Fluorescent immunoimage processing method and device based on hybrid processing

Abstract

The application relates to the technical field of image processing, and discloses a fluorescence immunization image processing method and device based on mixed processing, which comprises the following steps: performing image processing on a fluorescence immunization image to obtain a first processing image corresponding to the fluorescence immunization image and a background image; determining a target image according to the first processing image and the background image; analyzing brightness information of the background image; performing a region segmentation operation on the target image according to the brightness information to obtain a recognition region set; determining a fluorescence point recognition strategy of the recognition region set; and recognizing the recognition region set according to the fluorescence point recognition strategy to obtain fluorescence point information. It can be seen that, by implementing the application, the fluorescence immunization image can be recognized by a mixed recognition strategy combining a brightness threshold algorithm and an artificial intelligence recognition strategy, the recognition efficiency and accuracy are improved, the required calculation capacity for fluorescence point recognition is reduced, the hardware cost is reduced, the image is pretreated, and the anti-noise capability is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image processing, in particular to a fluorescence immunization image processing method and device based on hybrid processing. BACKGROUND

[0002] With the progress of life science and medical diagnosis technology, fluorescence immunization imaging method has shown important value in pathogen detection, tumor positioning and cell mechanism research due to its high specificity and visualization advantage. However, the practical application of this technology is still limited by two key bottlenecks, one of which is the strict requirement of the imaging system for signal-to-noise ratio, and the other is the high requirement for image algorithm.

[0003] In the prior art, the signal-to-noise ratio can be improved by improving the optical system, hardware shielding filter and the like; however, the existing image processing algorithm still has significant defects, among which, although the deep learning algorithm based on artificial intelligence can extract features under low signal-to-noise ratio conditions, it needs to rely on high-performance processors and long-time model training, resulting in high equipment cost and long development cycle, and the traditional brightness threshold cutting method (such as OpenCV algorithm) has lower hardware requirements, but is easily affected by non-uniform background noise, and is prone to misjudgment of core area as signal and misjudgment of edge signal as noise, which seriously affects the detection accuracy. Therefore, it is particularly important to propose a technical solution capable of improving the processing accuracy of fluorescence immunization images and having high noise immunity and low hardware cost. SUMMARY

[0004] The present application provides a fluorescence immunization image processing method and device based on hybrid processing, which can improve the processing accuracy of fluorescence immunization images and has high noise immunity and low hardware cost.

[0005] In order to solve the above technical problems, the present application discloses a fluorescence immunization image processing method based on hybrid processing, which comprises:

[0006] The fluorescence immunization image is subjected to image processing to obtain a processing image set corresponding to the fluorescence immunization image, wherein the processing image set comprises a first processing image and a background image;

[0007] According to the first processing image and the background image, a target image corresponding to the fluorescence immunization image is determined;

[0008] The brightness information corresponding to the background image is analyzed, and a region segmentation operation is performed on the target image according to the brightness information to obtain a recognition region set, wherein the recognition region set comprises a first recognition region, a second recognition region and a third recognition region;

[0009] determine a fluorescent spot identification strategy corresponding to the set of identification regions, and identify the set of identification regions according to the fluorescent spot identification strategy to obtain fluorescent spot information of the fluorescent immunoimage, the fluorescent spot identification strategy including a first strategy corresponding to the first identification region, a mixed identification strategy corresponding to the second identification region, and a second strategy corresponding to the third identification region.

[0010] As an optional implementation, in the first aspect of the present application, the identification of the set of identification regions according to the fluorescent spot identification strategy to obtain the fluorescent spot information of the fluorescent immunoimage includes:

[0011] identifying the first identification region according to the first strategy to obtain first fluorescent spot information corresponding to the first identification region, the first strategy including a brightness threshold cutting identification strategy;

[0012] identifying the second identification region according to the mixed identification strategy to obtain second fluorescent spot information corresponding to the second identification region;

[0013] identifying the third identification region according to the second strategy to obtain third fluorescent spot information corresponding to the third identification region, the second strategy including an artificial intelligence identification strategy;

[0014] determining the fluorescent spot information of the fluorescent immunoimage according to the first fluorescent spot information, the second fluorescent spot information, and the third fluorescent spot information.

[0015] As an optional implementation, in the first aspect of the present application, the identification of the second identification region according to the mixed identification strategy to obtain the second fluorescent spot information corresponding to the second identification region includes:

[0016] identifying the second identification region according to the brightness threshold cutting identification strategy to obtain first sub-fluorescent spot information corresponding to the second identification region;

[0017] array scanning the second identification region according to the artificial intelligence identification strategy to obtain second sub-fluorescent spot information corresponding to the second identification region;

[0018] determining the second fluorescent spot information corresponding to the second identification region according to the first sub-fluorescent spot information and the second sub-fluorescent spot information.

[0019] As an optional implementation, in the first aspect of the present application, the analysis of the brightness information corresponding to the background image and the execution of the region segmentation operation on the target image according to the brightness information to obtain the set of identification regions includes:

[0020] determine a fluorescence distribution region corresponding to the background image, and analyze brightness information of the fluorescence distribution region by a preset tool to obtain a brightness distribution curve of the background image;

[0021] determine a set of brightness change marker points in the brightness distribution curve, and determine a set of recognition regions of the target image according to the set of brightness change marker points, the set of brightness change marker points including a first brightness marker point and a second brightness marker point.

[0022] As an optional implementation, in the first aspect of the present application, the determining of the set of recognition regions of the target image according to the set of brightness change marker points includes:

[0023] determine a center point of the target image, and determine a first distance corresponding to the first brightness marker point and a second distance corresponding to the second brightness marker point according to the center point of the image and the set of brightness change marker points, the second distance being greater than the first distance;

[0024] determine a circular region, a first annular region and a second annular region with the center point of the image as the center in the target image according to the center point of the image, the first distance and the second distance, and determine the circular region as the first recognition region, the first annular region as the second recognition region and the second annular region as the third recognition region.

[0025] As an optional implementation, in the first aspect of the present application, the image processing of the fluorescence immunoassay image to obtain a set of processing images corresponding to the fluorescence immunoassay image includes:

[0026] performing pixel conversion processing on the fluorescence immunoassay image to obtain a conversion image corresponding to the fluorescence immunoassay image;

[0027] adjust the contrast of the conversion image according to a preset contrast adjustment rule to obtain a first processing image;

[0028] processing the fluorescence immunoassay image by a Gaussian blur algorithm to obtain a background image corresponding to the fluorescence immunoassay image.

[0029] As an optional implementation, in the first aspect of the present application, the pixel conversion processing on the fluorescence immunoassay image to obtain a conversion image corresponding to the fluorescence immunoassay image includes:

[0030] performing grayscale processing on the fluorescence immunoassay image to obtain a grayscale image corresponding to the fluorescence immunoassay image;

[0031] performing black-and-white conversion processing on the grayscale image to obtain a conversion image corresponding to the fluorescence immunoassay image;

[0032] And, after the first processing image and the background image are determined according to the first processing image and the background image, the method further comprises:

[0033] Based on a preset imaging system, the target image is deconvoluted to obtain a processed target image.

[0034] The second aspect of the application discloses a fluorescence immunoassay image processing device based on mixed processing, which comprises:

[0035] The first processing module is used for image processing of the fluorescence immunoassay image to obtain a set of processing images corresponding to the fluorescence immunoassay image, and the set of processing images comprises a first processing image and a background image.

