimensional image module configured to process the to-be-processed intravascular ultrasound two-dimensional image based on the target affine transformation parameter to obtain a target intravascular ultrasound two-dimensional image corresponding to the to-be-processed intravascular ultrasound two-dimensional image; an intravascular three-dimensional image module configured to perform three-dimensional reconstruction based on the target intravascular ultrasound two-dimensional image corresponding to each of the intravascular ultrasound two-dimensional images to obtain an intravascular ultrasound three-dimensional image of the imaging object; The target parameter determination module is specifically configured to: determine a back propagation gradient of the image distance, adjust a parameter value of an affine transformation parameter based on the back propagation gradient, wherein the adjustment of the affine transformation parameter is that, after the back propagation gradient is determined, the affine transformation parameter is updated along a negative gradient; perform affine transformation of the to-be-processed intravascular ultrasound two-dimensional image based on the adjusted affine transformation parameter to obtain a new transformed intravascular ultrasound two-dimensional image; determine an image distance between the new transformed intravascular ultrasound two-dimensional image and the reference intravascular ultrasound two-dimensional image as a new image distance, and determine a new back propagation gradient corresponding to the new image distance; iteratively perform the above operations, and when the new back propagation gradient satisfies an iteration stop condition, associate the affine transformation parameter associated with the new back propagation gradient that satisfies the iteration stop condition as the target affine transformation parameter; The affine transformation parameters comprise a first direction transformation parameter, a second direction transformation parameter and an angle transformation parameter, the first direction transformation parameter is used for adjusting the offset of a pixel in a first direction, the second direction transformation parameter is used for adjusting the offset of a pixel in a second direction, and the angle transformation parameter is used for adjusting the offset of a pixel value in an angle, and the first direction and the second direction are perpendicular to each other. The target parameter determination module is specifically configured to: determine a first direction propagation gradient of the image distance on the first direction transformation parameter, a second direction propagation gradient of the image distance on the second direction transformation parameter, and an angle propagation gradient of the image distance on the angle transformation parameter, respectively; use the first direction propagation gradient, the second direction propagation gradient and the angle propagation gradient as the back propagation gradient of the image distance.

7. A computer device, comprising: The device comprises: one or more processors; a storage device for storing one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement the image processing method according to any one of claims 1-5.

8. A computer-readable storage medium having stored thereon a computer program, characterized in that, The program is executed by the processor to implement the image processing method according to any one of claims 1-5.

Images