Nailfold microcirculation imaging control and image recognition and analysis method and system

Abstract

The invention relates to the technical field of image recognition processing, and provides a nail fold microcirculation imaging control and image recognition analysis method and system. Based on the scattering characteristic information of the imaging illumination light detection of the nail fold area, the attribute information of the illumination abnormal area is determined, and the illumination light projection modulation is performed; the nail fold area is image recognized, and the capillary image set of the target sub-area is obtained, so as to realize the directional image acquisition of the capillaries in the nail fold area; the vascular morphology is recognized for all the capillary images, the abnormal vascular parts therein are calibrated, and the abnormal vascular parts sub-image sets are obtained by classification, so as to reduce the recognition workload and the error recognition rate; the abnormal vascular parts sub-image set is subjected to time evolution analysis, and the blood circulation characteristic information is obtained, so as to generate the microcirculation model of the nail fold area, realize the global microcirculation characterization of the nail fold area, predict the abnormal microcirculation evolution result, comprehensively and correctly reflect the real microcirculation state of the nail fold area, and improve the credibility of the nail fold microcirculation recognition.

Description

Technical Field

[0001] The invention relates to the technical field of image recognition processing, and in particular to a nailfold microcirculation imaging control and image recognition analysis method and system. Background Art

[0002] The nail fold refers to the skin fold covering the root of the nail. The surface of the nail fold area is covered with squamous epithelium, in which there are papillae formed by the protrusions of the skin dermis. Each papilla area generally has a capillary, which is very close to the epidermis, making the nail fold area an ideal area for observing microcirculation. The existing technology predicts the microcirculation state of the corresponding area by performing microscopic imaging of the nail fold area to identify the blood flow of the capillaries inside it. Due to the large number of capillaries inside the nail fold area, it is necessary to perform microscopic imaging of a large number of capillaries to fully grasp the microcirculation state of the nail fold area, which increases the workload of identifying the microcirculation state of the nail fold. At the same time, in the process of imaging and analyzing a large number of capillaries, it is easy to have misidentification and missed identification, which cannot fully and correctly reflect the true microcirculation state of the nail fold area, reduce the credibility of nail fold microcirculation identification, and cannot provide a reliable basis for subsequent diagnosis. Summary of the invention

[0003] In view of the defects of the prior art, the present invention provides a method and system for nail fold microcirculation imaging control and image recognition and analysis, which determines the attribute information of the abnormal illumination area based on the scattering characteristic information of the imaging illumination light detected in the nail fold area, and performs illumination light projection modulation to ensure that the entire nail fold area can be properly illuminated to avoid interference with subsequent image recognition; performs image recognition on the nail fold area to obtain a capillary image set of the target sub-area, and realizes directional image acquisition of the capillaries inside the nail fold area; performs vascular morphology recognition on all capillary images, calibrates the abnormal vascular parts therein, and classifies them to obtain a sub-image set of the abnormal vascular parts, which is convenient for subsequent recognition of only the abnormal vascular parts, reducing the recognition workload and the error recognition rate; and performs time evolution analysis on the sub-image set of the abnormal vascular parts to obtain blood circulation characteristic information, thereby generating a microcirculation model of the nail fold area, realizing global microcirculation characterization of the nail fold area, and predicting the abnormal evolution results of the microcirculation, comprehensively and correctly reflecting the real microcirculation state of the nail fold area, and improving the credibility of nail fold microcirculation recognition.

[0004] The present invention provides a nail fold microcirculation imaging control and image recognition and analysis method, comprising the following steps:

[0005] Step S1, performing scattering detection of imaging illumination light on the nail fold area to obtain scattering characteristic information of the nail fold area to the imaging illumination light; determining illumination abnormal area attribute information of the nail fold area based on the scattering characteristic information; and performing illumination light projection modulation based on the illumination abnormal area attribute information;

[0006] Step S2, performing a first image recognition on the nail fold area to obtain distribution feature information of a target sub-area of ​​the nail fold area; based on the distribution feature information of the target sub-area, performing a second image recognition on the nail fold area to obtain a capillary image set corresponding to the target sub-area;

[0007] Step S3, performing vascular morphology recognition on all capillary images under the capillary image set to obtain vascular contour feature information corresponding to each of the capillary images; based on the vascular contour feature information, calibrating the abnormal vascular part of each capillary image; and then classifying all abnormal vascular parts under all capillary images to obtain a plurality of abnormal vascular part sub-image sets;

[0008] Step S4, performing time evolution analysis on each abnormal vascular site sub-image set to obtain blood circulation characteristic information of the corresponding abnormal vascular site; generating a microcirculation model of the nail fold area based on the blood circulation characteristic information of all abnormal vascular sites; and analyzing the microcirculation model to predict the abnormal evolution result of the microcirculation in the nail fold area.

[0009] In one embodiment disclosed in the present application, in the step S1, scattering detection of imaging illumination light is performed on the nail fold area to obtain scattering characteristic information of the nail fold area to the imaging illumination light; based on the scattering characteristic information, attribute information of the abnormal illumination area of ​​the nail fold area is determined; based on the attribute information of the abnormal illumination area, illumination light projection modulation is performed, including:

[0010] Performing visual detection on the scattered light formed by the nail fold area under the projection of the reference imaging illumination light to obtain a scattered light distribution image; analyzing the scattered light distribution image to obtain scattered light distribution characteristic information of the spatial area where the nail fold area is located; wherein the scattered light distribution characteristic information includes scattered light intensity distribution information of the spatial area at different orientations;

[0011] Based on the scattered light distribution characteristic information, all sub-regions under the nail fold area that form abnormal scattering intensity for the reference imaging illumination light are determined, and the distribution position information of all sub-regions in the nail fold area is used as the attribute information of the abnormal illumination area; based on the attribute information of the abnormal illumination area, the intensity of the area corresponding to the projection cross-section of the reference imaging illumination light is modulated.

[0012] In one embodiment disclosed in the present application, in the step S2, the nail fold region is subjected to a first image recognition to obtain target sub-region distribution feature information of the nail fold region; based on the target sub-region distribution feature information, the nail fold region is subjected to a second image recognition to obtain a capillary image set corresponding to the target sub-region, including:

[0013] Under the illumination of the projected modulated illumination light, the nail fold area is scanned and photographed to obtain a global image of the nail fold area; pixel texture recognition is performed on the global image to obtain the distribution position information of all dermal papilla areas under the nail fold area, which is used as the distribution feature information of the target sub-area;

[0014] Based on the distribution position information of all dermal papillary areas, each dermal papillary area is dynamically and directionally photographed to obtain a dynamic image corresponding to each dermal papillary area; the dynamic image is frame-by-frame identified to obtain a number of capillary images corresponding to all capillaries under each dermal papillary area at different times, thereby forming a capillary image set.

