An intelligent detection, analysis and management system for red blood cell in-and-out storage based on machine vision

The red blood cell intelligent detection system addresses manual subjectivity and automation challenges by employing machine vision and robotic clamping for accurate and efficient quality assessment and sorting.

CN118649910BActive Publication Date: 2024-12-06SUZHOU YIMAN BIOTECHNOLOGY CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411147313.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-21
Publication Date
2024-12-06
Estimated Expiration
2044-08-21

AI Technical Summary

Technical Problem

Existing red blood cell storage and warehousing management relies heavily on manual inspection, leading to subjective judgments, irregular storage of bags, and difficulty in automated detection due to bag shape irregularity and damage concerns.

Method used

A red blood cell intelligent detection and analysis system using machine vision, incorporating a scanning information entry module, image data acquisition, feature detection and analysis, qualification assessment, and sorting module, along with robotic clamping and visual positioning to automate the detection process.

Benefits of technology

Improves the accuracy and consistency of red blood cell quality detection, reduces subjectivity, and enhances the efficiency of outbound management by using visual cameras and robotic manipulation, while facilitating automated clamping and image-based evaluation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118649910B_ABST
    Figure CN118649910B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of red blood cell entry and exit detection and analysis management, and specifically discloses a red blood cell entry and exit intelligent detection and analysis management system based on machine vision. The present invention clamps a red blood cell box to a code scanning position by a manipulator gripper, scans a barcode of a red blood cell blood bag in the red blood cell box, and after completing the code scanning, clamps the red blood cell box to an image acquisition position by a manipulator gripper, acquires an image of the red blood cell blood bag in the red blood cell box, binds the image of the red blood cell blood bag with the barcode, and keeps a photo as a backup, which is helpful to solve later detection deviation disputes; according to the image of the red blood cell blood bag, extracts the outer packaging characteristics and inner red blood cell sample characteristics of the red blood cell blood bag, judges whether the red blood cell blood bag is qualified, and sorts the red blood cell blood bag by a manipulator gripper, thereby realizing automated red blood cell quality detection, reducing the subjectivity and randomness of manual judgment, and thus improving the efficiency and level of red blood cell entry and exit detection management.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of red blood cell in-and-out warehouse detection, analysis and management, and in particular to a red blood cell in-and-out warehouse intelligent detection, analysis and management system based on machine vision. Background Art

[0002] The entry and exit of red blood cells is a key link in blood management, which mainly refers to the entire process of storage, distribution and use of blood products, especially red blood cells. This process is of great significance to ensure the effective use of medical resources, improve patient treatment effects and ensure medical safety. Therefore, the management of the entry and exit of red blood cells is the key to blood safety and effective use, and plays an indispensable role in improving the quality and efficiency of medical services and protecting public health.

[0003] There are some deficiencies in the existing red blood cell storage and warehousing management: 1. Subjectivity and randomness of manual judgment: The existing red blood cell storage and warehousing management mainly adopts manual inspection, and the qualification of manual inspection of red blood cell samples depends on the personal experience and subjective judgment of the operator. Different operators may have different judgment criteria and observation angles, and lack of unified quantitative standards, which can easily lead to deviations in the results. In addition, the lack of photos also increases the controversy and possibility of disputes in the test results.

[0004] 2. Irregular storage and operational limitations of red blood cell bags: Red blood cell storage bags are usually soft bags with blood braids on top for easy detection and sampling, which results in irregular shapes of the bags when stored, and blood bags are not allowed to be directly clamped or subjected to damaging operations, making automated detection difficult to achieve.

[0005] In summary, the existing red blood cell quality inspection and warehousing management still rely on manual visual operations and cannot be achieved through mechanical or automated systems. Summary of the invention

[0006] In view of the above problems, the present invention proposes a red blood cell entry and exit intelligent detection and analysis management system based on machine vision, which realizes the function of red blood cell entry and exit detection and analysis management.

[0007] The technical solution adopted by the present invention to solve its technical problem is: the present invention provides a red blood cell in-and-out intelligent detection and analysis management system based on machine vision, including: a red blood cell scanning information entry module: used to clamp the red blood cell box in the red blood cell frame on the red blood cell out-of-warehouse conveyor belt to the set scanning position through a robotic arm clamp, and further scan the barcode on the surface of the red blood cell blood bag in the red blood cell box through a scanner, read the basic information of the red blood cells, and enter it into the red blood cell information library.

[0008] Red blood cell image data acquisition module: After the red blood cell bag is scanned, the robot gripper clamps the red blood cell box to the set image acquisition position, and further collects the image set of the red blood cell bag in the red blood cell box through the visual camera, binds the image set of the red blood cell bag with the barcode, and enters it into the red blood cell information library.

[0009] Red blood cell feature detection and analysis module: used to obtain the outer packaging feature information of the red blood cell blood bag and the red blood cell sample feature information in the red blood cell blood bag based on the image set of the red blood cell blood bag, wherein the outer packaging feature information includes the degree of damage of the blood bag and the degree of damage of the label, and the red blood cell sample feature information in the red blood cell blood bag includes the static stratification conformity, blood color conformity and the area of ​​clots, bubbles and flocs in the blood.