[0036] The first determination module is used for determining a target image corresponding to the fluorescence immunoassay image according to the first processing image and the background image.

[0037] The analysis module is used for analyzing the brightness information corresponding to the background image, and performing a region segmentation operation on the target image according to the brightness information to obtain a set of recognition regions, wherein the set of recognition regions comprises a first recognition region, a second recognition region and a third recognition region.

[0038] The second determination module is used for determining a fluorescence point recognition strategy corresponding to the set of recognition regions, and identifying the set of recognition regions according to the fluorescence point recognition strategy to obtain fluorescence point information of the fluorescence immunoassay image, wherein the fluorescence point recognition strategy comprises a first strategy corresponding to the first recognition region, a mixed recognition strategy corresponding to the second recognition region and a second strategy corresponding to the third recognition region.

[0039] As an optional implementation, in the second aspect of the application, the second determination module identifies the set of recognition regions according to the fluorescence point recognition strategy to obtain the fluorescence point information of the fluorescence immunoassay image, and the manner comprises:

[0040] The first recognition region is identified according to the first strategy to obtain first fluorescence point information corresponding to the first recognition region, and the first strategy comprises a brightness threshold cutting recognition strategy.

[0041] The second recognition region is identified according to the mixed recognition strategy to obtain second fluorescence point information corresponding to the second recognition region.

[0042] The third recognition region is identified according to the second strategy to obtain third fluorescence point information corresponding to the third recognition region, and the second strategy comprises an artificial intelligence recognition strategy.

[0043] According to the first fluorescent point information, the second fluorescent point information and the third fluorescent point information, the fluorescent point information of the fluorescent immune image is determined.

[0044] As an optional implementation, in the second aspect of the present application, the second determination module determines the second recognition region according to the mixed recognition strategy, and the manner of obtaining the second fluorescent point information corresponding to the second recognition region specifically includes:

[0045] According to the brightness threshold cutting recognition strategy, the second recognition region is recognized to obtain the first sub-fluorescent point information corresponding to the second recognition region;

[0046] According to the artificial intelligence recognition strategy, the second recognition region is array scanned to obtain the second sub-fluorescent point information corresponding to the second recognition region;

[0047] According to the first sub-fluorescent point information and the second sub-fluorescent point information, the second fluorescent point information corresponding to the second recognition region is determined.

[0048] As an optional implementation, in the second aspect of the present application, the analysis module analyzes the brightness information corresponding to the background image, and according to the brightness information, performs region segmentation operation on the target image to obtain a recognition region set, and the manner of specifically includes:

[0049] The fluorescent distribution region corresponding to the background image is determined, and the brightness information of the fluorescent distribution region is analyzed by a preset tool to obtain the brightness distribution curve of the background image;

[0050] A set of brightness change marker points in the brightness distribution curve is determined, and according to the set of brightness change marker points, a set of recognition regions of the target image is determined, and the set of brightness change marker points includes a first brightness marker point and a second brightness marker point.

[0051] As an optional implementation, in the second aspect of the present application, the analysis module determines the set of recognition regions of the target image according to the set of brightness change marker points, and the manner specifically includes:

[0052] The image center point of the target image is determined, and according to the image center point and the set of brightness change marker points, a first distance corresponding to the first brightness marker point and a second distance corresponding to the second brightness marker point are determined, and the second distance is greater than the first distance;

[0053] According to the image center point, the first distance, and the second distance, a circular region, a first annular region, and a second annular region are determined in the target image with the image center point as the center, and the circular region is determined as the first identification region, the first annular region is determined as the second identification region, and the second annular region is determined as the third identification region.

[0054] As an optional implementation, in the second aspect of the present application, the manner in which the first processing module performs image processing on the fluorescent immunoimage to obtain a processing image set corresponding to the fluorescent immunoimage specifically includes:

[0055] performing pixel conversion processing on the fluorescent immunoimage to obtain a conversion image corresponding to the fluorescent immunoimage;

[0056] adjusting the contrast of the conversion image according to a preset contrast adjustment rule to obtain a first processing image;

[0057] performing processing on the fluorescent immunoimage through a Gaussian blur algorithm to obtain a background image corresponding to the fluorescent immunoimage.

[0058] As an optional implementation, in the second aspect of the present application, the manner in which the first processing module performs pixel conversion processing on the fluorescent immunoimage to obtain a conversion image corresponding to the fluorescent immunoimage specifically includes:

[0059] performing grayscale processing on the fluorescent immunoimage to obtain a grayscale image corresponding to the fluorescent immunoimage;

[0060] performing black-and-white conversion processing on the grayscale image to obtain a conversion image corresponding to the fluorescent immunoimage;

[0061] and the device further includes:

[0062] a second processing module configured to, after the first determination module determines a target image corresponding to the fluorescent immunoimage according to the first processing image and the background image, perform deconvolution processing on the target image based on a preset imaging system to obtain a processed target image.

[0063] A third aspect of the present application discloses another fluorescent immunoimage processing device based on mixed processing, which comprises:

[0064] a memory storing executable program codes;

[0065] a processor coupled with the memory;

[0066] The processor invokes the executable program code stored in the memory to execute the mixed processing based fluorescent immunoimage processing method disclosed in the first aspect of the present application.

[0067] The fourth aspect of the present application discloses a computer storage medium, which stores computer instructions, and when the computer instructions are invoked, the mixed processing based fluorescent immunoimage processing method disclosed in the first aspect of the present application is executed.

[0068] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0069] In the embodiments of the present application, the fluorescent immunoimage is subjected to image processing to obtain a processing image set corresponding to the fluorescent immunoimage, the processing image set including a first processing image and a background image, a target image corresponding to the fluorescent immunoimage is determined according to the first processing image and the background image, the brightness information corresponding to the background image is analyzed, and a region segmentation operation is performed on the target image according to the brightness information to obtain a recognition region set, a fluorescent point recognition strategy corresponding to the recognition region set is determined, and the recognition region set is recognized according to the fluorescent point recognition strategy to obtain the fluorescent point information of the fluorescent immunoimage. It can be seen that, by implementing the present application, the fluorescent immunoimage can be subjected to partition recognition based on the mixed recognition strategy combining the brightness threshold algorithm and the artificial intelligence recognition strategy, the recognition efficiency and accuracy are improved, the computing capability required for fluorescent point recognition is reduced, the hardware cost is reduced, the image is preprocessed, and the noise resistance is improved. BRIEF DESCRIPTION OF DRAWINGS

[0070] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative effort.

[0071] Figure 1 is a flowchart of a mixed processing based fluorescent immunoimage processing method disclosed by the embodiments of the present application;

[0072] Figure 2 is a flowchart of another mixed processing based fluorescent immunoimage processing method disclosed by the embodiments of the present application;

[0073] Figure 3 is a schematic diagram of a brightness distribution curve disclosed by the embodiments of the present application;

[0074] Figure 4 is a structural schematic diagram of a mixed processing based fluorescent immunoimage processing device disclosed by the embodiments of the present application;

[0075] Figure 5 is another structure schematic diagram of a fluorescence immunoassay image processing device based on hybrid processing disclosed by the embodiment of the present application.

