Method, device and system for evaluating image splicing precision of biological tissue imaging, medium and product

Abstract

The invention relates to a method, a device and a system for evaluating image splicing precision of biological tissue imaging, a medium and a product. The method comprises the following steps: acquiring a plurality of imaging data obtained by imaging a calibration object by imaging equipment; carrying out image splicing on the multiple pieces of imaging data by utilizing a to-be-evaluated image splicing method to obtain a spliced image; determining gray level distribution data of an area where the calibration object is located in the spliced image; determining gray level change characteristics according to the gray level distribution data; and evaluating the splicing precision of the image splicing method based on the gray change feature and the minimum interval between the plurality of identifiers. According to the invention, the problem that there is no evaluation scheme for the splicing precision of the image splicing algorithm can be solved, and the splicing precision of the image splicing algorithm can be accurately evaluated.

Description

Technical Field

[0001] The present disclosure relates to the field of image processing, and in particular to an evaluation method, device, system, medium, and product for image stitching accuracy of biological tissue imaging. Background Art

[0002] With the development of image processing technology, image data processing methods have been widely used in various fields, especially for post-processing of imaging data collected by equipment.

[0003] Taking the biomedical field as an example, optical microscopes can be used to observe biological tissue structures, and combined with image data acquisition equipment to store imaging data, in order to conduct in-depth analysis of tissue cell arrangement, structural characteristics, and connection relationships. With the development of microscope technology, non-invasive three-dimensional imaging microscopes with depth imaging capabilities have been widely used in medical and brain science research. For example, Optical Coherence Tomography (OCT) has shown great value in ophthalmology and dentistry. For example, Polarization-Sensitive Optical Coherence Tomography (PSOCT) combined with polarization detection technology plays a unique role in the field of brain imaging based on the birefringence of tissues.

[0004] However, due to the limited imaging field of view, a single image often cannot show the full picture of the tissue. Therefore, in order to obtain comprehensive and high-resolution imaging, in some cases, the tissue needs to be imaged multiple times, and then the image stitching algorithm is used to stitch the data of multiple images to achieve complete and high-resolution imaging of the tissue. In this process, the final imaging quality is closely related to the stitching accuracy of the image stitching algorithm. However, there is currently a lack of methods to evaluate the stitching accuracy of the image stitching algorithm, which makes it difficult to ensure the accuracy of image stitching and the reliability of the stitched image. Summary of the invention

[0005] The present disclosure provides a method, device, system, medium, and product for evaluating the image stitching accuracy of biological tissue imaging, so as to at least solve the problem that there is a lack of an evaluation scheme for the stitching accuracy of an image stitching algorithm in the related art. The technical solution of the present disclosure is as follows: According to a first aspect of the present disclosure, there is provided a method for evaluating the image stitching accuracy of biological tissue imaging. The evaluation method includes: obtaining a plurality of imaging data obtained by an imaging device imaging a calibration object, wherein each imaging data includes at least a part of the calibration object, the calibration object has a plurality of identifiers, there is an interval between adjacent identifiers among the plurality of identifiers, and the imaging device is capable of imaging biological tissue; using an image stitching method to be evaluated, stitching the plurality of imaging data to obtain a stitched image; determining gray-scale distribution data of a region where the calibration object is located in the stitched image; determining a gray-scale change feature according to the gray-scale distribution data; and evaluating the stitching accuracy of the image stitching method based on the gray-scale change feature and the minimum interval between the plurality of identifiers.

[0006] Optionally, the plurality of imaging data are obtained by the following method: determining a plurality of imaging positions of the calibration object according to the imaging field of view of the imaging device, a preset imaging redundancy overlap amount, and a preset imaging order, wherein the imaging redundancy overlap amount represents the overlap amount between imaging regions at adjacent imaging positions; and according to the plurality of imaging positions, moving the calibration object to the plurality of imaging positions in sequence in a fixed field of view of the imaging device according to the imaging order, and imaging the fixed field of view by the imaging device after each movement of the calibration object to obtain the plurality of imaging data.

[0007] Optionally, the determining the gray-scale change feature according to the gray-scale distribution data includes: determining target gray-scale data corresponding to a target region of the stitched image in the gray-scale distribution data; and determining the gray-scale change feature according to the target gray-scale data, wherein the target region includes the stitching seam between the imaging data and the overlapping region between the imaging data.

[0008] Optionally, the target gray-scale data is determined by the following method: determining a reference data position corresponding to a reference position of the calibration object in the gray-scale distribution data; and determining the target gray-scale data corresponding to the target region in the gray-scale distribution data based on the reference data position and the positional relationship between the reference position and the target region in the stitched image.

[0009] Optionally, the gray-scale change feature is determined by the following method: determining a plurality of feature positions of the gray-scale change, wherein the feature positions include the full width at half maximum position; determining the feature distance between every two adjacent feature positions; and determining the gray-scale change feature based on the feature distance.

[0010] Optionally, the minimum interval between the plurality of identifiers is less than or equal to the minimum resolution interval that can be recognized by the stitched image stitched by using the image stitching method.

[0011] According to a second aspect of the present disclosure, there is provided an apparatus for evaluating the image stitching accuracy of biological tissue imaging. The evaluation apparatus includes: an acquisition unit configured to acquire a plurality of imaging data obtained by an imaging device imaging a calibration object, wherein each imaging data includes at least a part of the calibration object, the calibration object has a plurality of identifiers, there is an interval between adjacent identifiers among the plurality of identifiers, and the imaging device is capable of imaging biological tissue; a stitching unit configured to use an image stitching method to be evaluated to perform image stitching on the plurality of imaging data to obtain a stitched image; a first determination unit configured to determine the gray-scale distribution data of the region where the calibration object is located in the stitched image; a second determination unit configured to determine the gray-scale change characteristics according to the gray-scale distribution data; and an evaluation unit configured to evaluate the stitching accuracy of the image stitching method based on the gray-scale change characteristics and the minimum interval between the plurality of identifiers.

