A data processing system based on infrared images

Abstract

The present invention provides a data processing system based on infrared images. First, the infrared image to be processed is compared with the corresponding normal infrared image model to determine the associated abnormality type. Then, the infrared image to be processed is compared with the infrared image model corresponding to the associated abnormality type to obtain the correlation value between the infrared image to be processed and each associated abnormality type. Therefore, it is possible to know not only whether there is an abnormality in the infrared image to be processed, but also the type of the abnormality.

Description

Technical Field

[0001] The present invention relates to the general field of image data processing or generation technology, and in particular to a data processing system based on infrared images. Background Art

[0002] At present, with the widespread application of infrared thermal imaging technology in the medical field, the target infrared image is generally compared with the standard infrared image data set to determine whether the target infrared image has any abnormality. However, this comparison can only know that there is an abnormality, but cannot know what kind of abnormality it is. Summary of the invention

[0003] In view of the above technical problems, the technical solution adopted by the present invention is:

[0004] An embodiment of the present invention provides a data processing system based on infrared images, the system comprising a processor and a database and a memory connected to the processor in communication, the memory storing a computer program and a first abnormal type association table and a second abnormal type association table, the database comprising a normal image library and an abnormal image library, the normal image library storing n normal image groups divided according to grouping information, the grouping information at least comprising age range, gender, region and shooting time type, the grouping information corresponding to any two normal image groups is different, wherein each normal image group comprises m normal infrared images shot according to m body positions; the abnormal image The image library stores Q abnormal image sub-libraries divided according to abnormality types, each abnormal image sub-library includes k abnormal image folders divided according to abnormality levels, each abnormal image folder includes n abnormal image groups divided according to grouping information, and the grouping information corresponding to any two abnormal image groups is different, wherein each abnormal image group includes m abnormal infrared images taken according to m body parts; each row of the first abnormal type association table includes the ID, location information and the corresponding associated abnormal type ID of the corresponding acupuncture point, and each row of the second abnormal type association table includes the ID, location information and the corresponding associated abnormal type ID of the corresponding region of interest.

[0005] The processor is used to execute a computer program to implement the following steps:

[0006] S100, obtaining an infrared image group to be processed and corresponding basic information, and based on the basic information corresponding to the infrared image group to be processed, obtaining a corresponding normal image group from the normal image library as a target normal image group; the infrared image group to be processed includes m infrared images, and the basic information includes the age, gender, and region of the object corresponding to the infrared image to be processed and the shooting time corresponding to the infrared image group to be processed.

[0007] S200, comparing the temperature within the target area of ​​the i-th infrared image to be processed in the infrared image group to be processed with the temperature within the target area of ​​the corresponding normal infrared image in the target normal image group to obtain pixel points with abnormal temperature within the target area of ​​the i-th infrared image to be processed, and obtaining an abnormal pixel point set; the value of i ranges from 1 to m.

[0008] S300, based on the abnormal pixel point set of the i-th infrared image to be processed, obtain the corresponding abnormal area A i ={A i1 , A i2 , ..., A ij , ..., A if(i)}, A ij is the jth abnormal region in the target region of the i-th infrared image to be processed, j ranges from 1 to f(i), and f(i) is the number of abnormal regions in the target region of the i-th infrared image to be processed;

[0009] S400, based on A ij The corresponding location information and the location information of the acupuncture points and the region of interest are obtained from the first abnormality type association table and the second abnormality type association table, respectively, as A ij The associated exception type ET ij ={ET 1 ij , E.T. 2 ij ,……,ET r ij ,……,ET h(i，j) ij}, and associate the exception type with the union ET i =ET i1 ∪ET i2 ∪……ET ij ∪……ET if(i) As the target associated anomaly type of the i-th infrared image to be processed; ET r ij A ij The ID of the rth associated exception type, r ranges from 1 to h(i, j), where h(i, j) is A ij The number of associated exception types.

[0010] S500, compare the i-th infrared image to be processed with the corresponding abnormal infrared images in the abnormal image group corresponding to the corresponding target associated abnormal type, and obtain a sorted table of associated abnormal information of the i-th infrared image to be processed; wherein each associated abnormal information includes a corresponding abnormal type, abnormal level and association value.

