Verification method and device, electronic equipment and storage medium

Abstract

The invention discloses a verification method and device, electronic equipment and a storage medium. The method comprises the following steps: on the basis of obtaining feature region images of a plurality of continuous slice images, determining a target adjustment amount of a region of interest of a current number compared with a region of interest of a slice image with an initial number based on similarity and difference between the feature region images of the slice image with the current number and the feature region image of the slice image with the initial number, the current serial number is a serial number between the starting serial number and the ending serial number; adjusting the region of interest of the currently numbered slice image based on the target adjustment amount to obtain a target scanning region, and scanning the target scanning region after adjustment to obtain a region-of-interest image of the currently numbered slice image; and comparing the region-of-interest images of the slice images of the current number and the initial number, determining a target verification result of the region-of-interest image of the slice image of the current number, and visually displaying the target verification result. According to the method, the credit collection result of slice position registration can be visually displayed, so that whether errors exist in position registration or not can be found in time.

Description

Technical Field

[0001] Embodiments of the present invention relate to the field of image processing technology, and in particular to a verification method, device, electronic device, and storage medium. Background Art

[0002] Slice 3D reconstruction technology can be understood as cutting a sample into serial slices, scanning these slices in batches to obtain a high-resolution 2D image sequence, and then using computer algorithms to overlay these 2D images and reconstruct them into a 3D model to reveal the spatial structural characteristics of the sample. This technology has important applications in fields such as biomedical research.

[0003] To achieve high-quality 3D reconstruction results, the scanned images of different slices must maintain the same spatial positional relationship as much as possible. However, in existing technologies, batch scanning of slices requires operators to manually adjust the position of each slice to achieve image registration. This method can introduce human errors into the scanned images, and these errors cannot be detected in a timely manner. Summary of the Invention

[0004] The present invention provides a verification method, device, electronic device and storage medium, which can visually display the acceptance judgment results of slice position alignment, so as to promptly detect whether there is an error in the position alignment.

[0005] In a first aspect, an embodiment of the present invention provides a verification method, including:

[0006] Determining a target adjustment amount of a region of interest of a slice image with a current number compared to the slice image with a starting number based on similarities and differences between the characteristic region images of the slice image with a current number and the slice image with a starting number, wherein the current number is a number between the starting number and the ending number;

[0007] adjusting the region of interest of the current numbered slice image based on the target adjustment amount to obtain a target scanning region, and obtaining an image of the region of interest of the current numbered slice image by scanning the target scanning region after adjustment;

[0008] Compare the region of interest images of the current numbered slice image with the starting numbered slice image, determine a target verification result of the region of interest image of the current numbered slice image, and visually display the target verification result.

[0009] In a second aspect, an embodiment of the present invention provides a verification device, including:

[0010] a determination module configured to determine, based on the obtained characteristic region images of a plurality of consecutive slice images, a target adjustment amount of the region of interest of the slice image with the current number compared to the slice image with the starting number based on similarities and differences between the characteristic region images of the slice image with the current number and the slice image with the starting number, wherein the current number is a number between the starting number and the ending number;

[0011] An adjustment module, configured to adjust the region of interest of the current numbered slice image based on the target adjustment amount to obtain a target scanning region, and obtain an image of the region of interest of the current numbered slice image by scanning the target scanning region after the adjustment;

[0012] The verification module is used to compare the region of interest images of the current numbered slice image with the starting numbered slice image, determine a target verification result of the region of interest image of the current numbered slice image, and visually display the target verification result.

[0013] In a third aspect, an embodiment of the present invention provides an electronic device, including:

[0014] at least one processor; and

[0015] a memory communicatively connected to the at least one processor; wherein,

[0016] The memory stores a computer program executable by the at least one processor. The computer program is executed by the at least one processor to enable the at least one processor to perform the method according to the first aspect.

[0017] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in the first aspect.

[0018] The technical solution of the embodiment of the present invention is to obtain characteristic region images of multiple consecutive slice images, determine the target adjustment amount of the region of interest of the current numbered slice image compared to the starting numbered slice image based on the similarity and difference between the characteristic region images of the current numbered slice image and the starting numbered slice image, where the current number is a number between the starting number and the ending number; adjust the region of interest of the current numbered slice image based on the target adjustment amount to obtain a target scanning region, and after adjustment, obtain the region of interest image of the current numbered slice image through the target scanning region; compare the region of interest images of the current numbered slice image with the starting numbered slice image, determine the target verification result of the region of interest image of the current numbered slice image, and visualize the target verification result. This solution can automatically adjust the position of each slice layer to achieve position registration of different slice layers. After the position of each slice layer is adjusted, a region of interest image is scanned to obtain the region of interest image, and the scanned region of interest image is judged and the judgment result is visually displayed, so that the operator can perform subsequent processing based on the judgment result and can promptly detect whether there is an error in the automatic position registration.

[0019] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0021] Figure 1 This is a flowchart of a verification method provided according to the first embodiment of the present invention;

[0022] Figure 2 This is a flowchart of a verification method provided according to the second embodiment of the present invention;

[0023] Figure 3 is a schematic diagram of a comparison of slice images provided according to the third embodiment of the present invention;

[0024] Figure 4 is a schematic diagram of a region selection operation in a slice image provided according to a third embodiment of the present invention;

[0025] Figure 5 is a schematic diagram of a feature region image provided according to the third embodiment of the present invention;

[0026] Figure 6This is a schematic structural diagram of a verification device provided according to a fourth embodiment of the present invention;

[0027] Figure 7 It is a schematic structural diagram of an electronic device implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0028] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0029] It should be noted that the terms "first," "second," and the like in the present invention are used to distinguish similar objects and are not necessarily used to describe a specific order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having," as well as any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to these processes, methods, products, or apparatus.

[0030] Example 1

[0031] Figure 1 This is a flowchart of a verification method according to Embodiment 1 of the present invention. This embodiment is applicable to verifying the results of slice position registration. The method can be performed by a verification device, which can be implemented in software and / or hardware and integrated into an electronic device. Furthermore, the electronic device includes, but is not limited to, computers, laptops, and the like.

[0032] like Figure 1 As shown, the method includes:

[0033] S110. Based on the characteristic area images obtained from a plurality of consecutive slice images, determine the target adjustment amount of the region of interest of the slice image with the current number compared with the slice image with the starting number based on the similarity and difference between the characteristic area images of the slice image with the current number and the slice image with the starting number, where the current number is a number between the starting number and the ending number.

[0034] In the embodiment of the present invention, the slice may be a biological slice, or a slice in other slice scanning scenarios, which is not limited here.

[0035] In practical applications, slices from different layers of the same sample are typically placed sequentially on the sample stage of a scanning electron microscope. A panoramic image of the sample stage can be obtained (usually optical imaging), and the slices to be scanned are selected based on the panoramic image, such as 500 slices to be scanned. During slice selection, the large color contrast between the slices and the sample stage background color allows the edges of the selected slices to be automatically identified based on this contrast (see patent CN119599867A for details). This allows each slice to be numbered (for subsequent scanning operations in the order of the numbers) and positioned overall (for the sample stage to automatically move the slice corresponding to a certain number under the objective lens).

[0036] Typically, the slices are numbered from top to bottom and from left to right. Therefore, the slices can be arranged in several columns on the sample stage from the upper layer to the lower layer and from top to bottom and from left to right, and the slices can be numbered in the order of arrangement. Then, the scanning operation can be performed in the order of the slice numbers to obtain multiple continuous slice images.