[0015] In one embodiment disclosed in the present application, in the step S3, all capillary images under the capillary image set are subjected to vascular morphology recognition to obtain vascular contour feature information corresponding to each of the capillary images; based on the vascular contour feature information, abnormal vascular locations are calibrated for each capillary image; and all abnormal vascular locations under all capillary images are classified to obtain a plurality of abnormal vascular location sub-image sets, including:

[0016] Performing vascular wall shape recognition on all capillary images in the capillary image set to obtain vascular contour feature information corresponding to each of the capillary images; comparing the vascular contour feature information with reference contour feature information to obtain bending curvature offset information of the capillary; and calibrating abnormal vascular locations on the capillary images based on the bending curvature offset information;

[0017] Based on the position information of all abnormal blood vessel parts under all capillary images, the abnormal blood vessel parts in the same position are classified and processed to obtain several abnormal blood vessel part sub-image sets; wherein, all the abnormal blood vessel part sub-images contained in each abnormal blood vessel part sub-image set correspond to the same abnormal blood vessel part.

[0018] In one embodiment disclosed in the present application, in step S4, a time evolution analysis is performed on each abnormal vascular site sub-image set to obtain blood circulation characteristic information of the corresponding abnormal vascular site; a microcirculation model of the nail fold region is generated based on the blood circulation characteristic information of all abnormal vascular sites; and the microcirculation model is analyzed to predict the abnormal evolution result of the microcirculation in the nail fold region, including:

[0019] Based on the generation time information corresponding to all abnormal blood vessel part sub-images contained in each abnormal blood vessel part sub-image set, all abnormal blood vessel part sub-images are subjected to time evolution analysis to obtain blood circulation flow characteristic information of the corresponding abnormal blood vessel part; wherein the blood circulation flow characteristic information includes flow velocity information of different blood components inside the corresponding abnormal blood vessel part;

[0020] Based on the blood circulation flow characteristic information of all abnormal blood vessels and their distribution position information in the nail fold area, a microcirculation model of the nail fold area is generated; the microcirculation model is analyzed to predict the blood vessel position and time of microcirculation blockage in the nail fold area, which is used as the result of the abnormal microcirculation evolution.

[0021] The present invention also provides a nail fold microcirculation imaging control and image recognition and analysis system, comprising:

[0022] A nail fold area scattering feature recognition module is used to detect the scattering of the imaging illumination light in the nail fold area to obtain scattering feature information of the nail fold area to the imaging illumination light;

[0023] An illumination light modulation module, used to determine the illumination abnormality region attribute information of the nail fold region based on the scattering characteristic information; and perform illumination light projection modulation based on the illumination abnormality region attribute information;

[0024] A first image recognition module, used for performing first image recognition on the nail fold area to obtain target sub-area distribution feature information of the nail fold area;

[0025] A second image recognition module, configured to perform second image recognition on the nail fold region based on the target sub-region distribution feature information, and obtain a capillary image set corresponding to the target sub-region;

[0026] A blood vessel contour recognition module is used to perform blood vessel morphology recognition on all capillary images in the capillary image set to obtain blood vessel contour feature information corresponding to all capillary images;

[0027] The abnormal blood vessel part identification module is used to calibrate the abnormal blood vessel part of each capillary image based on the blood vessel contour feature information; and then classify all abnormal blood vessel parts under all capillary images to obtain a plurality of abnormal blood vessel part sub-image sets;

[0028] A blood circulation feature recognition module is used to perform time-evolution analysis on each abnormal blood vessel part sub-image set to obtain blood circulation feature information of the corresponding abnormal blood vessel part;

[0029] The microcirculation abnormality prediction module is used to generate a microcirculation model of the nail fold area based on the blood circulation characteristic information of all abnormal blood vessel parts; and analyze the microcirculation model to predict the evolution result of the microcirculation abnormality in the nail fold area.

[0030] In one embodiment disclosed in the present application, the nail fold region scattering feature recognition module is used to perform scattering detection of the nail fold region for imaging illumination light to obtain scattering feature information of the nail fold region for imaging illumination light, including:

[0031] Performing visual detection on the scattered light formed by the nail fold area under the projection of the reference imaging illumination light to obtain a scattered light distribution image; analyzing the scattered light distribution image to obtain scattered light distribution characteristic information of the spatial area where the nail fold area is located; wherein the scattered light distribution characteristic information includes scattered light intensity distribution information of the spatial area at different orientations;

[0032] The illumination light modulation module is used to determine the illumination abnormal area attribute information of the nail fold area based on the scattering characteristic information; and perform illumination light projection modulation based on the illumination abnormal area attribute information, including:

[0033] Based on the scattered light distribution characteristic information, all sub-regions under the nail fold area that form abnormal scattering intensity for the reference imaging illumination light are determined, and the distribution position information of all sub-regions in the nail fold area is used as the attribute information of the abnormal illumination area; based on the attribute information of the abnormal illumination area, the intensity of the area corresponding to the projection cross-section of the reference imaging illumination light is modulated.

[0034] In one embodiment disclosed in the present application, the first image recognition module is used to perform first image recognition on the nail fold area to obtain target sub-area distribution feature information of the nail fold area, including:

[0035] Under the illumination of the projected modulated illumination light, the nail fold area is scanned and photographed to obtain a global image of the nail fold area; pixel texture recognition is performed on the global image to obtain the distribution position information of all dermal papilla areas under the nail fold area, which is used as the distribution feature information of the target sub-area;

[0036] The second image recognition module is used to perform second image recognition on the nail fold region based on the target sub-region distribution feature information to obtain a capillary image set corresponding to the target sub-region, including:

[0037] Based on the distribution position information of all dermal papillary areas, each dermal papillary area is dynamically and directionally photographed to obtain a dynamic image corresponding to each dermal papillary area; the dynamic image is frame-by-frame identified to obtain a number of capillary images corresponding to all capillaries under each dermal papillary area at different times, thereby forming a capillary image set.