[0011] S600 , visually displaying the associated abnormal information sorting table corresponding to the m infrared images in the infrared image group to be processed.

[0012] The present invention has at least the following beneficial effects:

[0013] The infrared image-based data processing system provided by the embodiment of the present invention first compares the infrared image to be processed with the corresponding normal infrared image to determine the associated abnormality type, and then compares the infrared image to be processed with the infrared images corresponding to the associated abnormality types to obtain the correlation value between the infrared image to be processed and each associated abnormality type, thereby not only knowing whether there is an abnormality in the infrared image to be processed, but also knowing the type of the abnormality.

[0014] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0016] Figure 1 A flowchart of a computer program executed by an infrared image-based data processing system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0017] 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 those skilled in the art without creative work are within the scope of protection of the present invention.

[0018] An embodiment of the present invention provides a data processing system based on infrared images. The system includes a processor, and a database and a memory that are communicatively connected to the processor.

[0019] The memory stores a computer program and a first abnormality type association table and a second abnormality type association table, wherein each row of the first abnormality type association table includes the ID of the corresponding acupuncture point, location information and the ID of the corresponding associated abnormality type, and each row of the second abnormality type association table includes the ID of the corresponding area of ​​interest, location information and the ID of the corresponding associated abnormality type.

[0020] In the embodiment of the present invention, the acupoints are divided based on the theory of traditional Chinese medicine, and can be set based on actual needs, for example, they can include head and face acupoints, conception meridian acupoints, trunk acupoints, upper limb acupoints, lower limb acupoints and foot acupoints, etc. The region of interest can be divided into human body regions based on actual needs, for example, it can be divided based on Western medical human anatomy and traditional Chinese medicine theory, and can be set based on actual needs, for example, the division of internal organs based on traditional Chinese medicine theory and the division of heart, brain, joints, muscles and other regions based on Western medical human anatomy.

[0021] In the embodiment of the present invention, the ID of the acupoint may be the name or mark of the acupoint, and the ID of the region of interest may be the name or mark of the region of interest. The location information of the acupoint may be the location coordinates of the acupoint in the image, specifically the location coordinates of the included pixel points. The location information of the region of interest may be the location coordinates of the region of interest in the image, specifically the location coordinates of the included pixel points.

[0022] In the embodiment of the present invention, the x-axis of the coordinate system where the infrared image is located may be the lateral direction of the human body, the y-axis may be the longitudinal direction of the human body, and the origin of the coordinate system may be the lower left corner of the image.

[0023] In an embodiment of the present invention, the abnormality type may be a type obtained based on the physical function status, in one example, may be a disease type, and the specific disease type may be set according to actual needs. The ID of the abnormality type may be the name or identification of the abnormality type.

[0024] Furthermore, the database includes a normal image library and an abnormal image library, the normal image library stores n normal image groups divided according to grouping information, the grouping information includes at least age range, gender, region and shooting time type, the grouping information corresponding to any two normal image groups is different, wherein each normal image group includes m normal infrared images shot according to m body positions; the abnormal image library stores Q abnormal image sub-libraries divided according to abnormality types, each abnormal image sub-library includes k abnormal image folders divided according to abnormality levels, each abnormal image folder includes n abnormal image groups divided according to grouping information, the grouping information corresponding to any two abnormal image groups is different, wherein each abnormal image group includes m abnormal infrared images shot according to m body positions.

[0025] In an exemplary embodiment of the present invention, the age range includes 0-6 years old, 7-12 years old, 13-15 years old, 16-18 years old, 19-35 years old, 36-45 years old, 46-55 years old, 56-65 years old, 66-75 years old, 76-85 years old and 86-100 years old; the gender includes male and female; the region may include East China, South China, North China, Central China, Southwest China, Northwest China and Northeast China; the shooting time type is obtained based on any one of the four seasons, 24 solar terms and 72 Hous, and the specific division method can be set based on actual needs. For example, if it is divided based on the four seasons, the shooting time type includes four types: spring, summer, autumn and winter. If divided based on the 24 solar terms, the shooting time types include 24 types, namely, Beginning of Spring, Rain Water, Waking of Insects, Spring Equinox, Tomb-Sweeping Day, Grain Rain, Beginning of Summer, Grain Full, Grain in Ear, Summer Solstice, Minor Heat, Major Heat, Beginning of Autumn, End of Heat, White Dew, Autumnal Equinox, Cold Dew, Frost Descent, Beginning of Winter, Minor Snow, Major Snow, Winter Solstice, Minor Cold, and Major Cold. If divided based on the 72 solar terms, the shooting time types include 72 types, namely, 72 solar terms. It is known to those skilled in the art that the specific division method of the 72 solar terms may be the prior art.