[0012] According to a third aspect of the present disclosure, there is provided an evaluation system for the image stitching accuracy of biological tissue imaging. The evaluation system includes a movable stage, a calibration object, and a computing device. The computing device receives imaging data obtained by an imaging device imaging the calibration object placed on the movable stage. The imaging device is capable of imaging biological tissue. The computing device includes a processor and a memory for storing instructions executable by the processor. Wherein, when the instructions executable by the processor are run by the processor, the processor is caused to execute the method for evaluating the image stitching accuracy of biological tissue imaging according to the present disclosure.

[0013] According to a fourth aspect of the present disclosure, there is provided a computer-readable storage medium, which when the instructions in the computer-readable storage medium are executed by a processor of a computing device, enables the computing device to execute the method for evaluating the image stitching accuracy of biological tissue imaging according to the present disclosure.

[0014] According to a fifth aspect of the present disclosure, there is provided a computer program product including computer-executable instructions, which when executed by at least one processor, implement the method for evaluating the image stitching accuracy of biological tissue imaging according to the present disclosure.

[0015] The technical solutions provided by the present disclosure at least bring the following beneficial effects: According to the present disclosure, a plurality of imaging data obtained by an imaging device imaging a calibration object can be acquired, image stitching can be performed using an image stitching method to be evaluated to obtain a stitched image, and the gray-scale distribution data of the region where the calibration object is located in the stitched image can be determined, and the gray-scale change characteristics can be determined. Thus, the stitching accuracy of the image stitching method can be evaluated based on the gray-scale change characteristics and the minimum interval of the calibration object. In this way, the stitching accuracy of the image stitching algorithm can be accurately evaluated, which helps to ensure that the accuracy and reliability of the imaging data are not impaired due to image stitching errors, and provides guidance for subsequent processing or application of the stitched image.

[0016] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings herein are incorporated into the specification and constitute a part of this specification, showing embodiments consistent with the present disclosure, and are used together with the specification to explain the principles of the present disclosure, and do not constitute an improper limitation to the present disclosure.

[0018] Figure 1 is a schematic flowchart of a method for evaluating the stitching accuracy of image stitching of biological tissue imaging according to an exemplary embodiment of the present disclosure.

[0019] Figure 2 is a schematic diagram of a calibration object in a method for evaluating the stitching accuracy of image stitching of biological tissue imaging according to an exemplary embodiment of the present disclosure.

[0020] Figure 3 is a schematic diagram of an example of a system for evaluating the stitching accuracy of image stitching of biological tissue imaging according to an exemplary embodiment of the present disclosure.

[0021] Figure 4 is a schematic diagram of a plurality of imaging data in a method for evaluating the stitching accuracy of image stitching of biological tissue imaging according to an exemplary embodiment of the present disclosure.

[0022] Figure 5 is a schematic diagram of an image stitched using an image stitching method to be tested in a method for evaluating the stitching accuracy of image stitching of biological tissue imaging according to an exemplary embodiment of the present disclosure.

[0023] Figure 6 is a schematic diagram of gray-scale distribution data in a method for evaluating the stitching accuracy of image stitching of biological tissue imaging according to an exemplary embodiment of the present disclosure.

[0024] Figure 7 is a schematic flowchart of a step of determining target gray-scale data in a method for evaluating the stitching accuracy of image stitching of biological tissue imaging according to an exemplary embodiment of the present disclosure.

[0025] Figure 8 It is a diagram showing a list of full width at half maximum data measured in an evaluation method for image stitching accuracy of biological tissue imaging according to an exemplary embodiment of the present disclosure.

[0026] Figure 9 It is a schematic flowchart of an example of an evaluation method for image stitching accuracy of biological tissue imaging according to an exemplary embodiment of the present disclosure.

[0027] Figure 10 It is a schematic diagram showing the evaluation result of the accuracy of an image stitching method to be measured in an evaluation method for image stitching accuracy of biological tissue imaging according to an exemplary embodiment of the present disclosure.

[0028] Figure 11 It is a schematic block diagram of an evaluation system device for image stitching accuracy of biological tissue imaging according to an exemplary embodiment of the present disclosure.

[0029] Figure 12 It is a schematic block diagram of an evaluation system for image stitching accuracy of biological tissue imaging according to an exemplary embodiment of the present disclosure. Detailed implementation manners

[0030] In order to enable those of ordinary skill in the art to better understand the technical solutions of the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings.

[0031] It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to describe a specific order or sequence. It should be understood that such used data can be interchanged under appropriate circumstances so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0032] It should be noted here that "at least one of several items" in the present disclosure all represents three types of parallel situations including "any one of the several items", "any combination of several items", and "all of the several items". For example, "including at least one of A and B" includes the following three parallel situations: (1) including A; (2) including B; (3) including A and B. Another example, "executing at least one of step one and step two" means the following three parallel situations: (1) executing step one; (2) executing step two; (3) executing step one and step two.

[0033] As mentioned above, in the related art, there is a lack of methods for evaluating the stitching accuracy of image stitching algorithms, which makes it difficult to ensure the accuracy of image stitching and the reliability of the stitched images.

[0034] Specifically, taking OCT and PSOCT as examples, due to the limited imaging field of view, a single image often cannot show the full picture of the tissue. The higher the resolution of the imaging objective, the smaller the imaging field of view is usually. Therefore, in order to obtain comprehensive and high-resolution imaging, it is necessary to use an image stitching algorithm to stitch the data of multiple images to achieve complete and high-resolution imaging of the tissue. Therefore, multiple continuous imaging of biological slices and then stitching the images has become a key step in obtaining complete biological tissue structure information.

[0035] However, during the imaging process, due to the accuracy limitations of the imaging instrument and the mobile device itself, there is inevitably a certain deviation in the imaging position, resulting in a certain error between the stitched image and the true morphology of the tissue. The existence of these errors will lead to the loss of key information, errors in structural features or connection relationships in the further analysis of tissue images, and may even lead to wrong judgments and conclusions based on the stitched images, which will not only waste a lot of resources and time, but also have a negative impact on the development of the biomedical field.

[0036] Therefore, quantitative measurement of image stitching accuracy is very necessary, and it is urgent to develop a method for quantitative evaluation of image stitching accuracy.

[0037] In view of the above, exemplary embodiments of the present disclosure provide a method, device and system for evaluating image stitching accuracy of biological tissue imaging, a computer-readable storage medium and a computer program product, which can solve or at least alleviate the above problems.