[0010] Red blood cell qualification assessment module: used to analyze the qualification index of the red blood cell bag based on the outer packaging characteristic information of the red blood cell bag and the red blood cell sample characteristic information in the red blood cell bag.

[0011] Red blood cell sorting and outbound module: It is used to judge whether the red blood cell bag is qualified according to the qualified index of the red blood cell bag. If it is qualified, the robot gripper grabs the red blood cell box where the red blood cell bag is located for outbound transportation. If it is unqualified, the robot gripper grabs the red blood cell box where the red blood cell bag is located and places it in the designated area.

[0012] Database: used to store historical sample image sets of unqualified red blood cell blood bags.

[0013] Based on the above embodiments, the specific working process of the red blood cell scanning information entry module is: according to the preset principles, the manipulator is arranged on one side of the red blood cell outbound conveyor belt, and the scanner and the visual camera are arranged on the other side of the red blood cell outbound conveyor belt.

[0014] When the red blood cell frame on the red blood cell outbound conveyor belt moves into the set sensing area, the red blood cell outbound conveyor belt pauses, and the robotic gripper clamps the red blood cell box in the red blood cell frame on the red blood cell outbound conveyor belt to the set scanning position, and further scans the barcode on the surface of the red blood cell blood bag in the red blood cell box through the scanner to read the basic information of the red blood cells and enter it into the red blood cell information database.

[0015] On the basis of the above embodiments, the specific working process of the red blood cell image data acquisition module is as follows: after the red blood cell bag completes code scanning, the red blood cell box is clamped to the set image acquisition position by the robot gripper, and the robot gripper is adjusted to various set postures, and further the visual camera is used to collect images of the red blood cell bag in the red blood cell box at various angles, to construct an image set of the red blood cell bag in the red blood cell box, to bind the image set of the red blood cell bag to the barcode, and to enter the red blood cell information library.

[0016] On the basis of the above-mentioned embodiment, the specific working process of the red blood cell feature detection and analysis module includes: extracting a set of historical unqualified red blood cell blood bag sample images stored in the database, screening out red blood cell blood bag sample images with wrinkles on the surface, scratches on the surface and bleeding on the surface, and recording them as wrinkled red blood cell blood bag sample images, scratched red blood cell blood bag sample images and bleeding red blood cell blood bag sample images respectively.

[0017] According to the image set of the red blood cell blood bag, the surface image of the red blood cell blood bag is obtained, and the surface image of the red blood cell blood bag is compared with the wrinkled red blood cell blood bag sample image, the scratched red blood cell blood bag sample image and the bleeding red blood cell blood bag sample image respectively to obtain the total area of ​​the wrinkled area on the red blood cell blood bag surface, the total length of the surface scratches and the total area of ​​the surface bleeding area, which are recorded as , , .

[0018] By analyzing the formula Get the damage degree of the red blood cell blood bag ,in , , They represent the influence factors corresponding to the preset unit wrinkle area, unit scratch length and unit bleeding area, respectively. .

[0019] On the basis of the above-mentioned embodiment, the specific working process of the red blood cell feature detection and analysis module also includes: D1: according to the surface image of the red blood cell blood bag, obtain the red blood cell blood bag surface label image, compare the red blood cell blood bag surface label image with the preset red blood cell blood bag surface label reference image, and obtain the area of ​​each damaged area of ​​the red blood cell blood bag surface label, which is recorded as , Indicates The number of damaged areas, .

[0020] According to the preset principle, the weight factors corresponding to each sub-region of the red blood cell blood bag surface label are set, and the surface label sub-regions to which each damaged region of the red blood cell blood bag surface label belongs are obtained. The weight factors corresponding to each damaged region of the red blood cell blood bag surface label are screened and recorded as .

[0021] D2: Use image processing technology to obtain the grayscale values ​​of the red blood cell blood bag surface label image. If a grayscale value of the red blood cell blood bag surface label image does not belong to the preset reference range of the grayscale value of the red blood cell blood bag surface label image, then the grayscale value is recorded as an abnormal grayscale value. The abnormal grayscale values ​​of the red blood cell blood bag surface label image are counted, and the area of ​​the region where the abnormal grayscale value of the red blood cell blood bag surface label image is located is obtained, and it is recorded as the area of ​​each dirty area of ​​the red blood cell blood bag surface label, and expressed as , Indicates The number of dirty areas, .

[0022] Get the surface label sub-area to which each dirty area of ​​the red blood cell blood bag surface label belongs, filter out the weight factor corresponding to each dirty area of ​​the red blood cell blood bag surface label, and record it as .

[0023] D3: By analyzing the formula:

[0024] , get the label damage degree of the red blood cell bag , , where represent the preset weights of damaged area and dirty area respectively, , , They represent the preset thresholds of damaged area and dirty area respectively.