[0038] In one embodiment disclosed in the present application, the blood vessel contour recognition module is used to perform blood vessel morphology recognition on all capillary images in the capillary image set to obtain blood vessel contour feature information corresponding to all capillary images, including:

[0039] Performing vascular wall shape recognition on all capillary images in the capillary image set to obtain vascular contour feature information corresponding to all capillary images;

[0040] The abnormal blood vessel location identification module is used to calibrate the abnormal blood vessel location of each capillary image based on the blood vessel contour feature information; and then classify all abnormal blood vessel locations under all capillary images to obtain a plurality of abnormal blood vessel location sub-image sets, including:

[0041] Comparing the blood vessel contour feature information with the reference contour feature information to obtain the bending curvature offset information of the capillary; and calibrating the abnormal blood vessel position of the capillary image based on the bending curvature offset information;

[0042] Based on the position information of all abnormal blood vessel parts under all capillary images, the abnormal blood vessel parts in the same position are classified and processed to obtain several abnormal blood vessel part sub-image sets; wherein, all the abnormal blood vessel part sub-images contained in each abnormal blood vessel part sub-image set correspond to the same abnormal blood vessel part.

[0043] In one embodiment disclosed in the present application, the blood circulation feature recognition module is used to perform time-evolution analysis on each abnormal blood vessel part sub-image set to obtain blood circulation feature information of the corresponding abnormal blood vessel part, including:

[0044] Based on the generation time information corresponding to all abnormal blood vessel part sub-images contained in each abnormal blood vessel part sub-image set, all abnormal blood vessel part sub-images are subjected to time evolution analysis to obtain blood circulation flow characteristic information of the corresponding abnormal blood vessel part; wherein the blood circulation flow characteristic information includes flow velocity information of different blood components inside the corresponding abnormal blood vessel part;

[0045] The microcirculation abnormality prediction module is used to generate a microcirculation model of the nail fold area based on the blood circulation characteristic information of all abnormal blood vessel parts; and analyze the microcirculation model to predict the evolution result of the microcirculation abnormality in the nail fold area, including:

[0046] Based on the blood circulation flow characteristic information of all abnormal blood vessels and their distribution position information in the nail fold area, a microcirculation model of the nail fold area is generated; the microcirculation model is analyzed to predict the blood vessel position and time of microcirculation blockage in the nail fold area, which is used as the result of the abnormal microcirculation evolution.

[0047] Compared with the prior art, the nailfold microcirculation imaging control and image recognition and analysis method and system determine the attribute information of the abnormal illumination area based on the scattering characteristic information of the imaging illumination light detection of the nailfold area, thereby performing illumination light projection modulation to ensure that the entire nailfold area can be properly illuminated and avoid interference with subsequent image recognition; image recognition is performed on the nailfold area to obtain a capillary image set of the target sub-area, thereby realizing directional image acquisition of the capillaries inside the nailfold area; vascular morphology recognition is performed on all capillary images, and the abnormal vascular sites therein are calibrated, thereby classifying and obtaining a sub-image set of the abnormal vascular sites, so as to facilitate subsequent recognition of only the abnormal vascular sites, thereby reducing the recognition workload and the error recognition rate; time-evolution analysis is also performed on the sub-image set of the abnormal vascular sites to obtain blood circulation characteristic information, thereby generating a microcirculation model of the nailfold area, realizing global microcirculation characterization of the nailfold area, and predicting the abnormal evolution results of the microcirculation, comprehensively and correctly reflecting the real microcirculation state of the nailfold area, and improving the credibility of nailfold microcirculation recognition.

[0048] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.

[0049] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0051] Figure 1 A schematic diagram of the process flow of the nailfold microcirculation imaging control and image recognition and analysis method provided by the present invention;

[0052] Figure 2 A schematic diagram of the framework of the nailfold microcirculation imaging control and image recognition and analysis system provided by the present invention. DETAILED DESCRIPTION

[0053] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0054] See also Figure 1 , is a flow chart of a nail fold microcirculation imaging control and image recognition and analysis method provided by an embodiment of the present invention. The nail fold microcirculation imaging control and image recognition and analysis method comprises:

[0055] Step S1, performing scattering detection of imaging illumination light on the nail fold area to obtain scattering characteristic information of the nail fold area to the imaging illumination light; determining illumination abnormal area attribute information of the nail fold area based on the scattering characteristic information; and performing illumination light projection modulation based on the illumination abnormal area attribute information;

[0056] Step S2, performing a first image recognition on the nail fold region to obtain distribution feature information of a target sub-region of the nail fold region; based on the distribution feature information of the target sub-region, performing a second image recognition on the nail fold region to obtain a capillary image set corresponding to the target sub-region;

[0057] Step S3, performing vascular morphology recognition on all capillary images under the capillary image set to obtain vascular contour feature information corresponding to each of the capillary images; based on the vascular contour feature information, calibrating the abnormal vascular part of each capillary image; and then classifying all abnormal vascular parts under all capillary images to obtain a plurality of abnormal vascular part sub-image sets;

[0058] Step S4, performing time evolution analysis on each abnormal vascular site sub-image set to obtain blood circulation characteristic information of the corresponding abnormal vascular site; generating a microcirculation model of the nail fold area based on the blood circulation characteristic information of all abnormal vascular sites; and analyzing the microcirculation model to predict the abnormal evolution result of the microcirculation in the nail fold area.

[0059] The nailfold microcirculation imaging control and image recognition and analysis method determines the attribute information of the abnormal illumination area based on the scattering characteristic information of the imaging illumination light detection of the nailfold area, and performs illumination light projection modulation to ensure that the entire nailfold area can be properly illuminated to avoid interference with subsequent image recognition; the nailfold area is image recognized to obtain a capillary image set of the target sub-area, and directional image acquisition of the capillaries inside the nailfold area is realized; the vascular morphology of all capillary images is recognized, and the abnormal vascular parts are calibrated, so as to obtain a sub-image set of the abnormal vascular parts by classification, so as to facilitate subsequent recognition of only the abnormal vascular parts, reduce the recognition workload and the error recognition rate; the abnormal vascular part sub-image set is also analyzed in time evolution to obtain blood circulation characteristic information, so as to generate a microcirculation model of the nailfold area, realize the global microcirculation characterization of the nailfold area, and predict the abnormal evolution results of the microcirculation, comprehensively and correctly reflect the real microcirculation state of the nailfold area, and improve the credibility of nailfold microcirculation recognition.