[0038] In a first aspect of an exemplary embodiment of the present disclosure, a method for evaluating image stitching accuracy of biological tissue imaging is provided.

[0039] The method for evaluating the image stitching accuracy of biological tissue imaging according to the exemplary embodiment of the present disclosure can be applied to scenarios where users interact with software. For example, the software can be loaded on a user terminal, and the user can input evaluation instructions for the image stitching method to be evaluated on the user terminal. The user terminal can evaluate the stitching accuracy of the image stitching method by executing the method for evaluating the image stitching accuracy of biological tissue imaging according to the exemplary embodiment of the present disclosure.

[0040] Specifically, the user terminal can obtain a plurality of imaging data obtained by an imaging device imaging a calibration object, where each imaging data includes at least a part of the calibration object, the calibration object has a plurality of identifiers, and there is an interval between adjacent identifiers among the plurality of identifiers; the plurality of imaging data can be subjected to image stitching by using an image stitching method to be evaluated to obtain a stitched image; the gray-scale distribution data of the area where the calibration object is located in the stitched image can be determined; the gray-scale change characteristics can be determined according to the gray-scale distribution data; and the stitching accuracy of the image stitching method can be evaluated based on the gray-scale change characteristics and the minimum interval between the plurality of identifiers.

[0041] The above user terminal can be, for example, a workstation, a laptop computer, etc. However, the implementation scenario of the above method is only an example scenario. The method for evaluating the image stitching accuracy of biological tissue imaging according to the exemplary embodiments of the present disclosure can also be applied to other application scenarios. For example, a user can also request an evaluation of the image stitching accuracy from a server through a network on a user terminal (such as a mobile phone, a desktop computer, a tablet computer, etc.). The server can complete the request by executing the method for evaluating the image stitching accuracy of biological tissue imaging according to the exemplary embodiments of the present disclosure. Here, the server can be an independent server, a server cluster, a cloud computing platform or a virtualization center.

[0042] The method for evaluating the image stitching accuracy of biological tissue imaging according to the exemplary embodiments of the present disclosure can quickly realize the quantitative evaluation of the image stitching accuracy of biological tissue imaging, help ensure that the accuracy and reliability of imaging data are not damaged due to image stitching errors, and provide guidance for subsequent processing or application of the stitched image.

[0043] As Figure 1 shown, the method for evaluating the image stitching accuracy of biological tissue imaging can include the following steps: In step S110, a plurality of imaging data obtained by an imaging device imaging a calibration object can be obtained.

[0044] Here, the imaging device can image biological tissues. The imaging device can be, for example, but not limited to, optical microscopes such as OCT and PSOCT, electron microscopes, etc. It can be used, for example, to image biological tissue sections in the field of biomedicine. Since such an imaging device has a high imaging resolution and a small imaging field of view, the demand for image stitching and the requirement for stitching accuracy are also relatively high. Different from stitching images in the general field (such as the level of human eye resolution), during the imaging process, the imaging device may require magnifying the object to be photographed by dozens or even hundreds of times with high clarity. Therefore, a more accurate and reliable method is needed to evaluate its image stitching method. The evaluation method of the embodiments of the present disclosure can better cope with the image stitching evaluation of such precision imaging of microscopic imaging devices. Here, the biological tissue section can be, for example, a section of an ex vivo tissue or organ of a non-living animal, a local tissue section of a plant body, etc.

[0045] In this step, each imaging data can include at least a part of the calibration object, and the calibration object has a plurality of identifiers, and there is an interval between adjacent identifiers among the plurality of identifiers. For example, the plurality of imaging data can be different, and the position, pose, and photographed part of the calibration object in at least a part of the imaging data can all be different. Subsequently, using the image stitching method, these imaging data can be stitched together to obtain a complete calibration object.

[0046] The minimum interval between the identifiers of the calibration object (or it can also be called the minimum resolution interval) can refer to the minimum distance among the distances between adjacent identifiers. The identifiers can be, for example, calibration lines with equally spaced distributions in multiple directions. In this way, the stitching accuracy of the image stitching method in multiple directions can be evaluated, improving the accuracy of the evaluation. For example, Figure 2 shows a unit calibration object diagram according to an embodiment of the present disclosure. As Figure 2 shown, the calibration object can be, for example, a calibration ruler in the shape of a "plus" sign or a black and white chessboard, and calibration lines are marked in two mutually perpendicular directions. In addition, the calibration lines in each direction can mark multiple units of scales, such as millimeter (mm) scales, micrometer (μm) scales, etc. In Figure 2 , the minimum interval (1DIV) is 0.01 mm, that is, the resolution of the calibration object is 10 μm. The calibration object can be, for example, a calibration ruler with scales. For example, the lengths of the calibration object in two mutually perpendicular directions (such as the X direction and the Y direction) can both be 1 mm, and its resolution can both be 10 μm, that is, the distance between adjacent identifiers is 10 μm.

[0047] As an example, in step S110, multiple imaging data can be obtained by moving the calibration object or the imaging objective of the imaging device, and during the process of collecting these imaging data, the calibration object can be imaged at consecutive positions with a certain degree of redundant overlap.

[0048] For example, the above-mentioned multiple imaging data can be obtained in the following manner: According to the imaging field of view of the imaging device, a preset imaging redundant overlap amount, and a preset imaging sequence, determine multiple imaging positions of the calibration object; according to the multiple imaging positions, move the calibration object to the multiple imaging positions in sequence in the fixed field of view of the imaging device according to the imaging sequence, and image the fixed field of view with the imaging device after each movement of the calibration object to obtain multiple imaging data. Here, the imaging redundant overlap amount can represent the overlap amount between the imaging regions at adjacent imaging positions.

[0049] Specifically, during the imaging process, imaging of multiple consecutive positions of a calibration object such as a unit calibration scale can be achieved by moving a precision moving stage to obtain multiple imaging data. Here, imaging of multiple consecutive positions can be achieved by setting corresponding imaging position points according to the imaging field of view.

