Image registration method and device, computer device and medium

Abstract

Embodiments of the present application disclose a kind of image registration method, device, computer equipment and medium, belong to image processing technical field.The method includes: obtaining image group, the image group includes reference image and target image;Using first registration mode, according to the target image registration of the reference image, obtain first alternative registration image;In response to first overlap degree is not greater than reference overlap degree, using second registration mode, according to the target image registration of the reference image, obtain second alternative registration image;In response to second overlap degree is greater than the reference overlap degree, the second alternative registration image is determined as the registration image corresponding to the target image.The method uses multiple registration modes to determine registration image together, more widely applicable, can accurately register different types of images, improve the accuracy and success rate of image registration.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the technical field of image processing, and particularly relate to an image registration method and device, a computer device, and a medium. BACKGROUND

[0002] Image registration refers to a process of matching different images of a same object, and is widely applied in fields such as image three-dimensional reconstruction, remote sensing image analysis, and medical image processing. For example, in the field of medical image processing, different CT (Computed Tomography) images of a same object collected at different times are subjected to image registration, which can determine the change of the object. However, the image registration manner in the related art is not accurate enough, and is prone to registration failure. SUMMARY

[0003] Embodiments of the present application provide an image registration method, device, computer device, and medium, which improve the accuracy and success rate of image registration. The technical solution is as follows:

[0004] In one aspect, an image registration method is provided, which includes:

[0005] obtaining an image group, the image group including a reference image and a target image, the reference image and the target image being images of a target object at different time points;

[0006] using a first registration manner to register the target image according to the reference image, to obtain a first candidate registration image;

[0007] in response to a first overlap degree being not greater than a reference overlap degree, using a second registration manner to register the target image according to the reference image, to obtain a second candidate registration image, the first registration manner being different from the second registration manner, the first overlap degree referring to an overlap degree between the first candidate registration image and the reference image;

[0008] in response to a second overlap degree being greater than the reference overlap degree, determining the second candidate registration image as a registration image corresponding to the target image, the second overlap degree referring to an overlap degree between the second candidate registration image and the reference image.

[0009] In another aspect, an image registration device is provided, which includes:

[0010] an image group obtaining module configured to obtain an image group, the image group including a reference image and a target image, the reference image and the target image being images of a target object at different time points;

[0011] The first registration module is used to register the target image with the reference image using a first registration method to obtain a first candidate registration image;

[0012] The second registration module is used to register the target image according to the reference image in response to the first overlap degree not being greater than the reference overlap degree, and to obtain a second candidate registration image by adopting a second registration method. The first registration method is different from the second registration method. The first overlap degree refers to the overlap degree between the first candidate registration image and the reference image.

[0013] The registration image determination module is used to determine the second candidate registration image as the registration image corresponding to the target image in response to the second overlap being greater than the reference overlap, wherein the second overlap refers to the overlap between the second candidate registration image and the reference image.

[0014] In one possible implementation, the registration image determination module is further configured to determine the first candidate registration image as the registration image corresponding to the target image in response to the first overlap being greater than the reference overlap.

[0015] In another possible implementation, the first registration module is further configured to, in response to the first overlap not being greater than the reference overlap and the first registration count being less than the reference count, continue to execute the step of using the first registration method to register the target image according to the reference image to obtain a first candidate registration image, until the first overlap obtained by the last registration is greater than the reference overlap, or the first registration count is equal to the reference count, wherein the first registration count refers to the number of times the target image is registered using the first registration method.

[0016] In another possible implementation, the second registration module is configured to, in response to the first overlap obtained by the last registration not being greater than the reference overlap, and the first registration number being equal to the reference number, use the second registration method to register the target image according to the reference image to obtain a second alternative registration image.

[0017] In another possible implementation, the second registration module is further configured to, in response to the second overlap not being greater than the reference overlap and the second registration count being less than the reference count, continue to execute the step of using the second registration method to register the target image according to the reference image to obtain a second candidate registration image, until the second overlap obtained by the last registration is greater than the reference overlap, or the second registration count is equal to the reference count, wherein the first registration count refers to the number of times the target image is registered using the second registration method.

[0018] In another possible implementation, the device further includes a third registration module.

[0019] The registration image determination module is further configured to, in response to the fact that the second overlap obtained from the last registration is not greater than the reference overlap, and the second registration number is equal to the reference number, determine the maximum overlap among the plurality of first overlaps and plurality of second overlaps obtained from the registration, and determine the candidate registration image corresponding to the maximum overlap as the registration image corresponding to the target image; or...

[0020] The third registration module is configured to, in response to the fact that the second overlap obtained from the last registration is not greater than the reference overlap and the second registration number is equal to the reference number, adopt a third registration method to register the target image according to the reference image to obtain a third alternative registration image. The third registration method is different from the first registration method and the second registration method.

[0021] In another possible implementation, the registration method includes at least two of the following:

[0022] Based on the reference positions of multiple points in the reference image, multiple points in the target image are moved to obtain a registered image.

[0023] Based on the reference positions of multiple locations in the reference image, multiple locations in the target image are mapped to the target space to obtain a registered image;

[0024] The image registration model is invoked to register the target image with the reference image, resulting in a registered image.

[0025] In another possible implementation, the reference image and the target image are computed tomography (CT) images of the brain, and the apparatus further includes:

[0026] The preprocessing module is used to perform de-skulling on the original reference image and target image to obtain the processed reference image and target image.

[0027] In another possible implementation, the reference image and the target image are computed tomography (CT) images of the brain, and the apparatus further includes:

[0028] The preprocessing module is also used to perform desquamation on the registered image to obtain a processed registered image.

[0029] In another possible implementation, the reference image and the target image are computed tomography (CT) images of the brain, and the apparatus further includes:

[0030] The preprocessing module is also used to determine the absorption parameters corresponding to each location point in the original reference image and the target image, respectively. The absorption parameters are used to represent the degree of absorption of rays by the location point.

[0031] The preprocessing module is further configured to adjust the absorption parameters that are less than the first reference parameter among the multiple absorption parameters corresponding to the reference image and the target image to the first reference parameter, and adjust the absorption parameters that are greater than the second reference parameter among the multiple absorption parameters to the second reference parameter, so as to obtain the adjusted reference image and the adjusted target image.

[0032] In another possible implementation, the image group includes multiple images, and the registration image determination module is further configured to, after determining the registration images corresponding to the multiple image groups, determine the maximum overlap among the multiple overlaps corresponding to the multiple registration images, and determine the registration image corresponding to the maximum overlap as the registration image corresponding to the target image.

[0033] On the other hand, a computer device is provided, the computer device including a processor and a memory, the memory storing at least one computer program, the at least one computer program being loaded and executed by the processor to perform the operations performed in the image registration method as described above.

[0034] On the other hand, a computer-readable storage medium is provided that stores at least one computer program, which is loaded and executed by a processor to perform the operations performed in the image registration method described above.

[0035] On the other hand, a computer program product or computer program is provided, the computer program product or computer program including computer program code stored in a computer-readable storage medium, a processor of a computer device reading the computer program code from the computer-readable storage medium, the processor executing the computer program code, causing the computer device to perform the operations performed in the image registration method described above.