[0076] Figure 6 is another structure schematic diagram of a fluorescence immunoassay image processing device based on hybrid processing disclosed by the embodiment of the present application. DETAILED DESCRIPTION

[0077] In order for those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0078] The terms "first", "second", and the like in the specification and claims of the present application and the above-mentioned drawings are used to distinguish different objects, not to describe a specific order. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, device, product, or end including a series of steps or units is not limited to the listed steps or units, but can optionally include steps or units not listed, or can optionally include other steps or units inherent to the process, method, product, or end.

[0079] Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the present application. The appearance of the phrase in various places in the specification does not necessarily all refer to the same embodiment, nor is it necessarily mutually exclusive of other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0080] The present application discloses a fluorescence immunoassay image processing method and device based on hybrid processing, which can perform partition recognition on fluorescence immunoassay images based on a hybrid recognition strategy combining a brightness threshold algorithm with an artificial intelligence recognition strategy, improve recognition efficiency and accuracy, reduce the computing power required for fluorescence point recognition, reduce hardware costs, and improve noise resistance by preprocessing the images. The following will be described in detail.

[0081] Embodiment one

[0082] Please refer to Figure 1 , Figure 1 is a flowchart of a fluorescence immunoassay image processing method based on hybrid processing disclosed by the embodiment of the present application. Among them,Figure 1 The described hybrid processing-based fluorescence immunoimage processing method can be applied to a hybrid processing-based fluorescence immunoimage processing device, wherein the hybrid processing-based fluorescence immunoimage processing device can include an intelligent server or an intelligent platform for processing a fluorescence immunoimage, and the intelligent server includes a local server or a cloud server, and embodiments of the present application are not limited. Figure 1 As shown, the hybrid processing-based fluorescence immunoimage processing method can include the following operations:

[0083] 101. Image processing is performed on the fluorescence immunoimage to obtain a set of processing images corresponding to the fluorescence immunoimage, and the set of processing images includes a first processing image and a background image.

[0084] In embodiments of the present application, the image processing performed on the fluorescence immunoimage can include one or more of image grayscale processing, pixel conversion processing, contrast adjustment processing, and background extraction processing, and the set of processing images obtained after processing can include the first processing image and the background image, wherein the first processing image can include an image obtained after pixel conversion processing is performed on the fluorescence immunoimage to improve the signal-to-noise ratio of the fluorescence immunoimage for the first time, and the present application is not limited.

[0085] 102. A target image corresponding to the fluorescence immunoimage is determined based on the first processing image and the background image.

[0086] In embodiments of the present application, the first processing image can be subtracted from the background image based on a Boolean algorithm to obtain the target image corresponding to the fluorescence immunoimage to improve the signal-to-noise ratio of the fluorescence immunoimage for the second time, and the present application is not limited.

[0087] 103. Brightness information corresponding to the background image is analyzed, and a region segmentation operation is performed on the target image based on the brightness information to obtain a set of recognition regions.

[0088] In embodiments of the present application, the region segmentation operation on the target image is based on the signal-to-noise ratio, and since the signal-to-noise ratio is highly correlated with the brightness of the excitation light and the background, the region segmentation operation can be performed on the target image by analyzing the brightness information of the background image, and the region segmentation operation can be performed on the target image based on the brightness information to obtain a set of recognition regions, wherein the set of recognition regions can include a first recognition region, a second recognition region, and a third recognition region, and specifically, the first signal-to-noise ratio of the first recognition region, the second signal-to-noise ratio of the second recognition region, and the third signal-to-noise ratio of the third recognition region obtained by performing the region segmentation operation on the target image based on the signal-to-noise ratio of the target image decrease in turn, and the present application is not limited.

[0089] 104, determine the fluorescent point recognition strategy corresponding to the recognition region set, and identify the recognition region set according to the fluorescent point recognition strategy to obtain the fluorescent point information of the fluorescent immunoimage.

[0090] In the embodiment of the application, optionally, the signal-to-noise ratio of each recognition region in the recognition region set can be used to determine the fluorescent point recognition strategy of the recognition region, the fluorescent point recognition strategy includes a first strategy corresponding to a first recognition region, a mixed recognition strategy corresponding to a second recognition region, and a second strategy corresponding to a third recognition region, wherein the first signal-to-noise ratio of the first recognition region, the second signal-to-noise ratio of the second recognition region, and the third signal-to-noise ratio of the third recognition region decrease in turn, the first strategy can include a brightness threshold cutting recognition strategy, and the second strategy can include an artificial intelligence recognition strategy. Specifically, the first signal-to-noise ratio of the first recognition region is good, and a brightness threshold algorithm can be used to identify the fluorescent points in the region, the third signal-to-noise ratio of the third recognition region is poor, and an artificial intelligence recognition strategy can be used to identify the fluorescent points in the region, and the second signal-to-noise ratio of the second recognition region is between the first signal-to-noise ratio and the second signal-to-noise ratio. A mixed recognition strategy combining the brightness threshold algorithm and the artificial intelligence recognition strategy can be used to identify the fluorescent points in the region, and the application does not limit it.

[0091] In the embodiment of the application, compared with identifying the entire fluorescent immunoimage by using an artificial intelligence recognition strategy, the area of the region that needs to be identified in the third recognition region is only 19% of the normal area, the area of the region that needs to be mixed identified in the second recognition region is 17%, and the total area that needs to be involved is 38%, which greatly reduces the required computing power, and the brightness threshold cutting strategy only identifies the region with strong signal-to-noise ratio, thereby ensuring the identification efficiency and accuracy.

[0092] It can be seen that the implementation Figure 1 The described fluorescent immunoimage processing method based on mixed processing can perform image processing on the fluorescent immunoimage to obtain a processing image set corresponding to the fluorescent immunoimage, the processing image set includes a first processing image and a background image, a target image corresponding to the fluorescent immunoimage is determined according to the first processing image and the background image, the brightness information corresponding to the background image is analyzed, and a region segmentation operation is performed on the target image according to the brightness information to obtain a recognition region set, a fluorescent point recognition strategy corresponding to the recognition region set is determined, and the recognition region set is identified according to the fluorescent point recognition strategy to obtain the fluorescent point information of the fluorescent immunoimage. The mixed recognition strategy combining the brightness threshold algorithm and the artificial intelligence recognition strategy can be used to identify the fluorescent immunoimage, which improves the identification efficiency and accuracy, reduces the computing power required for fluorescent point identification, reduces the hardware cost, and improves the anti-noise ability by preprocessing the image.

[0093] In an optional embodiment, the identifying the set of identification regions according to the fluorescent point identification strategy to obtain the fluorescent point information of the fluorescent immunoimage can include the following operations:

[0094] The first identification region is identified according to the first strategy to obtain the first fluorescent point information corresponding to the first identification region, and the first strategy includes a brightness threshold cutting identification strategy.

[0095] The second identification region is identified according to the mixed identification strategy to obtain the second fluorescent point information corresponding to the second identification region.

[0096] The third identification region is identified according to the second strategy to obtain the third fluorescent point information corresponding to the third identification region, and the second strategy includes an artificial intelligence identification strategy.

[0097] The fluorescent point information of the fluorescent immunoimage is determined according to the first fluorescent point information, the second fluorescent point information and the third fluorescent point information.