[0037] In a series of slice images, the first slice image obtained corresponds to the starting number, and the last slice image obtained corresponds to the ending number. The current number is the number of the slice image for which the slice adjustment amount is currently being determined. The current number is any number between the starting number and the ending number.

[0038] Each slice image may include two types of regions, one is the feature region and the other is the region of interest. The positions of these two types of regions in each slice image may be predetermined. For details on how to determine the positions of these two types of regions in each slice image, please refer to the subsequent embodiments.

[0039] A feature region can be an area in a slice image that has a distinct characteristic or feature. Scanning a feature region can produce a feature region image. A region of interest (ROI) can be an area of interest in a slice image, such as an area to be reconstructed in a scanning operation. Scanning a ROI can produce a ROI image. Each slice image can contain one feature region, in which case the features of feature regions in different slice images are consistent. Each slice image can also contain multiple feature regions, in which case the number of feature regions in different slice images is the same and the features correspond to each other.

[0040] This step is performed after determining the feature regions and regions of interest for each slice image, obtaining the feature region images for each slice image, and not obtaining the region of interest image for the currently numbered slice image. Specifically, the similarity between the feature region images for the currently numbered slice image and the starting numbered slice image for the same feature region is determined. If the determined similarity exceeds a set threshold, the feature region image for the currently numbered slice image is determined to have passed verification for the feature region. Based on the difference between the verified feature region image and the feature region image corresponding to the starting numbered slice image, a target adjustment amount for the region of interest of the currently numbered slice image compared to the starting numbered slice image is determined. The difference in the feature region images refers to the offset and rotation of the feature region, and the target adjustment amount can be the offset and rotation of the region of interest of the currently numbered slice image compared to the starting numbered slice image.

[0041] S120 , adjusting the region of interest of the current numbered slice image based on the target adjustment amount to obtain a target scanning region, and obtaining a region of interest image of the current numbered slice image by scanning the target scanning region after adjustment.

[0042] In this step, after determining a target adjustment amount for the ROI of the current slice image relative to the starting slice image, the ROI of the current slice image is adjusted by the target adjustment amount, and the adjusted ROI is determined as the target scanning area. The target scanning area is then scanned to obtain a ROI image corresponding to the current slice image for 3D reconstruction. Adjusting the ROI of the current slice image can be understood as adjusting the rotation and offset of the current slice.

[0043] S130 , comparing the region of interest images of the current numbered slice image with the image of the starting numbered slice image, determining a target verification result of the region of interest image of the current numbered slice image, and visually displaying the target verification result.

[0044] In this step, the region of interest images of the current numbered slice image and the starting numbered slice image can be compared to determine whether the similarity between the two exceeds the set threshold. Based on the judgment result, it is determined whether the region of interest image of the current numbered slice image has passed the verification, and the result of whether the verification is passed is used as the target verification result, and the target verification result is visually displayed.

[0045] There is no limitation on the method of visually displaying the target verification results. For example, the target verification results can be displayed in the form of text on the interactive interface, such as displaying "The image verification of the region of interest of the current numbered slice image is qualified", or displaying "The image verification of the region of interest of the current numbered slice image is unqualified".

[0046] The technical solution of the embodiment of the present invention is to obtain characteristic region images of multiple consecutive slice images, determine the target adjustment amount of the region of interest of the current numbered slice image compared to the starting numbered slice image based on the similarity and difference between the characteristic region images of the current numbered slice image and the starting numbered slice image, where the current number is a number between the starting number and the ending number; adjust the region of interest of the current numbered slice image based on the target adjustment amount to obtain a target scanning region, and after adjustment, obtain the region of interest image of the current numbered slice image through the target scanning region; compare the region of interest images of the current numbered slice image with the starting numbered slice image, determine the target verification result of the region of interest image of the current numbered slice image, and visualize the target verification result. This solution can automatically adjust the position of each slice layer to achieve position registration of different slice layers. After the position of each slice layer is adjusted, a region of interest image is scanned to obtain the region of interest image, and the scanned region of interest image is judged and the judgment result is visually displayed, so that the operator can perform subsequent processing based on the judgment result and can promptly detect whether there is an error in the automatic position registration.

[0047] Example 2

[0048] Figure 2 This is a flowchart of a verification method provided according to Example 2 of the present invention. This embodiment is based on the above-mentioned Example 1, and further refines the target adjustment amount of the region of interest of the current numbered slice image compared with the starting numbered slice image based on the similarity and difference of the feature region images of the current numbered slice image and the starting numbered slice image, wherein each slice image includes multiple feature regions, and the number of feature regions of different slice images is the same and the feature correspondence is consistent; and further refines the target verification result of the region of interest image of the current numbered slice image by comparing the region of interest images of the current numbered slice image and the starting numbered slice image.

[0049] like Figure 2 As shown, the method includes:

[0050] S111 , based on obtaining characteristic region images of a plurality of consecutive slice images, taking each characteristic region as a characteristic region to be verified, and determining a second similarity between the characteristic region images of the current numbered slice image and the starting numbered slice image for the characteristic region to be verified.

[0051] In this step, any method capable of determining the similarity between two images may be used to determine the second similarity between the current numbered slice image and the starting numbered slice image for the feature region to be verified, which is not limited here.

[0052] S112: If the second similarity exceeds a second threshold, determine that the current numbered slice image has passed the feature region image verification of the feature region to be verified, and determine the feature region to be verified as a target feature region; otherwise, the verification fails.

[0053] In this step, if the second similarity exceeds the second threshold, the current numbered slice image is determined to have passed verification with the feature region image of the feature region to be verified, and the feature region to be verified is determined as the target feature region, so that the qualified feature region image is used for subsequent determination of the target adjustment amount. Otherwise, the current numbered slice image fails verification with the feature region image of the feature region to be verified, and the unqualified feature region image is not used for subsequent determination of the target adjustment amount. The second threshold is not limited.

[0054] Optionally, the result of whether the current numbered slice image is qualified for the feature region image of the feature region to be verified may be displayed visually, and the specific manner of the visual display is not limited.

[0055] S113 , determining a target adjustment amount of the region of interest of the current numbered slice image compared with the starting numbered slice image based on a difference between the feature region images of the target feature region and the starting numbered slice image.

[0056] That is, based on the difference between the characteristic area images of the target characteristic area between the current number and the starting number slice image, the offset and rotation amount of the target characteristic area of the current number compared with the starting number slice image are determined, and then the target adjustment amount of the region of interest of the current number compared with the starting number slice image is determined according to the determined offset and rotation amount of the target characteristic area.

[0057] In one embodiment, determining a target adjustment amount of the region of interest of the current numbered slice image compared with the starting numbered slice image based on a difference between the feature region images of the target feature region and the starting numbered slice image includes:

[0058] Determining a first adjustment amount of the current numbered slice image relative to the starting numbered slice image for the target feature region based on a difference between the feature region images of the current numbered slice image and the starting numbered slice image for the target feature region;

[0059] predicting a second adjustment amount of the current number slice image relative to the starting number slice image for the target feature region based on a difference between the starting number slice image and the ending number slice image for the target feature region;

[0060] A target adjustment amount of the region of interest of the slice image with the current number compared to the slice image with the starting number is determined according to the first adjustment amount and the second adjustment amount.

[0061] Specifically, the first adjustment amount is determined by:

[0062] If there is only one target feature area, the difference between the feature area images of the target feature area between the current number and the starting number slice image is compared to determine the offset and rotation of the target feature area with respect to the current number slice image compared to the starting number slice image, and the offset and rotation of the target feature area are determined as the first adjustment amount of the target feature area with respect to the current number slice image compared to the starting number slice image.

