Matching block search method, apparatus and storage medium in image registration

Abstract

The application relates to a matching block search method, device and storage medium in image registration, wherein the matching block search method in image registration comprises the following steps: acquiring a reference image, a to-be-matched image, control point information and preset search features, wherein the preset search features comprise a preset matching size; determining an initial matching block according to the control point information and the preset matching size; determining image information of the initial matching block, and determining target search features according to the image information; determining a reference matching block in the reference image according to the control point information and the target search features, and searching for a target matching block corresponding to the reference matching block in the to-be-matched image. Through the application, when the method is applied to X-ray contrast image registration, the problem that the existing X-ray contrast image registration algorithm cannot adaptively determine the size or search range of a matching block can be solved.

Description

Technical Field

[0001] This application relates to the field of image processing, and in particular to a matching block search method, apparatus and storage medium in image registration. Background Technology

[0002] In X-ray angiography, it is often necessary to register images acquired at different times. For example, Digital Subtraction Angiography (DSA) is an irreplaceable vascular visualization tool in clinical cardiovascular diagnosis and treatment due to its high resolution and contrast. Typically, DSA first takes continuous X-ray images of the area of ​​interest to the patient and fixes a frame before contrast agent injection as a mask. Then, a contrast agent, such as iodine, is injected into the patient, and X-rays are taken again to obtain real-time filling images (iodine, etc., entering the blood vessels and leaving a unique image after X-ray exposure, called a filling image). Finally, the mask is subtracted from the filling image, theoretically yielding an image containing only the blood vessels. However, since the mask and filling images are acquired at different times, unavoidable factors such as patient movement, breathing, heartbeat, and visceral peristalsis can cause inaccurate alignment between corresponding pixels of the mask and filling images. The subtracted image often contains numerous motion artifacts, reducing the diagnostic value of DSA. Therefore, registration of the mask and filling images is necessary.

[0003] In X-ray imaging registration, a set of control points needs to be determined based on the mask first. Then, the corresponding coordinate points in the film are determined as the starting point, and the size of the matching block and its search range are set. The most similar matching block is searched within the search range. The accuracy of the matching block search plays a crucial role in the image registration and subtraction effect.

[0004] The size and search range of the matching block are preset by the user based on experience. Existing X-ray image registration algorithms cannot adaptively determine the size and search range of the matching block. If the matching block size is set too large, it may be impossible to find the optimal matching block and displacement. If the matching block size is set too small, incorrect matching blocks and displacements may be found. If the matching block search range is set too large, it increases the computational load and the possibility of search errors. If the matching block search range is set too small, more similar matching blocks may be missed.

[0005] There is currently no effective solution to the problem that X-ray imaging registration algorithms cannot adaptively determine the size of the matching block or the search range in related technologies. Summary of the Invention

[0006] This invention provides a matching block search method, apparatus, and storage medium for image registration, to solve the problem that X-ray imaging registration algorithms in related technologies cannot adaptively determine the size and search range of the matching block.

[0007] In a first aspect, the present invention provides a matching block search method in image registration, the method comprising:

[0008] Acquire a reference image, an image to be matched, control point information, and preset search features, wherein the preset search features include a preset matching size;

[0009] The initial matching block is determined based on the control point information and the preset matching size;

[0010] Determine the image information of the initial matching block, and determine the target search features based on the image information;

[0011] Based on the control point information and target search features, a baseline matching block is determined in the baseline image, and a target matching block corresponding to the baseline matching block is searched in the image to be matched.

[0012] In some embodiments, the target search feature includes target matching size;

[0013] The step of determining the image information of the initial matching block and determining the target search features based on the image information includes:

[0014] Determine the information entropy of the initial matching block;

[0015] The target matching size is determined based on the information entropy.

[0016] In some embodiments, determining the target matching size based on the information entropy includes:

[0017] Based on the preset adjustment relationship of the preset matching size according to the information entropy, the preset matching size is adjusted by the information entropy to obtain the target matching size.

[0018] In some embodiments, determining the target matching size based on the information entropy includes:

[0019] The target matching size is determined by the information entropy based on a preset mapping relationship between the information entropy and the target matching size.

[0020] In some embodiments, the target search feature includes a target search range;

[0021] The step of determining the initial matching block based on the control point information and the preset matching size includes:

[0022] Based on the control point information and the preset matching size, a first initial matching block is determined in the reference image, and a second initial matching block is determined in the image to be matched;

[0023] The step of determining the image information of the initial matching block and determining the target search features based on the image information includes:

[0024] Determine the similarity between the first initial matching block and the second initial matching block;

[0025] The target search range is determined based on the similarity.