[0036] The beneficial effects of the technical solutions provided in this application include at least the following:

[0037] The methods, apparatus, computer equipment, and media provided in this application embodiment, when the image obtained by registration using the first registration method does not meet the requirements, continue to use the second registration method for registration until a registered image that meets the requirements is obtained. Compared with related technologies that only use one registration method, this application uses multiple registration methods to jointly determine the registered image, which has a wider range of applications and can accurately register different types of images, thereby improving the accuracy and success rate of image registration. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 This is a flowchart of an image registration method provided in an embodiment of this application;

[0040] Figure 2 This is a flowchart of another image registration method provided in the embodiments of this application;

[0041] Figure 3 This is a flowchart of another image registration method provided in the embodiments of this application;

[0042] Figure 4 This is a schematic diagram of an image registration network provided in an embodiment of this application;

[0043] Figure 5 This is a schematic diagram of a CT image provided in an embodiment of this application;

[0044] Figure 6 This is a schematic diagram of another image registration network provided in an embodiment of this application;

[0045] Figure 7 This is a schematic diagram of a registration image provided in an embodiment of this application;

[0046] Figure 8 This is a flowchart of another image registration method provided in the embodiments of this application;

[0047] Figure 9 This is a schematic diagram of the structure of an image registration device provided in an embodiment of this application;

[0048] Figure 10 This is a schematic diagram of another image registration device provided in an embodiment of this application;

[0049] Figure 11This is a schematic diagram of the structure of a terminal provided in an embodiment of this application;

[0050] Figure 12 This is a schematic diagram of the structure of a server provided in an embodiment of this application. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the implementation methods of this application will be further described in detail below with reference to the accompanying drawings.

[0052] It is understood that the terms "first," "second," etc., used in this application may be used to describe various concepts herein, but unless otherwise stated, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of this application, the first degree of overlap may be referred to as the second degree of overlap, and the second degree of overlap may be referred to as the first degree of overlap.

[0053] As used in this application, the terms "at least one," "multiple," "each," and "any" have the following meanings: at least one includes one, two, or more; multiple includes two or more; each refers to each of the corresponding multiple; and any refers to any one of the multiple. For example, multiple registered images include three registered images, and each registered image refers to every single one of the three registered images, while any refers to any one of the three registered images, which could be the first, the second, or the third.

[0054] To facilitate understanding of the embodiments of this application, the keywords involved in the embodiments of this application will be explained first:

[0055] CT images: Computed tomography images are produced by scanning a layer of the human body with X-rays of a certain thickness, receiving the X-rays that pass through the layer, and processing the received X-rays to obtain CT images.

[0056] Registration: refers to the matching of different images of the same object, including geometric correction, projection transformation and scaling.

[0057] BET (Brain Extraction Tool) is a deskull removal algorithm that processes images containing skull regions to remove skull areas from the images.

[0058] Rigid body transformation: In three-dimensional space, the rotation and translation of a geometric object.

[0059] Affine transformation: In geometry, it refers to the process of performing a geometric transformation and translation on a vector space to transform it into another vector space.

[0060] Artificial intelligence (AI) is the theory, methods, technology, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results. In other words, AI is a comprehensive technology within computer science that attempts to understand the essence of intelligence and produce a new kind of intelligent machine that can react in a way similar to human intelligence. AI studies the design principles and implementation methods of various intelligent machines, enabling them to possess the functions of perception, reasoning, and decision-making.

[0061] Artificial intelligence (AI) is a comprehensive discipline encompassing a wide range of fields, including both hardware and software technologies. Fundamental AI technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies primarily include computer vision, speech processing, natural language processing, and machine learning / deep learning.

[0062] Computer vision (CV) is a science that studies how to enable machines to "see." More specifically, it refers to machine vision, which uses cameras and computers to replace human eyes in tasks such as target recognition, tracking, and measurement, and further performs image processing to create images more suitable for human observation or transmission to instruments. As a scientific discipline, computer vision studies related theories and technologies, attempting to build artificial intelligence systems capable of extracting information from images or multidimensional data. Computer vision technologies typically include image processing, image recognition, image semantic understanding, image retrieval, OCR (Optical Character Recognition), video processing, video semantic understanding, video content / behavior recognition, 3D object reconstruction, 3D technology, virtual reality, augmented reality, simultaneous localization and mapping (SLAM), and common biometric recognition technologies such as facial recognition and fingerprint recognition.

[0063] Machine learning (ML) is a multidisciplinary field involving probability theory, statistics, approximation theory, convex analysis, and algorithm complexity theory. It specifically studies how computers can simulate or implement human learning behavior to acquire new knowledge or skills and reorganize existing knowledge structures to continuously improve their performance. Machine learning is the core of artificial intelligence and the fundamental way to endow computers with intelligence; its applications span all areas of artificial intelligence. Machine learning and deep learning typically include techniques such as artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and learn-by-doing.

[0064] With the research and advancement of artificial intelligence (AI) technology, AI is being studied and applied in various fields, such as smart homes, smart wearable devices, virtual assistants, smart speakers, smart marketing, autonomous driving, drones, robots, smart healthcare, and smart customer service. It is believed that with the development of technology, AI will be applied in more fields and play an increasingly important role.

[0065] The image registration method provided in this application is applied to a computer device, which is a terminal or a server. Optionally, the terminal is a computer, mobile phone, tablet computer, or other terminal. The server is an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms.

[0066] Figure 1 This is a flowchart illustrating an image registration method provided in an embodiment of this application. The execution subject of this embodiment is a computer device. See also... Figure 1 The method includes the following steps:

[0067] 101. Obtain an image group, which includes a reference image and a target image.

[0068] In this context, the reference image and the target image are images of the target object at different points in time. The target object can be a person, animal, plant, or other object. The target image is the image to be registered, and the reference image is the image used as a reference during registration; that is, the target image is registered according to the various position points in the reference image. For example, the target image is an image of the target object at the current time, and the reference image is an image of the target object at the same time one month prior to the current time.

[0069] Optionally, the reference image and the target image are images acquired by a computer device or transmitted by another computer device. Alternatively, the reference image and the target image are images acquired using different image acquisition methods. For example, the reference image and the target image are images taken with a camera, or CT images obtained by X-ray scanning, or acquired using other methods.

[0070] In one possible implementation, the reference image and the target image are preprocessed images, such as those subjected to filtering, image enhancement, or de-skeletalization. Different preprocessing methods can be used for different reference images and target images, or for different application scenarios.

[0071] 102. Using the first registration method, the target image is registered according to the reference image to obtain the first candidate registration image.

[0072] In this embodiment of the application, multiple registration methods can be used to register the target image. This embodiment of the application does not restrict which registration method is used first or which registration method is used later in the image registration process.

[0073] In one possible implementation, multiple registration methods include:

[0074] The first method involves moving multiple points in the target image based on their reference positions in the reference image to obtain a registered image. Both the reference and target images contain multiple points, and the movement of these points includes translation, rotation, and other similar actions.

[0075] In one possible implementation, an initial registration matrix is ​​obtained, which is used to register the target image. During the registration process of the target image using the initial registration matrix, each element in the initial registration matrix is ​​adjusted according to the position of each location point in the reference image so that each adjusted element can represent the position of the corresponding location point in the reference image. The adjusted initial registration matrix is ​​then multiplied with the target image to obtain the registered image.