[0026] Those skilled in the art know that, when the number of age ranges, genders, regions and shooting time types is determined, the corresponding number of combinations n can be obtained based on the calculation method of permutations and combinations, and any method of determining the corresponding number of combinations n based on the number of age ranges, genders, regions and shooting time types falls within the scope of protection of the present invention.

[0027] In an embodiment of the present invention, each normal image group can be obtained based on W corresponding normal infrared image samples. The specific acquisition method can be the existing technology, for example, it can be obtained based on the method disclosed in patent document CN202211204405.8. Similarly, each abnormal image group can be obtained based on W corresponding abnormal infrared image samples, and the specific acquisition method can be the same as the acquisition method of the normal image group.

[0028] In an illustrative embodiment of the present invention, four body positions may be included, namely, the whole body front position, the whole body back position, the whole body left side position and the whole body right side position. Among them, the whole body front position meets the following conditions: standing upright facing the camera, legs spread as wide as shoulders, eyes looking straight ahead, arms hanging naturally at an angle of 30° with the torso, arms located on both sides of the body, palms facing the camera, and five fingers open. The whole body back position meets the following conditions: standing upright with the back to the camera, legs spread as wide as shoulders, eyes looking straight ahead, arms hanging naturally at an angle of 30° with the torso, arms located on both sides of the body, palms facing away from the camera, and five fingers open. The whole body left side position meets the following conditions: standing in the front position, turning left on the spot; the left leg moves forward half a step, the right leg moves back half a step, the left arm extends forward 45°, the right arm swings back 45°, the palms of both hands face inward, and the five fingers open. The right side of the whole body meets the following conditions: standing in front, turning right on the spot; right leg forward half a step, left leg back half a step; right arm extended 45 degrees, left arm swung back 45 degrees, palms facing inward, fingers open. Image requirements: portrait centered, symmetrical, clear focus, complete display. The distance from both feet to the bottom edge of the scanning display frame is no less than 1 cm.

[0029] In an embodiment of the present invention, the abnormality level can be divided based on the abnormality degree of the abnormality type. For example, it can be divided based on the severity of the disease, and may include mild, mild-moderate, moderate, moderate-severe, severe and other levels. It may also include level one, level two, level three, level four and other levels arranged in order from low to high severity.

[0030] Furthermore, the processor is used to execute a computer program to implement the following steps:

[0031] S100, obtaining an infrared image group to be processed and corresponding basic information, and based on the basic information corresponding to the infrared image group to be processed, obtaining a corresponding normal image group from the normal image library as a target normal image group; the infrared image group to be processed includes m infrared images, and the basic information includes the age, gender, region and solar term when the infrared image group to be processed corresponds to the object.

[0032] In the embodiment of the present invention, the infrared image to be processed can be obtained by using an existing infrared shooting device.

[0033] Those skilled in the art know that, based on the basic information corresponding to the infrared image group to be processed, obtaining the corresponding target normal image group from the normal image library may be an existing technology. For example, if the age of the subject corresponding to the infrared image group to be processed is 20 years old, the gender is male, the region is North China, and the corresponding shooting time type is the beginning of spring, then the normal image group corresponding to the grouping information of 19-35 years old, male, Hebei Province, and the beginning of spring is obtained from the normal image library as the target normal image group of the infrared image group to be processed.

[0034] S200, comparing the temperature within the target area of ​​the i-th infrared image to be processed in the infrared image group to be processed with the temperature within the target area of ​​the corresponding normal infrared image in the target normal image group to obtain pixel points with abnormal temperature within the target area of ​​the i-th infrared image to be processed, and obtaining an abnormal pixel point set; the value of i ranges from 1 to m.