[0025] On the basis of the above-mentioned embodiment, the specific working process of the red blood cell feature detection and analysis module also includes: according to the image set of the red blood cell blood bag, obtaining the blood static stratification image in the red blood cell blood bag, and comparing it with the preset blood static stratification sample images, to obtain the blood static stratification sample image with the greatest similarity to the blood static stratification image in the red blood cell blood bag, and record it as the matching blood static stratification sample image of the red blood cell blood bag.

[0026] The static stratification conformity corresponding to each blood static stratification sample image is set, and the static stratification conformity corresponding to the matching blood static stratification sample image of the red blood cell blood bag is screened and obtained, which is recorded as the static stratification conformity of the red blood cell blood bag and expressed as .

[0027] Based on the above embodiment, the specific working process of the red blood cell characteristic detection and analysis module also includes: obtaining the blood color of the red blood cell blood bag, setting the corresponding conformity of various blood colors, screening and obtaining the blood color conformity of the red blood cell blood bag, and recording it as .

[0028] The historical unqualified red blood cell blood bag sample image set stored in the database is extracted, and the red blood cell blood bag sample images containing clots, bubbles and flocs in the blood are screened out respectively, and they are recorded as clot red blood cell blood bag sample images, bubble red blood cell blood bag sample images and floc red blood cell blood bag sample images, respectively.

[0029] According to the image set of red blood cell blood bag, the blood image in the red blood cell blood bag is obtained, and compared with the sample images of the red blood cell blood bag with clots, the sample images of the red blood cell blood bag with bubbles, and the sample images of the red blood cell blood bag with flocs, respectively, and the areas of clots, bubbles, and flocs in the blood of the red blood cell blood bag are analyzed and recorded as , , .

[0030] Based on the above embodiment, the specific working process of the red blood cell qualification evaluation module includes: and label damage Substitute into the analysis formula , get the outer packaging qualification coefficient of the red blood cell blood bag ,in Indicates the correction factor of the preset outer packaging qualification coefficient, , They represent the thresholds of the preset blood bag damage degree and label damage degree, respectively. , They represent the weights of the preset blood bag damage degree and label damage degree, .

[0031] The static stratification compliance of the red blood cell blood bag , Blood color conformity and the area of ​​clots, bubbles, and flocs in the blood , , Substitute into the analysis formula , get the qualified coefficient of the red blood cell sample in the red blood cell bag ,in It indicates the correction factor of the qualified coefficient of the red blood cell sample in the preset red blood cell blood bag. represents a natural constant, , , They respectively represent the influencing factors of the preset unit clot area, unit bubble area and unit flocculent area.

[0032] Based on the above embodiment, the specific working process of the red blood cell qualification evaluation module also includes: and the red blood cell sample qualification coefficient in the red blood cell blood bag Substitute into the analysis formula , get the qualified index of red blood cell blood bag ,in , They represent the weights of the preset outer packaging qualification coefficient and the red blood cell sample qualification coefficient in the red blood cell blood bag, respectively. .

[0033] On the basis of the above-mentioned embodiment, the specific working process of the red blood cell sorting and outbound module is as follows: the qualified index of the red blood cell blood bag is compared with the preset qualified index threshold value. If the qualified index of the red blood cell blood bag is greater than the preset qualified index threshold value, the red blood cell blood bag is qualified, and the robot gripper grasps the red blood cell box where the red blood cell blood bag is located for outbound transportation. Otherwise, the red blood cell blood bag is unqualified, and the robot gripper grasps the red blood cell box where the red blood cell blood bag is located and places it in the designated area.

[0034] Compared with the prior art, the intelligent detection, analysis and management system for red blood cell entry and exit based on machine vision described in the present invention has the following beneficial effects: 1. With the help of visual cameras and visual technology, the present invention collects images of red blood cell blood bags and extracts image features to evaluate whether the red blood cell blood bags to be shipped out are qualified and sorted, which can improve the accuracy and consistency of red blood cell sample quality detection, reduce the subjectivity and randomness of manual judgment, and thus improve the level and efficiency of red blood cell outbound detection management. At the same time, the collected images of the red blood cell blood bags are saved for subsequent viewing, which helps to resolve later detection deviation disputes.

[0035] 2. The present invention realizes automatic clamping of red blood cells through a manipulator and a clamping claw, in conjunction with a red blood cell frame, a red blood cell box and visual positioning technology. The red blood cell box realizes the function of standing up the red blood cell blood bag, and the red blood cell frame separates the red blood cell box to facilitate clamping by the manipulator, thereby facilitating automated red blood cell quality detection and improving the efficiency of red blood cell outbound management. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0037] Figure 1 This is a system module connection diagram of the present invention.

[0038] Figure 2 It is a scene schematic diagram of the present invention.

[0039] Figure 3 It is a schematic diagram of the red blood cell box of the present invention.

[0040] Figure numerals: 1. Red blood cell outbound conveyor belt; 2. Red blood cell frame; 3. Red blood cell box; 4. Visual camera; 5. Barcode scanner; 6. Robotic gripper; 7. Robotic arm. DETAILED DESCRIPTION

[0041] 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.