[0050] The imaging redundant overlap amount and the total number of imaging can be configured according to the imaging field of view in the used imaging device, the imaging resolution for target image acquisition by the imaging device, and the imaging range of the calibration object, and based on the imaging redundant overlap amount and the total number of imaging, calculate the imaging sequence and imaging position coordinates. Here, the imaging field of view and imaging resolution can be preset according to actual needs, and the imaging range of the calibration object can be such a range that the range of the spliced image is larger than the calibration object, so that the spliced image can contain the complete calibration object. In addition, the imaging sequence can be set according to actual needs, such as imaging in a zigzag pattern, etc. Here, mathematical or geometric methods can be used to determine the imaging redundant overlap amount, the total number of imaging, the imaging sequence, and the imaging position coordinates. For example, in the case of known imaging field of view, imaging resolution, and imaging range of the calibration object, the imaging redundant overlap amount (such as the overlap size between two imaging) and the total number of imaging can be calculated, and thus the coordinates of each imaging position can be calculated based on the imaging redundant overlap amount and the total number of imaging. In the case of known imaging position coordinates, the imaging sequence can be determined according to a preset method such as the shortest path method.

[0051] In addition, the imaging sequence can be determined according to the image evaluation method to be evaluated, however, the embodiments of the present disclosure do not particularly limit the imaging sequence. In addition, the imaging position coordinates, the redundant overlap amount of adjacent position imaging, and the imaging field of view size involved in the imaging parameters described herein can be related to the image evaluation method to be evaluated to splice the imaging data within a single layer of the chip using the image evaluation method to be evaluated.

[0052] For example, the above-mentioned movement process can be achieved by Figure 3 an evaluation system for the image stitching accuracy of biological tissue imaging according to an embodiment of the present disclosure as shown. Specifically, as Figure 3 shown, the system 300 may include a moving stage 310, a calibration object 320, and a computing device 330. Among them, the calibration object 320 can be placed on the moving stage 310. The moving stage 310 can be located below the imaging device 100 and is movable, for example, it can move in the vertical direction and two mutually perpendicular horizontal directions, so as to drive the calibration object 320 to move relative to the imaging device 100. Here, the moving stage 310 can be a stage capable of precise movement, which may include a stage control system (not shown). The imaging device 10 may include an imaging controller 11, a reference arm 12, and an objective lens 13. The computing device 330 can be used to execute the evaluation method described herein. The computing device 330 can communicate bidirectionally with the biological tissue imaging device 100 and the stage control system 310, for example, it can monitor the imaging trigger signal, imaging position information, imaging parameters, and imaging status. In addition, the computing device 330 is adapted to the communication methods supported by the actually used biological tissue imaging device 100 and the stage control system.

[0053] In addition, the system may further include a server 340. The server 340 can be communicatively connected to the computing device 330. For example, the server 340 can be used to store a plurality of imaging data.

[0054] Although the above describes an example implementation of collecting imaging data, the embodiments of the present disclosure are not limited thereto, and the collection of a plurality of imaging data can also be achieved through other systems or structures.

[0055] In addition, as an example, a parameter configuration module (for example, set in the computing device) can be used to set imaging parameters. For example, the imaging field of view size can be set to 3 mm, and the imaging redundant overlap amount can be set to 2 mm. According to the set imaging redundant overlap amount and imaging field of view, continuous imaging position coordinates can be generated. For example, for the imaging data as Figure 4 shown, the total number of scans can be set to 4 times according to the scanning positions of 2 in each row and each column, and the column scanning order with priority in the Y direction can be used to perform tomographic (C-scan) imaging at 4 consecutive positions, and the imaging XY plane at each position can be obtained as Figure 4 shown.

[0056] Here, the imaging redundant overlap amount can refer to the overlap amount in the current moving direction. For example, when the calibration object moves along the X direction, the imaging redundant overlap amount can be the overlapping length of the imaging data before and after the movement in the X direction.

[0057] In the above method, multi-position imaging data can be collected by moving the calibration object, which simplifies the process of imaging data collection and avoids moving the imaging device.

[0058] In step S120, an image stitching method to be evaluated can be used to stitch multiple imaging data to obtain a stitched image.

[0059] In this step, the image stitching method to be evaluated can be any image stitching method as long as it can stitch the imaging data.

[0060] For example, the image stitching method to be evaluated can be used to perform data preprocessing, data fusion and stitching on the obtained multiple consecutive position imaging data according to the imaging order and imaging parameters such as imaging redundancy overlap, and then obtain the stitched image, as shown in Figure 5 shown. In addition, stitching the imaging data can, for example, refer to in-chip stitching to obtain a single-layer in-chip image after stitching.

[0061] As an example, the resolution of the calibration object can be higher than or equal to the resolution that the image stitching method can achieve. For example, the minimum interval between multiple marks can be less than or equal to the minimum resolution interval that the stitched image obtained by using the image stitching method can be recognized (or processed), or rather, the distance between adjacent marks of the calibration object can be less than or equal to the minimum distance that the image stitching method can recognize (or process). In this way, the accuracy and reliability of the accuracy evaluation of the image stitching method can be ensured. Here, the resolution that the stitched image obtained by using the image stitching method can achieve, or the minimum resolution interval that can be recognized (or processed), or the minimum distance that can be recognized (or processed) can be known or predetermined, which is the performance of the image stitching method itself and can be determined when the method is proposed. Therefore, when evaluating the image stitching method to be evaluated by adopting the embodiments of the present disclosure, the resolution that the stitched image obtained by using this method can achieve, or the minimum resolution interval that the stitched image can be recognized (or processed), or the minimum distance that can be recognized (or processed) can be known.

[0062] In addition, the imaging data can be data that has not been subjected to optical correction processing, and the stitched image can be an image in which the stitching seam area has not been subjected to fusion processing. In this way, as shown in Figure 5 shown, there are multiple ( Figure 5 4 in the figure) gray-scale contrast displays that are different and stitching traces in the stitched image. However, the embodiments of the present disclosure are not limited thereto. In the case where the imaging data is subjected to optical correction or the stitching seam area is subjected to fusion processing, the evaluation method of the present disclosure can still be adopted.

[0063] In step S130, the gray-scale distribution data of the region where the calibration object is located in the spliced image can be determined.

[0064] In one example, the region where the calibration object is located in the spliced image can be recognized by means of image recognition (such as existing image recognition algorithms).

[0065] In another example, the gray-scale calculation range input by the user can also be received, and this range can be determined as the region where the calibration object is located in the spliced image. For example, the user can use the straight-line tool to determine the gray-scale calculation range in the above Figure 5 For example, mark points A to B.