[0060] Preferably, in step S1, the scattering detection of the imaging illumination light is performed on the nail fold area to obtain the scattering characteristic information of the imaging illumination light in the nail fold area; based on the scattering characteristic information, the illumination abnormal area attribute information of the nail fold area is determined; based on the illumination abnormal area attribute information, the illumination light projection modulation is performed, including:

[0061] Performing visual detection on the scattered light formed by the nail fold area under the projection of the reference imaging illumination light to obtain a scattered light distribution image; analyzing the scattered light distribution image to obtain scattered light distribution characteristic information of the spatial area where the nail fold area is located; wherein the scattered light distribution characteristic information includes scattered light intensity distribution information of the spatial area at different directions;

[0062] Based on the scattered light distribution characteristic information, all sub-regions under the nail fold area that form abnormal scattering intensity to the reference imaging illumination light are determined, and the distribution position information of all sub-regions in the nail fold area is used as the attribute information of the abnormal illumination area; based on the attribute information of the abnormal illumination area, the intensity of the area corresponding to the projection cross-section of the reference imaging illumination light is modulated.

[0063] In the above technical solution, the nail fold area is covered with epidermal tissue. In order to accurately and clearly shoot the nail fold area hidden under the epidermal tissue, it is necessary to project corresponding imaging illumination light onto the nail fold area. When the imaging illumination light is irradiated onto the nail fold area, the epidermal tissue of the nail fold area will generate scattered light to the illumination light irradiated thereon. If the intensity of the scattered light in a certain spatial orientation is too large, the shooting of the nail fold area will be overexposed, thereby affecting the shooting clarity of the nail fold area. To this end, a visual inspection is performed on the scattered light formed in the nail fold area under the projection of the reference imaging illumination light to obtain a scattered light distribution image, which represents the scattered light intensity distribution image formed in three-dimensional space when the corresponding epidermal tissue in the nail fold area scatters the reference imaging illumination light under the irradiation of the reference imaging illumination light. The scattered light distribution image is then analyzed to obtain scattered light intensity distribution information at different orientations in the three-dimensional space where the nail fold area is located. This can refine the determination of the scattered light intensity at each orientation, providing a reliable basis for subsequent judgment on whether the scattered light interferes with the clarity of the image shooting of the nail fold area; wherein the reference imaging illumination light may be, but is not limited to, illumination light with a preset cross-sectional light intensity distribution. Furthermore, based on the scattered light distribution characteristic information, all sub-regions under the nail fold area that form abnormal scattering intensity to the reference imaging illumination light are determined, for example, the scattered light intensity in a certain direction is compared with a preset intensity threshold, if the scattered light intensity in a certain direction is greater than the preset intensity threshold, then the sub-region corresponding to the projection of the direction on the nail fold area is determined as the sub-region that forms abnormal scattering intensity to the reference imaging illumination light, and the distribution position information of all the sub-regions in the nail fold area is used as the attribute information of the illumination abnormality area, so as to perform intensity modulation on the area corresponding to the projection cross-section of the reference imaging illumination light, so as to perform local light intensity modulation on the cross-sectional dimension of the reference imaging illumination light, wherein the local light intensity modulation can be achieved by electro-optical devices, acousto-optical devices or magneto-optical devices, which belongs to the conventional technical means in the field and will not be described in detail here.

[0064] Preferably, in step S2, a first image recognition is performed on the nail fold region to obtain distribution feature information of a target sub-region of the nail fold region; based on the distribution feature information of the target sub-region, a second image recognition is performed on the nail fold region to obtain a capillary image set corresponding to the target sub-region, including:

[0065] Under the illumination of the projected modulated illumination light, the nail fold area is scanned and photographed to obtain a global image of the nail fold area; pixel texture recognition is performed on the global image to obtain the distribution position information of all dermal papilla areas under the nail fold area, which is used as the distribution feature information of the target sub-area;

[0066] Based on the distribution position information of all dermal papillary areas, each dermal papillary area is dynamically and directionally photographed to obtain a dynamic image corresponding to each dermal papillary area; the dynamic image is frame-by-frame recognized to obtain a number of capillary images corresponding to all capillaries under each dermal papillary area at different times, thereby forming a capillary image set.

[0067] In the above technical solution, after the projection modulation of the imaging illumination light is completed, the imaging illumination light is directly used to illuminate the nail fold area, so that the illumination light can pass through the epidermal tissue above the nail fold area to reach the nail fold area. At this time, the nail fold area is scanned and photographed to obtain a global image of the nail fold area, and then the pixel texture recognition is performed on the global image to obtain the distribution position information of all the dermal papilla areas under the nail fold area, so as to locate and identify each dermal papilla area. Since the dermal papilla area and the capillaries inside the nail fold area correspond to each other, based on the distribution position information of all the dermal papilla areas, each dermal papilla area is dynamically directional photographed, so that dynamic directional photography of the capillaries can be achieved. The dynamic image corresponding to each dermal papilla area is also frame-by-frame identified to obtain a number of capillary images corresponding to all the capillaries under each dermal papilla area at different times, so as to form a capillary image set, so that the appearance and blood transmission of the capillaries in the corresponding time interval can be accurately and finely characterized.

[0068] Preferably, in step S3, all capillary images under the capillary image set are subjected to vascular morphology recognition to obtain vascular contour feature information corresponding to each of the capillary images; based on the vascular contour feature information, abnormal vascular sites are calibrated for each capillary image; and all abnormal vascular sites under all capillary images are classified to obtain a plurality of abnormal vascular site sub-image sets, including:

[0069] Performing vascular wall shape recognition on all capillary images under the capillary image set to obtain vascular contour feature information corresponding to each of the capillary images; comparing the vascular contour feature information with the reference contour feature information to obtain the bending curvature offset information of the capillary; and calibrating the abnormal vascular position of the capillary image based on the bending curvature offset information;

[0070] Based on the position information of all abnormal blood vessel parts under all capillary images, the abnormal blood vessel parts in the same position are classified and processed to obtain several abnormal blood vessel part sub-image sets; wherein, all the abnormal blood vessel part sub-images contained in each abnormal blood vessel part sub-image set correspond to the same abnormal blood vessel part.