[0098] In the optional embodiment, optionally, the fluorescent point identification strategy includes the first strategy corresponding to the first identification region, the mixed identification strategy corresponding to the second identification region and the second strategy corresponding to the third identification region, wherein the first signal-to-noise ratio of the first identification region, the second signal-to-noise ratio of the second identification region and the third signal-to-noise ratio of the third identification region decrease in turn, the first strategy can include the brightness threshold cutting identification strategy, the second strategy can include the artificial intelligence identification strategy, specifically, the first signal-to-noise ratio of the first identification region is good, the brightness threshold algorithm can be used to identify the fluorescent points in the region, the third signal-to-noise ratio of the third identification region is poor, the artificial intelligence identification strategy can be used to identify the fluorescent points in the region, and the second signal-to-noise ratio of the second identification region is between the first signal-to-noise ratio and the second signal-to-noise ratio, the mixed identification strategy combining the brightness threshold algorithm and the artificial intelligence identification strategy can be used to identify the fluorescent points in the region, wherein the mixed identification strategy can include the mixed identification strategy mainly using the brightness threshold cutting identification strategy and secondarily using the artificial intelligence identification strategy.

[0099] It can be seen that the optional embodiment can identify the first identification area according to the brightness threshold cutting identification strategy, obtain the first fluorescent point information corresponding to the first identification area, identify the second identification area according to the hybrid identification strategy, obtain the second fluorescent point information corresponding to the second identification area, identify the third identification area according to the artificial intelligence identification strategy, obtain the third fluorescent point information corresponding to the third identification area, and determine the fluorescent point information of the fluorescent immune image according to the first fluorescent point information, the second fluorescent point information and the third fluorescent point information. The fluorescent immune image can be partitioned and identified based on the hybrid identification strategy combining the brightness threshold algorithm and the artificial intelligence identification strategy, the identification efficiency and accuracy are improved, the computing power required for fluorescent point identification is reduced, and the hardware cost is reduced.

[0100] In another optional embodiment, identifying the second identification area according to the hybrid identification strategy to obtain the second fluorescent point information corresponding to the second identification area can include the following operations:

[0101] Identifying the second identification area according to the brightness threshold cutting identification strategy to obtain the first sub-fluorescent point information corresponding to the second identification area;

[0102] Array scanning the second identification area according to the artificial intelligence identification strategy to obtain the second sub-fluorescent point information corresponding to the second identification area;

[0103] Determining the second fluorescent point information corresponding to the second identification area according to the first sub-fluorescent point information and the second sub-fluorescent point information.

[0104] In the optional embodiment, optionally, the second identification area can be identified according to the brightness threshold cutting identification strategy to obtain the first sub-fluorescent point information corresponding to the second identification area. Since the second signal-to-noise ratio of the second identification area changes from high to low and is in an excessive area, the brightness of part of the fluorescent points is lower than the background brightness of the intermediate area, and the fluorescent points cannot be identified by the brightness threshold, so the result of the first sub-fluorescent point information obtained is usually smaller than the actual result, which is not limited in the embodiment.

[0105] In the optional embodiment, optionally, the second identification area can be array scanned according to the artificial intelligence identification strategy to obtain the second sub-fluorescent point information corresponding to the second identification area. After array scanning the second identification area according to the artificial intelligence identification strategy, the fluorescent points with low fluorescent brightness and difficult to identify and count in the first sub-fluorescent point information can be supplemented and identified, the identification efficiency and accuracy are improved, which is not limited in the embodiment.

[0106] It can be seen that the optional embodiment can identify the second identification area according to the brightness threshold cutting identification strategy, obtain the first sub-fluorescent point information corresponding to the second identification area, array scan the second identification area according to the artificial intelligence identification strategy, obtain the second sub-fluorescent point information corresponding to the second identification area, and determine the second fluorescent point information corresponding to the second identification area according to the first sub-fluorescent point information and the second sub-fluorescent point information. In the transition area with gradually changing signal-to-noise ratio, identification is performed based on the brightness threshold cutting identification strategy, and supplementary confirmation is performed based on the artificial intelligence identification strategy, thereby improving the fluorescent point identification efficiency and accuracy of the area, reducing the required calculation capability for fluorescent point identification, and reducing the hardware cost.

[0107] In still another optional embodiment, the image processing of the fluorescent immunoimage to obtain a set of processing images corresponding to the fluorescent immunoimage can include the following operations:

[0108] The pixel conversion processing of the fluorescent immunoimage to obtain a conversion image corresponding to the fluorescent immunoimage;

[0109] Adjusting the contrast of the conversion image according to a preset contrast adjustment rule to obtain a first processing image;

[0110] Processing the fluorescent immunoimage by a Gaussian blur algorithm to obtain a background image corresponding to the fluorescent immunoimage.

[0111] In the optional embodiment, the pixel conversion processing of the fluorescent immunoimage to obtain a conversion image corresponding to the fluorescent immunoimage can be performed, wherein the pixel conversion processing can include grayscale processing and black-and-white conversion processing. Then, the contrast of the conversion image is adjusted according to a preset contrast adjustment rule to obtain a first processing image. Specifically, the contrast of the conversion image can be set to 150%, so that the black in the conversion image is close to infinite black, and the white is limited, thereby improving the signal-to-noise ratio of the image for the first time. The present embodiment is not limited in this regard.

[0112] In the optional embodiment, the fluorescent immunoimage can be processed by a Gaussian blur algorithm to obtain a background image corresponding to the fluorescent immunoimage. The present embodiment is not limited in this regard.

[0113] It can be seen that the optional embodiment can perform pixel conversion processing of the fluorescent immunoimage to obtain a conversion image corresponding to the fluorescent immunoimage, adjust the contrast of the conversion image according to a preset contrast adjustment rule to obtain a first processing image, and process the fluorescent immunoimage by a Gaussian blur algorithm to obtain a background image corresponding to the fluorescent immunoimage. The fluorescent immunoimage can be preprocessed to reduce the influence of noise and improve the accuracy of region division and subsequent fluorescent point identification accuracy.

[0114] In yet another optional embodiment, the pixel conversion processing of the fluorescent immunization image to obtain the conversion image corresponding to the fluorescent immunization image can include the following operations:

[0115] The gray processing of the fluorescent immunization image to obtain the gray image corresponding to the fluorescent immunization image;

[0116] The black and white conversion processing of the gray image to obtain the conversion image corresponding to the fluorescent immunization image;

[0117] And, after determining the target image corresponding to the fluorescent immunization image according to the first processed image and the background image, the method further includes:

[0118] Based on the preset imaging system, the target image is deconvolved to obtain the processed target image.

[0119] In the optional embodiment, the gray processing of the fluorescent immunization image to obtain the gray image corresponding to the fluorescent immunization image, i.e. the conversion of the colored fluorescent immunization image to the gray image, or the direct use of the black and white camera to obtain the image, is optional and is not limited in the embodiment.

[0120] In the optional embodiment, after obtaining the gray image, the black and white conversion processing of the gray image to obtain the conversion image corresponding to the fluorescent immunization image is optional, i.e. the conversion of the fluorescent signal in the gray image from white to black and the conversion of the corresponding background from black to white to obtain the conversion image corresponding to the fluorescent immunization image, which is not limited in the embodiment.

[0121] In the optional embodiment, after determining the target image corresponding to the fluorescent immunization image according to the first processed image and the background image, the deconvolution processing of the target image based on the preset imaging system to obtain the processed target image is optional, i.e. the deconvolution processing of the target image by the preset imaging system (PSF) to restore the distortion of the outside fluorescent point caused by imaging, effectively reducing the recognition difficulties such as adhesion and overlap caused by distortion to restore the true shape of the image, which is not limited in the embodiment.