[0035] In S200, if the absolute value of the difference between the temperature of a pixel point in the target area of ​​the i-th infrared image to be processed and the temperature of the corresponding pixel point in the target area of ​​the corresponding normal infrared image in the target normal image group is greater than the set temperature threshold, the pixel point is determined to be an abnormal pixel point.

[0036] In the embodiment of the present invention, the target area may be a human body area, which may be identified based on a human body area identification model. The set temperature threshold may be an empirical value.

[0037] S300, based on the abnormal pixel point set of the i-th infrared image to be processed, obtain the corresponding abnormal area A i ={A i1 , A i2 , ..., A ij , ..., A if(i)}, A ij is the jth abnormal area in the target area of ​​the i-th infrared image to be processed, j ranges from 1 to f(i), and f(i) is the number of abnormal areas in the target area of ​​the i-th infrared image to be processed.

[0038] Further, in an exemplary embodiment, S300 may specifically include:

[0039] S301, acquiring isolated pixel points in an abnormal pixel point set of an i-th infrared image to be processed, and deleting the acquired isolated pixel points from the abnormal pixel point set to obtain an intermediate abnormal pixel point set.

[0040] In the embodiment of the present invention, an isolated pixel point refers to an abnormal pixel point that is not directly connected to the pixel point. Specifically, for any pixel point (x, y) in the abnormal pixel point set, if the pixel point (x, y-1), pixel point (x, y+1), pixel point (x-1, y), pixel point (x+1, y), pixel point (x-1, y+1), pixel point (x+1, y+1), pixel point (x-1, y-1), pixel point (x+1, y-1) are not abnormal pixels, then the pixel point is an abnormal pixel point.

[0041] S302, based on the intermediate abnormal pixel set, obtain the corresponding initial abnormal area AI i ={AI i1 , AIi2 ,……,AI is ,……,AI ig(i)}, AI is is the sth initial abnormal region in the target region of the i-th infrared image to be processed, s ranges from 1 to g(i), and g(i) is the number of initial abnormal regions in the target region of the i-th infrared image to be processed.

[0042] Those skilled in the art know that obtaining the corresponding initial abnormal region based on the intermediate abnormal pixel set may be an existing technology. For example, it may specifically include:

[0043] S1, set counter C=1;

[0044] S2, if the current intermediate abnormal pixel set is not empty, the first abnormal pixel in the current intermediate abnormal pixel set is used as the initial pixel; otherwise, the current control program is exited to obtain the corresponding initial abnormal area.

[0045] S3, acquiring pixel points connected to the initial pixel point from the current intermediate abnormal pixel point set to form a first connected pixel point subset.

[0046] S4, obtain the pixel points connected to each pixel point in the f-th connected pixel point from the current intermediate abnormal pixel point set to form the f+1-th connected pixel point subset. If the f+1-th connected pixel point subset is not empty, it means that there may still be connected pixel points, and execute S5. Otherwise, it means that there are no adjacent pixel points, and execute S6. The initial value of f is 1.

[0047] S5, set f=f+1, and execute S4.

[0048] S6, taking the union of the 1st connected pixel subset to the f+1th connected pixel subset as the Cth abnormal area; executing S7;

[0049] S7, set C=C+1, and execute S2.

[0050] S303, if DI is <D0, indicating that the number of abnormal pixels is too small, which may be noise, then AI is As the invalid abnormal area and delete it from the initial abnormal area, the initial abnormal area with the invalid abnormal area deleted is obtained as A i DI is For AI is The corresponding number of pixels, D0 is the set pixel number threshold.

[0051] In the embodiment of the present invention, D0 can be set based on actual needs, and can be an empirical value, for example, 3 to 5. In another exemplary embodiment, D0 can be the number of pixels included in an acupuncture point.

[0052] Furthermore, in another exemplary embodiment of the present invention, S303 may also be implemented based on the following steps:

[0053] S304, if DI is <D0 or W / L<d, indicating that the number of abnormal pixels is too small, or, although the number of abnormal pixels is greater than D0, but the abnormal pixels are strip-shaped, then AI is As the invalid abnormal area and delete it from the initial abnormal area, the initial abnormal area with the invalid abnormal area deleted is obtained as A i DI is For AI is The number of corresponding pixels, D0 is the set pixel number threshold; W is AI is The width of the bounding box, L is AI is The length of the bounding box, d is the set ratio threshold.