[0063] If there are multiple target feature areas, the difference between the feature area images of each target feature area between the current number and the starting number slice image is compared to determine the offset and rotation amount of each target feature area compared with the starting number slice image, and the offset and rotation amount corresponding to each of the multiple target feature areas are determined as the first adjustment amount for the target feature area compared with the starting number slice image.

[0064] Among them, the offset and rotation of the corresponding target feature area are determined by comparing the two feature area images. Reference can be made to patent CN114663686B or other disclosed registration technologies based on feature point detection. The registration technology based on feature point detection achieves image alignment by extracting key feature points (such as corners, edges, etc.) in the image and calculating the geometric transformation matrix between the two images. The implementation effect of the registration technology based on feature point detection is highly correlated with the image contrast. Therefore, the scanning contrast of the electron microscope can be adjusted, and the scanning mode with the largest contrast can be selected to obtain the feature area image.

[0065] The second adjustment amount is determined as follows:

[0066] The difference between the characteristic region images of the target characteristic region and the starting number slice images is compared to determine the regional adjustment amount for the target characteristic region. The method for determining the regional adjustment amount is basically the same as the method for determining the first adjustment amount described above and will not be repeated here.

[0067] Based on the determined regional adjustment amount, a second adjustment amount for the target feature region is predicted for the slice image with the current number compared to the slice image with the starting number. The second adjustment amount can be determined by ((M-1) / (N-1))×regional adjustment amount, where M is the current number and N is the total number of slice images. If there is only one target feature region, the regional adjustment amount involved in the calculation is the offset or rotation amount of the target feature region, and the second adjustment amount includes the offset and rotation amount for the target feature region with the current number compared to the slice image with the starting number. If there are multiple target feature regions, the regional adjustment amount involved in the calculation is the offset or rotation amount of any of the multiple target feature regions, and the second adjustment amount includes the offset and rotation amount corresponding to each of the multiple target feature regions with the current number compared to the slice image with the starting number.

[0068] For example, N is 500, M is 2, and the ending number is offset by 0.1 μm and rotated 1° clockwise for a certain target feature area compared to the starting number slice image. It can be predicted that the current number is offset by 0.1 μm / 499 and rotated 1° / 499 clockwise for the target feature area compared to the starting number slice image.

[0069] The target adjustment amount is determined as follows:

[0070] An offset difference between two offset amounts in the first adjustment amount and the second adjustment amount for the same target feature area is determined, and a rotation difference between two rotation amounts in the first adjustment amount and the second adjustment amount for the same target feature area is determined.

[0071] If there is one target feature area, the offset difference of the target feature area is determined as the offset of the region of interest of the current numbered slice image compared with the starting numbered slice image, and the rotation difference of the target feature area is determined as the rotation of the region of interest of the current numbered slice image compared with the starting numbered slice image.

[0072] If there are multiple target feature areas, the offset of the region of interest of the current numbered slice image compared with the starting numbered slice image is determined based on the difference in the offset amounts corresponding to the multiple target feature areas, and the rotation amount of the region of interest of the current numbered slice image compared with the starting numbered slice image is determined based on the difference in the rotation amounts corresponding to the multiple target feature areas.

[0073] The offset and rotation amount of the region of interest of the slice image with the current number compared to the slice image with the starting number are determined as the target adjustment amount of the region of interest of the slice image with the current number compared to the slice image with the starting number.

[0074] In one embodiment, there are multiple target feature regions, and determining the target adjustment amount of the region of interest of the slice image with the current number compared to the slice image with the starting number according to the first adjustment amount and the second adjustment amount includes:

[0075] Taking each target feature region as a feature region to be calculated, determining a first offset and a first rotation amount for the feature region to be calculated in the first adjustment amount, and a second offset and a second rotation amount for the feature region to be calculated in the second adjustment amount;

[0076] Determine an offset difference corresponding to the feature area to be calculated based on the first offset and the second offset, and determine a rotation difference corresponding to the feature area to be calculated based on the first rotation and the second rotation;

[0077] According to the offset difference and the rotation difference corresponding to each target feature region, a target adjustment amount of the region of interest of the slice image with the current number compared with the slice image with the starting number is determined.

[0078] That is, each target feature area is used as the feature area to be calculated, and the offset and rotation amount corresponding to the feature area to be calculated in the first adjustment amount are determined as the first offset and the first rotation amount, respectively; the offset and rotation amount corresponding to the feature area to be calculated in the second adjustment amount are determined as the second offset and the second rotation amount, respectively.

[0079] The difference between the first offset and the second offset is determined as the offset difference corresponding to the feature area to be calculated, and the difference between the first rotation amount and the second rotation amount is determined as the rotation difference corresponding to the feature area to be calculated.

[0080] According to the difference in offsets corresponding to each target feature area, the offset of the region of interest of the current numbered slice image compared with the starting numbered slice image is determined; according to the difference in rotations corresponding to each target feature area, the rotation of the region of interest of the current numbered slice image compared with the starting numbered slice image is determined; the offset and rotation of the region of interest of the current numbered slice image compared with the starting numbered slice image are determined as the target adjustment amount.

[0081] In one embodiment, determining a target adjustment amount of the region of interest of the slice image with the current number compared to the slice image with the starting number according to the offset difference and the rotation difference corresponding to each target feature region includes:

[0082] The target adjustment amount is determined by combining the average of the offset differences corresponding to each target feature area and the average of the rotation differences corresponding to each target feature area; or

[0083] The target adjustment amount is determined by combining a weighted average of the offset differences corresponding to each target feature region and a weighted average of the rotation differences corresponding to each target feature region.

[0084] That is, there are two calculation methods for the target adjustment amount, the first one is a calculation method with a lower calculation amount, and the second one is a calculation method with a higher calculation amount.

[0085] A calculation method with lower computational complexity is to take the average of the offset differences corresponding to each target feature area and determine it as the offset of the region of interest between the current number and the starting number slice image; to take the average of the rotation differences corresponding to each target feature area and determine it as the rotation of the region of interest between the current number and the starting number slice image; the offset and rotation of the region of interest are the target adjustment amounts.

[0086] A calculation method with a higher amount of calculation is to take the weighted average of the offset differences corresponding to each target feature area, and determine it as the offset of the region of interest between the current number and the starting number slice image; take the weighted average of the rotation differences corresponding to each target feature area, and determine it as the rotation of the region of interest between the current number and the starting number slice image; the offset and rotation of the region of interest are the target adjustment amount.

[0087] In one embodiment, the weight of the weighted average of the offset differences corresponding to each target feature area is determined based on the distance between the center of the region of interest in the current numbered slice image and the center of each target feature area; the weight of the weighted average of the rotation differences corresponding to each target feature area is determined based on the rotation amount for each target feature area in the second adjustment amount.

[0088] In practical applications, when taking the weighted average of the offset differences corresponding to each target feature area, the distance between the center of the region of interest in the current numbered slice image and the center of each target feature area can be determined. The offset weight corresponding to the target feature area that is far away can be set to be small, and the offset weight corresponding to the target feature area that is close can be set to be large. This is not limited here.

[0089] When taking the weighted average of the differences in the rotation amounts corresponding to each target feature area, the rotation amount for each target feature area in the second adjustment amount can be determined. The weight of the rotation amount corresponding to the target feature area with a large rotation amount can be set to be small, and the weight of the rotation amount corresponding to the target feature area with a small rotation amount can be set to be large. This is not limited here.