[0066] In this step, public image processing software and library functions such as Fiji, OpenCV library, matplotlib library, and Plotly library can be used to implement gray-scale calculation to obtain the gray-scale distribution data. As an example, the gray-scale distribution data can be a gray-scale distribution graph displayed in the form of a curve or a chart.

[0067] For example, the public image processing software Fiji can be used to open the spliced image, and the display effect of the image can be adjusted through the menu bar "Image--Adjust--Brightness&Contrast". The gray-scale distribution graph of the selected area can be drawn using the Fiji menu "Analyze--Plot Profile" as the above gray-scale distribution data.

[0068] In addition, in some cases, the imaging data of the imaging device does not include a scale. In this regard, before determining the gray-scale distribution data, this evaluation method can also include: setting a scale in the spliced image. Here, in one example, the scale can be set in the spliced image according to the actual calibration object by means of image identification (such as existing image object identification algorithms). In another example, the scale input by the user can also be received. For example, the user can draw a straight line on the spliced image, set the image display scale through "Analyze--Set Scale" of the software Fiji, and draw a scale on the image through "Analyze--Tools--Scale Bar". As Figure 5 shown, the scale can be, for example, the scale of the scale on the calibration object except for the smallest scale. The scale can be used to calibrate the correspondence between the gray-scale distribution and the position on the actual calibration object. Here, the scale can represent the proportional relationship between the size in the spliced image and the size of the actual object being photographed. For example, it can be the actual size represented by one pixel in the spliced image, providing a quantitative size reference for the image, which can provide a reference for the subsequent standardized measurement of length.

[0069] Figure 6An example of grayscale distribution data is shown, where the vertical axis represents the grayscale value and the horizontal axis represents the distance, which corresponds to the distance on the calibration object. For example, the starting point of the horizontal axis can be the starting point of the calibration object in one direction, the ending point of the horizontal axis can be the ending point of the calibration object in this direction, and the unit of the horizontal axis can correspond to the length unit of the calibration object. Figure 6 The grayscale distribution diagram shown, for example, can be obtained by calculating the grayscale using the software Fiji.

[0070] In step S140, the grayscale change characteristics of the grayscale change can be determined according to the grayscale distribution data.

[0071] Since the calibration object has multiple identifiers, after converting the stitched image into grayscale distribution data, the change in the grayscale value will also show grayscale change characteristics. For example, the grayscale value has grayscale change characteristic changes along the arrangement direction of the identifiers.

[0072] As an example, this step S140 may include: determining target grayscale data corresponding to the target area of the stitched image in the grayscale distribution data; and determining the grayscale change characteristics of the grayscale change according to the target grayscale data.

[0073] Here, the target area can be within the area where the calibration object is located, and it can include the stitching seam between the imaging data and the overlapping area between the imaging data. For example, as Figure 6 shown, area I is the grayscale curve corresponding to the stitching seam at two adjacent positions, and areas II and III are the stitching of the overlapping areas respectively. Here, by considering the stitching seam and the overlapping area, the image areas that may have stitching deviations or low stitching accuracy can be focused on, and the influence of non-stitching areas on the overall stitching accuracy evaluation can be excluded. However, the embodiments of the present disclosure are not limited thereto. In step S140, the grayscale change characteristics of the grayscale change in the entire area where the calibration object is located or the entire range in any stitching direction can also be determined.

[0074] As an example, as Figure 7 shown, the target grayscale data can be determined in the following manner: In step S710, the reference data position corresponding to the reference position of the calibration object can be determined in the grayscale distribution data.

[0075] Here, the reference position can be, for example, the end identifier of the calibration object (such as the starting end or the ending end), such as Figure 5 point A or point B in, or any other identifier of the calibration object. The reference position can be specified by the user. For example, the user can mark the reference position in the stitched image.

[0076] In this step, for example, according to the reference position, the corresponding reference data position in the gray-scale distribution data can be determined. For example, when the reference position is the start-end or end-end identifier, the reference data position can be the first gray-scale data or the last gray-scale data in the gray-scale distribution data.

[0077] In step S720, based on the reference data position and the positional relationship between the reference position and the target area in the stitched image, the target gray-scale data corresponding to the target area can be determined in the gray-scale distribution data.

[0078] In the case where the reference position and the target area are known, the relative positional relationship between the reference position and the target area can be determined by means such as image recognition algorithms or receiving user input. For example, in Figure 6 , taking the reference position as the start-end identifier (i.e., corresponding to point A in the stitched image) as an example, the reference Figure 5 , area I containing the stitching seam is the stitching seam closest to the start-end identifier (point A in the stitched image), and the relative distance between the two can be determined according to the scale. Therefore, the stitching seam gray-scale data closest to the reference data position (for example, at the above relative distance from the reference data position) can be searched for in the gray-scale distribution data as the target gray-scale data.

[0079] In this way, the target gray-scale data of the representative target area can be determined more accurately, thereby improving the accuracy of the stitching evaluation result.

[0080] In the above process, whether it is for the gray-scale distribution data of the entire range of the calibration object in a certain stitching direction or for the target gray-scale distribution data of the target area, as an example, the gray-scale change characteristics can be determined in the following way: determining multiple characteristic positions of the gray-scale change; determining the characteristic distance between every two adjacent characteristic positions; and determining the gray-scale change characteristics based on the characteristic distance.

[0081] Here, the characteristic positions can include, for example, the half-peak width positions, and the characteristic distance can be the half-peak width. Specifically, the half-peak width point positions can be determined according to the gray-scale distribution data. For example, for the measurement points shown in Figure 6 , two adjacent measurement points are successively determined, and the half-peak width measurement is realized through software such as Fiji to obtain Figure 8 the data list shown. The measured half-peak width value is essentially the distance between adjacent calibration scale lines, and the corresponding measurement unit is μm. In addition, the gray-scale peak can also be searched for in the gray-scale distribution data such as the gray-scale value map, the half-peak height of each gray-scale peak is calculated, and the two points corresponding to the half-peak height are found in the gray-scale map, and the distance between these two points is measured as the half-peak width, that is, the above characteristic distance.