[0026] In some embodiments, the preset search feature further includes a preset search range;

[0027] Determining the target search range based on the similarity includes:

[0028] Based on the preset adjustment relationship of the similarity to the preset search range, the preset search range is adjusted by the similarity to obtain the target search range.

[0029] In some embodiments, determining the target search range based on the similarity includes:

[0030] The target search range is determined based on the preset mapping relationship between the similarity and the target search range.

[0031] In some embodiments, determining the similarity between the first initial matching block and the second initial matching block includes:

[0032] The similarity is determined based on relevant information between the first initial matching block and the second initial matching block; wherein, the relevant information includes at least one of the following:

[0033] Mutual information, normalized mutual information, correlation coefficient, normalized correlation coefficient, maximum subtraction histogram energy, sum of squared differences, structural similarity.

[0034] Secondly, the present invention provides a matching block search device for image registration, the device comprising:

[0035] The information acquisition module is used to acquire a reference image, an image to be matched, control point information, and preset search features, wherein the preset search features include a preset matching size;

[0036] The first determining module is used to determine the initial matching block based on the control point information and the preset matching size;

[0037] The second determining module is used to determine the image information of the initial matching block and determine the target search features based on the image information.

[0038] The matching search module is used to determine a reference matching block in the reference image based on the control point information and target search features, and to search for a target matching block in the image to be matched that corresponds to the reference matching block.

[0039] Thirdly, the present invention provides an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the matching block search method in image registration described in the first aspect above.

[0040] In a fourth aspect, the present invention provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the matching block search method in image registration as described in the first aspect above.

[0041] Compared with related technologies, the matching block search method, apparatus, and storage medium provided in this invention for image registration adjust preset search features to obtain target search features based on the image information of the initial matching block. These target search features are more suitable for the current search process than the user-defined preset search features, thereby improving the efficiency and accuracy of matching block search. When this method is applied to X-ray contrast image registration, it can solve the problem that existing X-ray contrast image registration algorithms cannot adaptively determine the size or search range of the matching block.

[0042] Details of one or more embodiments of this application are set forth in the following drawings and description to make other features, objects and advantages of this application more readily apparent. Attached Figure Description

[0043] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0044] Figure 1 This is a hardware structure block diagram of a terminal that executes the matching block search method in image registration according to the present invention;

[0045] Figure 2 This is a flowchart of the matching block search method in image registration in an embodiment of the present invention;

[0046] Figure 3 This is a flowchart of a matching block search method in image registration according to one embodiment of the present invention;

[0047] Figure 4This is a flowchart of a matching block search method in image registration according to one embodiment of the present invention;

[0048] Figure 5 It is the reference image in the matching block search method of this invention embodiment;

[0049] Figure 6 This refers to the image to be matched in the matching block search method of this invention.

[0050] Figure 7 This is a structural block diagram of the matching block search device in image registration in an embodiment of the present invention. Detailed Implementation

[0051] To better understand the purpose, technical solution, and advantages of this application, the application is described and explained below in conjunction with the accompanying drawings and embodiments.

[0052] Unless otherwise defined, the technical or scientific terms used in this application shall have the general meaning as understood by one of ordinary skill in the art to which this application pertains. Words such as “a,” “an,” “an,” “the,” “the,” and “these,” used in this application, do not indicate quantitative limitation and may be singular or plural. The terms “comprising,” “including,” “having,” and any variations thereof used in this application are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that comprises a series of steps or modules (units) is not limited to the listed steps or modules (units) but may include steps or modules (units) not listed, or may include other steps or modules (units) inherent to such processes, methods, products, or devices. The terms “connected,” “linked,” and “coupled,” used in this application, are not limited to physical or mechanical connections but may include electrical connections, whether direct or indirect. The term “multiple” used in this application refers to two or more. The "and / or" operator describes the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: A alone, A and B simultaneously, and B alone. Typically, the character " / " indicates that the objects before and after it are in an "or" relationship. The terms "first," "second," and "third," etc., used in this application are merely for distinguishing similar objects and do not represent a specific ordering of the objects.