[0122] It can be seen that the implementation of the optional embodiment can perform the gray processing of the fluorescent immunization image to obtain the gray image corresponding to the fluorescent immunization image, the black and white conversion processing of the gray image to obtain the conversion image corresponding to the fluorescent immunization image, which can improve the signal-to-noise ratio of the image, reduce the image noise, improve the accuracy and reliability of the subsequent region division and fluorescent point recognition, and perform the deconvolution processing of the target image based on the preset imaging system to obtain the processed target image, restore the distortion of the outside fluorescent point caused by imaging, effectively reduce the recognition difficulties such as adhesion and overlap caused by distortion to restore the true shape of the image.

[0123] Embodiment two

[0124] Please refer to Figure 2 , Figure 2 is a flowchart of a fluorescence immunoassay image processing method based on hybrid processing disclosed by the embodiments of the present application. Wherein, Figure 2 The fluorescence immunoassay image processing method based on hybrid processing described can be applied to a fluorescence immunoassay image processing device based on hybrid processing, wherein the fluorescence immunoassay image processing device based on hybrid processing can include an intelligent server or an intelligent platform for processing fluorescence immunoassay images, and the intelligent server includes a local server or a cloud server, which is not limited by the embodiments of the present application. As Figure 2 The fluorescence immunoassay image processing method based on hybrid processing can include the following operations:

[0125] 201, image processing of the fluorescence immunoassay image is performed to obtain a set of processing images corresponding to the fluorescence immunoassay image, and the set of processing images includes a first processing image and a background image.

[0126] 202, according to the first processing image and the background image, a target image corresponding to the fluorescence immunoassay image is determined.

[0127] 203, a fluorescence distribution region corresponding to the background image is determined, and the brightness information of the fluorescence distribution region is analyzed by a preset tool to obtain a brightness distribution curve of the background image.

[0128] In the embodiments of the present application, optionally, the basis for the region segmentation operation of the target image is the signal-to-noise ratio, which is highly related to the brightness of the excitation light and the background, and thus can be obtained by brightness analysis of the background image. Specifically: the fluorescence distribution region corresponding to the background image can be determined, specifically, the background image can be imported into a target software (such as imageJ), the fluorescence distribution region in the background image is framed based on the framing function, the brightness information of the fluorescence distribution region is analyzed by a preset tool to obtain a brightness distribution curve of the background image, wherein the plotprofile function under the Analyze function module can be used to obtain the brightness distribution curve of different regions, at this time the brightness has been converted into a gray value, which is not limited by the present application.

[0129] 204, a set of brightness change marker points in the brightness distribution curve is determined, and a set of recognition regions of the target image is determined according to the set of brightness change marker points.

[0130] In the embodiments of the present application, optionally, as Figure 3 shown, Figure 3 is a schematic diagram of a brightness distribution curve disclosed by the embodiments of the present application, wherein a set of brightness change marker points in the brightness distribution curve can be determined, and the set of brightness change marker points includes a first brightness marker point and a second brightness marker point, that is Figure 3Points (1) and (2) in the image can be used to determine the set of recognition regions of the target image based on the set of marker points with brightness changes. Specifically, Figure 3 The area to the right of the midpoint (2) has the highest and most stable brightness, i.e. the highest and most stable signal-to-noise ratio, representing the first recognition area. The area from point (2) to point (1) has a brightness that decreases from high to low, i.e. the signal-to-noise ratio that decreases from high to low, representing the transition area, i.e. the second recognition area. The area from the origin to point (1) has the lowest brightness, i.e. the lowest signal-to-noise ratio, representing the third recognition area. This invention does not limit this area.

[0131] 205. Determine the fluorescence dot recognition strategy corresponding to the set of recognition regions, and recognize the set of recognition regions according to the fluorescence dot recognition strategy to obtain the fluorescence dot information of the fluorescence immunoassay image.

[0132] In this embodiment of the invention, for other descriptions of steps 201, 202 and 205, please refer to the detailed description of steps 101-103 in Embodiment 1 of the invention. These descriptions will not be repeated in this embodiment of the invention.

[0133] It is evident that implementation Figure 2 The described hybrid processing-based fluorescence immunoassay image processing method can process fluorescence immunoassay images to obtain a set of processed images corresponding to the fluorescence immunoassay image. The set of processed images includes a first processed image and a background image. Based on the first processed image and the background image, the target image corresponding to the fluorescence immunoassay image is determined, the fluorescence distribution region corresponding to the background image is determined, and the brightness information of the fluorescence distribution region is analyzed by a preset tool to obtain the brightness distribution curve of the background image. The set of brightness change markers in the brightness distribution curve is determined, and the set of recognition regions of the target image is determined based on the set of brightness change markers. The method can use the brightness change of the background image to replace the analysis of the signal-to-noise ratio of the target image, improving the convenience and accuracy of region segmentation of the target image. The method determines the fluorescence point recognition strategy corresponding to the set of recognition regions, and recognizes the set of recognition regions according to the fluorescence point recognition strategy to obtain the fluorescence point information of the fluorescence immunoassay image. The method can perform partition recognition of fluorescence immunoassay images based on a hybrid recognition strategy that combines a brightness threshold algorithm and an artificial intelligence recognition strategy, improving recognition efficiency and accuracy while reducing the computing power required for fluorescence point recognition, reducing hardware costs, and improving noise resistance by preprocessing the image.

[0134] In an optional embodiment, determining the set of recognition regions for the target image based on the set of brightness change marker points may include the following operations:

[0135] Determine the image center point of the target image, and based on the image center point and the set of brightness change marker points, determine the first distance corresponding to the first brightness marker point and the second distance corresponding to the second brightness marker point, wherein the second distance is greater than the first distance;

[0136] According to the image center point, the first distance, and the second distance, a circular region, a first annular region, and a second annular region are determined in the target image, with the image center point as the center, the circular region is determined as the first identification region, the first annular region is determined as the second identification region, and the second annular region is determined as the third identification region.

[0137] In the optional embodiment, the distribution region of the fluorescent points of the fluorescent immunoimage can be approximately a circle, and the processed background image is also a circle, so that the image center point of the target image can be determined, the image center point is the center of the circular background image, the first distance corresponding to the first luminance marker point and the second distance corresponding to the second luminance marker point can be determined according to the image center point and the set of luminance change marker points, the second distance is greater than the first distance, and specifically, as shown in Figure 3 , the origin in Figure 3 may represent an edge point of the circular background image, that is, Figure 3 The distance from the origin in Figure 3 to the image center point is the radius R, and the first distance can represent the distance from the point (2) in Figure 3 to the image center point, and specifically can be 0.8R, and the second distance can represent the distance from the point (1) in Figure 3 to the image center point, and specifically can be 0.9R, which is not limited in the embodiment.

[0138] In the optional embodiment, the distribution region of the fluorescent points of the fluorescent immunoimage can be approximately a circle, and the processed background image is also a circle, so that the image center point of the target image can be determined, the image center point is the center of the circular background image, the first distance corresponding to the first luminance marker point and the second distance corresponding to the second luminance marker point can be determined according to the image center point and the set of luminance change marker points, the second distance is greater than the first distance, and specifically, as shown in

[0139] It can be seen that the optional embodiment can determine the image center point of the target image, and determine the first distance corresponding to the first brightness marking point and the second distance corresponding to the second brightness marking point according to the image center point and the set of brightness change marking points, the second distance is greater than the first distance, and the circular region, the first annular region and the second annular region with the image center point as the center are determined in the target image according to the image center point, the first distance and the second distance, and the circular region is determined as the first identification region, the first annular region is determined as the second identification region, and the second annular region is determined as the third identification region, which can determine different regions in the target image based on the brightness change marking points of the background image, improve the convenience and accuracy of the region segmentation operation on the target image, and further improve the accuracy and efficiency of the partition identification of the fluorescent immune image.

