WO2022036972A1

WO2022036972A1 - Image segmentation method and apparatus, and electronic device and storage medium

Info

Publication number: WO2022036972A1
Application number: PCT/CN2020/138131
Authority: WO
Inventors: 韩泓泽; 刘星龙; 黄宁; 孙辉; 张少霆
Original assignee: 上海商汤智能科技有限公司
Priority date: 2020-08-17
Filing date: 2020-12-21
Publication date: 2022-02-24
Also published as: TW202209254A; CN111899268A; CN111899268B

Abstract

An image segmentation method and apparatus, and an electronic device and a storage medium. The method comprises: performing prediction on a pixel, which belongs to a target object, in an image to be processed, so as to obtain a preliminarily segmented image corresponding to the image to be processed (S11); and according to edge information of the target object in the image to be processed, adjusting, in the preliminarily segmented image, a pixel value of the pixel, which is predicted not to belong to the target object, in a closed area included in an edge of the target object, so as to obtain a first segmentation result corresponding to the image to be processed (S12).

Description

Image segmentation method and device, electronic device and storage medium

CROSS-REFERENCE TO RELATED APPLICATIONS

The present disclosure is based on a Chinese patent application with application number 202010827077.1 and an application date of August 17, 2020, and claims the priority of the Chinese patent application, the entire contents of which are incorporated herein by reference.

technical field

The present disclosure relates to the technical field of image processing, and in particular, to an image segmentation method and device, an electronic device and a storage medium.

Background technique

Image segmentation refers to the technology and process of dividing an image into several specific regions with unique properties and proposing objects of interest. Image segmentation is a key step from image processing to image analysis. Image segmentation methods in the related art are mainly divided into the following categories: threshold-based segmentation methods, region-based segmentation methods, edge-based segmentation methods, and specific theory-based segmentation methods.

SUMMARY OF THE INVENTION

The present disclosure provides an image segmentation method and device, an electronic device and a storage medium.

According to an aspect of the present disclosure, an image segmentation method is provided, comprising:

Predict the pixels belonging to the target object in the image to be processed, and obtain a preliminary segmented image corresponding to the image to be processed;

According to the edge information of the target object in the to-be-processed image, in the preliminary segmented image, the pixel values of the predicted pixels that do not belong to the target object in the enclosed area included in the edge of the target object are determined. Adjustment is performed to obtain a first segmentation result corresponding to the image to be processed.

By predicting the pixels belonging to the target object in the image to be processed, a preliminary segmented image corresponding to the image to be processed is obtained, and according to the edge information of the target object in the image to be processed, in the preliminary segmented image, the In the enclosed area included in the edge of the target object, the predicted pixel values of the pixels that do not belong to the target object are adjusted to obtain the first segmentation result corresponding to the to-be-processed image, so that a more accurate and robust image can be obtained. Split result.

In a possible implementation manner, in the preliminary segmented image, the predicted pixel value of the pixel belonging to the target object is a first preset value, and the predicted pixel value of the pixel not belonging to the target object is the first preset value. the second preset value;

According to the edge information of the target object in the image to be processed, in the preliminary segmented image, in the closed area included in the edge of the target object, the predicted pixels that do not belong to the target object are analyzed. The pixel value is adjusted to obtain the first segmentation result corresponding to the to-be-processed image, including:

Adjusting the pixel value of the closed area whose pixel value is the second preset value in the preliminary segmented image to the first preset value, to obtain a filled preliminary segmented image;

According to the edge information of the target object in the to-be-processed image, the pixel values of the filled preliminary segmented image are adjusted to obtain the first segmentation result corresponding to the to-be-processed image.

In this implementation manner, a filled preliminary segmented image is obtained by adjusting the pixel value of the closed area whose pixel value is the second preset value in the preliminary segmented image to the first preset value, thereby enabling the The first segmentation result corresponding to the processed image covers the inside of the organ of the target object, for example, covers the lung parenchyma such as the lung, the inside of the digestive tract (eg, the gastrointestinal tract), and the like. That is, by adopting the above-mentioned implementation manner, the missing holes in the target object (for example, in the human body) after image segmentation can be filled. By adjusting the pixel values of the filled preliminary segmented image according to the edge information of the target object in the to-be-processed image, the first segmentation result corresponding to the to-be-processed image can be obtained, thereby reducing the The background part in the image to be processed (ie the part that does not belong to the target object) is divided into the probability of belonging to the target object.

In a possible implementation manner, adjusting the pixel value of the enclosed area with the pixel value of the second preset value in the preliminary segmented image to the first preset value to obtain the filled preliminary segmentation images, including:

Splicing edges of preset widths around the preliminary segmented images to obtain a preliminarily segmented image after splicing, wherein the pixel values of the pixels of the edges of the spliced preset widths are the second preset values;

Select the pixel of the image edge of the preliminarily segmented image after the splicing as a seed point, and perform a flood filling operation on the preliminarily segmented image after the splicing to obtain the preliminarily segmented image after the filling.

By splicing the edge of the preset width around the preliminary segmented image, and then selecting the pixel of the image edge of the preliminarily segmented image after splicing as the seed point, it can be ensured that the seed point of the flood filling operation belongs to the background part ( That is, the part that does not belong to the target object), so that the first segmentation result corresponding to the image to be processed can cover the inside of the organ of the target object, thereby obtaining a more accurate segmentation result.

In a possible implementation manner, the pixel value of the filled preliminary segmented image is adjusted according to the edge information of the target object in the to-be-processed image to obtain the corresponding pixel value of the to-be-processed image. Describe the first segmentation result, including:

According to the edge information of the target object in the image to be processed, determine the maximum connected domain included in the edge of the target object in the filled preliminary segmented image;

The pixel values of the pixels outside the maximum connected region in the filled preliminary segmented image are adjusted to the second preset value to obtain the first segmentation result corresponding to the to-be-processed image.

According to this implementation, false positive regions that are not connected to the target object can be eliminated, thereby greatly reducing the probability of erroneously classifying the background part as belonging to the target object, thereby improving the accuracy of image segmentation. For example, if the target object is a human body, according to this example, false positive regions that are not connected to the human body can be eliminated, thereby greatly reducing the probability that the background part (eg, bed board, etc.) is erroneously classified as belonging to the human body.

In a possible implementation manner, after obtaining the first segmentation result corresponding to the image to be processed, the method further includes:

acquiring an image adjacent to the to-be-processed image and a second segmentation result corresponding to the adjacent image;

According to the pixel value of the pixel at the same position in the image to be processed and the adjacent image, and the second segmentation result, adjust the first segmentation result to obtain a third segmentation result corresponding to the to-be-processed image .

According to this implementation manner, the continuity of the image to be processed and the second segmentation result can be ensured, thereby helping to obtain a smoother and more accurate three-dimensional segmentation result. For example, if the target object is a human body, the continuity of the image to be processed and the human body in the adjacent images can be ensured, thereby helping to obtain a smoother and more accurate three-dimensional human body segmentation result. For example, the segmentation result corresponding to each CT image in the CT image sequence can be obtained by using this implementation manner, thereby obtaining a smoother and more accurate three-dimensional human body segmentation result.

In a possible implementation manner, the first segmentation result is adjusted according to the pixel value of the pixel at the same position in the image to be processed and the adjacent image, and the second segmentation result, to obtain The third segmentation result corresponding to the image to be processed includes:

According to the pixels in the adjacent images that belong to the target object in the second segmentation result and have a difference value of the pixel value at the same position as the image to be processed is less than or equal to a third preset value, The first segmentation result is adjusted to obtain the third segmentation result corresponding to the image to be processed.

By adjusting the first segmentation result according to the pixels in the second segmentation result that belong to the target object and whose pixel values at the same position as the to-be-processed image differ by less than or equal to a third preset value, The third segmentation result corresponding to the image to be processed is obtained, so that the first segmentation corresponding to the image to be processed can be divided according to the segmentation result corresponding to the pixels in the adjacent images that are relatively related to the image to be processed The result is adjusted, thereby helping to improve the accuracy of the final segmentation result corresponding to the image to be processed.

In a possible implementation manner, according to the adjacent images, in the second segmentation result, the difference value of the pixel values belonging to the target object and at the same position as the image to be processed For pixels less than or equal to the third preset value, adjust the first segmentation result to obtain the third segmentation result corresponding to the image to be processed, including:

Obtain the first pixel set according to the pixel whose difference value of the pixel value at the same position in the image to be processed and the adjacent image is less than or equal to the third preset value;

obtaining a second pixel set according to the pixels belonging to the target object in the second segmentation result in the first pixel set;

Adjusting the pixels of the second pixel set in the first segmentation result to belong to the target object to obtain the third segmentation result corresponding to the image to be processed.

A first pixel set is obtained according to the pixels whose difference between the pixel values at the same position in the image to be processed and the adjacent image is less than or equal to a third preset value, and according to the first pixel set at the same position The pixels belonging to the target object in the second segmentation result are obtained to obtain a second pixel set, and the pixels of the second pixel set in the first segmentation result are adjusted to belong to the target object to obtain the to-be-processed The third segmentation result corresponding to the image, so that the first segmentation result corresponding to the image to be processed can be processed according to the pixels in the second segmentation result that belong to the target object and are relatively related to the image to be processed. adjustment, thereby helping to improve the accuracy of the final segmentation result corresponding to the image to be processed.

In one possible implementation,

Before predicting the pixels belonging to the target object in the image to be processed, the method further includes: training a neural network according to the training image and labeling data of the training image, wherein the labeling data of the training image includes the training image The true value of the pixels belonging to the target object in ;

The predicting the pixels belonging to the target object in the image to be processed to obtain a preliminary segmented image corresponding to the image to be processed includes: inputting the image to be processed into the neural network, and predicting the image to be processed through the neural network According to the information of the pixels belonging to the target object in the to-be-processed image, a preliminary segmented image corresponding to the to-be-processed image is obtained.

In this implementation manner, the part of the image to be processed that belongs to the target object is predicted by the neural network. For example, if the image to be processed is a CT image, and the target object is a human body, this implementation does not consider removing various bedboards in the CT image, that is, no longer focuses on the non-human body part, but focuses on The segmentation of the human body in the CT image can ensure the accuracy and robustness of the segmentation results under a large amount of special-shaped bed plate data. That is, even if the image to be processed contains a special-shaped bed plate, an accurate and robust segmentation result can be obtained by adopting this implementation manner.