[0053] The method embodiments provided in this example can be executed on a terminal, computer, or similar computing device. For example, it can run on a terminal. Figure 1 This is a hardware structure block diagram of a terminal that executes the matching block search method in image registration according to the present invention. For example... Figure 1 As shown, a terminal may include one or more ( Figure 1Only one is shown in the diagram. A processor 102 and a memory 104 for storing data are also included. The processor 102 may be, but is not limited to, a microprocessor (MCU) or a programmable logic device (FPGA). The terminal may also include a transmission device 106 for communication functions and an input / output device 108. Those skilled in the art will understand that… Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the terminal described above. For example, the terminal may also include components that are larger than... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown are illustrated.

[0054] The memory 104 can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to the matching block search method in image registration in this invention. The processor 102 executes various functional applications and data processing by running the computer programs stored in the memory 104, thereby implementing the methods described above. The memory 104 may include high-speed random access memory and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0055] The transmission device 106 is used to receive or send data via a network. This network includes a wireless network provided by the terminal's communication provider. In one example, the transmission device 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 can be a Radio Frequency (RF) module used for wireless communication with the Internet.

[0056] This invention provides a matching block search method in image registration. Figure 2 This is a flowchart of the matching block search method in image registration according to an embodiment of the present invention. Figure 2 As shown, the process includes the following steps:

[0057] Step S210: Obtain the reference image, the image to be matched, control point information, and preset search features, including preset matching size.

[0058] In this step, the two images to be registered are first acquired. Typically, during image registration, a first control point is set in the first image, and a matching block of a preset size is determined centered on the first control point. Then, a second control point is set in the second image at the same position as the first control point, and a search range is determined centered on the second control point. Finally, within this search range, a matching block with the highest similarity to the matching block in the first image and the same size is searched, and the center of this matching block is used as the matching point of the first control point, forming a control point pair. The multiple sets of control point pairs obtained are used for the final image transformation. The matching block is an image block. In this invention, the first and second images are defined as the reference image and the image to be matched, respectively. As described above, before registration, control point information and preset search features need to be acquired. The control point information refers to the positional information of control points in the two images. Control points are reference points selected on the images for establishing geometric transformation functions; in this invention, they are reference points used for image block matching searches. Generally, each set of control point information can contain two sets of coordinate data, namely the coordinates of the two control points in the control point pair on the two images being matched. For example, the control point information obtained in this step includes the coordinates of the first control point in the first image and the coordinates of the second control point corresponding to the first control point in the second image. Simultaneously, during the search process, a matching point corresponding to the first control point is determined in the second image, and the first control point and its corresponding matching block form a new control point pair. Therefore, after the search is completed, control point information for subsequent registration is generated, which includes the coordinates of the first control point in the first image and the coordinates of the matching point corresponding to the first control point in the second image. Search features mainly include the matching size and search range. Preset search features include a preset matching size and a preset search range. The preset matching size is the preset size of the matching block. For example, the preset matching size can be 100×100, where 100 refers to 100 pixels, meaning the length and width of the matching block each correspond to 100 pixels. A matching block of this size is a square image block containing 10,000 pixels.

[0059] Furthermore, when this method is applied to X-ray imaging registration, the mask without contrast agent can be designated as the reference image, and the film with contrast agent can be designated as the image to be registered.

[0060] Step S220: Determine the initial matching block based on the control point information and the preset matching size.

[0061] In this step, control points are first determined in the image based on control point information. This can be achieved by a computer device generating control points in the image based on coordinate information input by the user, or by the user directly determining the control point positions in the image. Then, an initial matching block with a preset matching size is determined in the image, centered on the control points. The initial matching block is the image block centered on the control points; that is, in the image to be matched, the initial matching block is located at the center of the search range. It should be noted that in different embodiments, the initial matching block will be determined in different images. In some embodiments, the initial matching block is determined only in the reference image or the image to be matched. In other embodiments, the initial matching block needs to be determined in both the reference image and the image to be matched.

[0062] Step S230: Determine the image information of the initial matching block, and determine the target search features based on the image information.