[0071] In the above technical solution, all capillary images under the capillary image set are respectively subjected to vascular wall shape recognition to obtain the vascular contour feature information corresponding to each of the capillary images. Under normal circumstances, the contour of the capillaries in the nail fold area is relatively straight. When there is a lesion in the nail fold area, the contour of the capillaries will become curved and messy. The vascular contour feature information is compared with the reference contour feature information to obtain the curvature offset information of the capillary. If the curvature offset value of a certain part of the capillary exceeds the preset offset threshold, the corresponding capillary part is determined as an abnormal vascular part, so that the potential lesion part of the capillary can be accurately and comprehensively identified. In addition, based on the position information of all abnormal vascular parts under all capillary images, the abnormal vascular parts in the same position are classified and processed to obtain a number of abnormal vascular part sub-image sets, so that all abnormal vascular part sub-images contained in each abnormal vascular part sub-image set correspond to the same abnormal vascular part, which is convenient for the subsequent recognition and prediction of the time change of the same abnormal vascular part.

[0072] Preferably, in step S4, a time evolution analysis is performed on each abnormal blood vessel part sub-image set to obtain blood circulation characteristic information of the corresponding abnormal blood vessel part; a microcirculation model of the nail fold area is generated based on the blood circulation characteristic information of all abnormal blood vessel parts; and the microcirculation model is analyzed to predict the abnormal evolution result of the microcirculation in the nail fold area, including:

[0073] Based on the generation time information corresponding to all abnormal blood vessel part sub-images contained in each abnormal blood vessel part sub-image set, all abnormal blood vessel part sub-images are subjected to time evolution analysis to obtain blood circulation flow characteristic information of the corresponding abnormal blood vessel part; wherein the blood circulation flow characteristic information includes flow velocity information of different blood components inside the corresponding abnormal blood vessel part;

[0074] Based on the blood circulation flow characteristic information of all abnormal blood vessels and their distribution position information in the nail fold area, a microcirculation model of the nail fold area is generated. The microcirculation model is analyzed to predict the blood vessel position and time of microcirculation blockage in the nail fold area, which is used as the result of the abnormal microcirculation evolution.

[0075] In the above technical solution, based on the generation time information corresponding to all abnormal vascular part sub-images contained in each abnormal vascular part sub-image set, all abnormal vascular part sub-images are subjected to time evolution analysis to obtain the flow velocity information of different blood components inside the corresponding abnormal vascular part (such as the flow velocity of red blood cells, white blood cells and platelets), and the blood circulation situation inside the abnormal vascular part is refined and identified. In addition, based on the blood circulation flow characteristic information of all abnormal vascular parts and the distribution position information in the nail fold area, a microcirculation model of the nail fold area is generated, so as to predict the blood vessel position and time of microcirculation blockage in the nail fold area, so as to serve as the abnormal evolution result of the microcirculation, so as to fully and correctly reflect the real microcirculation state of the nail fold area and improve the credibility of nail fold microcirculation identification.

[0076] See also Figure 2 , is a schematic diagram of the framework of the nail fold microcirculation imaging control and image recognition and analysis system provided by an embodiment of the present invention. The nail fold microcirculation imaging control and image recognition and analysis system includes:

[0077] A nail fold area scattering feature recognition module is used to detect the scattering of the imaging illumination light in the nail fold area to obtain the scattering feature information of the imaging illumination light in the nail fold area;

[0078] An illumination light modulation module is used to determine the illumination abnormality region attribute information of the nail fold region based on the scattering characteristic information; and perform illumination light projection modulation based on the illumination abnormality region attribute information;

[0079] A first image recognition module is used to perform a first image recognition on the nail fold area to obtain target sub-area distribution feature information of the nail fold area;

[0080] A second image recognition module is used to perform second image recognition on the nail fold region based on the distribution feature information of the target sub-region to obtain a capillary image set corresponding to the target sub-region;

[0081] A blood vessel contour recognition module is used to perform blood vessel morphology recognition on all capillary images under the capillary image set to obtain blood vessel contour feature information corresponding to all capillary images;

[0082] The abnormal blood vessel part identification module is used to calibrate the abnormal blood vessel part of each capillary image based on the blood vessel contour feature information; and then classify all abnormal blood vessel parts under all capillary images to obtain a number of abnormal blood vessel part sub-image sets;

[0083] A blood circulation feature recognition module is used to perform time-evolution analysis on each abnormal blood vessel part sub-image set to obtain blood circulation feature information of the corresponding abnormal blood vessel part;

[0084] The microcirculation abnormality prediction module is used to generate a microcirculation model of the nail fold area based on the blood circulation characteristic information of all abnormal vascular sites; and to analyze the microcirculation model to predict the evolution result of the microcirculation abnormality in the nail fold area.

[0085] The nailfold microcirculation imaging control and image recognition and analysis system determines the attribute information of the abnormal illumination area based on the scattering characteristic information of the imaging illumination light detection of the nailfold area, and performs illumination light projection modulation to ensure that the entire nailfold area can be properly illuminated to avoid interference with subsequent image recognition; performs image recognition on the nailfold area to obtain a capillary image set of the target sub-area, and realizes directional image acquisition of the capillaries inside the nailfold area; performs vascular morphology recognition on all capillary images, calibrates the abnormal vascular sites therein, and classifies them to obtain a sub-image set of abnormal vascular sites, which is convenient for subsequent recognition of only the abnormal vascular sites, reducing the recognition workload and the error recognition rate; and performs time-evolution analysis on the sub-image set of the abnormal vascular sites to obtain blood circulation characteristic information, thereby generating a microcirculation model of the nailfold area, realizing global microcirculation characterization of the nailfold area, and predicting the abnormal evolution of microcirculation, comprehensively and correctly reflecting the real microcirculation state of the nailfold area, and improving the credibility of nailfold microcirculation recognition.

[0086] Preferably, the nail fold region scattering feature recognition module is used to perform scattering detection of the imaging illumination light in the nail fold region to obtain scattering feature information of the imaging illumination light in the nail fold region, including:

[0087] Performing visual detection on the scattered light formed by the nail fold area under the projection of the reference imaging illumination light to obtain a scattered light distribution image; analyzing the scattered light distribution image to obtain scattered light distribution characteristic information of the spatial area where the nail fold area is located; wherein the scattered light distribution characteristic information includes scattered light intensity distribution information of the spatial area at different directions;

[0088] The illumination light modulation module is used to determine the illumination abnormal area attribute information of the nail fold area based on the scattering characteristic information; and perform illumination light projection modulation based on the illumination abnormal area attribute information, including:

[0089] Based on the scattered light distribution characteristic information, all sub-regions under the nail fold area that form abnormal scattering intensity to the reference imaging illumination light are determined, and the distribution position information of all sub-regions in the nail fold area is used as the attribute information of the abnormal illumination area; based on the attribute information of the abnormal illumination area, the intensity of the area corresponding to the projection cross-section of the reference imaging illumination light is modulated.