[0090] S120 , adjusting the region of interest of the current numbered slice image based on the target adjustment amount to obtain a target scanning region, and obtaining a region of interest image of the current numbered slice image by scanning the target scanning region after adjustment.

[0091] S131 , determining a first similarity between the region of interest image of the current numbered slice image and the slice image of the starting numbered slice image.

[0092] In this step, any method capable of determining the similarity between two images may be used to determine the first similarity between the region of interest images of the current numbered slice image and the starting numbered slice image, which is not limited here.

[0093] S132: If the first similarity exceeds a first threshold, determine that the target verification result of the region of interest image of the current numbered slice image indicates that the verification is qualified; otherwise, the target verification result indicates that the verification is unqualified.

[0094] In this step, if the first similarity exceeds a first threshold, the image of the region of interest of the currently numbered slice image is determined to have passed the verification, and the qualified verification result is used as the target verification result; otherwise, the unqualified verification result is used as the target verification result. The first threshold is not limited.

[0095] S133: Visually display the target verification result.

[0096] The technical solution of the embodiment of the present invention verifies the feature area image of the current numbered slice image, and uses the feature area image that passes the verification to determine the target adjustment amount of the region of interest of the current numbered slice image compared with the starting numbered slice image, so that the determined target adjustment amount is more accurate; the region of interest of the current numbered slice image is adjusted by the determined target adjustment amount, and the region of interest image of the current numbered slice image is scanned after adjustment, and the region of interest image of the current numbered slice image is verified, and the verification result is visually displayed, so that the operator can promptly discover whether there is an error in the automatic position alignment.

[0097] The following is a detailed description of how to determine the location of the feature region and the region of interest in each slice image:

[0098] In one embodiment, the method for determining the characteristic region and the region of interest of each slice image includes:

[0099] Based on the acquisition of a plurality of consecutive slice images, in response to a region selection operation, determining a region of interest and a feature region in a slice image with a starting number, and determining a region of interest and a feature region in a slice image with an ending number;

[0100] The region of interest and the characteristic region in the current numbered slice image are determined based on the region of interest and the characteristic region in the starting numbered slice image by a position mapping method.

[0101] The region selection operation can be an operation for selecting a region of interest or a feature region in the slice images with the start and end numbers. For example, it can be an operation for selecting a region in the slice image; another example is an operation for clicking a point in the slice image to select a rectangular region of a set size centered at that point.

[0102] In response to the region selection operation, the region of interest and the feature region can be selected in the starting number slice image; the selection of the region of interest in the starting number slice image can be based on actual application needs and is not limited here; the selection of the feature region in the starting number slice image can be performed by comparing the starting number and ending number slice images to select the feature region with obvious features common to the two slice images.

[0103] In one embodiment, the selection of the characteristic region in the starting numbered slice image at least satisfies the following conditions:

[0104] The contrast between the selected feature area and the background in the starting number slice image exceeds the contrast threshold, the edge recognition exceeds the recognition threshold, and the feature area is unique;

[0105] The selected feature region is a feature region common to the start number slice image and the end number slice image, and the structural similarity between the start number slice image and the end number slice image exceeds a similarity threshold, and the adjustment amount is lower than an adjustment amount threshold;

[0106] When multiple characteristic regions are selected, the multiple characteristic regions are in a non-clustered distribution state in the starting numbered slice image.

[0107] That is, in the starting number slice image, the selected feature area should have a pattern structure with obvious characteristics. The requirements for obvious characteristics are that the contrast between the pattern structure and the background is large (that is, the contrast between the selected feature area and the background exceeds the contrast threshold), the edge is easy to identify (that is, the edge recognition exceeds the recognition threshold), and it is significantly different from other structures nearby or in the entire slice area (that is, the pattern structure is unique and there is no repetition); the selected feature area is the feature area shared by the starting number and the ending number slice images, that is, the two slice images have basically the same structure, such as a special structure with small structural changes in different layers (that is, the structural similarity exceeds the similarity threshold); the position offset or rotation amplitude of the selected feature area in the starting number and the ending number slice images is as small as possible (that is, the adjustment amount is lower than the adjustment amount threshold); the selected feature areas are multiple and distributed as far as possible to avoid clustering together (that is, non-clustered distribution).

[0108] When the region of interest and the feature region in the starting number slice image are determined, in response to the region selection operation, a region that is substantially consistent in position and features with the region of interest of the starting number slice image can be selected in the ending number slice image as the region of interest of the ending number slice image; a region that is substantially consistent in position and features with the feature region of the starting number slice image can also be selected as the feature region of the ending number slice image.

[0109] Using a position mapping method, the edges of the starting slice image and the current slice image can be aligned, and the bounding box of the region of interest of the starting slice image can be copied to the current slice image to obtain the region of interest of the current slice image. The determination of the feature region in the current slice image is basically the same as the determination of the region of interest in the current slice image, and will not be further described here.

[0110] In one embodiment, before determining the region of interest and the feature region in the slice image with the termination number in response to the region selection operation, the method further includes:

[0111] The image feature automatic matching algorithm is used to generate the recommended region of interest and feature region in the ending slice image based on the region of interest and feature region in the starting slice image.

[0112] Because each slice is numbered and positioned as a whole, the position of the center point of the ROI in the starting slice image relative to the slice as a whole can be determined. Using an automatic image feature matching algorithm, based on the position of the center point of the ROI in the starting slice image relative to the slice as a whole and combined with a position mapping relationship, the position of the center point of the ROI in the ending slice image can be determined. In the ending slice image, an outward search is performed with the center point of the ROI in the ending slice image as the center to determine an area in the ending slice image that is similar to the ROI in the starting slice image, which is then used as the recommended ROI in the ending slice image. The determination of the characteristic area in the recommended ending slice image is essentially the same as the method for determining the ROI in the recommended ending slice image, and will not be repeated here.

[0113] When the recommended ROI and feature regions in the end-numbered slice image are automatically generated, the recommended ROI and feature regions can be manually confirmed or fine-tuned through region selection operations to determine the ROI and feature regions in the end-numbered slice image.

[0114] By manually confirming or fine-tuning the region selection operation of the operator after generating the recommended region of interest and feature region in the end number slice image, the determination of the region of interest and feature region of the end number slice image can be made more accurate, and it can also be verified whether the feature region selected by the operator in the start number slice image is appropriate.

[0115] If the pattern structure of the feature area in the automatically recommended ending number slice image is difficult to identify or changes too much compared to the pattern structure of the starting number slice image, the feature area selected by the operator in the starting number slice image may be incorrect; if there is a feature area with a repeated structure in the automatically recommended ending number slice image, the feature area selected by the operator in the starting number slice image may be incorrect.

[0116] As for the position offset or excessive rotation amplitude of the pattern structure with obvious characteristics, and the excessive clustering of multiple feature areas, it usually does not affect the determination of the region of interest and feature area in the terminal numbered slice image; however, the position offset or excessive rotation amplitude of the pattern structure with obvious characteristics will increase the amount of calculation when judging the position and rotation adjustment amount of the current numbered slice; if multiple feature areas are too clustered, it is possible that the feature area is rotated or offset, while the region of interest is not rotated or offset, but the rotation or offset of the feature area will misjudge that the region of interest has also been rotated or offset.

[0117] Example 3

[0118] The embodiment of the present invention is an illustrative description of the above embodiments. The embodiment of the present invention provides a method for accurately locating a region of interest in a slice scan, taking a biological slice as an example, and specifically includes the following contents:

[0119] First, multiple consecutive slice images of biological slices need to be acquired.