[0054] The technical effect of S304 is that it can eliminate the elongated area that may be caused by shooting reasons, etc., and can further improve the accuracy.

[0055] S400, based on A ij The corresponding location information and the location information of the acupuncture points and the region of interest are obtained from the first abnormality type association table and the second abnormality type association table, respectively, as A ij The associated exception type ET ij ={ET 1 ij , E.T. 2 ij ,……,ET r ij ,……,ET h(i，j) ij}, and associate the exception type with the union ET i =ET i1 ∪ET i2 ∪……ET ij ∪……ET if(i) As the target associated anomaly type of the i-th infrared image to be processed; ET r ij A ij The ID of the rth associated exception type, r ranges from 1 to h(i, j), where h(i, j) is A ij The number of associated exception types.

[0056] S500, compare the i-th infrared image to be processed with the corresponding abnormal infrared images in the abnormal image group corresponding to the corresponding target associated abnormal type, and obtain a sorted table of associated abnormal information of the i-th infrared image to be processed; wherein each associated abnormal information includes a corresponding abnormal type, abnormal level and association value.

[0057] Furthermore, S500 may specifically include:

[0058] S501, obtaining an abnormal infrared image IMG corresponding to the p-th abnormal image group corresponding to the e-th abnormal level corresponding to the c-th target-associated abnormal type in the target-associated abnormal type corresponding to the i-th infrared image to be processed p i Abnormal area A corresponding to ce p i ce; the value of c is 1 to y(i), y(i) is the number of target-associated anomaly types of the i-th infrared image to be processed, the value of e is 1 to k, and the value of p is 1 to n.

[0059] For each abnormal infrared image, a corresponding abnormal area is preset.

[0060] S502, based on A p i ce, obtain the corresponding area from the i-th infrared image to be processed as the reference abnormal area AR of the i-th infrared image to be processed i ={AR i1 , A.R. i2 ,……,AR iu ,……,AR iz};AR iu is the u-th reference abnormal region of the i-th infrared image to be processed, the value of u ranges from 1 to z, and z is the number of reference abnormal regions.

[0061] S503, obtain A ij and AR iu The overlap area AC ju i , if AC ju i If it is not empty, AC ju i As the target overlap area, and obtain AC ju i with AR iu The area ratio S ju i , get the target overlap area set of the i-th image to be processed and the overlap area with IMG p i ce associated area ratio set S pi ce={S p1 i ce, S p2 i ce,……,S pv i ce,……,S pH i ce},S pv i ce is S p i The vth area ratio in ce, v ranges from 1 to H, H is S p i The number of area ratios in ce.

[0062] In the embodiment of the present invention, A ij and AR iu The intersection of the pixels is taken as AC ju i .AC ju i with AR iu The area ratio S ju i AC ju i The corresponding number of pixels and AR iu The ratio of the corresponding number of pixels.

[0063] S504, based on the target overlap region set of the i-th infrared image to be processed, in IMG p i ce to obtain the corresponding regions as the comparison region set.

[0064] It is known to those skilled in the art that any target overlap region set based on the i-th infrared image to be processed, in IMG p i ce to obtain the corresponding area as a comparison area set all belong to the protection scope of the present invention.

[0065] S505, obtaining the difference between the characteristic temperature corresponding to each area in the target overlap area set and the characteristic temperature corresponding to the corresponding comparison area, and obtaining the difference between the i-th image to be processed and IMG p i ce associated temperature difference set △T p i ce={△T p1 i ce, △T p2 i ce,……,△T pa i ce,……,△T pGi ce}, △T pa i ce is △T p i The ath temperature difference in ce, a ranges from 1 to G, G is △T p i The amount of temperature difference in ce.

[0066] In the embodiment of the present invention, the characteristic temperature may be the highest temperature, the lowest temperature, the average temperature, the temperature difference, etc. Preferably, it may be the average temperature and the temperature difference.

[0067] S506, respectively for S p i ce and △T pa i ce is normalized to obtain the normalized area ratio set SN p i ce={SN p1 i ce, SN p2 i ce,……,SN pv i ce,……,SN pH i ce} and the normalized temperature difference set △TN pa i ce={△TN p1 i ce,△TN p2 i ce,……,△TN pa i ce,……,△TN pG i ce}, SN pv i ce is S pv i Normalized value of ce, △TN pa i ce is △T pa i Normalized value of ce.