[0140] Embodiment three

[0141] Please refer to Figure 4 , Figure 4 is a structure schematic diagram of a fluorescent immune image processing device based on hybrid processing disclosed by the embodiment of the present application. Wherein, Figure 4 The fluorescent immune image processing device based on hybrid processing described can include an intelligent server or an intelligent platform for processing a fluorescent immune image, and the intelligent server includes a local server or a cloud server, and the embodiment of the present application is not limited. As shown in Figure 4 The fluorescent immune image processing device based on hybrid processing can include:

[0142] The first processing module 301 is configured to perform image processing on the fluorescent immune image to obtain a set of processing images corresponding to the fluorescent immune image, and the set of processing images includes a first processing image and a background image.

[0143] The first determination module 302 is configured to determine a target image corresponding to the fluorescent immune image according to the first processing image and the background image.

[0144] The analysis module 303 is configured to analyze the brightness information corresponding to the background image, and perform a region segmentation operation on the target image according to the brightness information to obtain a set of identification regions, and the set of identification regions includes a first identification region, a second identification region and a third identification region.

[0145] The second determination module 304 is configured to determine a fluorescent point identification strategy corresponding to the set of identification regions, and identify the set of identification regions according to the fluorescent point identification strategy to obtain fluorescent point information of the fluorescent immune image, and the fluorescent point identification strategy includes a first strategy corresponding to the first identification region, a hybrid identification strategy corresponding to the second identification region and a second strategy corresponding to the third identification region.

[0146] It can be seen that the optional embodiment can determine the image center point of the target image, and determine the first distance corresponding to the first brightness marking point and the second distance corresponding to the second brightness marking point according to the image center point and the set of brightness change marking points, the second distance is greater than the first distance, and the circular region, the first annular region and the second annular region with the image center point as the center are determined in the target image according to the image center point, the first distance and the second distance, and the circular region is determined as the first identification region, the first annular region is determined as the second identification region, and the second annular region is determined as the third identification region, which can determine different regions in the target image based on the brightness change marking points of the background image, improve the convenience and accuracy of the region segmentation operation on the target image, and further improve the accuracy and efficiency of the partition identification of the fluorescent immune image. Figure 4The described fluorescence immunization image processing device based on hybrid processing can perform image processing on the fluorescence immunization image to obtain a processed image set corresponding to the fluorescence immunization image, the processed image set including a first processed image and a background image, determine a target image corresponding to the fluorescence immunization image according to the first processed image and the background image, analyze brightness information corresponding to the background image, and perform a region segmentation operation on the target image according to the brightness information to obtain a recognition region set, determine a fluorescence point recognition strategy corresponding to the recognition region set, and recognize the recognition region set according to the fluorescence point recognition strategy to obtain fluorescence point information of the fluorescence immunization image. The fluorescence immunization image can be partitioned and recognized based on a hybrid recognition strategy combining a brightness threshold algorithm and an artificial intelligence recognition strategy, the recognition efficiency and accuracy are improved, the required computing power for fluorescence point recognition is reduced, the hardware cost is reduced, and the image is preprocessed to improve the noise resistance.

[0147] In an optional embodiment, as shown in Figure 5 The second determination module 304 determines the fluorescence point information of the fluorescence immunization image according to the recognition of the recognition region set based on the fluorescence point recognition strategy. The specific manner includes:

[0148] According to the first strategy, the first recognition region is recognized to obtain first fluorescence point information corresponding to the first recognition region. The first strategy includes a brightness threshold cutting recognition strategy.

[0149] According to the hybrid recognition strategy, the second recognition region is recognized to obtain second fluorescence point information corresponding to the second recognition region.

[0150] According to the second strategy, the third recognition region is recognized to obtain third fluorescence point information corresponding to the third recognition region. The second strategy includes an artificial intelligence recognition strategy.

[0151] According to the first fluorescence point information, the second fluorescence point information, and the third fluorescence point information, the fluorescence point information of the fluorescence immunization image is determined.

[0152] It can be seen that the implementation Figure 5The described fluorescence immunization image processing device based on hybrid processing can identify the first identification area according to the brightness threshold cutting identification strategy, obtain the first fluorescence point information corresponding to the first identification area, identify the second identification area according to the hybrid identification strategy, obtain the second fluorescence point information corresponding to the second identification area, identify the third identification area according to the artificial intelligence identification strategy, and obtain the third fluorescence point information corresponding to the third identification area. According to the first fluorescence point information, the second fluorescence point information and the third fluorescence point information, the fluorescence point information of the fluorescence immunization image is determined, the fluorescence immunization image can be identified based on the hybrid identification strategy combining the brightness threshold algorithm and the artificial intelligence identification strategy, the identification efficiency and accuracy are improved, the computing power required for fluorescence point identification is reduced, and the hardware cost is reduced.

[0153] In another optional embodiment, as shown in Figure 5 The second determination module 304 identifies the second identification area according to the hybrid identification strategy, and the specific way of obtaining the second fluorescence point information corresponding to the second identification area includes:

[0154] According to the brightness threshold cutting identification strategy, the second identification area is identified to obtain the first sub-fluorescence point information corresponding to the second identification area;

[0155] According to the artificial intelligence identification strategy, the second identification area is array scanned to obtain the second sub-fluorescence point information corresponding to the second identification area;

[0156] According to the first sub-fluorescence point information and the second sub-fluorescence point information, the second fluorescence point information corresponding to the second identification area is determined.

[0157] It can be seen that the implementation Figure 5 The described fluorescence immunization image processing device based on hybrid processing can identify the second identification area according to the brightness threshold cutting identification strategy, obtain the first sub-fluorescence point information corresponding to the second identification area, array scan the second identification area according to the artificial intelligence identification strategy, obtain the second sub-fluorescence point information corresponding to the second identification area, and determine the second fluorescence point information corresponding to the second identification area according to the first sub-fluorescence point information and the second sub-fluorescence point information. In the transition region with gradually changing signal-to-noise ratio, identification is performed based on the brightness threshold cutting identification strategy, and supplementary confirmation is performed based on the artificial intelligence identification strategy, so as to improve the fluorescence point identification efficiency and accuracy of the region, while reducing the computing power required for fluorescence point identification and reducing the hardware cost.

[0158] In yet another optional embodiment, as shown in Figure 5 The analysis module 303 analyzes the brightness information corresponding to the background image, and the way of performing region segmentation operation on the target image according to the brightness information to obtain the identification area set includes:

[0159] determine a fluorescence distribution region corresponding to the background image, analyze luminance information of the fluorescence distribution region through a preset tool, and obtain a luminance distribution curve of the background image;

[0160] determine a set of luminance change marker points in the luminance distribution curve, and determine a set of recognition regions of the target image according to the set of luminance change marker points, the set of luminance change marker points including a first luminance marker point and a second luminance marker point.