[0120] In practical applications, biological slices belonging to different layers of the same sample are usually arranged in sequence and placed on the sample stage of a scanning electron microscope. For example, the biological slices are arranged in several columns from the upper layer to the lower layer, from top to bottom, and from left to right on the sample stage. A panoramic image of the sample stage is obtained (usually optical imaging) at a first magnification (the lowest or lower magnification), and the biological slices to be scanned are selected based on the panoramic image, such as selecting 500 biological slices to be scanned.

[0121] When selecting slices, since the slices have a large color contrast with the sample stage background color, the edges of the selected slices can be automatically identified through this contrast (see patent CN119599867A for details), so that each slice can be numbered (so that subsequent scanning operations can be performed in the order of the numbers) and positioned as a whole (so that the sample stage can automatically move the slice corresponding to a certain number under the objective lens).

[0122] Assume that an operator has selected 500 biological slides to be scanned. Slice images of the first through 500th biological slides need to be obtained at the second magnification, i.e., slice images of a plurality of consecutive biological slides are obtained. The scanning order may be in the order of the slide numbers or not, and is not limited here.

[0123] Secondly, the region of interest and feature region of the first slice image are determined.

[0124] In response to the region selection operation, the region of interest and characteristic regions of the first slice image are determined. The region of interest of the first slice image is the region to be 3D reconstructed during the scanning operation. The characteristic regions of the first slice image are determined by comparing the first and 500th slice images.

[0125] Figure 3 FIG. 1 is a schematic diagram of a comparison of slice images provided according to the third embodiment of the present invention. Figure 3 As shown, the 1st and 500th slice images are displayed on the same page, so that the operator can compare the two slice images and select the feature area.

[0126] The selection of the feature area of the first slice image must meet the following requirements:

[0127] There is a distinctive pattern structure in the feature area. The requirements for distinctive features are that the pattern structure has a large contrast with the background, the edges are easy to identify, and it is significantly different from other structures in the vicinity or the entire slice area (that is, the pattern structure is unique and there is no repetition); in the first and 500th images, the distinctive pattern structure has basically the same structure (that is, a special biological structure with little structural variation in different layers of a biological slice), and the higher the structural similarity, the better; in the first and 500th images, the position offset or rotation amplitude of the distinctive pattern structure is as small as possible; when there are multiple feature areas, they are distributed as far as possible to avoid clustering.

[0128] When using a scanning electron microscope to obtain biological slice images at a magnification of one million times, biological structures (i.e., characteristic regions) suitable as feature sites for image registration need to meet the following conditions: high stability, minimal morphological changes, easy identification, and widespread distribution. These structures are usually located inside cells or at the subcellular level and are relatively stable organs or tissues within cells. The following are several common microbial intracellular organs or tissues that are suitable as feature sites:

[0129] 1. Mitochondria

[0130] Mitochondria are the energy plants within cells. They have a double-membrane structure and are typically oval or rod-shaped. In areas where mitochondria are densely packed, such as muscle cells or neurons, their distribution and morphology are relatively stable. Mitochondria are numerous and widely distributed, making them easily captured by scanning electron microscopy. Their shape and position vary minimally across different slices, making them suitable as stable features. This makes them suitable for studying metabolically active cell types.

[0131] 2. Ribosomes

[0132] Ribosomes are small, granular structures within cells responsible for protein synthesis, approximately 20-30 nanometers in diameter. They can float freely in the cytoplasm or be attached to the endoplasmic reticulum. The size and morphology of ribosomes remain largely unchanged across different slices. Their high density creates significant contrast in SEM images. Due to their small size and sensitivity to environmental influences, ribosomes may need to be used in conjunction with other feature points.

[0133] 3. Centrosome

[0134] The centrosome is the microtubule-organizing center that plays a key role in cell division. It consists of two perpendicular centrioles surrounded by a dense protein matrix. The centrosome's unique and stable structure makes it easy to identify in serial sections. Its regular geometry makes it a suitable reference point for precise registration. It is well-suited for studying processes related to cell division.

[0135] 4. Golgi apparatus

[0136] The Golgi apparatus is a complex structure composed of multiple flat vesicles that are primarily involved in the processing and secretion of proteins and lipids. The vesicles of the Golgi apparatus are clearly visible and morphologically stable. Local structural variations are minimal across different slices, providing reliable registration. This makes it suitable for studying cells with active secretory functions.

[0137] 5. Endoplasmic reticulum

[0138] The endoplasmic reticulum (ER) is divided into rough ER and smooth ER. The former has numerous ribosomes attached to its surface, while the latter is smooth and free of particles. The network-like structure of the ER is highly coherent in serial sections. The distribution of ribosomes in the rough ER, in particular, provides additional markers. This makes it suitable for studying cells involved in protein synthesis and transport.

[0139] 6. Lysosomes

[0140] Lysosomes are the digestive "workshops" within cells, enclosed by a single membrane and containing a variety of hydrolytic enzymes. Lysosomes are relatively uniform in size and morphology, and are evenly distributed. Their position varies minimally across slices, making them suitable as stable features. They are well-suited for studying cellular autophagy or degradation processes.

[0141] 7. Microtubules and microfilaments

[0142] Microtubules and microfilaments are essential components of the cytoskeleton, composed of polymers of α- / β-tubulin and actin, respectively, forming the intracellular support and transport network. The fibrous structures of microtubules and microfilaments exhibit a high degree of continuity and regularity in serial sections. Their arrangement serves as a natural grid for precise positioning, making them suitable for studying the mechanisms that maintain cellular morphology and transport.

[0143] 8. Chromatin (structure within the nucleus)

[0144] Chromatin is a complex composed of deoxyribonucleic acid (DNA) and histone proteins, presenting as filamentous or blocky structures distributed within the cell nucleus. The distribution pattern of chromatin maintains a certain degree of consistency across different slices. Its complex morphology provides rich texture information, which can be used for feature matching. It is suitable for studying gene expression regulation within the cell nucleus.

[0145] 9. Cell membrane and its accessory structures

[0146] The cell membrane is composed of a phospholipid bilayer, which may have microvilli, corrugations, or other accessory structures. The cell membrane's edges are clearly defined and easily identified under a scanning electron microscope. The presence of specific accessory structures on the cell membrane, such as synaptic vesicles or microvilli, can further enhance registration accuracy. This technique is suitable for studying intercellular communication or substance exchange processes.

[0147] Figure 4 FIG. 1 is a schematic diagram of a region selection operation in a slice image according to the third embodiment of the present invention. Figure 4 As shown, in the first slice image, the region of interest P1 and three characteristic regions A, B, and C can be determined through a region selection operation.

[0148] Then, the region of interest and feature region of the 500th slice image are determined.

[0149] After the ROI and three characteristic regions are determined in the first slice image, the ROI can be automatically found and determined on the 500th slice image using an automatic image feature matching algorithm. Since each biological slice is numbered and positioned as a whole in the above, the position of the ROI center point in the first slice image relative to the entire slice is known. Based on this position mapping relationship, the position of the ROI center point in the 500th slice image can be determined. By searching outward from the ROI center point for image regions in the 500th slice image that are similar to the ROI in the first slice image, the automatically recommended region of interest for the 500th slice image can be found. The determination of the three characteristic regions in the 500th slice image is the same as the automatic ROI search method for the 500th slice image.