[0042] See also Figure 1-3 As shown, the present invention provides a red blood cell in-and-out intelligent detection and analysis management system based on machine vision, including a red blood cell code scanning information entry module, a red blood cell image data acquisition module, a red blood cell feature detection and analysis module, a red blood cell qualification assessment module, a red blood cell sorting and out-of-warehouse module and a database.

[0043] The red blood cell image data acquisition module is connected to the red blood cell scanning information entry module and the red blood cell feature detection and analysis module respectively, the red blood cell qualification assessment module is connected to the red blood cell feature detection and analysis module and the red blood cell sorting and storage module respectively, and the database is connected to the red blood cell feature detection and analysis module.

[0044] The red blood cell scanning information entry module is used to clamp the red blood cell box in the red blood cell frame on the red blood cell outbound conveyor belt to the set scanning position through the robotic gripper, and further scan the barcode on the surface of the red blood cell blood bag in the red blood cell box through the scanner to read the basic information of the red blood cells and enter it into the red blood cell information library.

[0045] Furthermore, the specific working process of the red blood cell scanning information entry module is: according to the preset principle, the manipulator is arranged on one side of the red blood cell outbound conveyor belt, and the scanner and the visual camera are arranged on the other side of the red blood cell outbound conveyor belt.

[0046] When the red blood cell frame on the red blood cell outbound conveyor belt moves into the set sensing area, the red blood cell outbound conveyor belt pauses, and the robotic gripper clamps the red blood cell box in the red blood cell frame on the red blood cell outbound conveyor belt to the set scanning position, and further scans the barcode on the surface of the red blood cell blood bag in the red blood cell box through the scanner to read the basic information of the red blood cells and enter it into the red blood cell information database.

[0047] As a preferred solution, the red blood cell outbound conveyor belt is paused for a set period of time, and during this period, the robot gripper grasps the red blood cell box in the red blood cell frame to scan the code and capture images in turn.

[0048] As a preferred solution, the set scanning position refers to the position where the robot gripper grabs the red blood cell box and moves to the position where the scanner can clearly scan the barcode on the surface of the red blood cell bag in the red blood cell box.

[0049] In a specific embodiment, the set scanning position is located directly in front of the scanner.

[0050] As a preferred solution, the robot gripper grasps the red blood cell box for scanning by setting the motion trajectory and gripping posture of the robot gripper in the robot control system, wherein the motion trajectory and gripping posture of the robot gripper can be adjusted multiple times until the scanning of the barcode on the surface of the red blood cell bag in the red blood cell box is completed.

[0051] As a preferred solution, the barcode on the surface of the red blood cell bag is unique, and the basic information of the red blood cells is bound to the barcode on the surface of the red blood cell bag when the red blood cells are collected. The barcode contains the basic information of the red blood cells, wherein the basic information of the red blood cells includes but is not limited to: blood station name, blood donation number, product code, blood type, blood variety, blood collection date, preparation date, expiration date, blood donor information and storage conditions, etc.

[0052] As a preferred solution, a label is attached to the surface of the red blood cell bag, and the label also contains basic information of the red blood cells.

[0053] As a preferred solution, the red blood cell box is made of transparent material.

[0054] As a preferred solution, the red blood cell blood bag is a soft bag, and a blood braid is provided on the top of the red blood cell blood bag for convenient detection and sampling.

[0055] As a preferred solution, the red blood cell bag is placed vertically in the red blood cell box with the blood braid facing upward.

[0056] As a preferred solution, the red blood cell box can realize the function of standing up the red blood cell bag, and the red blood cell frame separates the red blood cell box to facilitate the robot to grab it.

[0057] The red blood cell image data acquisition module is used to clamp the red blood cell box to a set image acquisition position through a robotic gripper after the red blood cell blood bag completes code scanning, further collect an image set of the red blood cell blood bag in the red blood cell box through a visual camera, bind the image set of the red blood cell blood bag with the barcode, and enter it into the red blood cell information library.

[0058] Furthermore, the specific working process of the red blood cell image data acquisition module is as follows: after the red blood cell blood bag completes code scanning, the red blood cell box is clamped to the set image acquisition position by the robot gripper, and the robot gripper is adjusted to various set postures, and further the visual camera is used to collect images of the red blood cell blood bag in the red blood cell box at various angles, to construct an image set of the red blood cell blood bag in the red blood cell box, to bind the image set of the red blood cell blood bag to the barcode, and to enter the red blood cell information library.

[0059] As a preferred solution, the set image acquisition position refers to the position where the robot gripper grabs the red blood cell box and moves it to a position where the visual camera can clearly capture the red blood cell bag in the red blood cell box.

[0060] It should be noted that the present invention realizes automatic clamping of red blood cells through a manipulator and a clamp, in conjunction with a red blood cell frame, a red blood cell box and visual positioning technology, wherein the red blood cell box realizes the function of standing up the red blood cell blood bag, and the red blood cell frame separates the red blood cell box to facilitate clamping by the manipulator, thereby facilitating automated red blood cell quality detection and improving the efficiency of red blood cell outbound management.