[0076] The second method involves mapping multiple location points in the target image to the target space based on their reference positions in the reference image, resulting in a registered image. In this method, any two location points are connected to form a vector. The multiple location points in the target image reside in a vector space, while the target space is a different vector space. Essentially, vector mapping is performed on the multiple vectors formed by the multiple location points in the target image to obtain the registered image.

[0077] Optionally, vector mapping refers to performing a linear transformation and a translation on a vector. For example, the following formula can be used to map multiple vectors in a target image:

[0078]

[0079] in, Represents the mapped vector. Let A represent the vector formed by any two points, and let A represent the coefficients of the linear transformation. Indicates the distance of translation.

[0080] The third method involves calling an image registration model to register the target image with a reference image, resulting in a registered image. The image registration model is either a machine learning model trained on a computer or a machine learning model sent to the computer from another device. The reference and target images are the inputs to the image registration model, and the registered image is its output.

[0081] The first registration method can be any of the above-mentioned registration methods.

[0082] 103. In response to the first overlap being no greater than the reference overlap, a second registration method is adopted, and the target image is registered according to the reference image to obtain a second alternative registration image.

[0083] The first overlap refers to the overlap between the first candidate registration image and the reference image. The overlap is used to represent the degree of overlap of the same area in the candidate registration image and the reference image. The larger the overlap, the better the image registration effect, and the smaller the overlap, the worse the image registration effect. The reference overlap is any value, and the value range of the reference overlap is 0-1.

[0084] The first overlap not being greater than the reference overlap means that the first candidate image does not meet the registration condition, and the target image needs to be re-registered. The registration condition refers to the overlap of the candidate image being greater than the reference overlap. When re-registering the target image, the second registration method used is different from the first registration method. Different registration methods are suitable for different types of images. Therefore, for a given image, using only one method may lead to inaccurate registration. Using multiple methods helps to find a suitable registration method for the target image, thereby improving the accuracy of image registration.

[0085] For example, if the target image and the reference image are taken from the same angle but at different times, it is difficult to ensure that the two angles are exactly the same. Therefore, simply rotating or translating the target image will match the reference image. In this case, the first image registration method described above can be used to accurately register the target image. If the target image and the reference image are taken from different angles and at different times, it is necessary to make more changes to the position points in the target image if you want to register the target image. In this case, the second or third image registration method described above can be used to accurately register the target image.

[0086] 104. In response to the second overlap being greater than the reference overlap, the second candidate registration image is determined as the registration image corresponding to the target image.

[0087] Here, the second overlap refers to the degree of overlap between the second candidate registration image and the reference image. If the second overlap is greater than the reference overlap, it means that the currently obtained second candidate registration image meets the registration conditions, and the second candidate registration image is determined as the registration image corresponding to the target image.

[0088] It should be noted that the embodiments in this application are only illustrated using two registration methods as examples. In another embodiment, if the candidate registration image obtained by the second registration method does not meet the registration requirements, a third registration method, a fourth registration method, etc., can still be used for registration. The third and fourth registration methods are different from the first and second registration methods.

[0089] The method provided in this application embodiment, when the image obtained by registration using the first registration method does not meet the requirements, continues to use the second registration method until a registered image that meets the requirements is obtained. Compared with related technologies that only use one registration method, this application uses multiple registration methods to jointly determine the registered image, which has a wider range of applications and can accurately register different types of images, thereby improving the accuracy and success rate of image registration.

[0090] The above Figure 1 In the illustrated embodiments, registration is performed once using only one registration method. If the candidate registration images obtained from the first registration do not meet the registration requirements, another registration method is used to continue registering the target image until a registration image that meets the registration requirements is obtained. In this embodiment, to avoid the randomness of the single registration process, the target object is registered multiple times using one registration method. If none of the candidate registration images obtained from multiple registrations meet the registration requirements, another registration method is used to continue registering the target object. The registration process of multiple registrations using one registration method is described below. Figure 2 The example shown.

[0091] Figure 2 This is a flowchart of another image registration method provided in an embodiment of this application. The execution subject of this embodiment is a computer device. See also... Figure 2 The method includes the following steps:

[0092] 201. Obtain an image set including a reference image and a target image.

[0093] Step 201 is the same as step 101 above, and will not be repeated here.

[0094] 202. Using the first registration method, the target image is registered according to the reference image to obtain the first candidate registration image, and the first overlap corresponding to the first candidate registration image is determined.

[0095] In this embodiment of the application, after obtaining the first candidate registration image, it is first determined whether the first overlap corresponding to the first candidate registration image is greater than the reference overlap. If the first overlap is greater than the reference overlap, step 203 is not executed, and the obtained first candidate registration image is directly determined as the registration image corresponding to the target image. If the first overlap is not greater than the reference overlap, step 203 is executed.

[0096] In one possible implementation, DICE (Dice similarity coefficient) is used to represent the overlap, and the first overlap is obtained using the following formula:

[0097]

[0098] Where s represents the first degree of overlap, X represents the target image, Y represents the first candidate registration image, |X| represents the area of ​​the target image, |Y| represents the area of ​​the first candidate registration image, and |X∩Y| represents the area of ​​the overlapping portion between the target image and the first candidate registration image.

[0099] 203. In response to the first overlap being no greater than the reference overlap and the first registration number being less than the reference number, continue to execute step 202 until the first overlap obtained by the last registration is no greater than the reference overlap and the first registration number is equal to the reference number, then execute step 204.

[0100] Here, the first registration count represents the number of times the target image is registered using the first registration method. Each time the computer device performs registration on the target image using the first registration method, the first registration count is incremented by 1. The reference count is any number of times. For example, the reference count could be 10 times, 20 times, etc.

[0101] In this embodiment, if the first overlap is not greater than the reference overlap, it is determined whether the current first registration count is less than the reference count. If it is equal to the reference count, it means that the maximum number of times the target image has been registered using the first registration method has been reached, and other registration methods need to be used to register the target image. In this case, step 202 is not executed, and step 204 is executed directly. If the first registration count is less than the reference count, it means that the maximum number of times the target image has been registered using the first registration method has not been reached, and the target image needs to be registered using the first registration method again. That is, step 202 is executed again.

[0102] In one possible implementation, when the first registration method is as described above... Figure 1 In the first image registration method shown in the embodiment, an initial registration matrix is ​​randomly obtained each time the target image is registered using the first registration method.

[0103] 204. Using the second registration method, the target image is registered according to the reference image to obtain a second candidate registration image, and the second overlap corresponding to the second candidate registration image is determined.

[0104] In this embodiment, after obtaining the second candidate registration image, it is first determined whether the second overlap corresponding to the second candidate registration image is greater than the reference overlap. If the second overlap is greater than the reference overlap, step 205 is not executed, and the obtained second candidate registration image is directly determined as the registration image corresponding to the target image. If the second overlap is not greater than the reference overlap, step 205 is executed.

[0105] The method for obtaining the second degree of overlap is similar to that for obtaining the first degree of overlap, and will not be repeated here.