[0090] In the above technical solution, the nail fold area is covered with epidermal tissue. In order to accurately and clearly shoot the nail fold area hidden under the epidermal tissue, it is necessary to project corresponding imaging illumination light onto the nail fold area. When the imaging illumination light is irradiated onto the nail fold area, the epidermal tissue of the nail fold area will generate scattered light to the illumination light irradiated thereon. If the intensity of the scattered light in a certain spatial orientation is too large, the shooting of the nail fold area will be overexposed, thereby affecting the shooting clarity of the nail fold area. To this end, a visual inspection is performed on the scattered light formed in the nail fold area under the projection of the reference imaging illumination light to obtain a scattered light distribution image, which represents the scattered light intensity distribution image formed in three-dimensional space when the corresponding epidermal tissue in the nail fold area scatters the reference imaging illumination light under the irradiation of the reference imaging illumination light. The scattered light distribution image is then analyzed to obtain scattered light intensity distribution information at different orientations in the three-dimensional space where the nail fold area is located. This can refine the determination of the scattered light intensity at each orientation, providing a reliable basis for subsequent judgment on whether the scattered light interferes with the clarity of the image shooting of the nail fold area; wherein the reference imaging illumination light may be, but is not limited to, illumination light with a preset cross-sectional light intensity distribution. Furthermore, based on the scattered light distribution characteristic information, all sub-regions under the nail fold area that form abnormal scattering intensity to the reference imaging illumination light are determined, for example, the scattered light intensity in a certain direction is compared with a preset intensity threshold, if the scattered light intensity in a certain direction is greater than the preset intensity threshold, then the sub-region corresponding to the projection of the direction on the nail fold area is determined as the sub-region that forms abnormal scattering intensity to the reference imaging illumination light, and the distribution position information of all the sub-regions in the nail fold area is used as the attribute information of the illumination abnormality area, so as to perform intensity modulation on the area corresponding to the projection cross-section of the reference imaging illumination light, so as to perform local light intensity modulation on the cross-sectional dimension of the reference imaging illumination light, wherein the local light intensity modulation can be achieved by electro-optical devices, acousto-optical devices or magneto-optical devices, which belongs to the conventional technical means in the field and will not be described in detail here.

[0091] Preferably, the first image recognition module is used to perform first image recognition on the nail fold area to obtain target sub-area distribution feature information of the nail fold area, including:

[0092] Under the illumination of the projected modulated illumination light, the nail fold area is scanned and photographed to obtain a global image of the nail fold area; pixel texture recognition is performed on the global image to obtain the distribution position information of all dermal papilla areas under the nail fold area, which is used as the distribution feature information of the target sub-area;

[0093] The second image recognition module is used to perform second image recognition on the nail fold region based on the target sub-region distribution feature information to obtain a capillary image set corresponding to the target sub-region, including:

[0094] Based on the distribution position information of all dermal papillary areas, each dermal papillary area is dynamically and directionally photographed to obtain a dynamic image corresponding to each dermal papillary area; the dynamic image is frame-by-frame recognized to obtain a number of capillary images corresponding to all capillaries under each dermal papillary area at different times, thereby forming a capillary image set.

[0095] In the above technical solution, after the projection modulation of the imaging illumination light is completed, the imaging illumination light is directly used to illuminate the nail fold area, so that the illumination light can pass through the epidermal tissue above the nail fold area to reach the nail fold area. At this time, the nail fold area is scanned and photographed to obtain a global image of the nail fold area, and then the pixel texture recognition is performed on the global image to obtain the distribution position information of all the dermal papilla areas under the nail fold area, so as to locate and identify each dermal papilla area. Since the dermal papilla area and the capillaries inside the nail fold area correspond to each other, based on the distribution position information of all the dermal papilla areas, each dermal papilla area is dynamically directional photographed, so that dynamic directional photography of the capillaries can be achieved. The dynamic image corresponding to each dermal papilla area is also frame-by-frame identified to obtain a number of capillary images corresponding to all the capillaries under each dermal papilla area at different times, so as to form a capillary image set, so that the appearance and blood transmission of the capillaries in the corresponding time interval can be accurately and finely characterized.

[0096] Preferably, the blood vessel contour recognition module is used to perform blood vessel morphology recognition on all capillary images in the capillary image set to obtain blood vessel contour feature information corresponding to all capillary images, including:

[0097] Performing vascular wall shape recognition on all capillary images in the capillary image set to obtain vascular contour feature information corresponding to all capillary images;

[0098] The abnormal blood vessel part identification module is used to calibrate the abnormal blood vessel part of each capillary image based on the blood vessel contour feature information; and then classify all abnormal blood vessel parts under all capillary images to obtain a number of abnormal blood vessel part sub-image sets, including:

[0099] Comparing the blood vessel contour feature information with the reference contour feature information to obtain the bending curvature offset information of the capillary; and calibrating the abnormal blood vessel part of the capillary image based on the bending curvature offset information;

[0100] Based on the position information of all abnormal blood vessel parts under all capillary images, the abnormal blood vessel parts in the same position are classified and processed to obtain several abnormal blood vessel part sub-image sets; wherein, all the abnormal blood vessel part sub-images contained in each abnormal blood vessel part sub-image set correspond to the same abnormal blood vessel part.

[0101] In the above technical solution, all capillary images under the capillary image set are respectively subjected to vascular wall shape recognition to obtain the vascular contour feature information corresponding to each of the capillary images. Under normal circumstances, the contour of the capillaries in the nail fold area is relatively straight. When there is a lesion in the nail fold area, the contour of the capillaries will become curved and messy. The vascular contour feature information is compared with the reference contour feature information to obtain the curvature offset information of the capillary. If the curvature offset value of a certain part of the capillary exceeds the preset offset threshold, the corresponding capillary part is determined as an abnormal vascular part, so that the potential lesion part of the capillary can be accurately and comprehensively identified. In addition, based on the position information of all abnormal vascular parts under all capillary images, the abnormal vascular parts in the same position are classified and processed to obtain a number of abnormal vascular part sub-image sets, so that all abnormal vascular part sub-images contained in each abnormal vascular part sub-image set correspond to the same abnormal vascular part, which is convenient for the subsequent recognition and prediction of the time change of the same abnormal vascular part.