[0061] The red blood cell feature detection and analysis module is used to obtain the outer packaging feature information of the red blood cell blood bag and the red blood cell sample feature information in the red blood cell blood bag based on the image set of the red blood cell blood bag, wherein the outer packaging feature information includes the damage degree of the blood bag and the damage degree of the label, and the red blood cell sample feature information in the red blood cell blood bag includes the static stratification conformity, the blood color conformity and the area of ​​clots, bubbles and flocs in the blood.

[0062] Furthermore, the specific working process of the red blood cell feature detection and analysis module includes: extracting a set of historical unqualified red blood cell blood bag sample images stored in a database, screening out red blood cell blood bag sample images with wrinkles on the surface, scratches on the surface, and bleeding on the surface, and recording them as wrinkled red blood cell blood bag sample images, scratched red blood cell blood bag sample images, and bleeding red blood cell blood bag sample images, respectively.

[0063] According to the image set of the red blood cell blood bag, the surface image of the red blood cell blood bag is obtained, and the surface image of the red blood cell blood bag is compared with the wrinkled red blood cell blood bag sample image, the scratched red blood cell blood bag sample image and the bleeding red blood cell blood bag sample image respectively to obtain the total area of ​​the wrinkled area on the red blood cell blood bag surface, the total length of the surface scratches and the total area of ​​the surface bleeding area, which are recorded as , , .

[0064] As a preferred solution, the total area of ​​the wrinkled area on the red blood cell blood bag surface, the total length of the surface scratches and the total area of ​​the surface bleeding area are obtained. The specific method is: the surface image of the red blood cell blood bag is compared with the sample image of the wrinkled red blood cell blood bag, and the area in the red blood cell blood bag surface image whose similarity with the sample image of the wrinkled red blood cell blood bag is greater than the set similarity threshold is obtained, and the area is recorded as the wrinkled area on the red blood cell blood bag surface, and the wrinkled areas on the red blood cell blood bag surface are counted to obtain the total area of ​​the wrinkled area on the red blood cell blood bag surface.

[0065] Compare the surface image of the red blood cell bag with the sample image of the scratched red blood cell bag.

[0066] The area in the red blood cell blood bag surface image whose similarity with the scratched red blood cell blood bag sample image is greater than the set similarity threshold is obtained, and it is recorded as the scratch area on the red blood cell blood bag surface. The scratch areas on the red blood cell blood bag surface are counted, and the longest length of each scratch area on the red blood cell blood bag surface is obtained and accumulated to obtain the total length of the scratches on the red blood cell blood bag surface.

[0067] Compare the surface image of the red blood cell bag with the sample image of the red blood cell bag with leakage.

[0068] The area in the red blood cell blood bag surface image whose similarity with the bleeding red blood cell blood bag sample image is greater than the set similarity threshold is obtained, and it is recorded as the bleeding area on the red blood cell blood bag surface. The bleeding areas on the red blood cell blood bag surface are counted to obtain the total area of ​​the bleeding areas on the red blood cell blood bag surface.

[0069] By analyzing the formula Get the damage degree of the red blood cell blood bag ,in , , They represent the influence factors corresponding to the preset unit wrinkle area, unit scratch length and unit bleeding area, respectively. .

[0070] Furthermore, the specific working process of the red blood cell feature detection and analysis module also includes: D1: according to the surface image of the red blood cell blood bag, the surface label image of the red blood cell blood bag is obtained, and the surface label image of the red blood cell blood bag is compared with the preset red blood cell blood bag surface label reference image to obtain the area of ​​each damaged area of ​​the red blood cell blood bag surface label, which is recorded as , Indicates The number of damaged areas, .

[0071] According to the preset principle, the weight factors corresponding to each sub-region of the red blood cell blood bag surface label are set, and the surface label sub-regions to which each damaged region of the red blood cell blood bag surface label belongs are obtained. The weight factors corresponding to each damaged region of the red blood cell blood bag surface label are screened and recorded as .

[0072] D2: Use image processing technology to obtain the grayscale values ​​of the red blood cell blood bag surface label image. If a grayscale value of the red blood cell blood bag surface label image does not belong to the preset reference range of the grayscale value of the red blood cell blood bag surface label image, then the grayscale value is recorded as an abnormal grayscale value. The abnormal grayscale values ​​of the red blood cell blood bag surface label image are counted, and the area of ​​the region where the abnormal grayscale value of the red blood cell blood bag surface label image is located is obtained, and it is recorded as the area of ​​each dirty area of ​​the red blood cell blood bag surface label, and expressed as , Indicates The number of dirty areas, .

[0073] Get the surface label sub-area to which each dirty area of ​​the red blood cell blood bag surface label belongs, filter out the weight factor corresponding to each dirty area of ​​the red blood cell blood bag surface label, and record it as .

[0074] D3: By analyzing the formula ,

[0075] Get the label damage degree of the red blood cell bag ,in , Respectively represent the weights of the preset damaged area and dirty area, , , They represent the preset thresholds of damaged area and dirty area respectively.