In one possible implementation,

The training image is an electronic computed tomography CT image;

The training of the neural network according to the training image and the labeled data of the training image includes: normalizing the pixel values of the training image according to a preset CT value range to obtain a normalized training image; The normalized training image and the labeled data of the training image train the neural network.

In this implementation manner, the pixel values of the training image are normalized according to a preset CT value range to obtain a normalized training image, and according to the normalized training image and the training image The labeled data of the image trains the neural network, thereby helping to reduce the computational load of the neural network and improve the convergence speed of the neural network.

According to the pixel value of the pixel at the same position in the image to be processed and the adjacent image, and the second segmentation result, the preliminary segmented image is adjusted to obtain a fourth segmentation result corresponding to the to-be-processed image.

In a possible implementation manner, according to the pixel value of the pixel at the same position in the image to be processed and the adjacent image, and the second segmentation result, the preliminary segmented image is adjusted to obtain the The fourth segmentation result corresponding to the image to be processed includes:

According to the pixels in the adjacent images that belong to the target object in the second segmentation result and have a difference value of the pixel value at the same position as the image to be processed is less than or equal to a third preset value, Adjust the preliminary segmented image to obtain a fourth segmentation result corresponding to the to-be-processed image.

In a possible implementation manner, according to the adjacent images, in the second segmentation result, the difference value of the pixel values belonging to the target object and at the same position as the image to be processed For pixels less than or equal to the third preset value, adjust the preliminary segmented image to obtain a fourth segmentation result corresponding to the to-be-processed image, including:

The pixels of the second pixel set in the preliminary segmented image are adjusted to belong to the target object, and a fourth segmentation result corresponding to the to-be-processed image is obtained.

According to an aspect of the present disclosure, an image segmentation apparatus is provided, comprising:

a first segmentation part, configured to predict pixels belonging to the target object in the to-be-processed image, and obtain a preliminary segmented image corresponding to the to-be-processed image;

The first adjustment part is configured to, according to the edge information of the target object in the to-be-processed image, in the preliminary segmented image, predict that in the enclosed area included in the edge of the target object, those that do not belong to the The pixel values of the pixels of the target object are adjusted to obtain a first segmentation result corresponding to the image to be processed.

The first adjustment module is used for:

According to the edge information of the target object in the to-be-processed image, the pixel values of the filled preliminary segmented image are adjusted to obtain a first segmentation result corresponding to the to-be-processed image.

In a possible implementation manner, the first adjustment module is used for:

Selecting the pixels of the image edge of the spliced preliminary segmented image as seed points, and performing a flood filling operation on the spliced preliminary segmented image to obtain a filled preliminary segmented image.

In a possible implementation manner, the first adjustment module is used for:

According to the edge information of the target object in the to-be-processed image, determine the maximum connected domain included in the edge of the target object in the filled preliminary segmented image;

In a possible implementation, the apparatus further includes:

a second acquisition module, configured to acquire an image adjacent to the to-be-processed image and a second segmentation result corresponding to the adjacent image;

A third adjustment module, configured to adjust the first segmentation result according to the pixel value of the pixel at the same position in the image to be processed and the adjacent image, and the second segmentation result, to obtain the to-be-processed image The third segmentation result corresponding to the image.

In a possible implementation manner, the third adjustment module is used for:

According to the pixels in the adjacent images that belong to the target object in the second segmentation result and have a difference value of the pixel value at the same position as the image to be processed is less than or equal to a third preset value, Adjust the first segmentation result to obtain a third segmentation result corresponding to the to-be-processed image.

In a possible implementation manner, the third adjustment module is used for:

Adjusting the pixels of the second pixel set in the first segmentation result to belong to the target object to obtain a third segmentation result corresponding to the image to be processed.

In one possible implementation,

The device further includes: a training module for training a neural network according to the training image and the labeling data of the training image, wherein the labeling data of the training image includes the true values of the pixels belonging to the target object in the training image. value;

The first segmentation module is used for: inputting the image to be processed into the neural network, and predicting the information of the pixels belonging to the target object in the image to be processed through the neural network; The pixel information of the object is obtained to obtain a preliminary segmented image corresponding to the to-be-processed image.

In one possible implementation,

The training image is an electronic computed tomography CT image;

The training module is used for: normalizing the pixel values of the training image according to a preset CT value range to obtain a normalized training image; according to the normalized training image and the training image The labeled data of the images trains the neural network.

The second segmentation part is configured to predict the pixels belonging to the target object in the to-be-processed image, and obtain a preliminary segmented image corresponding to the to-be-processed image;

a first acquiring part, configured to acquire an image adjacent to the to-be-processed image and a second segmentation result corresponding to the adjacent image;

The second adjustment part is configured to adjust the preliminary segmented image according to the pixel value of the pixel at the same position in the image to be processed and the adjacent image and the second segmentation result to obtain the image to be processed The corresponding fourth segmentation result.

In a possible implementation manner, the second adjustment module is used for:

According to the pixels in the adjacent images that belong to the target object in the second segmentation result and have a difference value of the pixel value at the same position as the image to be processed is less than or equal to a third preset value, The preliminary segmented image is adjusted to obtain a fourth segmentation result corresponding to the to-be-processed image.

In a possible implementation manner, the second adjustment module is used for:

Obtain a second pixel set according to the pixels belonging to the target object in the second segmentation result in the first pixel set;

According to an aspect of the present disclosure, there is provided an electronic device comprising: one or more processors; a memory configured to store executable instructions; wherein the one or more processors are configured to invoke the memory storage executable instructions to perform the above image segmentation method.

According to an aspect of the present disclosure, there is provided a computer-readable storage medium having computer program instructions stored thereon, the computer program instructions implementing the above method when executed by a processor.

In the embodiment of the present disclosure, a preliminary segmented image corresponding to the to-be-processed image is obtained by predicting the pixels belonging to the target object in the to-be-processed image, and according to the edge information of the target object in the to-be-processed image, in the preliminary segmented image In dividing the image, the pixel values of the predicted pixels that do not belong to the target object in the enclosed area included in the edge of the target object are adjusted to obtain the first segmentation result corresponding to the image to be processed. Get more accurate and robust segmentation results.

According to an aspect of the present disclosure, there is provided a computer program including computer-readable codes, where, when the computer-readable codes are executed in an electronic device, a processor in the electronic device executes the above-mentioned image segmentation method.

It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure.

Other features and aspects of the present disclosure will become apparent from the following detailed description of exemplary embodiments with reference to the accompanying drawings.

Description of drawings

The accompanying drawings, which are incorporated into and constitute a part of this specification, illustrate embodiments consistent with the present disclosure, and together with the description, serve to explain the technical solutions of the present disclosure.

1-1 is a schematic diagram 1 of an application scenario of an image segmentation method provided by an embodiment of the present disclosure;

1-2 is a second schematic diagram of an application scenario of an image segmentation method provided by an embodiment of the present disclosure;

2 is a flowchart of an image segmentation method provided by an embodiment of the present disclosure;

3 is a schematic diagram of a U-shaped convolutional neural network provided by an embodiment of the present disclosure;

4 is a schematic diagram of splicing edges of a preset width around a preliminary segmented image to obtain a preliminarily segmented image after splicing, according to an embodiment of the present disclosure;

5 is a flowchart of an image segmentation method provided by another embodiment of the present disclosure;

FIG. 6 is a block diagram of an image segmentation apparatus provided by an embodiment of the present disclosure;

FIG. 7 is another block diagram of an image segmentation apparatus provided by an embodiment of the present disclosure;

FIG. 8 is a block diagram of an electronic device 800 according to an embodiment of the present disclosure;

FIG. 9 is a block diagram of an electronic device 1900 according to an embodiment of the present disclosure.

detailed description

Various exemplary embodiments, features and aspects of the present disclosure will be described in detail below with reference to the accompanying drawings. The same reference numbers in the figures denote elements that have the same or similar functions. While various aspects of the embodiments are shown in the drawings, the drawings are not necessarily drawn to scale unless otherwise indicated.

The word "exemplary" is used exclusively herein to mean "serving as an example, embodiment, or illustration." Any embodiment described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments.

The term "and/or" in this article is only an association relationship to describe the associated objects, indicating that there can be three kinds of relationships, for example, A and/or B, it can mean that A exists alone, A and B exist at the same time, and A and B exist independently B these three cases. In addition, the term "at least one" herein refers to any combination of any one of the plurality or at least two of the plurality, for example, including at least one of A, B, and C, and may mean including from A, B, and C. Any one or more elements selected from the set of B and C.

In addition, in order to better illustrate the present disclosure, numerous specific details are set forth in the following detailed description. It will be understood by those skilled in the art that the present disclosure may be practiced without certain specific details. In some instances, methods, means, components and circuits well known to those skilled in the art have not been described in detail so as not to obscure the subject matter of the present disclosure.

In the process of performing Computed Tomography (CT) on the human body, the bed plate of the CT instrument will become an artifact in the scanned CT image sequence. This kind of artifact will cause great interference in the 3D visualization of the human body by the computer-aided software (ie, the 3D human body model obtained from the CT image sequence) and the subsequent processing. This is because bed boards of various shapes will block the human body during 3D visualization, and when the organs in the human body are segmented, some special-shaped bed boards outside the human body may be identified as false positives.

In the related art, when performing human body segmentation on a CT image, the bed board in the CT image is mainly removed through threshold and morphological operations, and the human body part in the CT image is retained. Usually, the shape of the bed plate, the CT value of the bed plate in the CT image, and the uniformity of the CT value of the bed plate in the CT image are significantly different from those of the human body, which can be removed by thresholding and morphological manipulation. However, for some abnormal situations, related technologies cannot obtain accurate segmentation results. For example, a curved cortical bed board that fits closely with the human body is closely fitted with the human body in the CT image, the demarcation is not obvious, and the CT value is relatively close, so it is difficult to separate it from the human body. For another example, there are baffles on both sides of the bed board, and the arms of the human body are placed outside the baffles, and the baffles are clamped on both sides of the body. In this case, it is also difficult for related technologies to segment the bed board from the CT image. Among them, CT value is a calculation unit for measuring the density of a local tissue or organ in the human body, also known as Hounsfield Unit (HU).