[0068] In the embodiment of the present invention, any existing normalization method can be used to normalize S p i ce and △T pa i ce is normalized.

[0069] S507, obtaining the correlation value R p i ce=k1×Sum(SN pi ce)+k2×Sum(△TN pa i ce) as the i-th image to be processed and IMG p i ce, k1 and k2 are the first setting coefficient and the second setting coefficient respectively, Sum(SN p i ce) is the area ratio and value, Sum(SN p i ce)=SN p1 i ce+SN p2 i ce+……+SN pv i ce+……+SN pH i ce,Sum(△TN pa i ce) is the temperature difference and value, Sum(△TN pa i ce)=△TN p1 i ce+△TN p2 i ce+……+△TN pa i ce+……+△TN pG i ce.

[0070] In the embodiment of the present invention, k1+k2=1, k1 and k2 can be determined based on the importance of the area ratio and the temperature difference, the more important, the larger the corresponding coefficient. In an exemplary embodiment, k1=k2=0.5.

[0071] S508, sorting the correlation values ​​corresponding to all target abnormality types corresponding to the i-th infrared image to be processed in descending order to obtain a corresponding correlation value sorting table, and obtaining a correlation abnormality information sorting table corresponding to the i-th infrared image to be processed based on the corresponding correlation value sorting table.

[0072] S600, outputting corresponding prompt information based on a sorting table of associated abnormal information corresponding to m infrared images in the infrared image group to be processed.

[0073] Furthermore, S600 may specifically include:

[0074] S601, adding the correlation values ​​corresponding to the associated abnormality information having the same abnormality type and abnormality level in the associated abnormality information ranking table corresponding to the m infrared images to be processed, so as to merge the associated abnormality information ranking table of the m infrared images to be processed, and obtain a merged associated abnormality information ranking table;

[0075] For example, if the associated abnormal information sorting table corresponding to each infrared image to be processed includes abnormal information of abnormal type B and abnormal level D, the association values ​​corresponding to the abnormal information are added as the new association value of the abnormal information.

[0076] S602, sorting the merged associated abnormal information sorting table in descending order of the degree of association, and using the sorted associated abnormal information sorting table as the target associated abnormal information sorting table, and visually displaying it.

[0077] In the embodiment of the present invention, the specific display mode can be set based on actual needs, for example, the top K abnormal information in the target-related abnormal information ranking table are displayed in the first set color, and the remaining abnormal information is displayed in the second color, etc. The first set color and the second set color are different colors.

[0078] Furthermore, the abnormal information ranked first can also be flashed.

[0079] In summary, the infrared image-based data processing system provided by the embodiment of the present invention first compares the infrared image to be processed with the corresponding normal infrared image to determine the associated abnormality type, and then compares the infrared image to be processed with the infrared images corresponding to the associated abnormality types to obtain the correlation value between the infrared image to be processed and each associated abnormality type, thereby not only knowing whether there is an abnormality in the infrared image to be processed, but also knowing the type of the abnormality.

[0080] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution disclosed in the present invention can be achieved, and this document does not limit this.