[0150] According to the automatic search results: On the one hand, manual confirmation or fine-tuning is still required. The more accurate the ROI and feature area determined in the 500th slice image, the more accurate the automatic selection of ROI and feature areas in subsequent slice images will be. On the other hand, it can verify whether the feature area selected by the operator in the first slice image is appropriate. If the selection is inappropriate, for example:

[0151] If the pattern structure is difficult to identify or the pattern structure changes too much, the feature area automatically selected in the 500th slice image may be incorrect; if there is a repeated structure, more than 3 feature areas may be determined in the 500th slice image.

[0152] As for the position offset or excessive rotation of the pattern structure with obvious characteristics, or the excessive clustering of multiple characteristic areas, it usually does not affect the automatic search and determination of the ROI and three characteristic areas on the 500th slice image; however, the position offset or excessive rotation of the pattern structure with obvious characteristics will increase the amount of calculation when judging the position and rotation adjustment amount of the intermediate slices (the 2nd to the 499th slices); if multiple characteristic areas are too clustered, it is possible that the characteristic area is rotated or offset, while the ROI is not rotated or offset, but the rotation or offset of the characteristic area will misjudge that the ROI has also been rotated or offset.

[0153] Then, the regions of interest and characteristic regions of the 2nd to 499th slice images are determined.

[0154] Using a position mapping method, the edges of the first slice image and the currently numbered slice image can be aligned, and the bounding box of the region of interest of the first slice image can be copied to the currently numbered slice image to obtain the region of interest of the currently numbered slice image. The determination of the feature region in the currently numbered slice image is essentially the same as the determination of the region of interest in the currently numbered slice image and will not be further described here. The currently numbered image is any one from 2 to 499.

[0155] Then, characteristic region images of the 1st to 500th slice images and region of interest images of the 1st and 500th slice images are acquired.

[0156] The above-described process of aligning the edge of the first slice image with the edge of the currently numbered slice image to determine the region of interest and feature region of the currently numbered slice image already involves positioning adjustment of the ROI, i.e., it is necessary to determine how to correctly place the currently numbered slice under the objective lens so that the objective lens can accurately observe the ROI or its vicinity. This involves the movement direction and distance of the sample stage (moving the currently numbered slice to the position of the first slice - under the objective lens), and also involves the rotation direction and amplitude of the sample stage (because there is an inevitable difference between the placement posture of the currently numbered slice and the first slice). The specific implementation of the above process can be found in patent CN119599867A.

[0157] However, the above positioning adjustment is a coarse adjustment, which is mainly based on the positioning adjustment of the image obtained at the first magnification (the lowest or lower magnification). At the third magnification of the actual scanning operation (the third magnification > the second magnification > the first magnification), the determined ROI area still has a large offset and rotation, resulting in the region of interest obtained by the final scan being unsuitable for three-dimensional reconstruction; the purpose of this application is to provide a semi-automatic slice ROI precise positioning method based on the coarse positioning slice.

[0158] Based on the first magnification, the positions and directions of the ROIs and three characteristic regions in the 1st to 500th slice images relative to the entire slice were obtained, and then the characteristic region images of the 1st to 500th slice images and the region of interest images of the 1st and 500th slice images were obtained by scanning at the second magnification.

[0159] Figure 5 is a schematic diagram of a feature region image provided according to the third embodiment of the present invention, Figure 5 For some exemplary explanations. Figure 5 As shown, each row may correspond to a slice image, and columns A, B, and C may correspond to feature region images of different feature regions of the slice image.

[0160] Then, a target adjustment amount of the region of interest of any one of the second to 499th slice images relative to the region of interest of the first slice image is determined.

[0161] Each characteristic region is used as a characteristic region to be verified, and a second similarity between the characteristic region image of the current number and the first slice image relative to the characteristic region to be verified is determined. If the second similarity exceeds a second threshold, the characteristic region image of the current number slice image is determined to have passed verification, and the characteristic region to be verified is determined to be the target characteristic region; otherwise, verification fails.

[0162] Assuming that the target feature areas are three areas A, B, and C, the registration technology based on feature point detection is used to determine that in the current numbered slice image, relative to the first slice image, the offsets of the three target feature areas A, B, and C are D1, D2, and D3, respectively, and the rotation amounts are R1, R2, and R3, respectively. That is, the first adjustment amount of the target feature area of the current numbered slice image compared to the starting numbered slice image is determined.

[0163] Based on the difference between the target feature areas of the 1st and 500th slice images, it is predicted that in the current numbered slice image, relative to the 1st slice image, the offsets of the three target feature areas A, B, and C are Δ1, Δ2, and Δ3, respectively, and the rotation amounts are θ1, θ2, and θ3, respectively. That is, the second adjustment amount of the target feature area of the current numbered slice image compared to the starting numbered slice image is predicted.

[0164] Determining a target adjustment amount of the ROI of the current numbered slice image compared to the first slice image based on the first adjustment amount includes the following two methods, wherein the target adjustment amount includes an offset x and a rotation y of the ROI.

[0165] (1) Calculation method with lower computational complexity:

[0166]

[0167] (2) Calculation method with higher computational load:

[0168]

[0169] Specifically, the weights a, b, and c can be obtained from the distance between the ROI center and the center of the target feature area A, B, and C, that is, the offset weight of the target feature area far from the ROI center is small, and the offset weight of the target feature area close to the ROI center is large. The specific functional relationship can be designed by yourself; the weights e, f, and g can be obtained from the predicted rotation amount of the target feature area A, B, and C, that is, the weight of the target feature area with a large predicted rotation amount is small, and the weight of the target feature area with a small predicted rotation amount is large. The specific functional relationship can be designed by yourself.

[0170] Then, based on the target adjustment amount of the region of interest of any slice image from the 2nd to the 499th relative to the region of interest of the 1st slice image, the region of interest of any slice image is adjusted to obtain a target scanning area, and after adjustment, the region of interest image of any slice image is obtained by scanning the target scanning area.

[0171] Finally, a first similarity between the region of interest image of any slice image from the 2nd to the 499th slice image and the region of interest image of the 1st slice image is determined; if the first similarity exceeds a first threshold, the target verification result of the region of interest image of any slice image is determined to indicate that the verification is qualified, otherwise the target verification result indicates that the verification is unqualified; and the target verification result is visually displayed.

[0172] The above description is for ease of understanding. In actual applications, the span of the 1st and 500th slice images is relatively large. The ROI and feature areas of the 1st, 100th, 200th, 300th, 400th and 500th images can be obtained in sequence with manual assistance; then the ROI and feature areas of the 2nd to 99th, 101st to 199th, 201st to 299th, 301st to 399th, and 401st to 499th slice images are automatically obtained; and then the target adjustment amount of the ROI of the 2nd to 99th, 101st to 199th, 201st to 299th, 301st to 399th, and 401st to 499th slice images are determined respectively, the region of interest images are obtained based on the target adjustment amount, the region of interest images are verified, and the verification results are visualized, so that the registration effect of each slice ROI image obtained is better.

[0173] This application provides a real-time correction mechanism for slice scanning positions. During the scanning process, the slice position can be dynamically adjusted to reduce errors such as offset and rotation. It has the ability to accurately locate the region of interest and has the following advantages:

[0174] 1. High degree of automation, realizing semi-automatic image registration and scanning operation, improving scanning efficiency and reducing the workload of operators.

[0175] 2. Existing registration methods often focus on the alignment of the entire image, while ignoring the fine-tuning of the local ROI, which may cause distortion of key structures in the three-dimensional reconstruction; the solution provided in this application focuses on the ROI and has a strong ability to accurately locate the ROI.