In order to solve the technical problems similar to those described above, the embodiments of the present disclosure provide an image segmentation method and device, an electronic device, and a storage medium. By predicting the pixels belonging to the target object in the to-be-processed image, the corresponding The preliminary segmented image, according to the edge information of the target object in the image to be processed, in the preliminary segmented image, in the closed area included in the edge of the target object, the predicted object does not belong to the target object The pixel value of the pixel is adjusted to obtain the first segmentation result corresponding to the to-be-processed image, so that a more accurate and robust segmentation result can be obtained.

The executing subject of the image segmentation method may be an image segmentation device. For example, the image segmentation method may be performed by a terminal device or a server or other processing device. The terminal device may be a user equipment (User Equipment, UE), a mobile device, a user terminal, a terminal, a cellular phone, a cordless phone, a personal digital assistant (Personal Digital Assistant, PDA), a handheld device, a computing device, a vehicle-mounted device, or a wearable devices, etc. In some possible implementations, the image segmentation method may be implemented by a processor invoking computer-readable instructions stored in a memory.

Next, application scenarios of the image segmentation method provided by the embodiments of the present disclosure will be described.

In a possible implementation manner, referring to schematic diagram 1 of an application scenario of an image segmentation method shown in FIG. 1-1 , the image segmentation device 10 may include a processing device 11 and an image acquisition device 12. In this way, the image segmentation device 11 can pass The image acquisition device 12 acquires the image to be divided, and the processing device 12 performs segmentation processing on the to-be-divided image to obtain a first segmentation result. For example, the image segmentation device may be implemented as a CT machine, and a CT image to be segmented is acquired by a CT scanner, and image segmentation processing is performed on the acquired CT image to be segmented.

In another possible implementation manner, referring to the second schematic diagram of an application scenario of an image segmentation method shown in FIGS. 1-2 , the image segmentation apparatus 10 can receive the real-time collected images to be segmented transmitted by other devices 13 through the network 14 . , in this way, the image segmentation device 10 can perform segmentation processing on the received image to be segmented to obtain the first segmentation result. For example, the image segmentation device can be implemented as a smart phone, and the smart phone can receive the CT image to be segmented sent by the CT machine through the network, so that the smart phone can perform image segmentation processing on the received CT image to be segmented.

Based on the above application scenario, an image segmentation method provided by an embodiment of the present disclosure is described. FIG. 1 shows a flowchart of the image segmentation method provided by an embodiment of the present disclosure. As shown in FIG. 2, the image segmentation method includes step S11 and step S12.

In step S11, pixels belonging to the target object in the image to be processed are predicted, and a preliminary segmented image corresponding to the image to be processed is obtained.

In this embodiment of the present disclosure, the image to be processed may represent an image that needs to be segmented. The to-be-processed image may be a two-dimensional image or a three-dimensional image. In a possible implementation, the image to be processed may be a medical image. For example, the image to be processed may be a CT image, an MRI (Magnetic Resonance Imaging, magnetic resonance imaging) image, and the like. Of course, the to-be-processed image can also be any image that needs to be segmented other than medical images.

In this embodiment of the present disclosure, the target object may represent an object that needs to be segmented. For example, the target object may be a human body, an animal body, an organ of a human body, an organ of an animal body, or the like.

In the embodiment of the present disclosure, it can be predicted whether each pixel in the image to be processed belongs to the target object. For example, the probability that each pixel in the image to be processed belongs to the target object can be predicted. For any pixel in the image to be processed, if the probability of the pixel belonging to the target object is greater than or equal to the preset threshold, it can be determined that the pixel belongs to the target object; if the probability of the pixel belonging to the target object is less than the preset threshold, it can be determined The pixel does not belong to the target object. For example, the preset threshold may be 0.5.

In the embodiment of the present disclosure, according to the predicted pixels belonging to the target object in the to-be-processed image, a binarized preliminary segmented image corresponding to the to-be-processed image can be obtained. The size of the preliminary segmented image may be the same as that of the image to be processed. For example, if the height of the image to be processed is H and the width is W, the height of the preliminary segmented image is also H and the width is W. In a possible implementation manner, in the preliminary segmented image, the predicted pixel value of the pixel belonging to the target object is a first preset value, and the predicted pixel value of the pixel not belonging to the target object is the first preset value. The second preset value, the first preset value is not equal to the second preset value. For example, for any pixel in the image to be processed, if the pixel is predicted to belong to the target object, the pixel value of the pixel in the preliminarily segmented image is the first preset value; if it is predicted that the pixel does not belong to the target object, the image is preliminarily segmented The pixel value of the pixel in is the second preset value. For example, the first preset value is 1 and the second preset value is 0, that is, the predicted pixel value of the pixel belonging to the target object in the preliminary segmented image is 1, and the predicted pixel value of the pixel not belonging to the target object is 0 . The embodiments of the present disclosure do not limit the values of the first preset value and the second preset value, as long as the first preset value and the second preset value are different. For another example, the first preset value may be 0, and the second preset value may be 255.

In a possible implementation manner, before predicting the pixels belonging to the target object in the image to be processed, the method further includes: training a neural network according to the training image and the labeled data of the training image, wherein the training The labeling data of the image includes the true value of the pixels belonging to the target object in the training image; the predicting the pixels belonging to the target object in the image to be processed, and obtaining a preliminary segmented image corresponding to the image to be processed, including: The processed image is input into the neural network, and the neural network is used to predict the information of the pixels belonging to the target object in the to-be-processed image; according to the information of the pixels belonging to the target object in the to-be-processed image, the to-be-processed image is obtained. The image corresponds to the preliminary segmented image.

As an example of this implementation, the labeled data of the training image may include a mask corresponding to the training image, and the size of the mask corresponding to the training image may be the same as the training image. If in the training image, the true value of any pixel belongs to the target object, in the mask corresponding to the training image, the pixel value of the pixel may be a first preset value, for example, the first preset value The value can be 1; if in the training image, the true value of the pixel does not belong to the target object, then in the mask corresponding to the training image, the pixel value of the pixel can be the second preset value, for example , the second preset value may be 0. Of course, the labeled data of the training image is not limited to be represented by a mask. For example, the labeled data of the training image may also be represented by a matrix, a table, or the like.

In this implementation manner, the training image may be input into the neural network, and the predicted segmentation result of the training image may be output via the neural network, wherein the predicted segmentation result of the training image may include The probability of each pixel belonging to the target object; according to the labeled data of the training image and the predicted segmentation result of the training image, the value of the loss function corresponding to the training image is obtained; according to the value of the loss function corresponding to the training image value to train the neural network.

As an example of this implementation, the value of the Dice loss function can be obtained according to the predicted segmentation result of the training image obtained by the neural network and the labeling data of the training image. For example, the predicted segmentation result of the training image obtained by the neural network is P, the labeled data of the training image is M, and the value of the Dice loss function

In other examples, a loss function such as a cross-entropy loss function may also be employed.

As an example of this implementation, the value of the loss function can be passed to each parameter of the neural network layer by layer through reverse derivation, and adaptive matrix estimation (Adaptive moment estimation, Adam) can be used (for example, the learning rate is 0.0003), Stochastic Gradient Descent (SGD) and other optimizers to update the parameters of the neural network.

As an example of this implementation manner, the information of pixels in the image to be processed that are predicted by the neural network and belong to the target object may include the probability that each pixel in the image to be processed belongs to the target object. In this example, the obtaining a preliminary segmented image corresponding to the to-be-processed image according to the information of the pixels belonging to the target object in the to-be-processed image may include: for any pixel, if the If the probability that the pixel belongs to the target object is greater than or equal to the preset threshold, the pixel value of the pixel in the preliminary segmented image corresponding to the image to be processed is the first preset value; if the pixel in the image to be processed belongs to the target object If the probability is less than the preset threshold, the pixel value of the pixel in the preliminary segmented image corresponding to the image to be processed is the second preset value.

As another example of this implementation manner, the information of the pixels belonging to the target object in the image to be processed predicted by the neural network may include position information of the pixels belonging to the target object in the image to be processed. In this example, the obtaining a preliminary segmented image corresponding to the to-be-processed image according to the information of the pixels belonging to the target object in the to-be-processed image may include: for any pixel, if the to-be-processed image is The position information of the pixel belonging to the target object includes the position of the pixel, then the pixel value of the pixel in the preliminary segmented image corresponding to the image to be processed is the first preset value; if the image to be processed belongs to the target object If the position information of the pixel does not include the position of the pixel, the pixel value of the pixel in the preliminary segmented image corresponding to the image to be processed is the second preset value.

As an example of this implementation, the neural network may be a deep learning-based neural network. For example, the neural network may be a U-shaped convolutional neural network. FIG. 3 shows a schematic diagram of a U-shaped convolutional neural network in an embodiment of the present disclosure. In Figure 3, the data flow is from left to right, and the U-shaped convolutional neural network includes a compression process and a decompression process. As shown in Figure 3, after the image to be processed can be cropped or scaled to a size of 512×512, the U-shaped convolutional neural network can be input, and the human body part in the image to be processed can be fitted by the U-shaped convolutional neural network. Finally, output the preliminary segmented image. In the example shown in Figure 3, four "convolution-regularization-activation-pooling" operations can be performed on the image to be processed, the number of image channels is doubled during each convolution, the image size is halved during pooling, and the image The number of channels is increased from 32 to 256, and the image size is reduced from 512 × 512 to 64 × 64; 4 more "upsampling-convolution-regularization-activation-convolution-regularization-activation" operations are performed to restore the image is the original size, in which, the feature maps of the same size in the previous compression process need to be merged before each upsampling, and the number of channels is halved for each convolution, where the merging can use the concatenate operation; The convolution and activation operations restore the number of image channels to 1 and normalize the image. Among them, "convolution-regularization-activation" can be replaced with residual block (Residual Block), depth convolution block (Inception Block), dense block (Dense Block), etc. Pooling can be either max pooling or average pooling, or it can be replaced by a convolutional layer with a stride of 2.

In one example, the training image is a two-dimensional CT image, and the neural network is a two-dimensional convolutional neural network.

As an example of this implementation, training images can be augmented. For example, the training image can be randomly scaled by a factor of 0.6 to 1.4, and then cropped from the center of the scaled image at a size of 512 × 512 to obtain training images of the same size at different scales. Correspondingly, do the same for the mask corresponding to the training image.