[0082] In this way, statistical values such as the average can be calculated for the obtained feature distances (such as Length in the full width at half maximum measurement table), and the gray-scale change features can be obtained. Here, by calculating statistical values such as the average, the influence of measurement errors on the accuracy measurement of the image stitching method can be excluded.

[0083] Through this method, the distribution interval of gray-scale values can be quantified, so as to compare it with the actual scale of the calibration object and realize the quantitative evaluation of the stitching accuracy.

[0084] In step S150, the stitching accuracy of the image stitching method can be evaluated based on the gray-scale change features and the minimum interval between multiple identifiers.

[0085] As an example, the gray-scale change features can be compared with the minimum interval between multiple identifiers to evaluate the stitching accuracy. The comparison methods can include, for example, ratio comparison, difference comparison, percentage comparison, etc.

[0086] For example, Figure 9 shows an example of the evaluation result of the accuracy of the image stitching method to be measured in the method for evaluating the image stitching accuracy of biological tissue imaging according to an exemplary embodiment of the present disclosure. If the gray-scale change feature is L 测 = 9.8258um, and the scale line spacing of the actual calibration object (which is the minimum interval between multiple identifiers) is L 真 = 10μm, then the stitching accuracy of the in-chip stitched image can be expressed by the following formula:

[0087] Among them, the percentage calculated based on the gray-scale change features and the minimum interval between multiple identifiers is the evaluated stitching accuracy.

[0088] As described above, due to the limitation of the imaging field of view size, image stitching has become an indispensable technical means to achieve comprehensive and high-resolution imaging of tissues. For fine structural features and connection relationships in biological tissues, high-precision image stitching is crucial. Therefore, it is very necessary to evaluate the image stitching algorithm reliably and quickly. According to the image stitching evaluation method of the embodiments of the present disclosure, the in-chip image stitching accuracy can be quantified and evaluated quickly and effectively, providing an intuitive evaluation for the image stitching method, providing accurate feedback for the optimization of the image stitching algorithm, significantly improving the optimization efficiency of the image stitching algorithm, shortening the development cycle, and saving time costs. In addition, this method is particularly applicable to the application scenario of high-resolution imaging of tissues in the biomedical field.

[0089] In addition, for the method for evaluating the image stitching accuracy according to the embodiments of the present disclosure, the measurement principle is intuitive and easy to understand, and the operation process is clear and simple. In terms of gray-scale calculation, this method does not require additional development of complex algorithms. Only by leveraging open-source image software and function libraries, the measurement of the gray-scale distribution map and the full width at half maximum of the stitched image can be achieved, enabling a rapid and quantitative evaluation of the stitching algorithm to be measured, and providing accurate quantitative feedback data for the optimization of the image stitching algorithm. Using this method can not only significantly shorten the optimization cycle of the image stitching algorithm, but also improve the quality of image stitching, providing data guarantee for in-depth analysis of high resolution.

[0090] Figure 10 Fig. shows an example of the method for evaluating the image stitching accuracy of biological tissue imaging according to an exemplary embodiment of the present disclosure.

[0091] As Figure 10 shown, in step S1001, the image acquisition software can be started, and in step S1002, imaging parameters such as the imaging field of view, imaging redundancy overlap degree, etc. can be set. In step S1003, continuous imaging position coordinates can be generated according to the imaging range and imaging parameters.

[0092] In step S1004, the biological tissue imaging device can be controlled to start imaging. In step S1005, the calibration object can be moved to the current imaging position, and in step S1006, it can be determined whether the current imaging has been completed. In response to the completion of the current imaging, in step S1007, it can be further determined whether all imaging in the current column has been completed; in response to the non-completion of the current imaging, step S1006 can be executed again to continue waiting for an instruction indicating the completion of all imaging in the current column. Here, this instruction can be issued by the biological tissue imaging device, for example.

[0093] In response to the completion of all imaging in the current column, in step S1008, it can be further determined whether all positions have been imaged; in response to the non-completion of all imaging in the current column, step S1005 can be returned to and the next imaging position in the imaging sequence can be used as the current imaging position.

[0094] In response to the completion of all positions, in step S1009, the image data can be calculated for subsequent processing. Here, the format of the imaging data obtained by the imaging device may be different from the format of the input image of the image stitching method. Therefore, in this step, the image data can be calculated to convert it into the format of the input image of the image stitching method; in response to the non-completion of all positions, step S1005 can be returned to and the next imaging position in the imaging sequence can be used as the current imaging position.

[0095] In step S1010, an image stitching software can be launched to stitch the imaging data to obtain a stitched image. And in step S1011, an open-source software or function library can be used to calculate the gray-scale distribution map of the stitched image. In step S1012, the full width at half maximum (FWHM) in the gray-scale map can also be measured.

[0096] In step S1013, the measured FWHM data can be saved, and statistical values such as the average value of the FWHM data can be calculated as the gray-scale change feature of the gray-scale distribution. In step S1014, the true value of the minimum interval of the identification of the calibration object can be compared with the measured gray-scale change feature to complete the accuracy measurement of the image stitching method.

[0097] According to the image stitching evaluation method of the embodiments of the present disclosure, for the problem that the errors existing in image stitching reduce the accuracy, reliability and credibility in the subsequent in-depth analysis of image data, it can be used to optimize the image stitching algorithm, especially in high-resolution imaging involving biological tissues.

[0098] In addition, according to the evaluation method of the image stitching accuracy of the embodiments of the present disclosure, it not only helps to optimize the stitching algorithm to ensure that the accuracy and reliability of the data are not impaired due to image stitching errors. At the same time, it can also prevent differences from occurring when different research teams obtain the same biological tissue structure information due to image stitching errors, thereby weakening the credibility of the research. In addition, the evaluation of the image stitching accuracy can also promote the continuous improvement of the imaging technology to a certain extent.

[0099] In addition, according to the evaluation method of the image stitching accuracy of the embodiments of the present disclosure, the implementation is simple. The existing algorithms for calculating the gray-scale distribution map or measuring the FWHM can be used to quickly realize the quantitative evaluation of the in-chip image stitching accuracy, thereby shortening the optimization cycle of the image stitching algorithm and improving the overall development efficiency.