[0063] In this step, after the initial image patch is determined, the image information within the initial image patch is analyzed. Based on the image information, it is determined whether to adjust the preset search features or generate new search features, i.e., to determine the target search features. The target search features include the target matching size and the target search range. Since the target search features are generated based on the image information of the initial matching patch during the current search process, they are more suitable for the current search process than the preset search features given by the user, thereby improving the search efficiency and accuracy of the matching patch. For example, the image information can be information entropy, which is a statistical form of image features that reflects the average amount of information in the image. When the information entropy of the initial matching patch in the reference image or the image to be matched is low, it indicates that the initial matching patch contains less information, which will reduce the matching accuracy between matching patches. In this case, the matching patch size can be increased to allow it to contain more information. Conversely, the matching patch size can be decreased. The adjusted matching patch size is the target matching size. Additionally, the image information can also be the similarity between the initial matching patch of the reference image and the initial matching patch of the image to be matched. A higher similarity indicates a higher degree of matching between the two initial matching blocks, meaning their relative positional deviation is small. In this case, the preset search range can be reduced, or a smaller search range can be generated to reduce computational load and speed up the search. Conversely, a lower similarity indicates a lower similarity, meaning the preset search range can be increased, or a larger search range can be generated. The adjusted search range or the newly generated search range is the target search range.

[0064] Step S240: Based on the control point information and target search features, determine the reference matching block in the reference image, and search for the target matching block corresponding to the reference matching block in the image to be matched.

[0065] In this step, after the target search features are determined, the matching block search process is executed. The target search features include the target matching size and the target search range. First, control points are determined in the reference image and the image to be matched. Then, a reference matching block with the target matching size is determined in the reference image centered on the control points. Next, the target search range is determined in the image to be matched centered on the control points, and the target matching block with the same size and most similarity to the reference matching block is searched within the target search range.

[0066] It should be further noted that, in some embodiments, when performing step S230 above, adjustments can be made only to some preset search features based on the image information of the initial matching block. For example, in some embodiments, only the preset matching size can be adjusted based on the image information, without adjusting the preset search range; in this case, the preset search range is directly used as the target search range. In other embodiments, only the preset search range can be adjusted based on the image information, without adjusting the preset matching size; in this case, the preset matching size is directly used as the target matching size.

[0067] Through the above steps, an initial matching block is first determined in the reference image and / or the image to be matched using control point information and preset search features. Then, based on the image information of the initial matching block, the preset search features are adjusted to generate a new matching size and / or search range, thereby obtaining the target search features. Finally, the matching block is searched based on the target search features and the pre-given preset search features. This invention obtains the target search features by adjusting the preset search features using the image information of the initial matching block. Compared to the user-given preset search features, the target search features are more suitable for the current search process, thereby improving the efficiency and accuracy of the matching block search. When this method is applied to X-ray contrast image registration, it can solve the problem that existing X-ray contrast image registration algorithms cannot adaptively determine the size or search range of the matching block.

[0068] Figure 3 This is a flowchart of a matching block search method in image registration according to one embodiment of the present invention. (Refer to...) Figure 3 In some embodiments, the target search features include the target matching size; step S230, determining the image information of the initial matching block, and determining the target search features based on the image information includes: step S231, determining the information entropy of the initial matching block; step S232, determining the target matching size based on the information entropy.

[0069] This embodiment primarily determines the target matching size based on the image information of the initial matching block. The first step is to determine the information entropy of the initial matching block. The initial matching block is a matching block centered on a control point and having a preset matching size. Initial matching blocks exist in both the reference image and the image to be matched. This step can determine the information entropy of the initial matching block in the reference image, the initial matching block in the image to be matched, or a combination of the information entropies of the initial matching blocks in both images, such as summing or averaging the information entropies of the two initial matching blocks. Information entropy is a statistical form of image feature, reflecting the average amount of information in an image. If the information entropy of the initial matching block is large, it indicates richer image information within the matching block, allowing for a smaller matching block size. Conversely, if the information entropy of the initial matching block is small, it indicates less image information within the matching block, allowing for a larger matching block size. For example, when the image is a human medical image, if the information entropy value is large, it means that the human structure information in the matching block is richer. For example, if the control point is located in an area with large artifacts such as the nasal cavity, bones, and diaphragm, the size of the matching block will be reduced. If the information entropy value is small, it means that the human structure information in the matching block is less. For example, if the control point is located in an area such as the edge of the skull or the edge of the human contour, the matching block contains more blank areas. Therefore, the size of the matching block will be increased to contain more human structure information.