[0102] Preferably, the blood circulation feature recognition module is used to perform time-evolution analysis on each abnormal blood vessel part sub-image set to obtain blood circulation feature information of the corresponding abnormal blood vessel part, including:

[0103] Based on the generation time information corresponding to all abnormal blood vessel part sub-images contained in each abnormal blood vessel part sub-image set, all abnormal blood vessel part sub-images are subjected to time evolution analysis to obtain blood circulation flow characteristic information of the corresponding abnormal blood vessel part; wherein the blood circulation flow characteristic information includes flow velocity information of different blood components inside the corresponding abnormal blood vessel part;

[0104] The microcirculation abnormality prediction module is used to generate a microcirculation model of the nail fold area based on the blood circulation characteristic information of all abnormal blood vessel parts; and analyze the microcirculation model to predict the evolution result of the microcirculation abnormality in the nail fold area, including:

[0105] Based on the blood circulation flow characteristic information of all abnormal blood vessels and their distribution position information in the nail fold area, a microcirculation model of the nail fold area is generated. The microcirculation model is analyzed to predict the blood vessel position and time of microcirculation blockage in the nail fold area, which is used as the result of the abnormal microcirculation evolution.

[0106] In the above technical solution, based on the generation time information corresponding to all abnormal vascular part sub-images contained in each abnormal vascular part sub-image set, all abnormal vascular part sub-images are subjected to time evolution analysis to obtain the flow velocity information of different blood components inside the corresponding abnormal vascular part (such as the flow velocity of red blood cells, white blood cells and platelets), and the blood circulation situation inside the abnormal vascular part is refined and identified. In addition, based on the blood circulation flow characteristic information of all abnormal vascular parts and the distribution position information in the nail fold area, a microcirculation model of the nail fold area is generated, so as to predict the blood vessel position and time of microcirculation blockage in the nail fold area, so as to serve as the abnormal evolution result of the microcirculation, so as to fully and correctly reflect the real microcirculation state of the nail fold area and improve the credibility of nail fold microcirculation identification.

[0107] It can be seen from the contents of the above embodiments that the nail fold microcirculation imaging control and image recognition and analysis method and system determine the attribute information of the abnormal illumination area based on the scattering characteristic information of the imaging illumination light detection of the nail fold area, thereby performing illumination light projection modulation to ensure that the entire nail fold area can be properly illuminated to avoid interference with subsequent image recognition; image recognition is performed on the nail fold area to obtain a capillary image set of the target sub-area, thereby realizing directional image acquisition of the capillaries inside the nail fold area; vascular morphology recognition is performed on all capillary images, and the abnormal vascular parts are calibrated, thereby classifying and obtaining a sub-image set of the abnormal vascular parts, so as to facilitate subsequent recognition of only the abnormal vascular parts, thereby reducing the recognition workload and the error recognition rate; and time evolution analysis is also performed on the sub-image set of the abnormal vascular parts to obtain blood circulation characteristic information, thereby generating a microcirculation model of the nail fold area, realizing global microcirculation characterization of the nail fold area, and predicting the abnormal evolution results of the microcirculation, comprehensively and correctly reflecting the real microcirculation state of the nail fold area, and improving the credibility of nail fold microcirculation recognition.

[0108] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for controlling nail fold microcirculation imaging and image recognition and analysis, characterized in that: It includes the following steps: Step S1, performing scattering detection of imaging illumination light on the nail fold area to obtain scattering characteristic information of the nail fold area to the imaging illumination light; determining illumination abnormal area attribute information of the nail fold area based on the scattering characteristic information; and performing illumination light projection modulation based on the illumination abnormal area attribute information; Step S2, performing a first image recognition on the nail fold area to obtain target sub-area distribution feature information of the nail fold area; based on the target sub-area distribution feature information, performing a second image recognition on the nail fold area to obtain a capillary image set corresponding to the target sub-area, including scanning and photographing the nail fold area under the illumination of the projected modulated illumination light to obtain a global image of the nail fold area; performing pixel texture recognition on the global image to obtain respective distribution position information of all dermal papilla areas under the nail fold area, which is used as the target sub-area distribution feature information; based on the respective distribution position information of all dermal papilla areas, performing dynamic directional photography on each dermal papilla area to obtain a dynamic image corresponding to each dermal papilla area; performing frame recognition on the dynamic image to obtain a number of capillary images corresponding to all capillaries under each dermal papilla area at different times, thereby forming a capillary image set; Step S3, performing vascular morphology recognition on all capillary images in the capillary image set to obtain vascular contour feature information corresponding to all capillary images; Based on the blood vessel contour feature information, each capillary image is calibrated for abnormal blood vessel locations; then all abnormal blood vessel locations under all capillary images are classified to obtain a plurality of abnormal blood vessel location sub-image sets; Step S4, performing time-evolution analysis on each abnormal blood vessel part sub-image set to obtain blood circulation characteristic information of the corresponding abnormal blood vessel part; Based on the blood circulation characteristic information of all abnormal vascular parts, a microcirculation model of the nail fold area is generated; and the microcirculation model is analyzed to predict the abnormal microcirculation evolution result of the nail fold area, including performing time evolution analysis on all abnormal vascular part sub-images based on the generation time information corresponding to all abnormal vascular part sub-images contained in each abnormal vascular part sub-image set to obtain the blood circulation flow characteristic information of the corresponding abnormal vascular part; wherein the blood circulation flow characteristic information includes the flow velocity information of different blood components inside the corresponding abnormal vascular part; based on the blood circulation flow characteristic information of all abnormal vascular parts and the distribution position information in the nail fold area, a microcirculation model of the nail fold area is generated; and the microcirculation model is analyzed to predict the blood vessel position and time of microcirculation blockage in the nail fold area as the abnormal microcirculation evolution result.

2. The nailfold microcirculation imaging control and image recognition and analysis method according to claim 1, characterized in that: In the step S1, the scattering detection of the imaging illumination light is performed on the nail fold area to obtain the scattering characteristic information of the nail fold area to the imaging illumination light; Based on the scattering characteristic information, determining the illumination abnormal area attribute information of the nail fold area; Based on the attribute information of the abnormal lighting area, illumination light projection modulation is performed, including: Performing visual detection on the scattered light formed by the nail fold area under the projection of the reference imaging illumination light to obtain a scattered light distribution image; analyzing the scattered light distribution image to obtain scattered light distribution characteristic information of the spatial area where the nail fold area is located; wherein the scattered light distribution characteristic information includes scattered light intensity distribution information of the spatial area at different orientations; Based on the scattered light distribution characteristic information, all sub-regions under the nail fold area that form abnormal scattering intensity for the reference imaging illumination light are determined, and the distribution position information of all sub-regions in the nail fold area is used as the attribute information of the abnormal illumination area; based on the attribute information of the abnormal illumination area, the intensity of the area corresponding to the projection cross-section of the reference imaging illumination light is modulated.