[0100] In a second aspect of the exemplary embodiments of the present disclosure, there is provided an apparatus for evaluating the image stitching accuracy of biological tissue imaging, as Figure 11 shown. The evaluation apparatus may include an acquisition unit 1110, a stitching unit 1120, a first determination unit 1130, a second determination unit 1140, and an evaluation unit 1150.

[0101] The acquisition unit 1110 is configured to acquire a plurality of imaging data obtained by an imaging device imaging a calibration object, where each imaging data includes at least a part of the calibration object, the calibration object has a plurality of identifications, there is an interval between adjacent identifications among the plurality of identifications, and the imaging device is capable of imaging biological tissues.

[0102] The stitching unit 1120 is configured to use the image stitching method to be evaluated to perform image stitching on the plurality of imaging data to obtain a stitched image.

[0103] The first determination unit 1130 is configured to determine the gray-scale distribution data of the area where the calibration object is located in the stitched image.

[0104] The second determination unit 1140 is configured to determine the gray-scale change characteristics of the gray-scale change according to the gray-scale distribution data.

[0105] The evaluation unit 1150 is configured to evaluate the stitching accuracy of the image stitching method based on the gray-scale change characteristics and the minimum interval between multiple identifiers.

[0106] As an example, multiple imaging data are obtained by the following method: according to the imaging field of view of the imaging device, the preset imaging redundant overlap amount, and the preset imaging order, determine multiple imaging positions of the calibration object, where the imaging redundant overlap amount represents the overlap amount between the imaging areas of adjacent imaging positions; according to the multiple imaging positions, move the calibration object to the multiple imaging positions in sequence in the fixed field of view of the imaging device, and after each movement of the calibration object, use the imaging device to image the fixed field of view to obtain multiple imaging data.

[0107] As an example, the second determination unit 1140 is configured to: in the gray-scale distribution data, determine the target gray-scale data corresponding to the target area of the stitched image; according to the target gray-scale data, determine the gray-scale change characteristics of the gray-scale change, where the target area includes the stitching seam between the imaging data and the overlapping area between the imaging data.

[0108] As an example, the second determination unit 1140 is configured to determine the target gray-scale data in the following way: determine the reference data position corresponding to the reference position of the calibration object in the gray-scale distribution data; based on the reference data position and the positional relationship between the reference position and the target area in the stitched image, determine the target gray-scale data corresponding to the target area in the gray-scale distribution data.

[0109] As an example, the second determination unit 1140 is configured to determine the gray-scale change characteristics of the gray-scale change in the following way: determine multiple characteristic positions of the gray-scale change, where the characteristic positions include the full width at half maximum positions; determine the characteristic distances between every two adjacent characteristic positions; based on the characteristic distances, determine the gray-scale change characteristics.

[0110] As an example, the minimum interval between multiple identifiers is less than or equal to the minimum resolution interval that can be recognized by the stitched image obtained by using the image stitching method.

[0111] Regarding the device in the above embodiments, the specific manner in which each unit performs operations has been described in detail in the embodiments related to the method. Each unit in the evaluation device for the image stitching accuracy of biological tissue imaging can execute the corresponding steps in the method according to the method for evaluating the image stitching accuracy of biological tissue imaging in the method embodiments of the first aspect above, and will not be elaborated here.

[0112] In the third aspect of the exemplary embodiments of the present disclosure, an evaluation system for the image stitching accuracy of biological tissue imaging is provided. As Figure 12 shown, the evaluation system 1200 may include a movable stage 1210, a calibration object 1220, and a computing device 1230. The computing device 1230 receives imaging data obtained by an imaging device imaging the calibration object placed on the movable stage, and the imaging device can image biological tissue. Here, the evaluation system 1200 may include an imaging device, and the imaging device may be disposed above the movable stage 1210; the evaluation system 1200 may also not include an imaging device. For example, an external imaging device may also be connected to the evaluation system 1200 to evaluate the image stitching accuracy.

[0113] The computing device 1230 includes a processor and a memory for storing processor-executable instructions. When the processor-executable instructions are run by the processor, the processor is caused to execute the method for evaluating the image stitching accuracy of biological tissue imaging according to the present disclosure.

[0114] As an example, the evaluation system may be the evaluation system as Figure 3 shown, which has been described in detail above and will not be elaborated here.

[0115] As an example, the computing device 1230 does not have to be a single device, and may also be any aggregate of devices or circuits that can execute the above instructions (or instruction sets) individually or jointly. The computing device 1230 may also be a part of an integrated control system or a system manager, or may be configured to interface with a local or remote (e.g., via wireless transmission) server.

[0116] In the computing device 1230, the processor may include a central processing unit (CPU), a graphics processing unit (GPU), a programmable logic device, a dedicated processor system, a microcontroller, or a microprocessor. As an example and not a limitation, the processor may also include an analog processor, a digital processor, a microprocessor, a multi-core processor, a processor array, a network processor, etc.

[0117] The processor can execute instructions or code stored in the memory, where the memory can also store data. The instructions and data can also be sent and received via the network interface device over the network, where the network interface device can employ any known transmission protocol. The memory can be integrated with the processor, for example, by arranging RAM or flash memory within an integrated circuit microprocessor, etc. Additionally, the memory can include separate devices, such as external disk drives, storage arrays, or other storage devices that can be used by any database system. The memory and the processor can be operatively coupled or can communicate with each other, for example, via I / O ports, network connections, etc., such that the processor can read files stored in the memory.

[0118] In addition, the computing device 1230 can also include a video display (such as a liquid crystal display) and a user interaction interface (such as a keyboard, a mouse, a touch input device, etc.). All components of the computing device 1230 can be connected to each other via a bus and / or a network.

[0119] In a fourth aspect of the exemplary embodiments of the present disclosure, a computer-readable storage medium is provided. When the instructions in the computer-readable storage medium are executed by a processor of a computing device, the computing device is enabled to execute a method for evaluating the image stitching accuracy of biological tissue imaging according to the embodiments of the present disclosure.