[0070] It should be noted that the calculation method for information entropy differs for different types of images. For example, information entropy for color images is calculated based on color information, while information entropy for grayscale images is calculated based on grayscale information. This invention uses grayscale images as an example to introduce a method for calculating information entropy. The formula for calculating the information entropy of matching block I is as follows:

[0071]

[0072] Where N represents the maximum value of grayscale, p i H(I) is the probability of the i-th gray value appearing in matching block I, determined by the gray histogram of matching block I. The range of information entropy is: 0 ≤ H(I) ≤ logN. The larger the information entropy H(I), the more information is contained in the matching block.

[0073] To better calculate information entropy, it can be normalized. The normalization formula is as follows:

[0074]

[0075] Where, H(I) Normalized This represents the normalized information entropy, with a value range of [0, 1].

[0076] Furthermore, there are two main ways to determine the target matching size using the information entropy of the initial matching block.

[0077] In one embodiment, step S232, determining the target matching size based on information entropy, includes: adjusting the preset matching size based on the preset adjustment relationship of the preset matching size according to information entropy, and obtaining the target matching size.

[0078] In this embodiment, the preset matching size is mainly adjusted based on the information entropy of the initial matching block to obtain the target matching size. When the information entropy is large, the preset matching size is reduced; when the information entropy is small, the preset matching size is increased. The preset adjustment relationship is as follows:

[0079]

[0080] Where A is the preset matching size, A ′ For the target matching size, H(I) Normalized This represents the normalized information entropy, and α is a preset control parameter used to control the adjustment range of the information entropy to the preset matching size. For example, α can be set to 2. In this case, when the normalized information entropy is 0.5, the target matching size is equal to the preset matching size. When the normalized information entropy is greater than 0.5, the target matching size is less than the preset matching size, and the larger the information entropy, the smaller the target matching size.

[0081] In another embodiment, step S232, determining the target matching size based on information entropy, includes: determining the target matching size based on a preset mapping relationship between information entropy and target matching size using information entropy.

[0082] Unlike the previous embodiment, the determination of the target matching size in this embodiment is independent of the preset matching size. After establishing a preset mapping relationship between information entropy and target matching size, the target matching size can be directly determined using the information entropy and this preset mapping relationship. For example, when the normalized information entropy is 0.5, the target matching size can be directly determined as 100×100; when the normalized information entropy is 0.6, the target matching size can be directly determined as 80×80.

[0083] Figure 4 This is a flowchart of a matching block search method in image registration according to one embodiment of the present invention. (Refer to...) Figure 4In some embodiments, the target search features include the target search range; step S220, determining the initial matching block based on control point information and a preset matching size, includes: step S221, determining a first initial matching block in a reference image and a second initial matching block in an image to be matched based on control point information and a preset matching size; step S230, determining the image information of the initial matching block and determining the target search features based on the image information includes: step S233, determining the similarity between the first initial matching block and the second initial matching block; step S234, determining the target search range based on the similarity.

[0084] This embodiment primarily determines the target search range based on the image information of the initial matching blocks. First, the information entropy of the initial matching blocks is determined. An initial matching block is a matching block centered on a control point and having a preset matching size; it exists in both the reference image and the image to be matched. This step requires calculating the similarity between the initial matching blocks in the reference image and the image to be matched. Therefore, a first initial matching block needs to be determined in the reference image, and a second initial matching block needs to be determined in the image to be matched. Methods for calculating similarity include, but are not limited to, Mutual Information (MI), Normalized Mutual Information (NMI), Correlation Coefficient (CC), Normalized Correlation Coefficient (NCC), Energy of the histogram of differences (EHD), Sum of Squared Differences (SSD), and Structural Similarity (SSIM), among other similarity measurement algorithms. In this step, one of the similarity measurement algorithms mentioned above can be selected to calculate the similarity between the first and second initial matching blocks. For example, the mutual information between the first and second initial matching blocks can be calculated, and the mutual information can be used to characterize the similarity between them. It should be noted that the similarity measurement algorithm based on maximum subtraction histogram energy is usually applied to digital subtraction angiography images, while other similarity measurement algorithms are more general. If the similarity between the first and second matching blocks is large, it means that the matching degree between the two matching blocks is high, which also means that the relative positional deviation between the two initial matching blocks is small. In this case, the search range can be narrowed, thereby reducing the amount of computation. If the similarity between the first and second matching blocks is small, it means that the matching degree between the two matching blocks is small, which also means that the relative positional deviation between the two initial matching blocks is large. In this case, the search range needs to be increased to ensure that the correct matching block is found and to avoid the limitation of a fixed search range.