[0076] Furthermore, the specific working process of the red blood cell feature detection and analysis module also includes: according to the image set of the red blood cell blood bag, obtaining the blood static stratification image in the red blood cell blood bag, and comparing it with the preset blood static stratification sample images, to obtain the blood static stratification sample image with the greatest similarity to the blood static stratification image in the red blood cell blood bag, and record it as the matching blood static stratification sample image of the red blood cell blood bag.

[0077] The static stratification conformity corresponding to each blood static stratification sample image is set, and the static stratification conformity corresponding to the matching blood static stratification sample image of the red blood cell blood bag is screened and obtained, which is recorded as the static stratification conformity of the red blood cell blood bag and expressed as .

[0078] As a preferred solution, when placed statically, the blood will be stratified into red blood cells and serum. Unqualified red blood cells will have fatty blood, hemolysis, etc. in the middle layer after static stratification.

[0079] Furthermore, the specific working process of the red blood cell characteristic detection and analysis module also includes: obtaining the blood color of the red blood cell blood bag, setting the compliance corresponding to various blood colors, screening and obtaining the blood color compliance of the red blood cell blood bag, and recording it as .

[0080] The historical unqualified red blood cell blood bag sample image set stored in the database is extracted, and the red blood cell blood bag sample images containing clots, bubbles and flocs in the blood are screened out respectively, and they are recorded as clot red blood cell blood bag sample images, bubble red blood cell blood bag sample images and floc red blood cell blood bag sample images, respectively.

[0081] According to the image set of red blood cell blood bag, the blood image in the red blood cell blood bag is obtained, and compared with the sample images of the red blood cell blood bag with clots, the sample images of the red blood cell blood bag with bubbles, and the sample images of the red blood cell blood bag with flocs, respectively, and the areas of clots, bubbles, and flocs in the blood of the red blood cell blood bag are analyzed and recorded as , , .

[0082] As a preferred solution, the method for analyzing the area of ​​clots, bubbles and floccules in the blood of the red blood cell blood bag is based on the same principle as the method for analyzing the total area of ​​wrinkles on the surface of the red blood cell blood bag.

[0083] The red blood cell qualification evaluation module is used to analyze the qualification index of the red blood cell bag according to the outer packaging characteristic information of the red blood cell bag and the red blood cell sample characteristic information in the red blood cell bag.

[0084] Furthermore, the specific working process of the red blood cell qualification assessment module includes: and label damage Substitute into the analysis formula , get the outer packaging qualification coefficient of the red blood cell blood bag ,in Indicates the correction factor of the preset outer packaging qualification coefficient, , They represent the thresholds of the preset blood bag damage degree and label damage degree, respectively. , They represent the weights of the preset blood bag damage degree and label damage degree, .

[0085] The static stratification compliance of the red blood cell blood bag , Blood color conformity and the area of ​​clots, bubbles, and flocs in the blood , , Substitute into the analysis formula , get the qualified coefficient of the red blood cell sample in the red blood cell bag ,in It indicates the correction factor of the qualified coefficient of the red blood cell sample in the preset red blood cell blood bag. represents a natural constant, , , They respectively represent the influencing factors of the preset unit clot area, unit bubble area and unit flocculent area.

[0086] Furthermore, the specific working process of the red blood cell qualification evaluation module also includes: and the red blood cell sample qualification coefficient in the red blood cell blood bag Substitute into the analysis formula , get the qualified index of red blood cell blood bag ,in , They represent the weights of the preset outer packaging qualification coefficient and the red blood cell sample qualification coefficient in the red blood cell blood bag, respectively. .

[0087] The red blood cell sorting and outbound module is used to determine whether the red blood cell blood bag is qualified according to the qualified index of the red blood cell blood bag. If it is qualified, the robot gripper grabs the red blood cell box where the red blood cell blood bag is located for outbound transportation. If it is unqualified, the robot gripper grabs the red blood cell box where the red blood cell blood bag is located and places it in the designated area.

[0088] Furthermore, the specific working process of the red blood cell sorting and outbound module is: comparing the qualified index of the red blood cell blood bag with the preset qualified index threshold; if the qualified index of the red blood cell blood bag is greater than the preset qualified index threshold, the red blood cell blood bag is qualified, and the robot gripper clamps the red blood cell box where the red blood cell blood bag is located for outbound transportation; otherwise, the red blood cell blood bag is unqualified, and the robot gripper clamps the red blood cell box where the red blood bag is located and places it in the designated area.

[0089] The database is used to store a historical sample image set of unqualified red blood cell blood bags.

[0090] It should be noted that the present invention, with the help of visual cameras and visual technology, collects images of red blood cell blood bags and extracts image features to evaluate whether the red blood cell blood bags to be shipped out are qualified and sorted, thereby improving the accuracy and consistency of red blood cell sample quality detection, reducing the subjectivity and randomness of manual judgment, thereby improving the level and efficiency of red blood cell shipping detection management. At the same time, the collected images of the red blood cell blood bags are saved for subsequent viewing, which helps to resolve later detection deviation disputes.