[0176] 3. The solution of this application has a good balance between computational efficiency and adjustment accuracy. By designing a special verification method, it not only improves the confidence of the feature area, but also the verification calculation amount of ROI is significantly lower than that of the conventional solution.

[0177] 4. The solution of this application is not limited to three-dimensional reconstruction of biological slices, but can also be applied to other slice batch scanning scenarios with similar requirements. However, this application makes full use of the characteristics of biological slice images (such as thin layer structure, tiny features, etc.), so it is more targeted in the application of three-dimensional reconstruction of biological slices.

[0178] Example 4

[0179] Figure 6 FIG. 4 is a schematic diagram of a structure of a verification device according to a fourth embodiment of the present invention. This embodiment is applicable to the case where the result of slice position registration is verified, such as Figure 6 As shown, the specific structure of the device includes:

[0180] a determination module 61 for determining, based on the obtained characteristic region images of a plurality of consecutive slice images, a target adjustment amount of the region of interest of the slice image with the current number compared to the slice image with the starting number based on similarities and differences between the characteristic region images of the slice image with the current number and the slice image with the starting number, wherein the current number is a number between the starting number and the ending number;

[0181] An adjustment module 62 is configured to adjust the region of interest of the current numbered slice image based on the target adjustment amount to obtain a target scanning region, and obtain an image of the region of interest of the current numbered slice image by scanning the target scanning region after the adjustment;

[0182] The verification module 63 is configured to compare the region of interest images of the current numbered slice image with the image of the starting numbered slice image, determine a target verification result of the region of interest image of the current numbered slice image, and visually display the target verification result.

[0183] The verification device provided in this embodiment determines, through a determination module, a target adjustment amount for the region of interest of the current slice image compared to the starting slice image based on the similarity and difference between the feature region images of the current slice image and the starting slice image, where the current number is a number between the starting and ending numbers. The adjustment module adjusts the region of interest of the current slice image based on the target adjustment amount to obtain a target scanning region. After adjustment, the region of interest image of the current slice image is scanned through the target scanning region. The verification module compares the region of interest images of the current slice image with the starting slice image to determine a target verification result for the region of interest image of the current slice image, and visually displays the target verification result. This solution can automatically adjust the position of each slice layer to achieve positional registration of different slice images. After the position of each slice layer is adjusted, a region of interest image is scanned to obtain a region of interest image. The scanned region of interest image is then judged and the judgment result is visually displayed, allowing the operator to perform subsequent processing based on the judgment result and promptly detect whether there is an error in the automatic position registration.

[0184] Furthermore, the verification module 63 is specifically configured to:

[0185] Determining a first similarity between the region of interest image of the current numbered slice image and the starting numbered slice image;

[0186] If the first similarity exceeds a first threshold, it is determined that the target verification result of the region of interest image of the current numbered slice image indicates that the verification is qualified; otherwise, the target verification result indicates that the verification is unqualified.

[0187] Furthermore, each slice image includes multiple feature regions, and the number of feature regions of different slice images is the same and the feature correspondence is consistent. The determination module 61 is specifically configured to:

[0188] Taking each characteristic region as a characteristic region to be verified, determining a second similarity between the current numbered slice image and the starting numbered slice image with respect to the characteristic region image of the characteristic region to be verified;

[0189] If the second similarity exceeds a second threshold, it is determined that the current numbered slice image has passed the feature region image verification of the feature region to be verified, and the feature region to be verified is determined as the target feature region; otherwise, the verification fails;

[0190] Based on the difference between the characteristic region images of the target characteristic region between the current numbered slice image and the starting numbered slice image, a target adjustment amount of the region of interest of the current numbered slice image compared with the starting numbered slice image is determined.

[0191] Furthermore, the determination module 61 is specifically configured to:

[0192] Determining a first adjustment amount of the current numbered slice image relative to the starting numbered slice image for the target feature region based on a difference between the feature region images of the current numbered slice image and the starting numbered slice image for the target feature region;

[0193] predicting a second adjustment amount of the current number slice image relative to the starting number slice image for the target feature region based on a difference between the starting number slice image and the ending number slice image for the target feature region;

[0194] A target adjustment amount of the region of interest of the slice image with the current number compared to the slice image with the starting number is determined according to the first adjustment amount and the second adjustment amount.

[0195] Furthermore, there are multiple target feature areas, and the determination module 61 is specifically configured to:

[0196] Taking each target feature region as a feature region to be calculated, determining a first offset and a first rotation amount for the feature region to be calculated in the first adjustment amount, and a second offset and a second rotation amount for the feature region to be calculated in the second adjustment amount;

[0197] Determine an offset difference corresponding to the feature area to be calculated based on the first offset and the second offset, and determine a rotation difference corresponding to the feature area to be calculated based on the first rotation and the second rotation;

[0198] According to the offset difference and the rotation difference corresponding to each target feature region, a target adjustment amount of the region of interest of the slice image with the current number compared with the slice image with the starting number is determined.

[0199] Furthermore, the determination module 61 is specifically configured to:

[0200] The target adjustment amount is determined by combining the average of the offset differences corresponding to each target feature area and the average of the rotation differences corresponding to each target feature area; or

[0201] The target adjustment amount is determined by combining a weighted average of the offset differences corresponding to each target feature region and a weighted average of the rotation differences corresponding to each target feature region.

[0202] Further, taking a weighted average of the offset differences corresponding to the target feature regions, the weight is determined based on the distance between the center of the region of interest in the current numbered slice image and the center of each target feature region;

[0203] The weight of the weighted average of the rotation amount differences corresponding to each target feature region is determined based on the rotation amount for each target feature region in the second adjustment amount.

[0204] Furthermore, the method for determining the characteristic region and the region of interest of each slice image includes:

[0205] Based on the acquisition of a plurality of consecutive slice images, in response to a region selection operation, determining a region of interest and a feature region in a slice image with a starting number, and determining a region of interest and a feature region in a slice image with an ending number;

[0206] The region of interest and the characteristic region in the current numbered slice image are determined based on the region of interest and the characteristic region in the starting numbered slice image by a position mapping method.

[0207] Furthermore, the selection of the characteristic region in the starting numbered slice image at least meets the following conditions:

[0208] The contrast between the selected feature area and the background in the starting number slice image exceeds the contrast threshold, the edge recognition exceeds the recognition threshold, and the feature area is unique;

[0209] The selected feature region is a feature region common to the start number slice image and the end number slice image, and the structural similarity between the start number slice image and the end number slice image exceeds a similarity threshold, and the adjustment amount is lower than an adjustment amount threshold;

[0210] When multiple characteristic regions are selected, the multiple characteristic regions are in a non-clustered distribution state in the starting numbered slice image.

[0211] Furthermore, before determining the region of interest and the feature region in the slice image with the termination number in response to the region selection operation, the method further includes:

[0212] The image feature automatic matching algorithm is used to generate the recommended region of interest and feature region in the ending slice image based on the region of interest and feature region in the starting slice image.

[0213] The verification device provided in the embodiment of the present invention can execute the verification method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0214] Example 5

[0215] Figure 7 is a schematic diagram of an electronic device that implements an embodiment of the present invention. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, mainframe computers, and other suitable computers. The components shown herein, their connections and relationships, and their functions are provided for example only and are not intended to limit the implementation of the present inventions described and / or claimed herein.

[0216] like Figure 7 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., which is communicatively connected to the at least one processor 11. The memory stores a computer program that can be executed by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. Various programs and data required for the operation of the electronic device 10 can also be stored in the RAM 13. The processor 11, ROM 12, and RAM 13 are connected to each other via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0217] Multiple components in the electronic device 10 are connected to the I / O interface 15, including an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0218] The processor 11 may be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the verification method.