As an example of this implementation, training images can be divided into training and validation sets. For example, training images can be split into training and validation sets in a 4:1 ratio.

As an example of this implementation, the neural network may be repeatedly trained using training images until the loss of the neural network on the validation set falls below 0.03.

Since the related technology uses operations such as morphology to segment the image, it is necessary to introduce a large number of hyperparameters, such as the threshold selected during binarization, the number of opening/closing operations, and the structure selected during erosion/dilation. For different human body parts (head, torso, hands, pedals), the threshold value needs to be changed to obtain normal segmentation results. In this implementation, the target object in the training image is segmented through a neural network, which can be widely used in similar tasks without setting hyperparameters, so the robustness is high.

As an example of this implementation, the training image is an electronic computed tomography CT image; the training of the neural network according to the training image and the labeled data of the training image includes: according to a preset CT value range, performing The pixel values of the training image are normalized to obtain a normalized training image; the neural network is trained according to the normalized training image and the labeled data of the training image.

In one example, the preset CT value range may be determined according to the CT value range of the target object. For example, if the target object is the human body, the preset CT value range may be set to [-500, 1200] according to the CT value range of the human body organs.

In one example, performing normalization processing on pixel values of the training image according to a preset CT value range to obtain a normalized training image, including: for any pixel in the training image, According to the preset CT value range, the pixel value of the pixel is preprocessed to obtain the preprocessed pixel value of the pixel, wherein the preprocessed pixel value of the pixel is within the preset range. The ratio of the first difference to the second difference is taken as the normalized pixel value of the pixel, wherein the first difference is equal to the preprocessed pixel value of the pixel The difference from the lower boundary value of the preset CT value range, the second difference is equal to the difference between the upper boundary value of the preset CT value range and the preprocessed pixel value of the pixel . For example, the preprocessed pixel value of the pixel is h, the lower boundary value of the preset CT value range is h _min , and the upper boundary value of the preset CT value range is h _max , then the The normalized pixel value of the pixel can be equal to

According to the normalized pixel value of each pixel of the training image, a normalized training image can be obtained. That is, in the normalized training image, the pixel value of any pixel is the normalized pixel value of the pixel.

Wherein, for any pixel in the training image, performing preprocessing on the pixel value of the pixel according to the preset CT value range to obtain the preprocessed pixel value of the pixel may include : For any pixel in the training image, if the pixel value of the pixel is smaller than the lower boundary value of the preset CT value range, the lower boundary value can be used as the preprocessed pixel value of the pixel ; If the pixel value of the pixel is greater than the upper boundary value of the preset CT value range, then the upper boundary value can be used as the preprocessed pixel value of the pixel; if the pixel value of the pixel is in Within the preset CT value range, the pixel value of the pixel may be used as the preprocessed pixel value of the pixel. For example, the preset CT value range is [-500, 1200], the lower boundary value of the preset CT value range is -500, and the upper boundary value of the preset CT value range is 1200. If the pixel value of a pixel in the training image is -505, you can use -500 as the preprocessed pixel value of the pixel; if the pixel value of a pixel in the training image is 1250, you can use 1200 as the pixel value after preprocessing. The preprocessed pixel value of the pixel; if the pixel value of a certain pixel in the training image is 800, 800 can be used as the preprocessed pixel value of the pixel.

In this example, according to the preset CT value range, the pixel values of the training image are normalized to obtain a normalized training image, and according to the normalized training image and the training image The labeled data is used to train the neural network, thereby helping to reduce the computational load of the neural network and improve the convergence speed of the neural network.

In step S12, according to the edge information of the target object in the to-be-processed image, in the preliminary segmented image, it is predicted that in the closed area included in the edge of the target object, the target object does not belong to the target object The pixel value of the pixel is adjusted to obtain the first segmentation result corresponding to the image to be processed.

In this embodiment of the present disclosure, an edge detection method may be used to determine the edge information of the target object in the image to be processed. For example, edge detection methods such as Canny algorithm and Sobel algorithm may be used to determine the edge information of the target object in the image to be processed. The edge information of the target object in the image to be processed may include position information of pixels belonging to the edge of the target object in the image to be processed.

In a possible implementation manner, the first segmentation result may be used as the final segmentation result corresponding to the image to be processed.

In the embodiment of the present disclosure, a preliminary segmented image corresponding to the to-be-processed image is obtained by predicting the pixels belonging to the target object in the to-be-processed image, and according to the edge information of the target object in the to-be-processed image, in the preliminary segmented image In dividing the image, the pixel values of the predicted pixels that do not belong to the target object in the enclosed area included in the edge of the target object are adjusted to obtain the first segmentation result corresponding to the image to be processed. When the target object is a human body or an animal body, the pixels inside the organs of the target object can also be segmented as belonging to the target object, so that a more accurate and robust segmentation result can be obtained. For example, if the image to be processed is a CT image, and the target object is a human body, the CT image is segmented by using the image segmentation method provided by the embodiment of the present disclosure, so that the human body part in the CT image can be accurately segmented, and the outside of the human body in the CT image can be accurately removed. interference (e.g. bed board, ventilator lines, head fixtures, etc.).

In a possible implementation manner, according to the edge information of the target object in the to-be-processed image, in the preliminary segmented image, the predicted value in the enclosed area included in the edge of the target object is Adjusting the pixel values of the pixels that do not belong to the target object to obtain the first segmentation result corresponding to the image to be processed, comprising: adjusting the pixel value in the preliminary segmented image to the value of the closed area of the second preset value. The pixel value is adjusted to the first preset value to obtain a filled preliminary segmented image; according to the edge information of the target object in the to-be-processed image, the pixel value of the filled preliminary segmented image is adjusted, A first segmentation result corresponding to the to-be-processed image is obtained.

Since the interior of some organs (such as the lungs, the interior of the digestive tract) contains air with low density, and the exterior of the target object (such as the human body) is also air, in the preliminary segmentation image, the interior of these organs may be segmented as belonging to the background part . In this implementation manner, a filled preliminary segmented image is obtained by adjusting the pixel value of the closed area whose pixel value is the second preset value in the preliminary segmented image to the first preset value, thereby enabling the The first segmentation result corresponding to the processed image covers the inside of the organ of the target object, for example, covers the lung parenchyma such as the lung, the inside of the digestive tract (eg, the gastrointestinal tract), and the like. That is, by adopting the above-mentioned implementation manner, the missing holes in the target object (for example, in the human body) after image segmentation can be filled. By adjusting the pixel values of the filled preliminary segmented image according to the edge information of the target object in the to-be-processed image, the first segmentation result corresponding to the to-be-processed image can be obtained, thereby reducing the The background part in the image to be processed (ie the part that does not belong to the target object) is divided into the probability of belonging to the target object.

As an example of this implementation, adjusting the pixel value of the enclosed area with the pixel value of the second preset value in the preliminary segmented image to the first preset value to obtain the filled preliminary segmented image , comprising: splicing edges of a preset width around the preliminary segmented image to obtain a preliminarily segmented image after splicing, wherein the pixel value of the pixels of the edge of the spliced preset width is the second preset value ; Select the pixel of the image edge of the preliminarily segmented image after the splicing as a seed point, and perform a flood filling operation on the preliminarily segmented image after the splicing to obtain the preliminary segmented image after the filling.

In this example, the preset width may be greater than or equal to 1 pixel. For example, the preset width may be 1 pixel. FIG. 4 shows a schematic diagram of splicing edges of a preset width around a preliminary segmented image to obtain a preliminarily segmented image after splicing. In the example shown in FIG. 4, the preset width is 1 pixel. As shown in FIG. 4 , edges with preset widths may be spliced around the preliminary segmented image. In other examples, a side with a preset width can also be spliced on one side, two sides or three sides of the preliminary segmented image.

In this example, the pixels of the image edge of the preliminarily segmented image after splicing may refer to the pixels on the edge of the preliminarily segmented image after splicing, for example, the uppermost pixel of the preliminarily segmented image after splicing pixel, bottommost pixel, leftmost pixel, rightmost pixel, etc. For example, the pixel in the upper left corner of the stitched preliminary segmented image may be used as the seed point.

In this example, by splicing edges of a preset width around the preliminary segmented image, and then selecting the pixels of the image edge of the preliminarily segmented image after splicing as seed points, the seeds of the flood filling operation can be guaranteed. The point belongs to the background part (ie the part that does not belong to the target object), so that the first segmentation result corresponding to the image to be processed can cover the inside of the organ of the target object, thereby obtaining a more accurate segmentation result.

As an example of this implementation, the pixel value of the filled preliminary segmented image is adjusted according to the edge information of the target object in the to-be-processed image to obtain the first corresponding to the to-be-processed image. The segmentation result includes: according to the edge information of the target object in the to-be-processed image, determining the maximum connected domain included in the edge of the target object in the preliminarily segmented image after filling; The pixel values of the pixels outside the maximum connected region in the segmented image are adjusted to the second preset value to obtain the first segmentation result corresponding to the to-be-processed image. According to this example, false positive regions that are not connected to the target object can be eliminated, thereby greatly reducing the probability of erroneously classifying the background part as belonging to the target object, thereby improving the accuracy of image segmentation. For example, if the target object is a human body, according to this example, false positive regions that are not connected to the human body can be eliminated, thereby greatly reducing the probability that the background part (eg, bed board, etc.) is erroneously classified as belonging to the human body.

In a possible implementation manner, after obtaining the first segmentation result corresponding to the image to be processed, the method further includes: acquiring images adjacent to the image to be processed and the adjacent images The corresponding second segmentation result; according to the pixel value of the pixel at the same position in the image to be processed and the adjacent image, and the second segmentation result, adjust the first segmentation result to obtain the to-be-processed The third segmentation result corresponding to the image.

In this implementation manner, the image adjacent to the image to be processed may be an image belonging to the same image sequence as the image to be processed and adjacent to the image to be processed. For example, the image to be processed is a CT image, and the adjacent images may be images belonging to the same CT image sequence as the image to be processed and adjacent to the image to be processed. The second segmentation result may refer to the final segmentation result corresponding to the adjacent images.

As an example of this implementation, the first segmentation result is adjusted according to the pixel value of the pixel at the same position in the to-be-processed image and the adjacent image, and the second segmentation result, to obtain the The third segmentation result corresponding to the image to be processed includes: according to the adjacent images, the second segmentation result belongs to the target object and is in the same position as the image to be processed. For pixels whose difference value is less than or equal to a third preset value, the first segmentation result is adjusted to obtain a third segmentation result corresponding to the image to be processed.