[0106] 205. In response to the second overlap being no greater than the reference overlap and the second registration number being less than the reference number, continue to execute step 204 until the second overlap obtained by the last registration is no greater than the reference overlap and the second registration number is equal to the reference number, then execute step 206.

[0107] The second registration number represents the number of times the target image is registered using the second registration method. Each time the computer device registers the target image using the second registration method, the second registration number is incremented by 1. The reference number is any number.

[0108] In this embodiment, if the second overlap is not greater than the reference overlap, it is determined whether the current second registration count is less than the reference count. If it is equal to the reference count, it means that the maximum number of times the target image has been registered using the second registration method has been reached, and other registration methods need to be used to register the target image, or a candidate registration image needs to be selected from multiple candidate registration images as the registration image corresponding to the target image. In this case, step 204 is not executed, and step 206 is executed directly. If the second registration count is less than the reference count, it means that the maximum number of times the target image has been registered using the second registration method has not been reached, and the target image needs to be registered using the second registration method again, that is, step 204 is executed again.

[0109] 206. Determine the maximum overlap among the multiple first overlaps and multiple second overlaps obtained from registration, and determine the candidate registration image corresponding to the maximum overlap as the registration image corresponding to the target image.

[0110] If the overlap of multiple candidate registration images does not meet the registration conditions, the largest overlap is selected from the obtained overlaps, and the candidate registration image corresponding to the largest overlap is determined as the registration image corresponding to the target image. Specifically, if the largest overlap is among multiple first overlaps, the corresponding first candidate registration image is determined as the registration image corresponding to the target image; if the largest overlap is among multiple second overlaps, the corresponding second candidate registration image is determined as the registration image corresponding to the target image.

[0111] In one possible implementation, step 206 is replaced by the following steps: In response to the second overlap obtained from the last registration not being greater than the reference overlap, and the second registration count being equal to the reference count, a third registration method is used to register the target image based on the reference image, resulting in a third candidate registration image. In response to the overlap corresponding to the third registration image being greater than the reference overlap, this third candidate registration image is determined as the registration image corresponding to the target image. That is, if the overlap corresponding to the multiple candidate registration images does not meet the registration conditions, other registration methods are continued to register the target image until a registration image that meets the registration conditions is obtained.

[0112] The method provided in this application embodiment, when the image obtained by multiple registrations using the first registration method does not meet the requirements, continues to use the second registration method until a registered image that meets the requirements is obtained. Compared with related technologies that only use one registration method, this method uses multiple registration methods to jointly determine the registered image, which has a wider range of applications and can accurately register different types of images. Moreover, by using each registration method for multiple registrations, the randomness of a single registration process can be avoided, thus improving the accuracy and success rate of image registration.

[0113] Furthermore, if a higher registration accuracy is required, the reference overlap and reference number can be set to larger values ​​to obtain a more accurate registered image. If a lower registration accuracy is required, the reference overlap and reference number can be set to smaller values ​​to obtain a registered image faster. The flexibility is improved by setting the reference overlap and reference number according to the requirements of the registration effect.

[0114] In one possible implementation, see [link to relevant documentation]. Figure 3 Taking rigid body registration and affine registration as examples, with each method performing a maximum of n registrations, the registration process of the target image is explained. The registration process is described in the following steps:

[0115] 301. Obtain the reference image and the target image.

[0116] 302. Rigid body registration is used to process the reference image and the target image to obtain the first candidate registration image.

[0117] 303. Determine whether the first overlap between the first candidate registration image and the reference image is greater than the reference overlap. If the first overlap is greater than the reference overlap, proceed to step 309. If the first overlap is not greater than the reference overlap, proceed to step 304.

[0118] 304. Determine whether the current first registration count is less than n. If the first registration count is less than n, then execute step 302 again. If the first registration count is equal to n, then execute step 305.

[0119] 305. Affine registration is used to process the reference image and the target image to obtain a second alternative registration image.

[0120] 306. Determine whether the second overlap between the second candidate registration image and the reference image is greater than the reference overlap. If the second overlap is greater than the reference overlap, proceed to step 309. If the second overlap is not greater than the reference overlap, proceed to step 307.

[0121] 307. Determine whether the current number of second registrations is less than n. If the number of second registrations is less than n, then execute step 305 again. If the number of second registrations is equal to n, then execute step 308.

[0122] 308. Determine the maximum overlap among the n first overlaps and n second overlaps, and obtain the candidate registration image corresponding to the maximum overlap.

[0123] 309. The currently obtained candidate registration image is determined as the registration image of the target image.

[0124] In one possible implementation, see [link to relevant documentation]. Figure 4 The image registration network shown includes an input module 401, a first registration module 402, a first overlap comparison module 403, a first number comparison module 404, a second registration module 405, a second overlap comparison module 406, a second number comparison module 407, a selection module 408, and an output module 409.

[0125] The input module 401 is used to execute step 301, the first registration module 402 is used to execute step 302, the first overlap comparison module 403 is used to execute step 303, the first number comparison module 404 is used to execute step 304, the second registration module 405 is used to execute step 305, the second overlap comparison module 406 is used to execute step 306, the second number comparison module 407 is used to execute step 307, the selection module 408 is used to execute step 308, and the output module 409 is used to execute step 309.

[0126] The above embodiments are all described using the processing of a group of images as an example. In one possible implementation, the group of images may include multiple images. When the group of images includes multiple images, multiple registration images are obtained. The maximum overlap among the multiple overlaps of the multiple registration images is determined, and the registration image corresponding to the maximum overlap is determined as the registration image corresponding to the target image.

[0127] In one possible implementation, the original reference image and the target image are processed using different image preprocessing methods to obtain processed reference images and target images, thus resulting in multiple image groups. One image group includes the original reference image and the target image, while the other image groups each include reference images and target images obtained by different image preprocessing methods.

[0128] In this embodiment, taking CT images of the brain as an example where both the reference image and the target image are brain images, the processing procedure for multiple image groups is described. See [link to documentation]. Figure 5 The CT image shown represents the skull region in white areas, and the area inside the white areas represents the brain region that needs to be registered. Therefore, the skull region does not need to be registered in the CT image. The skull region needs to be removed during the image registration process. Different preprocessing methods used to process the CT image may result in different image registration effects.

[0129] In one possible implementation, see [link to relevant documentation]. Figure 6 The image registration framework shown includes an input network, three different registration units, a selection network, and an output network. The first and second registration units both include a decranialization network and an image registration network. The difference is that in the first registration unit, the input network is connected to the decranialization network, and the decranialization network is connected to the image registration network; in the second registration unit, the input network is connected to the image registration network, and the image registration network is connected to the decranialization network. The third registration unit includes a brain-picking window module, an image registration network, and a decranialization network. The input network is connected to the brain-picking window module, the brain-picking window module is connected to the image registration network, and the image registration network is connected to the decranialization network.

[0130] The first registration unit processes the target and reference images as follows: First, a de-skullization network is used to de-skullize the original target and reference images, resulting in a de-skullized target and reference image. Then, an image registration network is used to register the de-skullized target and reference images, resulting in a registered image. The second registration unit processes the target and reference images as follows: First, an image registration network is used to register the original target and reference images, resulting in a registered image containing the skull region. Then, a de-skullization network is used to de-skullize the obtained registered image, resulting in the final registered image. The third registration unit processes the target and reference images as follows: First, a brain window extraction module is used to extract brain windows from the original target and reference images, resulting in a target and reference image with brain windows extracted. Then, an image registration network is used to register the target and reference images with brain windows extracted, resulting in a registered image. Finally, a de-skullization network is used to de-skullize the obtained registered image, resulting in the final registered image.