[0085] Furthermore, there are two main ways to determine the target search range based on the similarity between two initial matching blocks.

[0086] In one embodiment, the preset search features further include a preset search range; step S234, determining the target search range based on similarity includes: adjusting the preset search range based on similarity according to a preset adjustment relationship of similarity, thereby obtaining the target search range.

[0087] In this embodiment, the preset search range is mainly adjusted based on the similarity between two initial matching blocks to obtain the target search range. When the similarity is high, the preset search range is narrowed; when the similarity is low, the preset search range is increased. The preset adjustment relationship is as follows:

[0088]

[0089] Where B is the preset search range, B ′ Let NMI represent the similarity score, and β be a preset control parameter used to control the adjustment of the similarity score to the preset search range. From the above formula, it can be seen that the higher the similarity score, the smaller the target search range, and vice versa.

[0090] In another embodiment, step S234, determining the target search range based on similarity, includes: determining the target search range based on a preset mapping relationship between similarity and target search range.

[0091] Unlike the previous embodiment, the determination of the target search range in this embodiment is independent of the preset search range, therefore, the user does not need to pre-define the preset search range. After establishing a preset mapping relationship between similarity and the target search range, the target search range can be directly determined through similarity and this preset mapping relationship.

[0092] It should be noted that the search range in this invention can also be represented by a search step size. The search step size is defined as the distance between the matching block centered on the control point and the boundary of the search range. For example, if the search step size is set to 20, the value represents the number of pixels. In this case, extending 20 pixels outward from the matching block centered on the control point will yield the search range, which is 140×140 pixels.

[0093] Figure 5 This is the reference image in the matching block search method of this invention. Figure 6 This refers to the image to be matched in the matching block search method of this invention. (See reference...) Figure 5 and Figure 6The above method will be explained more intuitively below. First, a first control point 11 is determined in the reference image, and a second control point (not shown in the figure) is determined at the same position in the image to be matched. Then, using the first control point 11 and the second control point as the center, an initial matching block (not shown in the figure) is determined using a preset matching size, and the target matching size is determined based on the information entropy of the initial matching block. Next, using the first control point 11 as the center in the reference image, a reference matching block 21 is determined using the target matching size. And based on the similarity between the two initial matching blocks, a target search range 30 is determined in the image to be matched, centered on the second control point. Finally, the target matching block 22 most similar to the reference matching block 21 is found in the target search range, and the center point of the target matching block 22 is determined as the matching point 12 of the first control point 11. The first control point 11 and its matching point 12 constitute a set of matching point pairs, which are used for the final image transformation.

[0094] The following example, using the application of this method in X-ray imaging registration, will be used to specifically illustrate the matching block search method in this invention.

[0095] In one specific embodiment, the matching block search method is used to register a mask image and a contrast image in an X-ray imaging image. The matching block search method includes:

[0096] 1. Adaptively determine the size of the matching block (ROI).

[0097] First, set a fixed ROI size, for example, a square image patch with a side length of Size = 100. Then, calculate the information entropy of the image patch containing the control points in Contrast. Information entropy is a statistical form of image feature, reflecting the average amount of information in the image. The formula for calculating the information entropy of an image is as follows:

[0098]

[0099] Where N represents the maximum value of grayscale, p i H(I) is the probability of the i-th gray value appearing in image I, determined by the gray-level histogram of image I. The range of information entropy is: 0 ≤ H(I) ≤ logN. The larger the information entropy H(I), the more information is contained in the image.

[0100] Normalize the information entropy H(I):

[0101]

[0102] Where, H(I) Normalized This represents the normalized information entropy, with a value range of [0, 1].

[0103] According to H(I)Normalized The side length Size is adjusted based on the adaptive ROI. The adjusted Size is:

[0104]

[0105] Here, α is a fixed value, a control parameter determined empirically, used to adjust H(I). Normalized The adjustment range for Size.

[0106] If the information entropy H(I) is large, it means that the human body structure information in the image patch is richer. For example, if the control point is located in an area with large artifacts such as the nasal cavity, bones, or diaphragm, the size of the ROI should be reduced. If the information entropy H(I) is small, it means that the human body structure information in the image patch is less. For example, if the control point is located in an area such as the edge of the skull or the edge of the human body contour, the ROI contains more empty areas. Therefore, the size of the ROI should be increased to include more human body structure information.

[0107] 2. Adaptively determine the search step size.