[0091] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they shall all fall within the protection scope of the present invention.

Claims

1. A machine vision-based intelligent detection, analysis and management system for red blood cell storage and outflow, characterized in that: include: Red blood cell code scanning information entry module: used to clamp the red blood cell box in the red blood cell frame on the red blood cell outbound conveyor belt to the set code scanning position through the manipulator gripper, and further scan the barcode on the surface of the red blood cell bag in the red blood cell box through the barcode scanner to read the basic information of the red blood cells and enter it into the red blood cell information database; Red blood cell image data acquisition module: After the red blood cell bag is scanned, the robot gripper clamps the red blood cell box to the set image acquisition position, and further collects the image set of the red blood cell bag in the red blood cell box through the visual camera, binds the image set of the red blood cell bag with the barcode, and enters it into the red blood cell information database; Red blood cell feature detection and analysis module: used to obtain the outer packaging feature information of the red blood cell bag and the red blood cell sample feature information in the red blood cell bag according to the image set of the red blood cell bag, wherein the outer packaging feature information includes the damage degree of the blood bag and the damage degree of the label, and the red blood cell sample feature information in the red blood cell bag includes the static stratification conformity, blood color conformity and the area of ​​clots, bubbles and flocs in the blood; Red blood cell qualification evaluation module: used to analyze the qualification index of the red blood cell bag according to the outer packaging characteristic information of the red blood cell bag and the characteristic information of the red blood cell sample in the red blood cell bag; Red blood cell sorting and outbound module: used to judge whether the red blood cell bag is qualified according to the qualified index of the red blood cell bag. If it is qualified, the robot gripper grabs the red blood cell box where the red blood cell bag is located for outbound transportation. If it is unqualified, the robot gripper grabs the red blood cell box where the red blood cell bag is located and places it in the designated area. Database: used to store historical sample image sets of unqualified red blood cell blood bags; The specific working process of the red blood cell qualification assessment module includes: The damage degree of the red blood cell blood bag and label damage Substitute into the analysis formula , get the outer packaging qualification coefficient of the red blood cell blood bag ,in Indicates the correction factor of the preset outer packaging qualification coefficient, , They represent the thresholds of the preset blood bag damage degree and label damage degree, respectively. , They represent the weights of the preset blood bag damage degree and label damage degree, ; The static stratification compliance of the red blood cell blood bag , Blood color conformity and the area of ​​clots, bubbles, and flocs in the blood , , Substitute into the analysis formula: , get the qualified coefficient of the red blood cell sample in the red blood cell bag ,in It indicates the correction factor of the qualified coefficient of the red blood cell sample in the preset red blood cell blood bag. represents a natural constant, , , They represent the influencing factors of the preset unit clot area, unit bubble area and unit flocculent area respectively; The specific working process of the red blood cell qualification assessment module also includes: The outer packaging qualification coefficient of the red blood cell blood bag and the qualified coefficient of red blood cell samples in red blood cell blood bags Substitute into the analysis formula Get the qualified index of red blood cell blood bag ,in , They represent the weights of the preset outer packaging qualification coefficient and the red blood cell sample qualification coefficient in the red blood cell blood bag, respectively. .

2. According to claim 1, a machine vision-based intelligent detection, analysis and management system for red blood cell storage and outflow, characterized in that: The specific working process of the red blood cell scanning information entry module is as follows: According to the preset principle, the robot is arranged on one side of the red blood cell outbound conveyor belt, and the barcode scanner and visual camera are arranged on the other side of the red blood cell outbound conveyor belt; When the red blood cell frame on the red blood cell outbound conveyor belt moves into the set sensing area, the red blood cell outbound conveyor belt pauses, and the robotic gripper clamps the red blood cell box in the red blood cell frame on the red blood cell outbound conveyor belt to the set scanning position, and further scans the barcode on the surface of the red blood cell blood bag in the red blood cell box through the scanner to read the basic information of the red blood cells and enter it into the red blood cell information database.

3. According to claim 1, a machine vision-based intelligent detection, analysis and management system for red blood cell storage and outflow, characterized in that: The specific working process of the red blood cell image data acquisition module is as follows: After the red blood cell bag is scanned, the red blood cell box is clamped to the set image acquisition position by the robot gripper, and the robot gripper is adjusted to various set postures. The visual camera is further used to collect images of the red blood cell bag in the red blood cell box at various angles, and an image set of the red blood cell bag in the red blood cell box is constructed. The image set of the red blood cell bag is bound to the barcode and entered into the red blood cell information library.