[0131] The registration process of the image registration network for the target image is detailed above. Figure 2 The structure of the image registration network in the illustrated embodiment is detailed above. Figure 4 The desquamation network uses a machine learning model to identify the original target image and reference image, determine the skull region in the target image and reference image, remove the skull region, and obtain the target image and reference image after skull removal; the brain window module adjusts the absorption parameters corresponding to each position in the original target image and reference image to obtain the adjusted target image and reference image; the selection network is used to select a registration image from the registration images obtained from the three registration units; and the output network is used to output the selected registration image.

[0132] In one possible implementation, absorption parameters corresponding to each location point in the original reference image and the target image are determined respectively. Absorption parameters smaller than a first reference parameter are adjusted to the first reference parameter, and absorption parameters larger than a second reference parameter are adjusted to the second reference parameter, resulting in an adjusted reference image and an adjusted target image. The absorption parameters represent the degree of X-ray absorption at the location point, and the range of the absorption parameters is -1000HU to 1000HU, where HU (Hounsfield Units) is the unit of the absorption parameter. The first reference parameter is smaller than the second reference parameter, and both the first and second reference parameters are any parameters within the range of -1000 to 1000. For example, the first reference parameter is 40, and the second reference parameter is 80.

[0133] For example, using the above Figure 6The image registration framework shown performs image registration processing on the original target image and the reference image to obtain... Figure 7 The registered images are shown below. The two images on the left are the original reference image and the target image for registration. The first image is the original reference image, and the second image is the original target image. The two images on the right are the processed images. The first image is the processed reference image; during registration, only the skull region was removed from this reference image. Therefore, compared to the original reference image, the resulting reference image only has the skull region removed. The second image is the processed registered image, which shows that its shape, size, and orientation are basically the same as the reference image.

[0134] The method provided in this application can be applied to various scenarios. For example, in a medical setting, the method provided in this application registers CT images, resulting in a registered image. This registered image is only one basis for doctors to determine the patient's physical condition; it still needs to be combined with other relevant information or the patient's physical condition to determine the patient's diagnosis. For example, in image preprocessing scenarios in machine learning, the method provided in this application registers the original image, and the registered image is then used for subsequent image recognition, image segmentation, and other processing. The following is a brief explanation... Figure 8 The illustrated embodiments provide a detailed description of their application in medical settings.

[0135] Figure 8 This is a flowchart of another image registration method provided in this application embodiment. The execution subject of this application embodiment is a computer device; see [link to relevant documentation]. Figure 8 The method includes the following steps:

[0136] 801. Obtain the original reference CT image and the target CT image.

[0137] Using the above Figure 6 The image registration framework shown processes the reference CT image and the target CT image. Steps 802-803 are the processing steps for the first registration unit, steps 804-805 are the processing steps for the second registration unit, and steps 806-808 are the processing steps for the third registration unit.

[0138] 802. Perform craniotomy on the original reference CT image and the target CT image to obtain the craniotomy-removed CT image and the target CT image.

[0139] 803. An image registration network is used to register the target CT image after skull removal and the reference CT image to obtain a registered image.

[0140] 804. An image registration network is used to register the original target CT image and the reference CT image to obtain a registered image containing the skull region.

[0141] 805. Perform skull removal processing on the registered image containing the skull region to obtain the registered image after skull removal.

[0142] 806. Adjust the absorption parameters in the original target CT image and reference CT image to obtain the adjusted target CT image and reference CT image.

[0143] 807. An image registration network is used to register the adjusted target CT image and the reference CT image to obtain a registered image containing the skull region.

[0144] 808. Perform skull removal processing on the registered image containing the skull region to obtain the registered image after skull removal.

[0145] 809. Select one of the three registered images as the registered image corresponding to the target CT image.

[0146] The target CT image is a CT image of the patient's brain taken at the current time point, while the reference CT image is a CT image of the patient's brain taken before the current time point. Since the patient's brain tissue structure does not stretch or deform over time, if the patient has no abnormalities in the brain tissue, the registered image obtained by registering the target CT image should be consistent with the reference CT image. However, if the patient's brain tissue has lesions, the registered image obtained by registering the target CT image will also have significant differences from the reference CT image.

[0147] It should be noted that the embodiments of this application are only illustrated using CT images of the brain as an example. In another embodiment, it may be CT images of the lungs or other CT images.

[0148] Figure 9 This is a schematic diagram of the structure of an image registration device provided in an embodiment of this application. See also... Figure 9 The device includes:

[0149] The image group acquisition module 901 is used to acquire an image group, which includes a reference image and a target image. The reference image and the target image are images of the target object at different time points.

[0150] The first registration module 902 is used to register the target image with the reference image using the first registration method to obtain the first candidate registration image;

[0151] The second registration module 903 is used to register the target image with the reference image in response to the first overlap degree not being greater than the reference overlap degree, and to obtain the second candidate registration image by adopting the second registration method. The first registration method is different from the second registration method. The first overlap degree refers to the overlap degree between the first candidate registration image and the reference image.

[0152] The registration image determination module 904 is used to determine the second candidate registration image as the registration image corresponding to the target image in response to the second overlap being greater than the reference overlap. The second overlap refers to the overlap between the second candidate registration image and the reference image.

[0153] The apparatus provided in this application continues to use a second registration method to register images that do not meet the requirements when the image obtained by registration using the first registration method does not meet the requirements, until a registered image that meets the requirements is obtained. Compared with related technologies that only use one registration method, this application uses multiple registration methods to jointly determine the registered image, which has a wider range of applications and can accurately register different types of images, thereby improving the accuracy and success rate of image registration.

[0154] In one possible implementation, the registration image determination module 904 is further configured to determine the first candidate registration image as the registration image corresponding to the target image in response to the first overlap being greater than the reference overlap.

[0155] In another possible implementation, the first registration module 902 is further configured to continue executing the step of registering the target image with the reference image using the first registration method to obtain a first candidate registered image in response to the first overlap being no greater than the reference overlap and the first registration number being less than the reference number, until the first overlap obtained by the last registration is greater than the reference overlap, or the first registration number is equal to the reference number, wherein the first registration number refers to the number of times the target image is registered using the first registration method.

[0156] In another possible implementation, the second registration module 903 is used to register the target image according to the reference image in response to the fact that the first overlap obtained from the last registration is not greater than the reference overlap and the first registration number is equal to the reference number, thereby obtaining a second alternative registration image.

[0157] In another possible implementation, the second registration module 903 is further configured to continue executing the step of registering the target image with the reference image using the second registration method to obtain a second alternative registration image in response to the second overlap being no greater than the reference overlap and the second registration number being less than the reference number, until the second overlap obtained by the last registration is greater than the reference overlap, or the second registration number is equal to the reference number, where the first registration number refers to the number of times the target image is registered using the second registration method.

[0158] In another possible implementation, see Figure 10 The device also includes a third registration module 905.