[0219] In some embodiments, the verification method can be implemented as a computer program that is tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the verification method described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the verification method in any other appropriate manner (e.g., by means of firmware).

[0220] Various embodiments of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0221] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the computer program is executed by the processor, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The computer program may be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0222] In the context of the present invention, computer-readable storage media can be tangible media that can contain or store a computer program for use with an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. Computer-readable storage media can include but are not limited to electronic, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, computer-readable storage media can be machine-readable signal media. More specific examples of machine-readable storage media can include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0223] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0224] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0225] A computing system may include clients and servers. The clients and servers are typically remote from each other and typically interact via a communication network. This client-server relationship arises through computer programs running on the respective computers, creating a client-server relationship. The server may be a cloud server, also known as a cloud computing server or cloud host. This server is a hosting product within the cloud computing service ecosystem that addresses the management difficulties and limited scalability of traditional physical hosting and VPS services.

[0226] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in the present invention can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved. This is not limited herein.

[0227] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.

Claims

1. A verification method, characterized in that: include: Determining a target adjustment amount of a region of interest of a slice image with a current number compared to the slice image with a starting number based on similarities and differences between the characteristic region images of the slice image with a current number and the slice image with a starting number, wherein the current number is a number between the starting number and the ending number; adjusting the region of interest of the current numbered slice image based on the target adjustment amount to obtain a target scanning region, and obtaining an image of the region of interest of the current numbered slice image by scanning the target scanning region after adjustment; Compare the region of interest images of the current numbered slice image with the starting numbered slice image, determine a target verification result of the region of interest image of the current numbered slice image, and visually display the target verification result.

2. The method according to claim 1, characterized in that Comparing the region of interest images of the current numbered slice image with the starting numbered slice image to determine a target verification result of the region of interest image of the current numbered slice image includes: Determining a first similarity between the region of interest image of the current numbered slice image and the starting numbered slice image; If the first similarity exceeds a first threshold, it is determined that the target verification result of the region of interest image of the current numbered slice image indicates that the verification is qualified; otherwise, the target verification result indicates that the verification is unqualified.

3. The method according to claim 1, characterized in that Each slice image includes multiple feature regions, and different slice images have the same number of feature regions and consistent feature correspondence. Based on the similarity and difference between the feature region images of the current numbered slice image and the starting numbered slice image, determining the target adjustment amount of the region of interest of the current numbered slice image compared with the starting numbered slice image includes: Taking each characteristic region as a characteristic region to be verified, determining a second similarity between the current numbered slice image and the starting numbered slice image with respect to the characteristic region image of the characteristic region to be verified; If the second similarity exceeds a second threshold, it is determined that the current numbered slice image has passed the feature region image verification of the feature region to be verified, and the feature region to be verified is determined as the target feature region; otherwise, the verification fails; Based on the difference between the characteristic region images of the target characteristic region between the current numbered slice image and the starting numbered slice image, a target adjustment amount of the region of interest of the current numbered slice image compared with the starting numbered slice image is determined.

4. The method according to claim 3, characterized in that Determining a target adjustment amount of the region of interest of the current numbered slice image compared with the starting numbered slice image based on a difference between the feature region images of the target feature region and the starting numbered slice image includes: Determining a first adjustment amount of the current numbered slice image relative to the starting numbered slice image for the target feature region based on a difference between the feature region images of the current numbered slice image and the starting numbered slice image for the target feature region; predicting a second adjustment amount of the current number slice image relative to the starting number slice image for the target feature region based on a difference between the starting number slice image and the ending number slice image for the target feature region; A target adjustment amount of the region of interest of the slice image with the current number compared to the slice image with the starting number is determined according to the first adjustment amount and the second adjustment amount.

5. The method according to claim 4, characterized in that There are multiple target feature regions, and determining a target adjustment amount of the region of interest of the slice image with the current number compared to the slice image with the starting number according to the first adjustment amount and the second adjustment amount includes: Taking each target feature region as a feature region to be calculated, determining a first offset and a first rotation amount for the feature region to be calculated in the first adjustment amount, and a second offset and a second rotation amount for the feature region to be calculated in the second adjustment amount; Determine an offset difference corresponding to the feature area to be calculated based on the first offset and the second offset, and determine a rotation difference corresponding to the feature area to be calculated based on the first rotation and the second rotation; According to the offset difference and the rotation difference corresponding to each target feature region, a target adjustment amount of the region of interest of the slice image with the current number compared with the slice image with the starting number is determined.

6. The method according to claim 5, characterized in that Determining a target adjustment amount of the region of interest of the slice image with the current number compared with the slice image with the starting number according to the offset difference and the rotation difference corresponding to each target feature region includes: The target adjustment amount is determined by combining the average of the offset differences corresponding to each target feature area and the average of the rotation differences corresponding to each target feature area; or The target adjustment amount is determined by combining a weighted average of the offset differences corresponding to each target feature region and a weighted average of the rotation differences corresponding to each target feature region.

7. The method according to claim 6, characterized in that Taking a weighted average of the offset differences corresponding to the target feature regions, the weight is determined based on the distance between the center of the region of interest in the current numbered slice image and the center of each target feature region; The weight of the weighted average of the rotation amount differences corresponding to each target feature region is determined based on the rotation amount for each target feature region in the second adjustment amount.

8. The method according to claim 1, characterized in that Methods for determining the characteristic regions and regions of interest of each slice image include: Based on the acquisition of a plurality of consecutive slice images, in response to a region selection operation, determining a region of interest and a feature region in a slice image with a starting number, and determining a region of interest and a feature region in a slice image with an ending number; The region of interest and the characteristic region in the current numbered slice image are determined based on the region of interest and the characteristic region in the starting numbered slice image by a position mapping method.

9. The method according to claim 8, characterized in that The selection of feature areas in the starting number slice image must meet at least the following conditions: The contrast between the selected feature area and the background in the starting number slice image exceeds the contrast threshold, the edge recognition exceeds the recognition threshold, and the feature area is unique; The selected feature region is a feature region common to the start number slice image and the end number slice image, and the structural similarity between the start number slice image and the end number slice image exceeds a similarity threshold, and the adjustment amount is lower than an adjustment amount threshold; When multiple characteristic regions are selected, the multiple characteristic regions are in a non-clustered distribution state in the starting numbered slice image.

10. The method according to claim 8, characterized in that Before determining the region of interest and the feature region in the slice image with the termination number in response to the region selection operation, the method further includes: The image feature automatic matching algorithm is used to generate the recommended region of interest and feature region in the ending slice image based on the region of interest and feature region in the starting slice image.

11. A calibration device, characterized in that: include: a determination module configured to determine, based on the obtained characteristic region images of a plurality of consecutive slice images, a target adjustment amount of the region of interest of the slice image with the current number compared to the slice image with the starting number based on similarities and differences between the characteristic region images of the slice image with the current number and the slice image with the starting number, wherein the current number is a number between the starting number and the ending number; An adjustment module, configured to adjust the region of interest of the current numbered slice image based on the target adjustment amount to obtain a target scanning region, and obtain an image of the region of interest of the current numbered slice image by scanning the target scanning region after the adjustment; The verification module is used to compare the region of interest images of the current numbered slice image with the starting numbered slice image, determine a target verification result of the region of interest image of the current numbered slice image, and visually display the target verification result.

12. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor. The computer program is executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 10.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 10 is implemented.

Images