In this example, the difference between the pixel values of the adjacent image and the image to be processed at the same position may refer to the difference between the normalized pixel values of the adjacent image and the image to be processed at the same position . For example, the third preset value may be 0.1. Of course, it is also possible to compare the original pixel values of adjacent images and the image to be processed at the same position.

In this example, by adjusting the pixel according to the second segmentation result that belongs to the target object and has a pixel value at the same position as the image to be processed whose pixel value difference is less than or equal to a third preset value The first segmentation result is obtained, and the third segmentation result corresponding to the image to be processed is obtained, so that the image to be processed can be classified according to the segmentation result corresponding to the pixel relatively related to the image to be processed in the adjacent images. The corresponding first segmentation result is adjusted, thereby helping to improve the accuracy of the final segmentation result corresponding to the image to be processed. The segmentation result corresponding to any pixel in the adjacent images may refer to whether the pixel belongs to the target object in the second segmentation result.

In an example, according to the adjacent images, the difference between the pixel values of the target object in the second segmentation result and at the same position as the image to be processed is less than or equal to the first three preset values of pixels, adjusting the first segmentation result to obtain a third segmentation result corresponding to the image to be processed, including: according to the pixel value at the same position in the image to be processed and the adjacent image Pixels whose difference value is less than or equal to the third preset value are obtained to obtain a first pixel set; according to the pixels belonging to the target object in the second segmentation result of the first pixel set, a second pixel set is obtained; The pixels of the second pixel set in the first segmentation result are adjusted to belong to the target object, and a third segmentation result corresponding to the image to be processed is obtained.

In this example, the difference between the pixel values of any pixel in the first pixel set in the to-be-processed image and the adjacent image is less than or equal to a third preset value. The difference between the pixel values of any pixel in the second pixel set in the to-be-processed image and the adjacent image is less than or equal to a third preset value, and belongs to the target in the second segmentation result object. For example, if the first segmentation result is A, the first pixel set is B, and the second segmentation result is C, the third segmentation result may be S=A∪(B∩C).

In the above example, the first pixel set is obtained according to the pixels whose difference between the pixel values in the image to be processed and the adjacent images at the same position is less than or equal to the third preset value, and according to the first pixel set. A pixel is concentrated in the pixels belonging to the target object in the second segmentation result to obtain a second pixel set, and the pixels of the second pixel set in the first segmentation result are adjusted to belong to the target object, The third segmentation result corresponding to the image to be processed is obtained, whereby the pixels corresponding to the image to be processed can be classified according to the pixels in the second segmentation result that belong to the target object and are relatively related to the image to be processed. The first segmentation result is adjusted, thereby helping to improve the accuracy of the final segmentation result corresponding to the image to be processed.

In a possible implementation manner, the third segmentation result may be used as the final segmentation result corresponding to the image to be processed.

FIG. 5 shows another flowchart of the image segmentation method provided by the embodiment of the present disclosure. The executing subject of the image segmentation method may be an image segmentation device. For example, the image segmentation method may be performed by a terminal device or a server or other processing device. The terminal device may be a user equipment, a mobile device, a user terminal, a terminal, a cellular phone, a cordless phone, a personal digital assistant, a handheld device, a computing device, a vehicle-mounted device, a wearable device, or the like. In some possible implementations, the image segmentation method may be implemented by a processor invoking computer-readable instructions stored in a memory. As shown in FIG. 5 , the image segmentation method includes steps S41 to S43.

In step S41, pixels belonging to the target object in the image to be processed are predicted, and a preliminary segmented image corresponding to the image to be processed is obtained.

In step S42, an image adjacent to the to-be-processed image and a second segmentation result corresponding to the adjacent image are acquired.

In this embodiment of the present disclosure, the image adjacent to the to-be-processed image may be an image that belongs to the same image sequence as the to-be-processed image and is adjacent to the to-be-processed image. For example, the image to be processed is a CT image, and the adjacent images may be images belonging to the same CT image sequence as the image to be processed and adjacent to the image to be processed. The second segmentation result may refer to the final segmentation result corresponding to the adjacent images.

In step S43, according to the pixel value of the pixel at the same position in the image to be processed and the adjacent image, and the second segmentation result, the preliminary segmented image is adjusted to obtain a fourth corresponding image to be processed. Split result.

In the embodiment of the present disclosure, the continuity of the image to be processed and the second segmentation result can be ensured, thereby helping to obtain a smoother and more accurate three-dimensional segmentation result. For example, if the target object is a human body, the continuity of the image to be processed and the human body in the adjacent images can be ensured, thereby helping to obtain a smoother and more accurate three-dimensional human body segmentation result. For example, a segmentation result corresponding to each CT image in the CT image sequence can be obtained by using the embodiments of the present disclosure, thereby obtaining a smoother and more accurate three-dimensional human body segmentation result.

In a possible implementation manner, according to the pixel value of the pixel at the same position in the image to be processed and the adjacent image, and the second segmentation result, the preliminary segmented image is adjusted to obtain the The fourth segmentation result corresponding to the image to be processed includes: according to the adjacent images, the second segmentation result belongs to the target object and is in the same position as the image to be processed. For pixels whose difference value is less than or equal to the third preset value, the preliminary segmented image is adjusted to obtain a fourth segmentation result corresponding to the image to be processed.

In this implementation manner, the difference between the pixel values of the adjacent image and the image to be processed at the same position may refer to the normalized difference between the adjacent image and the image to be processed at the same position Difference of pixel values. For example, the third preset value may be 0.1. Of course, the original pixel values of the adjacent images and the image to be processed at the same position can also be compared.

In this implementation manner, according to the adjacent images, in the second segmentation result, the difference between the pixel values belonging to the target object and at the same position as the image to be processed is less than or equal to the first Pixels with three preset values, adjust the preliminary segmented image, and obtain a fourth segmentation result corresponding to the image to be processed. For the segmentation result, the preliminary segmented image corresponding to the to-be-processed image is adjusted, thereby helping to improve the accuracy of the final segmentation result corresponding to the to-be-processed image. The segmentation result corresponding to any pixel in the adjacent images may refer to whether the pixel belongs to the target object in the second segmentation result.

As an example of this implementation, in the adjacent images, the difference between the pixel values in the second segmentation result belonging to the target object and at the same position as the image to be processed is less than or a pixel equal to a third preset value, adjusting the preliminary segmented image to obtain a fourth segmentation result corresponding to the to-be-processed image, including: according to the to-be-processed image and the adjacent image at the same position; A pixel whose pixel value difference is less than or equal to a third preset value is obtained to obtain a first pixel set; and a second pixel set is obtained according to the pixels of the first pixel set that belong to the target object in the second segmentation result ; Adjust the pixels of the second pixel set in the preliminary segmented image to belong to the target object, and obtain a fourth segmentation result corresponding to the to-be-processed image.

In this example, a first pixel set is obtained according to a pixel whose difference between the pixel value of the image to be processed and the pixel value of the adjacent image at the same position is less than or equal to a third preset value. A pixel is concentrated in the pixels belonging to the target object in the second segmentation result to obtain a second pixel set, and the pixels of the second pixel set in the first segmentation result are adjusted to belong to the target object, The third segmentation result corresponding to the image to be processed is obtained, whereby the pixels corresponding to the image to be processed can be classified according to the pixels in the second segmentation result that belong to the target object and are relatively related to the image to be processed. The first segmentation result is adjusted, thereby helping to improve the accuracy of the final segmentation result corresponding to the image to be processed.

In a possible implementation manner, the fourth segmentation result may be used as the final segmentation result corresponding to the image to be processed.

In a possible implementation manner, after obtaining the fourth segmentation result corresponding to the to-be-processed image, the method further includes: according to the edge information of the target object in the to-be-processed image, in the In the fourth segmentation result, the pixel values of the predicted pixels not belonging to the target object in the enclosed area included in the edge of the target object are adjusted to obtain the fifth segmentation result corresponding to the image to be processed.

As an example of this implementation, according to the edge information of the target object in the to-be-processed image, in the fourth segmentation result, in the closed area included in the edge of the target object, the predicted Adjusting the pixel values of pixels that do not belong to the target object to obtain a fifth segmentation result corresponding to the to-be-processed image, including: closing the pixel value in the fourth segmentation result to the second preset value The pixel value of the area is adjusted to the first preset value, and a filled preliminary segmented image corresponding to the fourth segmentation result is obtained; according to the edge information of the target object in the to-be-processed image, the filled The pixel value of the preliminary segmented image is adjusted to obtain the fifth segmentation result corresponding to the to-be-processed image.

In an example, adjusting the pixel value of the enclosed area with the pixel value of the second preset value in the fourth segmentation result to the first preset value, to obtain a corresponding pixel value of the fourth segmentation result The filled preliminary segmented image includes: splicing edges of a preset width around the fourth segmentation result to obtain a fourth segmentation result after splicing, wherein the pixel values of the pixels of the spliced edges of the preset width are is the second preset value; select the pixels of the image edge of the fourth segmentation result after splicing as a seed point, and perform a flood filling operation on the fourth segmentation result after splicing to obtain the fourth segmentation The result corresponds to the padded preliminary segmented image.

In an example, according to the edge information of the target object in the to-be-processed image, in the fourth segmentation result, in the closed area included in the edge of the target object, the predicted object does not belong to the target object. adjusting the pixel values of the pixels of the target object to obtain a fifth segmentation result corresponding to the image to be processed, including: determining the filled preliminary segmentation according to the edge information of the target object in the image to be processed the maximum connected domain included in the edge of the target object in the image; adjust the pixel values of the pixels outside the maximum connected domain in the filled preliminary segmented image to the second preset value to obtain the The fifth segmentation result corresponding to the image to be processed.

Among them, "According to the edge information of the target object in the image to be processed, in the fourth segmentation result, in the closed area included in the edge of the target object, the predicted object that does not belong to the target object is The specific implementation method of adjusting the pixel value of the pixel to obtain the fifth segmentation result corresponding to the image to be processed” is the same as the above “According to the edge information of the target object in the image to be processed, in the preliminary segmentation In the image, adjust the pixel values of the predicted pixels that do not belong to the target object in the enclosed area included in the edge of the target object, and obtain the first segmentation result corresponding to the image to be processed". similar, and will not be repeated here.