[0108] First, set a fixed search step size, for example, Step is 20 pixels. Then, calculate the similarity between the image patches containing the control points in the Mask image and the image patches containing the control points in the Contrast image, such as Normalized Mutual Information (NMI).

[0109] The search step size Step is adaptively adjusted based on NMI. The adjusted Step is as follows:

[0110]

[0111] Here, β is a fixed value, a control parameter determined based on experience, used to adjust the magnitude of NMI's adjustment of the search step size Step.

[0112] If the similarity is high, it means that the Mask image patch where the control point is located matches the Contrast image patch well. In this case, the search step size needs to be reduced to reduce the amount of computation. If the similarity is low, it means that the Mask image patch where the control point is located matches the Contrast image patch poorly. In this case, the search step size needs to be increased to ensure that the correct image patch is found and to avoid the limitation of a fixed search step size.

[0113] Furthermore, in the process of adaptively determining the search step size, the methods for calculating image similarity include, but are not limited to: Mutual Information (MI), Normalized Mutual Information (NMI), Correlation Coefficient (CC), Normalized Correlation Coefficient (NCC), Energy of the histogram of differences (EHD), Sum of Squared Differences (SSD), Structural Similarity (SSIM), and other similarity measurement algorithms.

[0114] As can be seen from the above description, the matching block search method in image registration proposed in this invention can adaptively determine the ROI size and search step size based on image information. Compared with a fixed ROI size and search step size, it can better utilize computing resources and more accurately search for the correct matching block and displacement in the whole image, thereby producing a better registered image and significantly improving the quality of the subtraction image.

[0115] This embodiment also provides a matching block search device for image registration, which is used to implement the above embodiments and preferred embodiments, and will not be repeated as described above. The terms "module," "unit," "subunit," etc., used below can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0116] Figure 7 This is a structural block diagram of the matching block search device in image registration in an embodiment of the present invention, as shown below. Figure 7 As shown, the device includes:

[0117] The information acquisition module 510 is used to acquire the reference image, the image to be matched, control point information, and preset search features, including preset matching size.

[0118] The first determining module 520 is used to determine the initial matching block based on the control point information and the preset matching size;

[0119] The second determining module 530 is used to determine the image information of the initial matching block and determine the target search features based on the image information.

[0120] The matching search module 540 is used to determine the reference matching block in the reference image based on the control point information and the target search features, and to search for the target matching block corresponding to the reference matching block in the image to be matched.

[0121] Through the aforementioned modules, an initial matching block is first determined in the reference image and / or the image to be matched using control point information and preset search features. Then, based on the image information of the initial matching block, the preset search features are adjusted to generate a new matching size and / or search range, thereby obtaining the target search features. Finally, the matching block is searched based on the target search features and the pre-given preset search features. This invention obtains the target search features by adjusting the preset search features using the image information of the initial matching block. Compared to the user-given preset search features, the target search features are more suitable for the current search process, thereby improving the efficiency and accuracy of the matching block search. When this method is applied to X-ray contrast image registration, it can solve the problem that existing X-ray contrast image registration algorithms cannot adaptively determine the size or search range of the matching block.

[0122] It should be noted that the above modules can be functional modules or program modules, and can be implemented through software or hardware. For modules implemented through hardware, the above modules can reside in the same processor; or the above modules can be located in different processors in any combination.

[0123] This embodiment also provides an electronic device including a memory and a processor, the memory storing a computer program and the processor being configured to run the computer program to perform the steps in any of the above method embodiments.

[0124] Optionally, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor and the input / output device is connected to the processor.

[0125] Optionally, in this embodiment, the processor can be configured to perform the following steps via a computer program:

[0126] S1, acquire the reference image, the image to be matched, control point information, and preset search features, including preset matching size.

[0127] S2, determine the initial matching block based on the control point information and the preset matching size.

[0128] S3, determine the image information of the initial matching block, and determine the target search features based on the image information.

[0129] S4. Based on the control point information and target search features, determine the reference matching block in the reference image, and search for the target matching block corresponding to the reference matching block in the image to be matched.

[0130] It should be noted that the specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementations, and will not be repeated in this embodiment.

[0131] Furthermore, in conjunction with the matching block search method in image registration provided in the above embodiments, this embodiment can also provide a storage medium for implementation. This storage medium stores a computer program; when executed by a processor, this computer program implements any of the matching block search methods in image registration described in the above embodiments.