[0159] The registration image determination module 904 is further configured to, in response to the second overlap obtained from the last registration not being greater than the reference overlap and the second registration number being equal to the reference number, determine the maximum overlap among the multiple first overlaps and multiple second overlaps obtained from the registration, and determine the candidate registration image corresponding to the maximum overlap as the registration image corresponding to the target image; or,

[0160] The third registration module 905 is used to respond to the fact that the second overlap obtained from the last registration is not greater than the reference overlap and the second registration number is equal to the reference number, and to adopt the third registration method to register the target image according to the reference image to obtain the third alternative registration image. The third registration method is different from the first registration method and the second registration method.

[0161] In another possible implementation, the registration method includes at least two of the following:

[0162] Based on the reference positions of multiple points in the reference image, multiple points in the target image are moved to obtain the registered image;

[0163] Based on the reference positions of multiple points in the reference image, multiple points in the target image are mapped to the target space to obtain the registered image;

[0164] The image registration model is invoked to register the target image with the reference image, resulting in a registered image.

[0165] In another possible implementation, the reference image and the target image are computed tomography (CT) images of the brain, see [link to relevant documentation]. Figure 10 The device also includes:

[0166] The preprocessing module 906 is used to perform de-skulling on the original reference image and target image to obtain the processed reference image and target image.

[0167] In another possible implementation, the reference image and the target image are computed tomography (CT) images of the brain, see [link to relevant documentation]. Figure 10 The device also includes:

[0168] The preprocessing module 906 is also used to perform desquamation on the registered image to obtain the processed registered image.

[0169] In another possible implementation, the reference image and the target image are computed tomography (CT) images of the brain, see [link to relevant documentation]. Figure 10 The device also includes:

[0170] The preprocessing module 906 is also used to determine the absorption parameters corresponding to each location point in the original reference image and the target image, respectively. The absorption parameters are used to represent the degree of absorption of rays by the location point.

[0171] The preprocessing module 906 is further configured to adjust the absorption parameters that are smaller than the first reference parameter among the multiple absorption parameters corresponding to the reference image and the target image to the first reference parameter, and adjust the absorption parameters that are larger than the second reference parameter among the multiple absorption parameters to the second reference parameter, so as to obtain the adjusted reference image and the adjusted target image.

[0172] In another possible implementation, the image group includes multiple images. The registration image determination module 904 is further configured to determine the maximum overlap among multiple overlaps of the multiple registration images after determining the registration images corresponding to the multiple image groups, and determine the registration image corresponding to the maximum overlap as the registration image corresponding to the target image.

[0173] All of the above-mentioned optional technical solutions can be combined in any way to form the optional embodiments of this application, and will not be described in detail here.

[0174] It should be noted that the image registration device provided in the above embodiments is only illustrated by the division of the above functional modules when registering images. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above. In addition, the image registration device and the image registration method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0175] This application also provides a computer device, which includes a processor and a memory. The memory stores at least one computer program, which is loaded and executed by the processor to perform the operations performed in the image registration method of the above embodiments.

[0176] Optionally, the computer device is provided as a terminal. Figure 11This is a schematic diagram of the structure of a terminal 1100 provided in an embodiment of this application. The terminal 1100 can be a portable mobile terminal, such as a smartphone, tablet computer, MP3 player (Moving Picture Experts Group Audio Layer III), MP4 player (Moving Picture Experts Group Audio Layer IV), laptop computer, or desktop computer. The terminal 1100 may also be referred to as user equipment, portable terminal, laptop terminal, desktop terminal, or other names.

[0177] Terminal 1100 includes a processor 1101 and a memory 1102.

[0178] Processor 1101 may include one or more processing cores, such as a quad-core processor or an octa-core processor. Processor 1101 may be implemented using at least one hardware form selected from DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). Processor 1101 may also include a main processor and a coprocessor. The main processor, also known as a CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, processor 1101 may integrate a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, processor 1101 may also include an AI (Artificial Intelligence) processor, which is used to handle computational operations related to machine learning.

[0179] The memory 1102 may include one or more computer-readable storage media, which may be non-transitory. The memory 1102 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In some embodiments, the non-transitory computer-readable storage media in the memory 1102 are used to store at least one computer program, which is executed by the processor 1101 to implement the image registration method provided in the method embodiments of this application.

[0180] In some embodiments, the terminal 1100 may also optionally include a peripheral device interface 1103 and at least one peripheral device. The processor 1101, memory 1102, and peripheral device interface 1103 can be connected via a bus or signal line. Each peripheral device can be connected to the peripheral device interface 1103 via a bus, signal line, or circuit board. Specifically, the peripheral device includes at least one of the following: a radio frequency circuit 1104, a display screen 1105, a camera assembly 1106, an audio circuit 1107, a positioning assembly 1108, and a power supply 1109.

[0181] Peripheral device interface 1103 can be used to connect at least one I / O (Input / Output) related peripheral device to processor 1101 and memory 1102. In some embodiments, processor 1101, memory 1102 and peripheral device interface 1103 are integrated on the same chip or circuit board; in some other embodiments, any one or two of processor 1101, memory 1102 and peripheral device interface 1103 can be implemented on separate chips or circuit boards, which is not limited in this embodiment.

[0182] The radio frequency (RF) circuit 1104 is used to receive and transmit RF (Radio Frequency) signals, also known as electromagnetic signals. The RF circuit 1104 communicates with communication networks and other communication devices via electromagnetic signals. The RF circuit 1104 converts electrical signals into electromagnetic signals for transmission, or converts received electromagnetic signals back into electrical signals. Optionally, the RF circuit 1104 includes: an antenna system, an RF transceiver, one or more amplifiers, a tuner, an oscillator, a digital signal processor, a codec chipset, a user identity module card, etc. The RF circuit 1104 can communicate with other terminals via at least one wireless communication protocol. This wireless communication protocol includes, but is not limited to: the World Wide Web, metropolitan area networks, intranets, various generations of mobile communication networks (2G, 3G, 4G, and 5G), wireless local area networks, and / or WiFi (Wireless Fidelity) networks. In some embodiments, the RF circuit 1104 may also include circuitry related to NFC (Near Field Communication), which is not limited in this application.

[0183] Display screen 1105 is used to display a UI (User Interface). This UI may include graphics, text, icons, videos, and any combination thereof. When display screen 1105 is a touch display screen, it also has the ability to collect touch signals on or above its surface. These touch signals can be input as control signals to processor 1101 for processing. In this case, display screen 1105 can also be used to provide virtual buttons and / or a virtual keyboard, also known as soft buttons and / or a soft keyboard. In some embodiments, there may be one display screen 1105, disposed on the front panel of terminal 1100; in other embodiments, there may be at least two display screens, disposed on different surfaces of terminal 1100 or in a folded design; in still other embodiments, display screen 1105 may be a flexible display screen, disposed on a curved or folded surface of terminal 1100. Furthermore, display screen 1105 may be configured as a non-rectangular, irregular shape, i.e., a non-rectangular screen. The display screen 1105 can be made of materials such as LCD (Liquid Crystal Display) and OLED (Organic Light-Emitting Diode).