In a possible implementation manner, the fifth segmentation result may be used as the final segmentation result corresponding to the image to be processed.

It can be understood that each method embodiment mentioned in the present disclosure can be combined with each other to form a combined embodiment without violating the principle and logic. Due to space limitations, the present disclosure will not repeat them. Those skilled in the art can understand that, in the above method in the specific implementation manner, the specific execution order of each step should be determined by its function and internal logic.

The embodiments of the present disclosure are described below through a specific application scenario.

First, get the training image and the mask corresponding to the training image. The training image is a CT image of the human body. Here, the preset CT value range can be set to [-500, 1200] according to the CT values of all the tissues and organs of the human body. In this way, all the tissues and organs of the human body are covered.

Based on the preset CT value range, any pixel in the training image is preprocessed to obtain the preprocessed pixel value. Specifically, for any pixel in the training image, if the pixel value of the pixel is smaller than the lower boundary value of the preset CT value range, the lower boundary value may be used as the preprocessed value of the pixel. pixel value; if the pixel value of the pixel is greater than the upper boundary value of the preset CT value range, the upper boundary value can be used as the preprocessed pixel value of the pixel; if the pixel value of the pixel is If the value is within the preset CT value range, the pixel value of the pixel may be used as the preprocessed pixel value of the pixel. For example, if the pixel value of a pixel in the training image is -505, you can use -500 as the preprocessed pixel value of the pixel; if the pixel value of a pixel in the training image is 1250, you can use -500 as the pixel value after preprocessing. 1200 is used as the preprocessed pixel value of the pixel; if the pixel value of a certain pixel in the training image is 800, 800 can be used as the preprocessed pixel value of the pixel.

Then, according to the preset CT value range, the pixel value of any pixel of the training image is normalized to obtain a normalized training image. Here, formula (1) can be used to normalize the pixel value of any pixel of the training image:

Wherein, h is the preprocessed pixel value of the pixel, h _min is the lower boundary value of the preset CT value range, and h _max is the upper boundary value of the preset CT value range. In this way, by performing the above processing on each pixel in the training image, a normalized training image can be obtained.

Here, the normalized training images can be augmented. For example, the normalized training image can be randomly scaled by a factor of 0.6 to 1.4, and then cropped from the center of the scaled image at a size of 512 × 512 to obtain training images of the same size at different scales. Correspondingly, do the same for the mask corresponding to the training image.

Further, the normalized and augmented training images can be divided into a training set and a validation set. For example, the processed training images can be divided into training and validation sets in a 4:1 ratio.

In this way, the U-shaped convolutional neural network can be repeatedly trained by using the training set until the loss of the U-shaped convolutional neural network on the verification set drops below 0.03, and a trained U-shaped convolutional neural network is obtained.

In practical applications, the CT image to be processed is obtained, and the CT image to be processed is input into the U-shaped convolutional neural network after training, and the U-shaped convolutional neural network is used to predict whether the CT image to be processed belongs to Information of the pixels of the target object; according to the information of the pixels belonging to the target object in the CT image to be processed, a preliminary segmented image corresponding to the CT image to be processed is obtained.

After the preliminary segmented image is obtained, edges with a width of 1 pixel can be spliced around the preliminary segmented image to obtain a preliminarily segmented image after splicing; the pixel in the upper left corner of the preliminarily segmented image after splicing is selected as a seed point, and perform a flood filling operation on the preliminarily segmented image after splicing to obtain a preliminarily segmented image after filling. Here, according to the edge information of the target object in the CT image to be processed, the maximum connected domain included in the edge of the target object in the preliminarily segmented image after filling can be determined; The pixel values of the pixels outside the maximum connected region in the image are adjusted to the second preset value to obtain the first segmentation result corresponding to the CT image to be processed. After the first segmentation result is obtained, an image adjacent to the CT image to be processed and a second segmentation result corresponding to the adjacent image may be obtained.

Further, the first pixel set can be obtained according to the pixel whose difference between the CT image to be processed and the pixel value at the same position in the adjacent image is less than or equal to the third preset value; according to the first pixel Concentrating on the pixels belonging to the target object in the second segmentation result to obtain a second pixel set; adjusting the pixels of the second pixel set in the first segmentation result to belong to the target object to obtain the The third segmentation result corresponding to the CT image to be processed.

In addition, the present disclosure also provides image segmentation devices, electronic equipment, computer-readable storage media, and programs, all of which can be used to implement any image segmentation method provided by the present disclosure. For the corresponding technical solutions and technical effects, please refer to the corresponding section of the method. record, without further elaboration.

FIG. 6 shows a block diagram of an image segmentation apparatus provided by an embodiment of the present disclosure. As shown in FIG. 6 , the image segmentation device includes: a first segmentation part 51, configured to predict pixels belonging to the target object in the image to be processed, and obtain a preliminary segmented image corresponding to the image to be processed; a first adjustment part 52, It is configured to, according to the edge information of the target object in the to-be-processed image, in the preliminary segmented image, perform an analysis of the predicted pixels that do not belong to the target object in the enclosed area included in the edge of the target object. The pixel values are adjusted to obtain the first segmentation result corresponding to the image to be processed.

In a possible implementation manner, in the preliminary segmented image, the predicted pixel value of the pixel belonging to the target object is a first preset value, and the predicted pixel value of the pixel not belonging to the target object is the first preset value. the second preset value; the first adjustment part 52 is configured to adjust the pixel value of the closed area whose pixel value in the preliminary segmented image is the second preset value to the first preset value, to obtain The filled preliminary segmented image; according to the edge information of the target object in the to-be-processed image, the pixel values of the filled preliminary segmented image are adjusted to obtain a first segmentation result corresponding to the to-be-processed image.

In a possible implementation manner, the first adjustment part 52 is configured to splicing edges of a preset width around the preliminary segmented image to obtain a preliminarily segmented image after splicing, wherein the spliced preset The pixel value of the pixel of the side of the width is the second preset value; the pixel of the image edge of the spliced preliminary segmented image is selected as the seed point, and the flooded filling operation is performed on the spliced preliminary segmented image, Obtain the preliminarily segmented image after filling.

In a possible implementation manner, the first adjustment part 52 is configured to determine the edge of the target object in the filled preliminary segmented image according to the edge information of the target object in the image to be processed The maximum connected domain included; the pixel values of the pixels outside the maximum connected domain in the filled preliminary segmented image are adjusted to the second preset value to obtain the first segment corresponding to the image to be processed. result.

In a possible implementation manner, the apparatus further includes: a second acquisition part, configured to acquire an image adjacent to the image to be processed and a second segmentation result corresponding to the adjacent image; a third adjustment part, configured to adjust the first segmentation result according to the pixel value of the pixel at the same position in the to-be-processed image and the adjacent image, and the second segmentation result, to obtain the corresponding pixel value of the to-be-processed image The third segmentation result.

In a possible implementation manner, the third adjustment part is configured to be based on pixel values in the adjacent images that belong to the target object and are at the same position as the image to be processed in the second segmentation result For pixels whose difference value is less than or equal to a third preset value, adjust the first segmentation result to obtain a third segmentation result corresponding to the image to be processed.

In a possible implementation manner, the third adjustment part is configured to be smaller than or equal to a third preset value according to the difference between the pixel values in the image to be processed and the pixel values in the adjacent images at the same position. pixels to obtain a first pixel set; according to the first pixel set in the pixels belonging to the target object in the second segmentation result, a second pixel set is obtained; the second pixel set in the first segmentation result The pixels of the set are adjusted to belong to the target object, and a third segmentation result corresponding to the image to be processed is obtained.

In a possible implementation manner, the apparatus further includes: a training part configured to train a neural network according to a training image and labeling data of the training image, wherein the labeling data of the training image includes The true value of the pixels belonging to the target object; the first segmentation part 51 is configured to input the image to be processed into the neural network, and predict the information of the pixels belonging to the target object in the image to be processed through the neural network ; Obtain a preliminary segmented image corresponding to the to-be-processed image according to the information of the pixels belonging to the target object in the to-be-processed image.

In a possible implementation manner, the training image is an electronic computed tomography CT image; the training part is configured to perform normalization processing on the pixel values of the training image according to a preset CT value range, A normalized training image is obtained; the neural network is trained according to the normalized training image and the labeled data of the training image.

In the embodiment of the present disclosure, a preliminary segmented image corresponding to the to-be-processed image is obtained by predicting the pixels belonging to the target object in the to-be-processed image, and according to the edge information of the target object in the to-be-processed image, in the preliminary segmented image, the target Adjust the pixel values of the predicted pixels that do not belong to the target object in the enclosed area included in the edge of the object, and obtain the first segmentation result corresponding to the image to be processed, so that a more accurate and robust segmentation can be obtained. result.

FIG. 7 shows another block diagram of an image segmentation apparatus provided by an embodiment of the present disclosure. As shown in FIG. 7 , the image segmentation device includes: a second segmentation part 61, configured to predict pixels belonging to the target object in the image to be processed, and obtain a preliminary segmented image corresponding to the image to be processed; a first acquisition part 62, is configured to obtain the image adjacent to the image to be processed and the second segmentation result corresponding to the adjacent image; the second adjustment part 63 is configured to be the same as the adjacent image according to the image to be processed The pixel value of the pixel at the position and the second segmentation result are used to adjust the preliminary segmented image to obtain a fourth segmentation result corresponding to the to-be-processed image.

In a possible implementation manner, the second adjustment part 63 is configured to, according to the adjacent images, belong to the target object in the second segmentation result, and are in the same range as the to-be-processed image. If the difference between the pixel values at the same position is less than or equal to the third preset value, the preliminary segmented image is adjusted to obtain a fourth segmented result corresponding to the image to be processed.

In a possible implementation manner, the second adjustment part 63 is configured to be less than or equal to a third preset value according to the difference between the pixel values in the image to be processed and the pixel values in the adjacent images at the same position obtain the first pixel set; according to the pixels belonging to the target object in the second segmentation result, the second pixel set is obtained; the second pixel set in the preliminary segmented image is The pixels of the set are adjusted to belong to the target object, and a fourth segmentation result corresponding to the image to be processed is obtained.