3. The nailfold microcirculation imaging control and image recognition and analysis method according to claim 1, characterized in that: In the step S3, all capillary images under the capillary image set are subjected to vascular morphology recognition to obtain vascular contour feature information corresponding to each of the capillary images; based on the vascular contour feature information, abnormal vascular locations are calibrated for each capillary image; and all abnormal vascular locations under all capillary images are classified to obtain a plurality of abnormal vascular location sub-image sets, including: Performing vascular wall shape recognition on all capillary images in the capillary image set to obtain vascular contour feature information corresponding to each of the capillary images; comparing the vascular contour feature information with reference contour feature information to obtain bending curvature offset information of the capillary; and calibrating abnormal vascular locations on the capillary images based on the bending curvature offset information; Based on the position information of all abnormal blood vessel parts under all capillary images, the abnormal blood vessel parts in the same position are classified and processed to obtain several abnormal blood vessel part sub-image sets; wherein, all the abnormal blood vessel part sub-images contained in each abnormal blood vessel part sub-image set correspond to the same abnormal blood vessel part.

4. Nailfold microcirculation imaging control and image recognition and analysis system, characterized in that: include: A nail fold area scattering feature recognition module is used to detect the scattering of the imaging illumination light in the nail fold area to obtain scattering feature information of the nail fold area to the imaging illumination light; An illumination light modulation module, used to determine the illumination abnormality region attribute information of the nail fold region based on the scattering characteristic information; and perform illumination light projection modulation based on the illumination abnormality region attribute information; The first image recognition module is used to perform first image recognition on the nail fold area to obtain target sub-area distribution feature information of the nail fold area, including scanning and photographing the nail fold area under the illumination of the projected modulated illumination light to obtain a global image of the nail fold area; performing pixel texture recognition on the global image to obtain respective distribution position information of all dermal papilla areas under the nail fold area, and using this as the target sub-area distribution feature information; The second image recognition module is used to perform second image recognition on the nail fold area based on the target sub-area distribution feature information to obtain a capillary image set corresponding to the target sub-area, including dynamically directional shooting of each dermal papilla area based on the respective distribution position information of all dermal papilla areas to obtain a dynamic image corresponding to each dermal papilla area; performing frame recognition on the dynamic image to obtain a plurality of capillary images corresponding to all capillaries under each dermal papilla area at different times, thereby forming a capillary image set; A blood vessel contour recognition module is used to perform blood vessel morphology recognition on all capillary images in the capillary image set to obtain blood vessel contour feature information corresponding to all capillary images; The abnormal blood vessel part identification module is used to calibrate the abnormal blood vessel part of each capillary image based on the blood vessel contour feature information; and then classify all abnormal blood vessel parts under all capillary images to obtain a plurality of abnormal blood vessel part sub-image sets; A blood circulation feature recognition module is used to perform a time evolution analysis on each abnormal blood vessel part sub-image set to obtain blood circulation feature information of the corresponding abnormal blood vessel part, including performing a time evolution analysis on all abnormal blood vessel part sub-images based on the generation time information corresponding to all abnormal blood vessel part sub-images contained in each abnormal blood vessel part sub-image set to obtain blood circulation flow feature information of the corresponding abnormal blood vessel part; wherein the blood circulation flow feature information includes flow velocity information of different blood components inside the corresponding abnormal blood vessel part; The microcirculation abnormality prediction module is used to generate a microcirculation model of the nail fold area based on the blood circulation characteristic information of all abnormal blood vessel parts; and analyze the microcirculation model to predict the evolution result of the microcirculation abnormality in the nail fold area, including generating the microcirculation model of the nail fold area based on the blood circulation flow characteristic information of all abnormal blood vessel parts and the distribution position information in the nail fold area; analyze the microcirculation model to predict the blood vessel position and time of microcirculation blockage in the nail fold area as the evolution result of the microcirculation abnormality.

5. The nailfold microcirculation imaging control and image recognition and analysis system according to claim 4, characterized in that: The nail fold area scattering feature recognition module is used to perform scattering detection of the imaging illumination light in the nail fold area to obtain scattering feature information of the imaging illumination light in the nail fold area, including: Performing visual detection on the scattered light formed by the nail fold area under the projection of the reference imaging illumination light to obtain a scattered light distribution image; analyzing the scattered light distribution image to obtain scattered light distribution characteristic information of the spatial area where the nail fold area is located; wherein the scattered light distribution characteristic information includes scattered light intensity distribution information of the spatial area at different orientations; The illumination light modulation module is used to determine the illumination abnormal area attribute information of the nail fold area based on the scattering characteristic information; and perform illumination light projection modulation based on the illumination abnormal area attribute information, including: Based on the scattered light distribution characteristic information, all sub-regions under the nail fold area that form abnormal scattering intensity for the reference imaging illumination light are determined, and the distribution position information of all sub-regions in the nail fold area is used as the attribute information of the abnormal illumination area; based on the attribute information of the abnormal illumination area, the intensity of the area corresponding to the projection cross-section of the reference imaging illumination light is modulated.

6. The nailfold microcirculation imaging control and image recognition and analysis system according to claim 4, characterized in that: The blood vessel contour recognition module is used to perform blood vessel morphology recognition on all capillary images in the capillary image set to obtain blood vessel contour feature information corresponding to all capillary images, including: Performing vascular wall shape recognition on all capillary images in the capillary image set to obtain vascular contour feature information corresponding to all capillary images; The abnormal blood vessel location identification module is used to calibrate the abnormal blood vessel location of each capillary image based on the blood vessel contour feature information; and then classify all abnormal blood vessel locations under all capillary images to obtain a plurality of abnormal blood vessel location sub-image sets, including: Comparing the blood vessel contour feature information with the reference contour feature information to obtain the bending curvature offset information of the capillary; and calibrating the abnormal blood vessel position of the capillary image based on the bending curvature offset information; Based on the position information of all abnormal blood vessel parts under all capillary images, the abnormal blood vessel parts in the same position are classified and processed to obtain several abnormal blood vessel part sub-image sets; wherein, all the abnormal blood vessel part sub-images contained in each abnormal blood vessel part sub-image set correspond to the same abnormal blood vessel part.

Images