[0161] It can be seen that in the implementation Figure 5 The fluorescence immunoassay image processing device based on hybrid processing described herein can perform image processing on a fluorescence immunoassay image to obtain a set of processed images corresponding to the fluorescence immunoassay image, the set of processed images including a first processed image and a background image. According to the first processed image and the background image, a target image corresponding to the fluorescence immunoassay image is determined, a fluorescence distribution region corresponding to the background image is determined, and the luminance information of the fluorescence distribution region is analyzed through a preset tool to obtain a luminance distribution curve of the background image. A set of recognition regions of the target image is determined according to a set of luminance change marker points in the luminance distribution curve, which can replace the analysis of the signal-to-noise ratio of the target image by the luminance change of the background image, improve the convenience and accuracy of the region segmentation operation on the target image, determine a fluorescence point recognition strategy corresponding to the set of recognition regions, and identify the set of recognition regions according to the fluorescence point recognition strategy to obtain fluorescence point information of the fluorescence immunoassay image. The hybrid recognition strategy based on the combination of the luminance threshold algorithm and the artificial intelligence recognition strategy can be used to identify the fluorescence immunoassay image, which improves the recognition efficiency and accuracy while reducing the computing power required for fluorescence point recognition, reduces the hardware cost, and improves the anti-noise capability through image preprocessing.

[0162] In yet another optional embodiment, as Figure 5 shown, the specific manner in which the analysis module 303 determines the set of recognition regions of the target image according to the set of luminance change marker points includes:

[0163] determining an image center point of the target image, and determining a first distance corresponding to the first luminance marker point and a second distance corresponding to the second luminance marker point according to the image center point and the set of luminance change marker points, the second distance being greater than the first distance;

[0164] determining a circular region, a first annular region and a second annular region with the image center point as the center in the target image according to the image center point, the first distance and the second distance, and determining the circular region as the first recognition region, the first annular region as the second recognition region, and the second annular region as the third recognition region.

[0165] It can be seen that in the implementation Figure 5The described fluorescence immunization image processing device based on mixed processing can determine the image center point of the target image, determine the first distance corresponding to the first brightness marker point and the second distance corresponding to the second brightness marker point according to the image center point and the set of brightness change marker points, the second distance is greater than the first distance, determine the circular region, the first annular region and the second annular region with the image center point as the center in the target image according to the image center point, the first distance and the second distance, and determine the circular region as the first identification region, the first annular region as the second identification region and the second annular region as the third identification region, which can determine different regions in the target image based on the brightness change marker points of the background image, improve the convenience and accuracy of the region segmentation operation on the target image, and further improve the accuracy and efficiency of the partition identification of the fluorescence immunization image.

[0166] In yet another optional embodiment, as shown in Figure 5 The specific manner in which the first processing module 301 performs image processing on the fluorescence immunization image to obtain a set of processed images corresponding to the fluorescence immunization image includes:

[0167] Performing pixel conversion processing on the fluorescence immunization image to obtain a conversion image corresponding to the fluorescence immunization image;

[0168] Adjusting the contrast of the conversion image according to a preset contrast adjustment rule to obtain a first processed image;

[0169] Processing the fluorescence immunization image through a Gaussian blur algorithm to obtain a background image corresponding to the fluorescence immunization image.

[0170] As can be seen, the implementation Figure 5 The described fluorescence immunization image processing device based on mixed processing can perform pixel conversion processing on the fluorescence immunization image to obtain a conversion image corresponding to the fluorescence immunization image, adjust the contrast of the conversion image according to a preset contrast adjustment rule to obtain a first processed image, and process the fluorescence immunization image through a Gaussian blur algorithm to obtain a background image corresponding to the fluorescence immunization image, which can pre-process the fluorescence immunization image, reduce noise influence, improve region division accuracy and subsequent fluorescence point recognition accuracy.

[0171] In yet another optional embodiment, as shown in Figure 5 The specific manner in which the first processing module 301 performs pixel conversion processing on the fluorescence immunization image to obtain a conversion image corresponding to the fluorescence immunization image includes:

[0172] Performing grayscale processing on the fluorescence immunization image to obtain a grayscale image corresponding to the fluorescence immunization image;

[0173] Performing black and white conversion processing on the grayscale image to obtain a conversion image corresponding to the fluorescence immunization image;

[0174] And the fluorescence immunoimage processing device based on hybrid processing can further comprise:

[0175] The second processing module 305 is configured to, after the first determination module 302 determines the target image corresponding to the fluorescence immunoimage based on the first processed image and the background image, perform deconvolution processing on the target image based on a preset imaging system to obtain a processed target image.

[0176] It can be seen that the implementation Figure 6 The fluorescence immunoimage processing device based on hybrid processing described in the embodiment can perform gray scale processing on the fluorescence immunoimage to obtain a gray scale image corresponding to the fluorescence immunoimage, perform black and white conversion processing on the gray scale image to obtain a converted image corresponding to the fluorescence immunoimage, improve the signal-to-noise ratio of the image, reduce image noise, and improve the accuracy and reliability of subsequent region division and fluorescent point recognition, perform deconvolution processing on the target image based on a preset imaging system to obtain a processed target image, restore the distortion of the outer fluorescent point caused by imaging, effectively reduce the recognition difficulties such as sticking and overlapping caused by distortion, and restore the true shape of the image.

[0177] Embodiment four

[0178] Please refer to Figure 6 , Figure 6 is another structure diagram of the fluorescence immunoimage processing device based on hybrid processing disclosed in the embodiment of the application. As ​ shown, the fluorescence immunoimage processing device based on hybrid processing can comprise:

[0179] The memory 401 stores executable program codes;

[0180] The processor 402 is coupled to the memory 401;

[0181] The processor 402 invokes the executable program codes stored in the memory 401 to execute the steps in the fluorescence immunoimage processing method based on hybrid processing described in the embodiment one or the embodiment two of the application.

[0182] Embodiment five

[0183] The embodiment of the application discloses a computer storage medium, which stores computer instructions. When the computer instructions are invoked, the steps in the fluorescence immunoimage processing method based on hybrid processing described in the embodiment one or the embodiment two of the application are executed.

[0184] Embodiment six

[0185] The embodiment of the present application discloses a computer program product, which comprises a non-transitory computer readable storage medium storing a computer program, and the computer program is operable to make a computer execute steps in the hybrid processing based fluorescence immunoimage processing method described in embodiment one or embodiment two.

[0186] The apparatus embodiments described above are only schematic, wherein the modules described as separate components can or can not be physically separate, and the components displayed as modules can or can not be physical modules, i.e., can be located in one place, or can be distributed on multiple network modules. Part or all of the modules can be selected to achieve the purpose of the embodiment scheme according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0187] Through the specific description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be realized by means of software and the necessary general hardware platform, and of course, can also be realized by hardware. Based on such understanding, the above technical solutions can be embodied in the form of software products, and the computer software products can be stored in a computer readable storage medium, including a read-only memory (Read-Only Memory, ROM), a random access memory (Random Access Memory, RAM), a programmable read-only memory (Programmable Read-only Memory, PROM), an erasable programmable read-only memory (Erasable Programmable Read Only Memory, EPROM), a one-time programmable read-only memory (One-time Programmable Read-Only Memory, OTPROM), an electrically erasable programmable read-only memory (Electrically-Erasable Programmable Read-Only Memory, EEPROM), a compact disc read-only memory (Compact Disc Read-Only Memory, CD-ROM) or other optical disk storage, a magnetic disk storage, a magnetic tape storage, or any other computer readable medium that can be used to carry or store data.