[0081] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A data processing system based on infrared images, characterized in that: The system includes a processor, a database and a memory connected to the processor in communication, the memory stores a computer program and a first abnormal type association table and a second abnormal type association table, the database includes a normal image library and an abnormal image library, the normal image library stores n normal image groups divided according to grouping information, the grouping information at least includes age range, gender, region and shooting time type, and the grouping information corresponding to any two normal image groups is different, wherein each normal image group includes m normal infrared images shot according to m body parts; the abnormal image library stores Q abnormal image sub-libraries divided according to abnormal types, each abnormal image sub-library includes k abnormal image folders divided according to abnormal levels, each abnormal image folder includes n abnormal image groups divided according to grouping information, and the grouping information corresponding to any two abnormal image groups is different, wherein each abnormal image group includes m abnormal infrared images shot according to m body parts; each row of the first abnormal type association table includes the ID of the corresponding acupuncture point, the position information and the ID of the corresponding associated abnormal type, and each row of the second abnormal type association table includes the ID of the corresponding region of interest, the position information and the ID of the corresponding associated abnormal type; The processor is used to execute a computer program to implement the following steps: S100, obtaining an infrared image group to be processed and corresponding basic information, and based on the basic information corresponding to the infrared image group to be processed, obtaining a corresponding normal image group from the normal image library as a target normal image group; the infrared image group to be processed includes m infrared images, and the basic information includes the age, gender, and region of the object corresponding to the infrared image to be processed and the shooting time corresponding to the infrared image group to be processed; S200, comparing the temperature in the target area of ​​the i-th infrared image to be processed in the infrared image group to be processed with the temperature in the target area of ​​the corresponding normal infrared image in the target normal image group to obtain pixel points with abnormal temperature in the target area of ​​the i-th infrared image to be processed, and obtaining an abnormal pixel point set; the value of i ranges from 1 to m; S300, based on the abnormal pixel point set of the i-th infrared image to be processed, obtain the corresponding abnormal area A i ={A i1 , A i2 , ..., A ij , ..., A if（i） }, A ij is the jth abnormal region in the target region of the i-th infrared image to be processed, j ranges from 1 to f(i), and f(i) is the number of abnormal regions in the target region of the i-th infrared image to be processed; S400, based on A ij The corresponding location information and the location information of the acupuncture points and the region of interest are obtained from the first abnormality type association table and the second abnormality type association table, respectively, as A ij The associated exception type ET ij ={ET 1 ij , E.T. 2 ij ,……,ET r ij ,……,ET h（i，j） ij }, and associate the exception type with the union ET i =ET i1 ∪ET i2 ∪……ET ij ∪……ET if(i) As the target associated anomaly type of the i-th infrared image to be processed; ET r ij A ij The ID of the rth associated exception type, r ranges from 1 to h(i, j), h(i, j) is A ij The number of associated exception types; S500, comparing the i-th infrared image to be processed with the corresponding abnormal infrared images in the abnormal image group corresponding to the corresponding target associated abnormal type, and obtaining a sorted table of associated abnormal information of the i-th infrared image to be processed; wherein each associated abnormal information includes a corresponding abnormal type, an abnormal level and a correlation value; S600 , visually displaying the associated abnormal information sorting table corresponding to the m infrared images in the infrared image group to be processed.

2. The system according to claim 1, characterized in that In S200, if the absolute value of the difference between the temperature of a pixel point in the target area of ​​the i-th infrared image to be processed and the temperature of the corresponding pixel point in the target area of ​​the corresponding normal infrared image in the target normal image group is greater than the set temperature threshold, the pixel point is determined to be an abnormal pixel point.

3. The system according to claim 1, characterized in that S300 specifically includes: S301, acquiring isolated pixel points in an abnormal pixel point set of an i-th infrared image to be processed, and deleting the acquired isolated pixel points from the abnormal pixel point set to obtain an intermediate abnormal pixel point set; S302, based on the intermediate abnormal pixel set, obtain the corresponding initial abnormal area AI i ={AI i1 , AI i2 ,……,AI is ,……,AI ig（i） }, AI is is the sth initial abnormal region in the target region of the i-th infrared image to be processed, s ranges from 1 to g(i), and g(i) is the number of initial abnormal regions in the target region of the i-th infrared image to be processed; S303, if DI is <D0, then AI is As the invalid abnormal area and delete it from the initial abnormal area, the initial abnormal area with the invalid abnormal area deleted is obtained as A i DI is For AI is The corresponding number of pixels, D0 is the set pixel number threshold.

4. The system according to claim 3, characterized in that S303 is also implemented based on the following steps: S304, if DI is <D0 or W / L<d, then AI is As the invalid abnormal area and delete it from the initial abnormal area, the initial abnormal area with the invalid abnormal area deleted is obtained as A i DI is For AI is The number of corresponding pixels, D0 is the set pixel number threshold; W is AI is The width of the bounding box, L is AI is The length of the bounding box, d is the set ratio threshold.