[0184] The camera assembly 1106 is used to acquire images or videos. Optionally, the camera assembly 1106 includes a front-facing camera and a rear-facing camera. The front-facing camera is disposed on the front panel of the terminal, and the rear-facing camera is disposed on the back of the terminal. In some embodiments, there are at least two rear-facing cameras, which are any one of a main camera, a depth-sensing camera, a wide-angle camera, and a telephoto camera, to achieve background blurring by fusion of the main camera and the depth-sensing camera, panoramic shooting by fusion of the main camera and the wide-angle camera, VR (Virtual Reality) shooting, or other fusion shooting functions. In some embodiments, the camera assembly 1106 may also include a flash. The flash can be a single-color temperature flash or a dual-color temperature flash. A dual-color temperature flash refers to a combination of a warm light flash and a cool light flash, which can be used for light compensation at different color temperatures.

[0185] The audio circuit 1107 may include a microphone and a speaker. The microphone is used to collect sound waves from the user and the environment, converting the sound waves into electrical signals that are input to the processor 1101 for processing, or input to the radio frequency circuit 1104 for voice communication. For stereo sound acquisition or noise reduction purposes, multiple microphones may be used, each positioned at a different location on the terminal 1100. The microphone may also be an array microphone or an omnidirectional microphone. The speaker is used to convert electrical signals from the processor 1101 or the radio frequency circuit 1104 into sound waves. The speaker may be a conventional diaphragm speaker or a piezoelectric ceramic speaker. When the speaker is a piezoelectric ceramic speaker, it can convert electrical signals not only into audible sound waves but also into inaudible sound waves for purposes such as distance measurement. In some embodiments, the audio circuit 1107 may also include a headphone jack.

[0186] The positioning component 1108 is used to locate the current geographical location of the terminal 1100 in order to enable navigation or LBS (Location Based Service). The positioning component 1108 can be a positioning component based on the US GPS (Global Positioning System), China's BeiDou system, Russia's Granas positioning system, or the European Union's Galileo positioning system.

[0187] Power supply 1109 is used to power the various components in terminal 1100. Power supply 1109 can be AC ​​power, DC power, a disposable battery, or a rechargeable battery. When power supply 1109 includes a rechargeable battery, the rechargeable battery can be a wired rechargeable battery or a wireless rechargeable battery. A wired rechargeable battery is a battery that is charged via a wired line, and a wireless rechargeable battery is a battery that is charged via a wireless coil. The rechargeable battery can also be used to support fast charging technology.

[0188] In some embodiments, the terminal 1100 further includes one or more sensors 1110. The one or more sensors 1110 include, but are not limited to: an accelerometer 1111, a gyroscope 1112, a pressure sensor 1113, a fingerprint sensor 1114, an optical sensor 1115, and a proximity sensor 1116.

[0189] Accelerometer 1111 can detect the magnitude of acceleration along the three axes of a coordinate system established with terminal 1100. For example, accelerometer 1111 can be used to detect the components of gravitational acceleration along the three axes. Processor 1101 can control display screen 1105 to display the user interface in either a landscape or portrait view based on the gravitational acceleration signal acquired by accelerometer 1111. Accelerometer 1111 can also be used for collecting game or user motion data.

[0190] The gyroscope sensor 1112 can detect the orientation and rotation angle of the terminal 1100. The gyroscope sensor 1112 can work in conjunction with the accelerometer sensor 1111 to collect the user's 3D movements on the terminal 1100. Based on the data collected by the gyroscope sensor 1112, the processor 1101 can perform the following functions: motion sensing (e.g., changing the UI based on the user's tilt), image stabilization during shooting, game control, and inertial navigation.

[0191] The pressure sensor 1113 can be disposed on the side bezel of the terminal 1100 and / or on the lower layer of the display screen 1105. When the pressure sensor 1113 is disposed on the side bezel of the terminal 1100, it can detect the user's grip signal on the terminal 1100, and the processor 1101 can perform left / right hand recognition or quick operation based on the grip signal collected by the pressure sensor 1113. When the pressure sensor 1113 is disposed on the lower layer of the display screen 1105, the processor 1101 can control the operable controls on the UI interface based on the user's pressure operation on the display screen 1105. The operable controls include at least one of button controls, scroll bar controls, icon controls, and menu controls.

[0192] The fingerprint sensor 1114 is used to collect the user's fingerprint. The processor 1101 identifies the user's identity based on the fingerprint collected by the fingerprint sensor 1114, or the fingerprint sensor 1114 identifies the user's identity based on the collected fingerprint. When the user's identity is identified as trusted, the processor 1101 authorizes the user to perform relevant sensitive operations, including unlocking the screen, viewing encrypted information, downloading software, making payments, and changing settings. The fingerprint sensor 1114 can be located on the front, back, or side of the terminal 1100. When the terminal 1100 has physical buttons or a manufacturer's logo, the fingerprint sensor 1114 can be integrated with the physical buttons or manufacturer's logo.

[0193] An optical sensor 1115 is used to collect ambient light intensity. In one embodiment, the processor 1101 can control the display brightness of the display screen 1105 based on the ambient light intensity collected by the optical sensor 1115. Specifically, when the ambient light intensity is high, the display brightness of the display screen 1105 is increased; when the ambient light intensity is low, the display brightness of the display screen 1105 is decreased. In another embodiment, the processor 1101 can also dynamically adjust the shooting parameters of the camera assembly 1106 based on the ambient light intensity collected by the optical sensor 1115.

[0194] The proximity sensor 1116, also known as a distance sensor, is installed on the front panel of the terminal 1100. The proximity sensor 1116 is used to detect the distance between the user and the front of the terminal 1100. In one embodiment, when the proximity sensor 1116 detects that the distance between the user and the front of the terminal 1100 is gradually decreasing, the processor 1101 controls the display screen 1105 to switch from a screen-on state to a screen-off state; when the proximity sensor 1116 detects that the distance between the user and the front of the terminal 1100 is gradually increasing, the processor 1101 controls the display screen 1105 to switch from a screen-off state to a screen-on state.

[0195] Those skilled in the art will understand that Figure 11 The structure shown does not constitute a limitation on terminal 1100 and may include more or fewer components than shown, or combine certain components, or use different component arrangements.

[0196] Optionally, the computer device is provided as a server. Figure 12 This is a schematic diagram of a server structure provided in an embodiment of this application. The server 1200 can vary significantly due to different configurations or performance. It may include one or more Central Processing Units (CPUs) 1201 and one or more memories 1202. The memories 1202 store at least one computer program, which is loaded and executed by the processor 1201 to implement the methods provided in the various method embodiments described above. Of course, the server may also have wired or wireless network interfaces, a keyboard, and input / output interfaces for input and output. The server may also include other components for implementing device functions, which will not be elaborated upon here.

[0197] This application also provides a computer-readable storage medium storing at least one computer program, which is loaded and executed by a processor to implement the operations performed in the image registration method of the above embodiments.

[0198] This application also provides a computer program product or computer program, which includes computer program code stored in a computer-readable storage medium. A processor of a computer device reads the computer program code from the computer-readable storage medium and executes the computer program code, causing the computer device to perform the operations performed in the image registration method of the above embodiments.

[0199] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0200] The above are merely optional embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present application should be included within the protection scope of the present application.