4. According to claim 1, a machine vision-based intelligent detection, analysis and management system for red blood cell storage and outflow, characterized in that: The specific working process of the red blood cell characteristic detection and analysis module includes: Extract the historical unqualified red blood cell blood bag sample image set stored in the database, screen out the red blood cell blood bag sample images with wrinkles on the surface, scratches on the surface, and bleeding on the surface, and record them as wrinkled red blood cell blood bag sample images, scratched red blood cell blood bag sample images, and bleeding red blood cell blood bag sample images respectively; According to the image set of the red blood cell blood bag, the surface image of the red blood cell blood bag is obtained, and the surface image of the red blood cell blood bag is compared with the wrinkled red blood cell blood bag sample image, the scratched red blood cell blood bag sample image and the bleeding red blood cell blood bag sample image respectively to obtain the total area of ​​the wrinkled area on the red blood cell blood bag surface, the total length of the surface scratches and the total area of ​​the surface bleeding area, which are recorded as , , ; By analyzing the formula Get the damage degree of the red blood cell blood bag ,in , , They represent the influence factors corresponding to the preset unit wrinkle area, unit scratch length and unit bleeding area, respectively. .

5. According to claim 4, a machine vision-based intelligent detection, analysis and management system for red blood cell storage and outflow, characterized in that: The specific working process of the red blood cell characteristic detection and analysis module also includes: D1: According to the surface image of the red blood cell blood bag, obtain the surface label image of the red blood cell blood bag, compare the surface label image of the red blood cell blood bag with the preset reference image of the surface label of the red blood cell blood bag, and obtain the area of ​​each damaged area of ​​the surface label of the red blood cell blood bag, which is recorded as , Indicates The number of damaged areas, ; According to the preset principle, the weight factors corresponding to each sub-region of the red blood cell blood bag surface label are set, and the surface label sub-regions to which each damaged region of the red blood cell blood bag surface label belongs are obtained. The weight factors corresponding to each damaged region of the red blood cell blood bag surface label are screened and recorded as ; D2: Use image processing technology to obtain the grayscale values ​​of the red blood cell blood bag surface label image. If a grayscale value of the red blood cell blood bag surface label image does not belong to the preset reference range of the grayscale value of the red blood cell blood bag surface label image, then the grayscale value is recorded as an abnormal grayscale value. The abnormal grayscale values ​​of the red blood cell blood bag surface label image are counted, and the area of ​​the region where the abnormal grayscale value of the red blood cell blood bag surface label image is located is obtained, and it is recorded as the area of ​​each dirty area of ​​the red blood cell blood bag surface label, and expressed as , Indicates The number of dirty areas, ; Get the surface label sub-area to which each dirty area of ​​the red blood cell blood bag surface label belongs, filter out the weight factor corresponding to each dirty area of ​​the red blood cell blood bag surface label, and record it as ; D3: By analyzing the formula Get the label damage degree of the red blood cell bag ,in , Respectively represent the weights of the preset damaged area and dirty area, , , They represent the preset thresholds of damaged area and dirty area respectively.

6. The intelligent detection, analysis and management system for red blood cell storage and outbound delivery based on machine vision according to claim 5 is characterized by: The specific working process of the red blood cell characteristic detection and analysis module also includes: According to the image set of the red blood cell blood bag, a blood static stratification image in the red blood cell blood bag is obtained, and compared with each preset blood static stratification sample image, to obtain a blood static stratification sample image with the greatest similarity to the blood static stratification image in the red blood cell blood bag, and record it as a matching blood static stratification sample image of the red blood cell blood bag; The static stratification conformity corresponding to each blood static stratification sample image is set, and the static stratification conformity corresponding to the matching blood static stratification sample image of the red blood cell blood bag is screened and obtained, which is recorded as the static stratification conformity of the red blood cell blood bag and expressed as .

7. The intelligent detection, analysis and management system for red blood cell storage and outbound delivery based on machine vision according to claim 6 is characterized by: The specific working process of the red blood cell characteristic detection and analysis module also includes: Get the blood color of the red blood cell bag, set the corresponding degree of compliance of various blood colors, filter out the blood color compliance of the red blood cell bag, and record it as ; Extract the historical unqualified red blood cell blood bag sample image set stored in the database, and screen out the red blood cell blood bag sample images containing clots, bubbles and flocs in the blood, and record them as clot red blood cell blood bag sample images, bubble red blood cell blood bag sample images and floc red blood cell blood bag sample images respectively; According to the image set of red blood cell blood bag, the blood image in the red blood cell blood bag is obtained, and compared with the sample images of the red blood cell blood bag with clots, the sample images of the red blood cell blood bag with bubbles, and the sample images of the red blood cell blood bag with flocs, respectively, and the areas of clots, bubbles, and flocs in the blood of the red blood cell blood bag are analyzed and recorded as , , .

8. According to claim 1, a machine vision-based intelligent detection, analysis and management system for red blood cell storage and outflow, characterized in that: The specific working process of the red blood cell sorting out module is as follows: The qualified index of the red blood cell bag is compared with the preset qualified index threshold. If the qualified index of the red blood cell bag is greater than the preset qualified index threshold, the red blood cell bag is qualified, and the robot gripper grabs the red blood cell box where the red blood cell bag is located for outbound transportation. Otherwise, the red blood cell bag is unqualified, and the robot gripper grabs the red blood cell box where the red blood bag is located and places it in the designated area.