[0120] A computer-readable storage medium may, for example, be a memory including instructions. Optionally, the computer-readable storage medium may be: read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), electrically erasable programmable read-only memory (EEPROM), dynamic random access memory (DRAM), static random access memory (SRAM), flash memory, non-volatile memory, CD-ROM, CD-R, CD+R, CD-RW, CD+RW, DVD-ROM, DVD-R, DVD+R, DVD-RW, DVD+RW, DVD-RAM, BD-ROM, BD-R, BD-R LTH, BD-RE, Blu-ray or optical disc memory, hard disk drive (HDD), solid state drive (SSD), cartridge memory (such as multimedia card, secure digital (SD) card or extreme digital (XD) card), magnetic tape, floppy disk, magneto-optical data storage device, optical data storage device, hard disk, solid state disk, and any other device configured to store a computer program and any associated data, data files, and data structures in a non-transitory manner and provide the computer program and any associated data, data files, and data structures to a processor or computer such that the processor or computer can execute the computer program. The computer program in the above computer-readable storage medium can run in an environment deployed in computer devices such as clients, hosts, proxy devices, servers, etc. In addition, in one example, the computer program and any associated data, data files, and data structures are distributed on a networked computer system such that the computer program and any associated data, data files, and data structures are stored, accessed, and executed in a distributed manner by one or more processors or computers.

[0121] In a fifth aspect of the exemplary embodiments of the present disclosure, there is provided a computer program product including computer-executable instructions that, when executed by at least one processor, implement the method for evaluating the image stitching accuracy of biological tissue imaging according to the embodiments of the present disclosure.

[0122] Those skilled in the art will readily conceive of other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. The present disclosure is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include known common knowledge or conventional technical means in the technical field not disclosed by the present disclosure. The specification and embodiments are only to be considered as exemplary, and the true scope and spirit of the present disclosure are pointed out by the claims.

[0123] In addition, it should be noted that although several examples of each step are described above with reference to specific drawings, it should be understood that the embodiments of the present disclosure are not limited to the combinations given in the examples. Steps appearing in different drawings can be combined, and the execution order of each step can be changed, which will not be exhaustively listed here.

[0124] It should be understood that the present disclosure is not limited to the exact structures already described and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present disclosure is only limited by the appended claims.

Claims

1. An evaluation method for the image stitching accuracy of biological tissue imaging, characterized in that, The evaluation method includes: Obtaining a plurality of imaging data obtained by an imaging device imaging a calibration object, wherein each imaging data includes at least a part of the calibration object, the calibration object has a plurality of identifiers, there is an interval between adjacent identifiers among the plurality of identifiers, and the imaging device can image biological tissues; Using an image stitching method to be evaluated, stitching the plurality of imaging data to obtain a stitched image; Determining the gray-scale distribution data of the area where the calibration object is located in the stitched image; Determining the gray-scale change characteristics according to the gray-scale distribution data; Evaluating the stitching accuracy of the image stitching method based on the gray-scale change characteristics and the minimum interval between the plurality of identifiers.

2. The evaluation method according to claim 1, wherein The plurality of imaging data are obtained by the following method: According to the imaging field of view of the imaging device, a preset imaging redundancy overlap amount, and a preset imaging order, determining a plurality of imaging positions of the calibration object, wherein the imaging redundancy overlap amount represents the overlap amount between imaging areas at adjacent imaging positions; According to the plurality of imaging positions, sequentially moving the calibration object to the plurality of imaging positions in the fixed field of view of the imaging device according to the imaging order, and imaging the fixed field of view by the imaging device after each movement of the calibration object to obtain the plurality of imaging data.

3. The evaluation method according to claim 1, wherein The determining the gray-scale change characteristics according to the gray-scale distribution data includes: In the gray-scale distribution data, determining target gray-scale data corresponding to a target area of the stitched image; Determining the gray-scale change characteristics according to the target gray-scale data, wherein the target area includes the stitching seam between the imaging data and the overlapping area between the imaging data.

4. The evaluation method according to claim 3, wherein The target gray-scale data is determined by the following method: Determining a reference data position corresponding to a reference position of the calibration object in the gray-scale distribution data; Based on the reference data position and the positional relationship between the reference position and the target area in the stitched image, determining the target gray-scale data corresponding to the target area in the gray-scale distribution data.

5. The evaluation method according to claim 1 or 3, characterized in that, The gray-scale change characteristics are determined by the following method: Determining a plurality of characteristic positions of the gray-scale change, wherein the characteristic positions include the full width at half maximum position; Determining the characteristic distance between every two adjacent characteristic positions; Determining the gray-scale change characteristics based on the characteristic distance.

6. The evaluation method according to claim 1, wherein The minimum interval between the plurality of identifiers is less than or equal to the minimum resolution interval that can be recognized by the stitched image stitched by using the image stitching method.

7. An evaluation device for the image stitching accuracy of biological tissue imaging, characterized in that, The evaluation device includes: An acquisition unit configured to acquire a plurality of imaging data obtained by an imaging device imaging a calibration object, wherein each imaging data includes at least a part of the calibration object, the calibration object has a plurality of identifiers, there is an interval between adjacent identifiers among the plurality of identifiers, and the imaging device can image biological tissues; A stitching unit configured to use an image stitching method to be evaluated to stitch the plurality of imaging data to obtain a stitched image; A first determination unit, configured to determine grayscale distribution data of a region where the calibration object is located in the spliced image; A second determination unit, configured to determine grayscale change characteristics according to the grayscale distribution data; An evaluation unit, configured to evaluate the splicing accuracy of the image splicing method based on the grayscale change characteristics and the minimum interval between the multiple identifiers; 8. An evaluation system for the image stitching accuracy of biological tissue imaging, characterized in that, The evaluation system includes a movable stage, a calibration object, and a computing device; The computing device receives imaging data obtained by an imaging device imaging a calibration object placed on the movable stage, and the imaging device can image biological tissues; The computing device includes a processor and a memory for storing instructions executable by the processor. Wherein, when the instructions executable by the processor are run by the processor, the processor is caused to execute the method for evaluating the image splicing accuracy of biological tissue imaging according to any one of claims 1 to 6; 9. A computer-readable storage medium, characterized in that, When the instructions in the computer-readable storage medium are executed by the processor of the computing device, the computing device is enabled to execute the method for evaluating the image splicing accuracy of biological tissue imaging according to any one of claims 1 to 6; 10. A computer program product, comprising computer-executable instructions, characterized in that, The computer-executable instructions, when executed by at least one processor, implement the method for evaluating the image splicing accuracy of biological tissue imaging according to any one of claims 1 to 6.

Images