Claims

1. An image registration method, characterized in that, The method includes: Acquire an image group, which includes an original reference image and a target image, wherein the reference image and the target image are images of the target object at different time points; The original reference image and target image are subjected to desquamation processing to obtain a desquamated image and a target image; an image registration network is used to register the desquamated target image and the reference image to obtain a first registered image; The image registration network is used to register the original target image and the reference image to obtain a registered image containing the skull region; the registered image containing the skull region is then subjected to skull removal processing to obtain a second registered image after skull removal. The absorption parameters in the original target image and reference image are adjusted to obtain the adjusted target image and reference image; the image registration network is used to register the adjusted target image and reference image to obtain a registered image containing the skull region; the registered image containing the skull region is subjected to skull removal processing to obtain a third registered image after skull removal. The maximum overlap among the overlaps of the first, second, and third registration images is determined, and the registration image corresponding to the maximum overlap is determined as the registration image corresponding to the target image. The overlap refers to the overlap with the reference image.

2. The method according to claim 1, characterized in that, The process of registering any set of target images and reference images using the image registration network described above includes: Using a first registration method, the target image is registered based on the reference image to obtain a first candidate registration image; In response to the first overlap being no greater than the reference overlap, a second registration method is adopted to register the target image according to the reference image to obtain a second candidate registration image. The first registration method is different from the second registration method. The first overlap refers to the overlap between the first candidate registration image and the reference image. In response to a second overlap degree being greater than the reference overlap degree, the second candidate registration image is determined as the registration image corresponding to the target image, wherein the second overlap degree refers to the overlap degree between the second candidate registration image and the reference image.

3. The method according to claim 2, characterized in that, After registering the target image according to the reference image using the first registration method to obtain a first candidate registration image, the method further includes: In response to the first overlap being greater than the reference overlap, the first candidate registration image is determined as the registration image corresponding to the target image.

4. The method according to claim 2, characterized in that, After registering the target image according to the reference image using the first registration method to obtain a first candidate registration image, the method further includes: In response to the first overlap being no greater than the reference overlap and the first registration count being less than the reference count, the step of using the first registration method to register the target image according to the reference image to obtain a first candidate registration image continues until the first overlap obtained by the last registration is greater than the reference overlap, or the first registration count is equal to the reference count, wherein the first registration count refers to the number of times the target image is registered using the first registration method.

5. The method according to claim 4, characterized in that, In response to the first overlap being no greater than the reference overlap, a second registration method is adopted to register the target image according to the reference image, resulting in a second candidate registration image, including: In response to the fact that the first overlap obtained by the last registration is not greater than the reference overlap, and the first registration number is equal to the reference number, the second registration method is adopted to register the target image according to the reference image to obtain a second candidate registration image.

6. The method according to claim 4, characterized in that, In response to the first overlap being no greater than the reference overlap, a second registration method is adopted to register the target image according to the reference image, resulting in a second candidate registration image, including: In response to the second overlap being no greater than the reference overlap and the second registration count being less than the reference count, the step of using the second registration method to register the target image according to the reference image to obtain a second candidate registration image continues until the second overlap obtained by the last registration is greater than the reference overlap, or the second registration count is equal to the reference count, where the first registration count refers to the number of times the target image is registered using the second registration method.

7. The method according to claim 6, characterized in that, The method further includes: In response to the fact that the second overlap obtained from the last registration is not greater than the reference overlap, and the second registration number is equal to the reference number, the maximum overlap among the multiple first overlaps and multiple second overlaps obtained from the registration is determined, and the candidate registration image corresponding to the maximum overlap is determined as the registration image corresponding to the target image; or, In response to the fact that the second overlap obtained by the last registration is not greater than the reference overlap, and the second registration number is equal to the reference number, a third registration method is adopted to register the target image according to the reference image to obtain a third alternative registration image. The third registration method is different from the first registration method and the second registration method.

8. The method according to claim 2, characterized in that, Registration methods include at least two of the following: Based on the reference positions of multiple points in the reference image, multiple points in the target image are moved to obtain a registered image. Based on the reference positions of multiple locations in the reference image, multiple locations in the target image are mapped to the target space to obtain a registered image; The image registration model is invoked to register the target image with the reference image, resulting in a registered image.

9. The method according to claim 1, characterized in that, The reference image and the target image are computed tomography (CT) images of the brain. Adjusting the absorption parameters in the original target image and reference image to obtain the adjusted target image and reference image includes: The absorption parameters corresponding to each location point in the original reference image and the target image are determined respectively, and the absorption parameters are used to represent the degree of absorption of the rays by the location point; The absorption parameters of the reference image and the target image that are smaller than the first reference parameter are adjusted to the first reference parameter, and the absorption parameters of the reference image and the target image that are larger than the second reference parameter are adjusted to the second reference parameter, so as to obtain the adjusted reference image and the adjusted target image.

10. An image registration device, characterized in that, The device includes: An image group acquisition module is used to acquire an image group, which includes an original reference image and a target image, wherein the reference image and the target image are images of the target object at different time points; The registration image determination module is used for: The original reference image and target image are subjected to desquamation processing to obtain a desquamated image and a target image; an image registration network is used to register the desquamated target image and the reference image to obtain a first registered image; The image registration network is used to register the original target image and the reference image to obtain a registered image containing the skull region; the registered image containing the skull region is then subjected to skull removal processing to obtain a second registered image after skull removal. The absorption parameters in the original target image and reference image are adjusted to obtain the adjusted target image and reference image; the image registration network is used to register the adjusted target image and reference image to obtain a registered image containing the skull region; the registered image containing the skull region is subjected to skull removal processing to obtain a third registered image after skull removal. The maximum overlap among the overlaps of the first, second, and third registration images is determined, and the registration image corresponding to the maximum overlap is determined as the registration image corresponding to the target image. The overlap refers to the overlap with the reference image.

11. The apparatus according to claim 10, characterized in that, The registration image determination module includes: The first registration module is used to register the target image with the reference image using a first registration method to obtain a first candidate registration image; The second registration module is used to register the target image according to the reference image in response to the first overlap degree not being greater than the reference overlap degree, and to obtain a second candidate registration image by adopting a second registration method. The first registration method is different from the second registration method. The first overlap degree refers to the overlap degree between the first candidate registration image and the reference image. The registration image determination module is used to determine the second candidate registration image as the registration image corresponding to the target image in response to the second overlap being greater than the reference overlap. The second overlap refers to the overlap between the second candidate registration image and the reference image.

12. The apparatus according to claim 11, characterized in that, The first registration module is further configured to, in response to the first overlap not being greater than the reference overlap and the first registration number being less than the reference number, continue to execute the step of using the first registration method to register the target image according to the reference image to obtain a first candidate registration image, until the first overlap obtained by the last registration is greater than the reference overlap, or the first registration number is equal to the reference number, wherein the first registration number refers to the number of times the target image is registered using the first registration method.

13. A computer device, characterized in that, The computer device includes a processor and a memory, the memory storing at least one computer program, which is loaded and executed by the processor to perform the operations performed in the image registration method as described in any one of claims 1 to 9.

14. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores at least one computer program, which is loaded and executed by a processor to perform the operations performed in the image registration method as described in any one of claims 1 to 9.

Images