In a possible implementation manner, the apparatus further includes: a fourth adjustment part, configured to, according to the edge information of the target object in the image to be processed, in the fourth segmentation result, perform an adjustment on the target The pixel values of the predicted pixels not belonging to the target object in the enclosed area included in the edge of the object are adjusted to obtain a fifth segmentation result corresponding to the image to be processed.

In a possible implementation manner, the fourth adjustment part is configured to adjust the pixel value of the enclosed area whose pixel value is the second preset value in the fourth segmentation result to the first preset value value to obtain the filled preliminary segmented image corresponding to the fourth segmentation result; according to the edge information of the target object in the to-be-processed image, adjust the pixel values of the filled preliminary segmented image to obtain the the fifth segmentation result corresponding to the image to be processed.

In a possible implementation manner, the fourth adjustment part is configured to splicing edges of a preset width around the fourth segmentation result to obtain a spliced fourth segmentation result, wherein the spliced The pixel value of the pixel of the side of the width is set to the second preset value; the pixel of the image edge of the fourth segmentation result after the splicing is selected as the seed point, and the fourth segmentation result after the splicing is flooded A filling operation is performed to obtain a filled preliminary segmented image corresponding to the fourth segmentation result.

In a possible implementation manner, the fourth adjustment part is configured to determine, according to edge information of the target object in the image to be processed, where the edge of the target object in the filled preliminary segmented image is located. The maximum connected domain included; the pixel values of the pixels outside the maximum connected domain in the filled preliminary segmented image are adjusted to the second preset value to obtain the fifth segmentation result corresponding to the image to be processed .

In some embodiments, the functions or included parts of the apparatus provided by the embodiments of the present disclosure may be configured to execute the methods described in the above method embodiments, and the specific implementation and technical effects may refer to the above method embodiments. It is concise and will not be repeated here.

In the embodiments of the present disclosure and other embodiments, a "part" may be a part of a circuit, a part of a processor, a part of a program or software, etc., of course, a unit, a module or a non-modularity.

Embodiments of the present disclosure further provide a computer-readable storage medium, on which computer program instructions are stored, and when the computer program instructions are executed by a processor, the foregoing method is implemented. Wherein, the computer-readable storage medium may be a non-volatile computer-readable storage medium, or may be a volatile computer-readable storage medium.

An embodiment of the present disclosure further provides a computer program, including computer-readable code, when the computer-readable code is executed in an electronic device, the execution of the processor in the electronic device is configured to implement the above-mentioned image segmentation method.

Embodiments of the present disclosure further provide another computer program product for storing computer-readable instructions, which, when executed, cause the computer to perform the operations of the image segmentation method provided by any of the foregoing embodiments.

Embodiments of the present disclosure further provide an electronic device, including: one or more processors; a memory for storing executable instructions; wherein the one or more processors are configured to invoke executable instructions stored in the memory instruction to execute the above method.

The electronic device may be provided as a terminal, server or other form of device.

FIG. 8 shows a block diagram of an electronic device 800 provided by an embodiment of the present disclosure. For example, electronic device 800 may be a mobile phone, computer, digital broadcast terminal, messaging device, game console, tablet device, medical device, fitness device, personal digital assistant, etc. terminal.

8, an electronic device 800 may include one or more of the following components: a processing component 802, a memory 804, a power supply component 806, a multimedia component 808, an audio component 810, an input/output (I/O) interface 812, a sensor component 814 , and the communication component 816 .

The processing component 802 generally controls the overall operation of the electronic device 800, such as operations associated with display, phone calls, data communications, camera operations, and recording operations. The processing component 802 can include one or more processors 820 to execute instructions to perform all or some of the steps of the methods described above. Additionally, processing component 802 may include one or more sections that facilitate interaction between processing component 802 and other components. For example, processing component 802 may include a multimedia portion to facilitate interaction between multimedia component 808 and processing component 802.

Memory 804 is configured to store various types of data to support operation at electronic device 800 . Examples of such data include instructions for any application or method operating on electronic device 800, contact data, phonebook data, messages, pictures, videos, and the like. Memory 804 may be implemented by any type of volatile or nonvolatile storage device or combination thereof, such as static random access memory (SRAM), electrically erasable programmable read only memory (EEPROM), erasable Programmable Read Only Memory (EPROM), Programmable Read Only Memory (PROM), Read Only Memory (ROM), Magnetic Memory, Flash Memory, Magnetic or Optical Disk.

Power supply assembly 806 provides power to various components of electronic device 800 . Power supply components 806 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to electronic device 800 .

Multimedia component 808 includes a screen that provides an output interface between the electronic device 800 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from a user. The touch panel includes one or more touch sensors to sense touch, swipe, and gestures on the touch panel. The touch sensor may not only sense the boundaries of the touch or swipe action, but also detect the duration and pressure associated with the touch or swipe action. In some embodiments, the multimedia component 808 includes a front-facing camera and/or a rear-facing camera. When the electronic device 800 is in an operation mode, such as a shooting mode or a video mode, the front camera and/or the rear camera may receive external multimedia data. Each of the front and rear cameras can be a fixed optical lens system or have focal length and optical zoom capability.

Audio component 810 is configured to output and/or input audio signals. For example, audio component 810 includes a microphone (MIC) that is configured to receive external audio signals when electronic device 800 is in operating modes, such as calling mode, recording mode, and voice recognition mode. The received audio signal may be further stored in memory 804 or transmitted via communication component 816 . In some embodiments, audio component 810 also includes a speaker for outputting audio signals.

The I/O interface 812 provides an interface between the processing component 802 and a peripheral interface module, which may be a keyboard, a click wheel, a button, or the like. Buttons may include, but are not limited to: a home button, volume buttons, start button, and lock button.

Sensor assembly 814 includes one or more sensors for providing status assessment of various aspects of electronic device 800 . For example, the sensor assembly 814 can detect the on/off state of the electronic device 800, the relative positioning of the components, such as the display and the keypad of the electronic device 800, the sensor assembly 814 can also detect the electronic device 800 or one of the electronic device 800 Changes in the position of components, presence or absence of user contact with the electronic device 800 , orientation or acceleration/deceleration of the electronic device 800 and changes in the temperature of the electronic device 800 . Sensor assembly 814 may include a proximity sensor configured to detect the presence of nearby objects in the absence of any physical contact. Sensor assembly 814 may also include a light sensor, such as a complementary metal oxide semiconductor (CMOS) or charge coupled device (CCD) image sensor, for use in imaging applications. In some embodiments, the sensor assembly 814 may also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

Communication component 816 is configured to facilitate wired or wireless communication between electronic device 800 and other devices. The electronic device 800 can access a wireless network based on communication standards, such as wireless network (Wi-Fi), second generation mobile communication technology (2G), third generation mobile communication technology (3G), fourth generation mobile communication technology (4G) )/Long Term Evolution (LTE) of Universal Mobile Communications Technology, Fifth Generation Mobile Communications Technology (5G), or a combination thereof. In one exemplary embodiment, the communication component 816 receives broadcast signals or broadcast related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 816 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

In an exemplary embodiment, electronic device 800 may be implemented by one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable A programmed gate array (FPGA), controller, microcontroller, microprocessor or other element implementation is used to perform the above method.

In an exemplary embodiment, a non-transitory computer-readable storage medium is also provided, comprising a memory 804 of computer program instructions executable by the processor 820 of the electronic device 800 to perform the above-described method.

FIG. 9 shows a block diagram of an electronic device 1900 provided by an embodiment of the present disclosure. For example, the electronic device 1900 may be provided as a server. 9, electronic device 1900 includes processing component 1922, which further includes one or more processors, and a memory resource represented by memory 1932 for storing instructions executable by processing component 1922, such as applications. An application program stored in memory 1932 may include one or more portions each corresponding to a set of instructions. Additionally, the processing component 1922 is configured to execute instructions to perform the above-described methods.

The electronic device 1900 may also include a power supply assembly 1926 configured to perform power management of the electronic device 1900, a wired or wireless network interface 1950 configured to connect the electronic device 1900 to a network, and an input output (I/O) interface 1958 . The electronic device 1900 can operate based on an operating system stored in the memory 1932, such as a Microsoft server operating system (Windows ServerTM), a graphical user interface based operating system (Mac OS XTM) introduced by Apple, a multi-user multi-process computer operating system (UnixTM). ), Free and Open Source Unix-like Operating System (LinuxTM), Open Source Unix-like Operating System (FreeBSDTM) or similar.

In an exemplary embodiment, a non-volatile computer-readable storage medium is also provided, such as memory 1932 comprising computer program instructions executable by processing component 1922 of electronic device 1900 to perform the above-described method.

The present disclosure may be a system, method and/or computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement various aspects of the present disclosure.

A computer-readable storage medium may be a tangible device that can hold and store instructions for use by the instruction execution device. The computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples (non-exhaustive list) of computer readable storage media include: portable computer disks, hard disks, random access memory (RAM), read only memory (ROM), erasable programmable read only memory (EPROM) or flash memory), static random access memory (SRAM), portable compact disk read only memory (CD-ROM), digital versatile disk (DVD), memory sticks, floppy disks, mechanically coded devices, such as printers with instructions stored thereon Hole cards or raised structures in grooves, and any suitable combination of the above. Computer-readable storage media, as used herein, are not to be construed as transient signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (eg, light pulses through fiber optic cables), or through electrical wires transmitted electrical signals.

The computer readable program instructions described herein may be downloaded to various computing/processing devices from a computer readable storage medium, or to an external computer or external storage device over a network such as the Internet, a local area network, a wide area network, and/or a wireless network. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and/or edge servers. A network adapter card or network interface in each computing/processing device receives computer-readable program instructions from a network and forwards the computer-readable program instructions for storage in a computer-readable storage medium in each computing/processing device .

Computer program instructions for carrying out operations of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or instructions in one or more programming languages. Source or object code, written in any combination, including object-oriented programming languages, such as Smalltalk, C++, etc., and conventional procedural programming languages, such as the "C" language or similar programming languages. The computer readable program instructions may 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 implement. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (eg, using an Internet service provider through the Internet connect). In some embodiments, custom electronic circuits, such as programmable logic circuits, field programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), can be personalized by utilizing state information of computer readable program instructions. Computer readable program instructions are executed to implement various aspects of the present disclosure.

Aspects of the present disclosure are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the disclosure. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer readable program instructions.