[0188] It should be finally pointed out that: the fluorescent immunoimage processing method and device based on hybrid processing disclosed by the embodiment of the present application is only the preferred embodiment of the present application, and is only used to illustrate the technical solutions of the present application, but not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that; it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A fluorescent immunoimage processing method based on hybrid processing, characterized by, The method comprises: image processing of the fluorescent immunoimage to obtain a set of processed images corresponding to the fluorescent immunoimage, the set of processed images comprising a first processed image and a background image; determining a target image corresponding to the fluorescent immunoimage according to the first processed image and the background image; analyzing luminance information corresponding to the background image and performing a region segmentation operation on the target image according to the luminance information to obtain a set of recognition regions, the set of recognition regions comprising a first recognition region, a second recognition region, and a third recognition region; determining a fluorescent point recognition strategy corresponding to the set of recognition regions and recognizing the set of recognition regions according to the fluorescent point recognition strategy to obtain fluorescent point information of the fluorescent immunoimage, the fluorescent point recognition strategy comprising a first strategy corresponding to the first recognition region, a hybrid recognition strategy corresponding to the second recognition region, and a second strategy corresponding to the third recognition region; and the recognizing the set of recognition regions according to the fluorescent point recognition strategy to obtain the fluorescent point information of the fluorescent immunoimage comprises: recognizing the first recognition region according to the first strategy to obtain first fluorescent point information corresponding to the first recognition region, the first strategy comprising a luminance threshold cutting recognition strategy; recognizing the second recognition region according to the hybrid recognition strategy to obtain second fluorescent point information corresponding to the second recognition region; recognizing the third recognition region according to the second strategy to obtain third fluorescent point information corresponding to the third recognition region, the second strategy comprising an artificial intelligence recognition strategy; determining the fluorescent point information of the fluorescent immunoimage according to the first fluorescent point information, the second fluorescent point information, and the third fluorescent point information; and the recognizing the second recognition region according to the hybrid recognition strategy to obtain the second fluorescent point information corresponding to the second recognition region comprises: recognizing the second recognition region according to the luminance threshold cutting recognition strategy to obtain first sub-fluorescent point information corresponding to the second recognition region; array scanning the second recognition region according to the artificial intelligence recognition strategy to obtain second sub-fluorescent point information corresponding to the second recognition region; determining the second fluorescent point information corresponding to the second recognition region according to the first sub-fluorescent point information and the second sub-fluorescent point information.

2. The hybrid processing-based fluorescent immunoimage processing method according to claim 1, characterized by, The analyzing the luminance information corresponding to the background image and performing a region segmentation operation on the target image according to the luminance information to obtain a set of recognition regions comprises: determining a fluorescent distribution region corresponding to the background image and analyzing luminance information of the fluorescent distribution region through a preset tool to obtain a luminance distribution curve of the background image; determining a set of luminance change marker points in the luminance distribution curve and determining a set of recognition regions of the target image according to the set of luminance change marker points, the set of luminance change marker points comprising a first luminance marker point and a second luminance marker point.

3. The hybrid processing-based fluorescent immunoimage processing method according to claim 2, wherein, The method comprises the following steps: determining a center point of the target image, and determining a first distance corresponding to the first luminance mark point and a second distance corresponding to the second luminance mark point according to the center point and the set of luminance change mark points, the second distance being greater than the first distance; determining a circular region, a first annular region and a second annular region in the target image according to the center point, the first distance and the second distance, and determining the circular region as the first recognition region, the first annular region as the second recognition region and the second annular region as the third recognition region.

4. The hybrid processing-based fluorescent immunoimage processing method according to claim 1, characterized by, The method comprises the following steps: performing pixel conversion processing on the fluorescence immunoassay image to obtain a converted image corresponding to the fluorescence immunoassay image; adjusting the contrast of the converted image according to a preset contrast adjustment rule to obtain a first processed image; performing processing on the fluorescence immunoassay image through a Gaussian blur algorithm to obtain a background image corresponding to the fluorescence immunoassay image.

5. The hybrid processing-based fluorescent immunoimage processing method according to claim 4, wherein, The method comprises the following steps: performing grayscale processing on the fluorescence immunoassay image to obtain a grayscale image corresponding to the fluorescence immunoassay image; performing black-and-white conversion processing on the grayscale image to obtain a converted image corresponding to the fluorescence immunoassay image; and, after the target image corresponding to the fluorescence immunoassay image is determined according to the first processed image and the background image, the method further comprises: performing deconvolution processing on the target image based on a preset imaging system to obtain a processed target image.

6. A fluorescence immunoassay image processing apparatus based on hybrid processing, characterized by comprising: The device comprises: a first processing module configured to perform image processing on a fluorescence immunoassay image to obtain a set of processed images corresponding to the fluorescence immunoassay image, the set of processed images comprising a first processed image and a background image; a first determination module configured to determine a target image corresponding to the fluorescence immunoassay image according to the first processed image and the background image; an analysis module configured to analyze luminance information corresponding to the background image, and perform a region segmentation operation on the target image according to the luminance information to obtain a set of recognition regions, the set of recognition regions comprising a first recognition region, a second recognition region and a third recognition region; a second determination module configured to determine a fluorescence point recognition strategy corresponding to the set of recognition regions, and perform recognition on the set of recognition regions according to the fluorescence point recognition strategy to obtain fluorescence point information of the fluorescence immunoassay image, the fluorescence point recognition strategy comprising a first strategy corresponding to the first recognition region, a mixed recognition strategy corresponding to the second recognition region and a second strategy corresponding to the third recognition region; and, the second determination module performs recognition on the set of recognition regions according to the fluorescence point recognition strategy to obtain the fluorescence point information of the fluorescence immunoassay image in the following manner: According to the first strategy, the first identified region is identified to obtain first fluorescent point information corresponding to the first identified region, and the first strategy includes a brightness threshold cutting identification strategy; According to the mixed identification strategy, the second identified region is identified to obtain second fluorescent point information corresponding to the second identified region; According to the second strategy, the third identified region is identified to obtain third fluorescent point information corresponding to the third identified region, and the second strategy includes an artificial intelligence identification strategy; According to the first fluorescent point information, the second fluorescent point information and the third fluorescent point information, the fluorescent point information of the fluorescent immunoassay image is determined; And the second determination module identifies the second identified region according to the mixed identification strategy to obtain second fluorescent point information corresponding to the second identified region, and the manner specifically includes: According to the brightness threshold cutting identification strategy, the second identified region is identified to obtain first sub-fluorescent point information corresponding to the second identified region; According to the artificial intelligence identification strategy, the second identified region is array scanned to obtain second sub-fluorescent point information corresponding to the second identified region; According to the first sub-fluorescent point information and the second sub-fluorescent point information, the second fluorescent point information corresponding to the second identified region is determined.

7. A fluorescence immunoassay image processing apparatus based on hybrid processing, characterized by comprising: The device includes: a memory storing executable program codes; a processor coupled with the memory; the processor invokes the executable program codes stored in the memory to execute the mixed processing based fluorescent immunoassay image processing method according to any one of claims 1-5.

8. A computer storage medium, characterized in that The computer storage medium stores computer instructions, and the computer instructions are used to execute the mixed processing based fluorescent immunoassay image processing method according to any one of claims 1-5 when invoked.

Images