5. The system according to claim 1, characterized in that S500 specifically includes: S501, obtaining an abnormal infrared image IMG corresponding to the p-th abnormal image group corresponding to the e-th abnormal level corresponding to the c-th target-associated abnormal type in the target-associated abnormal type corresponding to the i-th infrared image to be processed p i Abnormal area A corresponding to ce p i ce; c ranges from 1 to y(i), y(i) is the number of target-related anomaly types of the i-th infrared image to be processed, e ranges from 1 to k, and p ranges from 1 to n; S502, based on A p i ce, obtain the corresponding area from the i-th infrared image to be processed as the reference abnormal area AR of the i-th infrared image to be processed i ={AR i1 , A.R. i2 ,……,AR iu ,……,AR iz };AR iu is the u-th reference abnormal region of the i-th infrared image to be processed, where u ranges from 1 to z and z is the number of reference abnormal regions; S503, obtain A ij and AR iu The overlap area AC ju i , if AC ju i If it is not empty, AC ju i As the target overlap area, and obtain AC ju i with AR iu The area ratio S ju i , get the target overlap area set of the i-th image to be processed and the overlap area with IMG p i ce associated area ratio set S p i ce={S p1 i ce, S p2 i ce,……,S pv i ce,……,S pH i ce},S pv i ce is S p i The vth area ratio in ce, v ranges from 1 to H, H is S p i The number of area ratios in ce; S504, based on the target overlap region set of the i-th infrared image to be processed, in IMG p i Get the corresponding area in ce as the comparison area set; S505, obtaining the difference between the characteristic temperature corresponding to each area in the target overlap area set and the characteristic temperature corresponding to the corresponding comparison area, and obtaining the difference between the i-th image to be processed and IMG p i ce associated temperature difference set △T p i ce={△T p1 i ce, △T p2 i ce,……,△T pa i ce , ..., △T pG i ce }, △T pa i ce △T p i The ath temperature difference in ce, a ranges from 1 to G, G is △T p i The number of temperature differences in ce; S506, respectively S p i ce and △T pa i ce Perform normalization to obtain the normalized area ratio set SN p i ce={SN p1 i ce, SN p2 i ce,……,SN pv i ce,……,SN pH i ce} and the normalized temperature difference set △TN pa i ce ={△TN p1 i ce,△TN p2 i ce,……,△TN paice ,……,△TN pG i ce }, SN pv i ce is S pv i Normalized value of ce, △TN pa i ce △T pa i ce The normalized value of S507, obtaining the correlation value R p i ce=k1×Sum(SN p i ce)+k2×Sum(△TN pa i ce ) as the i-th image to be processed and IMG p i ce, k1 and k2 are the first setting coefficient and the second setting coefficient respectively, Sum (SN p i ce) is the area ratio and value, Sum(SN p i ce)=SN p1 i ce+SN p2 i ce+……+SN pv i ce+……+SN pH i ce, Sum (△TN pa i ce ) is the temperature difference and value, Sum (△TN pa i ce ) = △TN p1 i ce+△TN p2 i ce+……+△TN pa i ce +……+△TN pG i ce ; S508, sorting the correlation values ​​corresponding to all target abnormality types corresponding to the i-th infrared image to be processed in descending order to obtain a corresponding correlation value sorting table, and obtaining a correlation abnormality information sorting table corresponding to the i-th infrared image to be processed based on the corresponding correlation value sorting table.

6. The system according to claim 5, characterized in that S600 specifically includes: S601, adding the correlation values ​​corresponding to the associated abnormality information having the same abnormality type and abnormality level in the associated abnormality information ranking table corresponding to the m infrared images to be processed, so as to merge the associated abnormality information ranking table of the m infrared images to be processed, and obtain a merged associated abnormality information ranking table; S602, sorting the merged associated abnormal information sorting table in descending order of the degree of association, and using the sorted associated abnormal information sorting table as the target associated abnormal information sorting table, and visually displaying it.

7. The system according to claim 1, characterized in that The age ranges include 0-6 years old, 7-12 years old, 13-15 years old, 16-18 years old, 19-35 years old, 36-45 years old, 46-55 years old, 56-65 years old, 66-75 years old, 76-85 years old and 86-100 years old; the gender includes male and female; the regions include East China, South China, North China, Central China, Southwest China, Northwest China and Northeast China; the shooting time type is obtained based on any one of the four seasons, 24 solar terms and 72 Hou.

8. The system according to claim 1, characterized in that It includes 4 body positions, namely, the whole body anterior position, the whole body posterior position, the whole body left position and the whole body right position.

Images