[0132] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0133] It should be understood that the specific embodiments described herein are merely illustrative of the application and not intended to limit it. All other embodiments derived by those skilled in the art based on the embodiments provided in this application without inventive effort are within the scope of protection of this application.

[0134] Obviously, the accompanying drawings are merely some examples or embodiments of this application. Those skilled in the art can apply this application to other similar situations based on these drawings without any creative effort. Furthermore, it is understood that although the work done in this development process may be complex and lengthy, for those skilled in the art, certain design, manufacturing, or production modifications made based on the technical content disclosed in this application are merely conventional technical means and should not be considered as insufficient disclosure of this application.

[0135] The term "embodiment" in this application refers to a specific feature, structure, or characteristic described in connection with an embodiment that may be included in at least one embodiment of this application. The appearance of this phrase in various places in the specification does not necessarily imply the same embodiment, nor does it imply that it is mutually exclusive with or independent of other embodiments. It will be clearly or implicitly understood by those skilled in the art that the embodiments described in this application may be combined with other embodiments without conflict.

[0136] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of patent protection. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the appended claims.

Claims

1. A matching block search method in image registration, characterized in that, The method includes: The system acquires a baseline image, a matching image, control point information, and preset search features, including a preset matching size; the baseline image and the matching image are medical images. The initial matching block is determined based on the control point information and the preset matching size; Determine the image information of the initial matching block, and determine the target search features based on the image information; the target search features include the target matching size and the target search range. Based on the control point information and target search features, a baseline matching block is determined in the baseline image, and a target matching block corresponding to the baseline matching block is searched in the image to be matched. The step of determining the initial matching block based on the control point information and the preset matching size includes: Based on the control point information and the preset matching size, a first initial matching block is determined in the reference image, and a second initial matching block is determined in the image to be matched; The step of determining the image information of the initial matching block and determining the target search features based on the image information includes: Determine the similarity between the first initial matching block and the second initial matching block; The target search range is determined based on the similarity. The step of determining the image information of the initial matching block and determining the target search features based on the image information further includes: Determine the information entropy of the initial matching block; determine the target matching size based on the information entropy.

2. The matching block search method in image registration according to claim 1, characterized in that, Determining the target matching size based on the information entropy includes: Based on the preset adjustment relationship between the information entropy and the preset matching size, the preset matching size is adjusted by the information entropy to obtain the target matching size.

3. The matching block search method in image registration according to claim 1, characterized in that, Determining the target matching size based on the information entropy includes: The target matching size is determined by the information entropy based on a preset mapping relationship between the information entropy and the target matching size.

4. The matching block search method in image registration according to claim 1, characterized in that, The preset search features also include a preset search range; Determining the target search range based on the similarity includes: Based on the preset adjustment relationship of the similarity to the preset search range, the preset search range is adjusted by the similarity to obtain the target search range.

5. The matching block search method in image registration according to claim 1, characterized in that, Determining the target search range based on the similarity includes: The target search range is determined based on the preset mapping relationship between the similarity and the target search range.

6. The matching block search method in image registration according to claim 1, characterized in that, Determining the similarity between the first initial matching block and the second initial matching block includes: The similarity is determined based on relevant information between the first initial matching block and the second initial matching block; wherein, the relevant information includes at least one of the following: Mutual information, normalized mutual information, correlation coefficient, normalized correlation coefficient, maximum subtraction histogram energy, sum of squared differences, structural similarity.

7. A matching block search device for image registration, characterized in that, The device includes: The information acquisition module is used to acquire a reference image, a matching image, control point information, and preset search features, wherein the preset search features include a preset matching size; the reference image and the matching image are medical images. The first determining module is used to determine an initial matching block based on the control point information and the preset matching size, including determining a first initial matching block in the reference image based on the control point information and the preset matching size, and determining a second initial matching block in the image to be matched. The second determining module is used to determine the image information of the initial matching block and determine the target search features based on the image information; the target search features include the target matching size and the target search range; wherein, determining the image information of the initial matching block and determining the target search features based on the image information includes: Determine the similarity between the first initial matching block and the second initial matching block; determine the target search range based on the similarity; Determine the information entropy of the initial matching block; determine the target matching size based on the information entropy; The matching search module is used to determine a reference matching block in the reference image based on the control point information and target search features, and to search for a target matching block in the image to be matched that corresponds to the reference matching block.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the matching block search method in image registration as described in any one of claims 1 to 6.

Images