These computer readable program instructions may be provided to a processor of a general purpose computer, special purpose computer or other programmable data processing apparatus to produce a machine that causes the instructions when executed by the processor of the computer or other programmable data processing apparatus , resulting in means for implementing the functions/acts specified in one or more blocks of the flowchart and/or block diagrams. These computer readable program instructions can also be stored in a computer readable storage medium, these instructions cause a computer, programmable data processing apparatus and/or other equipment to operate in a specific manner, so that the computer readable medium on which the instructions are stored includes An article of manufacture comprising instructions for implementing various aspects of the functions/acts specified in one or more blocks of the flowchart and/or block diagrams.

Computer readable program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other equipment to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other equipment to produce a computer-implemented process , thereby causing instructions executing on a computer, other programmable data processing apparatus, or other device to implement the functions/acts specified in one or more blocks of the flowcharts and/or block diagrams.

The flowchart and block diagrams in the Figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a section, segment, or portion of instructions that includes one or more functions for implementing the specified logical function(s) executable instructions. In some alternative implementations, the functions noted in the blocks may occur out of the order noted in the figures. For example, two blocks in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It is also noted that each block of the block diagrams and/or flowchart illustrations, and combinations of blocks in the block diagrams and/or flowchart illustrations, can be implemented in dedicated hardware-based systems that perform the specified functions or actions , or can be implemented in a combination of dedicated hardware and computer instructions.

The computer program product can be specifically implemented by hardware, software or a combination thereof. In an optional embodiment, the computer program product is embodied as a computer storage medium, and in another optional embodiment, the computer program product is embodied as a software product, such as a software development kit (Software Development Kit, SDK), etc. Wait.

Various embodiments of the present disclosure have been described above, and the foregoing descriptions are exemplary, not exhaustive, and not limiting of the disclosed embodiments. Numerous modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terminology used herein was chosen to best explain the principles of the various embodiments, the practical application or improvement over the technology in the marketplace, or to enable others of ordinary skill in the art to understand the various embodiments disclosed herein.

Industrial Applicability

In the embodiment of the present disclosure, a preliminary segmented image corresponding to the to-be-processed image is obtained by predicting the pixels belonging to the target object in the to-be-processed image; and according to the edge information of the target object in the to-be-processed image, the preliminary segmented image is In the process, the pixel values of the predicted pixels that do not belong to the target object in the enclosed area included in the edge of the target object are adjusted to obtain a first segmentation result corresponding to the image to be processed. In this way, when the target object is a human body or an animal body, the pixels inside the organs of the target object can also be segmented as belonging to the target object, so that a more accurate and robust segmentation result can be obtained.

Claims

An image segmentation method, comprising:

Predict the pixels belonging to the target object in the image to be processed, and obtain a preliminary segmented image corresponding to the image to be processed;

According to the edge information of the target object in the to-be-processed image, in the preliminary segmented image, the pixel values of the predicted pixels that do not belong to the target object in the enclosed area included in the edge of the target object are determined. Adjustment is performed to obtain a first segmentation result corresponding to the image to be processed.
The method according to claim 1, wherein, in the preliminary segmented image, the predicted pixel value of the pixel belonging to the target object is a first preset value, and the predicted pixel value of the pixel not belonging to the target object is the first preset value. The value is the second preset value;

According to the edge information of the target object in the image to be processed, in the preliminary segmented image, in the closed area included in the edge of the target object, the predicted pixels that do not belong to the target object are analyzed. The pixel value is adjusted to obtain the first segmentation result corresponding to the to-be-processed image, including:

Adjusting the pixel value of the closed area whose pixel value is the second preset value in the preliminary segmented image to the first preset value, to obtain a filled preliminary segmented image;

According to the edge information of the target object in the to-be-processed image, the pixel values of the filled preliminary segmented image are adjusted to obtain the first segmentation result corresponding to the to-be-processed image.
The method according to claim 2, wherein the pixel value of the closed area whose pixel value is the second preset value in the preliminary segmented image is adjusted to the first preset value to obtain the filled Preliminary segmented images, including:

Splicing edges of preset widths around the preliminary segmented images to obtain a preliminarily segmented image after splicing, wherein the pixel values of the pixels of the edges of the spliced preset widths are the second preset values;

Selecting the pixels of the image edges of the spliced preliminary segmented images as seed points, and performing a flood filling operation on the spliced preliminary segmented images to obtain the filled preliminary segmented images.
The method according to claim 2 or 3, wherein the pixel value of the filled preliminary segmented image is adjusted according to the edge information of the target object in the to-be-processed image to obtain the to-be-processed image. The first segmentation result corresponding to the image includes:

According to the edge information of the target object in the image to be processed, determine the maximum connected domain included in the edge of the target object in the filled preliminary segmented image;

The pixel values of the pixels outside the maximum connected region in the filled preliminary segmented image are adjusted to the second preset value to obtain the first segmentation result corresponding to the to-be-processed image.
The method according to any one of claims 1 to 4, wherein after obtaining the first segmentation result corresponding to the image to be processed, the method further comprises:

acquiring an image adjacent to the to-be-processed image and a second segmentation result corresponding to the adjacent image;

According to the pixel value of the pixel at the same position in the image to be processed and the adjacent image, and the second segmentation result, adjust the first segmentation result to obtain a third segmentation result corresponding to the to-be-processed image .
The method according to claim 5, wherein the first segmentation result is adjusted according to the pixel value of the pixel at the same position in the image to be processed and the adjacent image, and the second segmentation result , obtain the third segmentation result corresponding to the image to be processed, including:

According to the pixels in the adjacent images that belong to the target object in the second segmentation result and have a difference value of the pixel value at the same position as the image to be processed is less than or equal to a third preset value, The first segmentation result is adjusted to obtain the third segmentation result corresponding to the to-be-processed image.
The method according to claim 6, wherein, according to the adjacent images, in the second segmentation result, the pixel values that belong to the target object and are in the same position as the image to be processed For pixels whose difference value is less than or equal to the third preset value, adjust the first segmentation result to obtain the third segmentation result corresponding to the image to be processed, including:

Obtain the first pixel set according to the pixel whose difference value of the pixel value at the same position in the image to be processed and the adjacent image is less than or equal to the third preset value;

Obtain a second pixel set according to the pixels belonging to the target object in the second segmentation result in the first pixel set;

Adjusting the pixels of the second pixel set in the first segmentation result to belong to the target object to obtain the third segmentation result corresponding to the image to be processed.
The method according to any one of claims 1 to 7, wherein before predicting the pixels belonging to the target object in the image to be processed, the method further comprises: training according to the training image and the labeling data of the training image A neural network, wherein the labeled data of the training image includes the true value of pixels belonging to the target object in the training image;

The predicting the pixels belonging to the target object in the image to be processed to obtain a preliminary segmented image corresponding to the image to be processed includes: inputting the image to be processed into the neural network, and predicting the image to be processed through the neural network According to the information of the pixels belonging to the target object in the to-be-processed image, a preliminary segmented image corresponding to the to-be-processed image is obtained.
The method of claim 8, wherein,

The training image is an electronic computed tomography CT image;

The training of the neural network according to the training image and the labeled data of the training image includes: normalizing the pixel values of the training image according to a preset CT value range to obtain a normalized training image; The normalized training image and the labeled data of the training image train the neural network.
An image segmentation method, comprising:

Predict the pixels belonging to the target object in the image to be processed, and obtain a preliminary segmented image corresponding to the image to be processed;

acquiring an image adjacent to the to-be-processed image and a second segmentation result corresponding to the adjacent image;

According to the pixel value of the pixel at the same position in the image to be processed and the adjacent image, and the second segmentation result, the preliminary segmented image is adjusted to obtain a fourth segmentation result corresponding to the to-be-processed image.
The method according to claim 10, wherein, according to the pixel value of the pixel at the same position in the image to be processed and the adjacent image, and the second segmentation result, the preliminary segmented image is adjusted, Obtain the fourth segmentation result corresponding to the to-be-processed image, including:

According to the pixels in the adjacent images that belong to the target object in the second segmentation result and have a difference value of the pixel value at the same position as the image to be processed is less than or equal to a third preset value, Adjust the preliminary segmented image to obtain the fourth segmentation result corresponding to the to-be-processed image.
The method according to claim 11, wherein, according to the adjacent images, in the second segmentation result, the pixel values that belong to the target object and are in the same position as the image to be processed For pixels whose difference is less than or equal to the third preset value, adjust the preliminary segmented image to obtain the fourth segmentation result corresponding to the to-be-processed image, including:

Obtain the first pixel set according to the pixel whose difference value of the pixel value at the same position in the image to be processed and the adjacent image is less than or equal to the third preset value;

Obtain a second pixel set according to the pixels belonging to the target object in the second segmentation result in the first pixel set;

The pixels of the second pixel set in the preliminary segmented image are adjusted to belong to the target object, and the fourth segmentation result corresponding to the image to be processed is obtained.
An image segmentation device, comprising:

a first segmentation part, configured to predict pixels belonging to the target object in the to-be-processed image, and obtain a preliminary segmented image corresponding to the to-be-processed image;

The first adjustment part is configured to, according to the edge information of the target object in the to-be-processed image, in the preliminary segmented image, predict that in the enclosed area included in the edge of the target object, those that do not belong to the The pixel values of the pixels of the target object are adjusted to obtain a first segmentation result corresponding to the image to be processed.
An image segmentation device, comprising:

The second segmentation part is configured to predict the pixels belonging to the target object in the to-be-processed image, and obtain a preliminary segmented image corresponding to the to-be-processed image;

a first acquiring part, configured to acquire an image adjacent to the to-be-processed image and a second segmentation result corresponding to the adjacent image;

The second adjustment part is configured to adjust the preliminary divided image according to the pixel value of the pixel at the same position in the image to be processed and the adjacent image, and the second segmentation result to obtain the image to be processed The corresponding fourth segmentation result.
An electronic device comprising:

one or more processors;

memory configured to store executable instructions;

wherein the one or more processors are configured to invoke executable instructions stored in the memory to perform the method of any one of claims 1-12.
A computer-readable storage medium having computer program instructions stored thereon, the computer program instructions implementing the method of any one of claims 1 to 12 when executed by a processor.
A computer program, comprising computer-readable codes, when the computer-readable codes are executed in an electronic device, the processor in the electronic device implements the method described in any one of claims 1 to 12 when executed. method.