Safety control method and system of treatment cabinet based on face recognition

Abstract

The invention belongs to the technical field of face recognition, and provides a safety control method and system for a treatment cabinet based on face recognition, and the method comprises the following steps: irradiating a to-be-recognized face through a near-infrared light source, and obtaining a near-infrared light source picture; obtaining a near-infrared light source gray scale picture; according to the gray scale information of the near-infrared light source gray scale picture, calculating a coverage brightness range difference, and obtaining a face shielding object area through the coverage brightness range difference; carrying out image segmentation on the near-infrared light source picture according to the face shielding object region, obtaining a non-coverage surface picture region, and carrying out mimicry fuzzy restoration on the non-coverage surface picture region; and identifying the non-coverage surface picture area and the corresponding non-coverage surface picture area in the stored near-infrared light source picture information of the user. According to the safety control method provided by the embodiment of the invention, the working efficiency of face recognition in a medical emergency scene can be effectively improved by introducing a mimicry fuzzy repair algorithm.

Description

Technical Field

[0001] The present invention belongs to the technical field of facial recognition, and in particular relates to a safety control method and system for a treatment cabinet based on facial recognition. Background Art

[0002] In the medical industry, facial recognition technology is often used for medical staff identity verification, patient identity confirmation, and automatic activation of medical equipment. However, traditional facial recognition technology, such as patent number CN119131860A, entitled "A facial recognition system and anesthesia cabinet using the system", has a significantly reduced recognition effect when medical staff wear protective equipment such as masks and goggles. For example, especially in high-pressure environments such as emergency and surgery, medical staff often need to wear protective equipment such as masks and goggles. These protective equipment will seriously block facial features, resulting in the inability of facial recognition systems to accurately and quickly authenticate identity simply through eye distance.

[0003] When a facial recognition system faces an obstruction, it usually directly captures a complete facial image and compares it with the information in the database. However, when medical staff wear protective gear, the obstruction will cause some feature information of the image to be missing or severely deformed, resulting in a significant decrease in recognition accuracy. In addition, traditional facial recognition methods usually prioritize improving the security of the system and try to reduce the risk of misidentification, but ignore the tight work pace of medical staff and the need for rapid response. In high-pressure environments such as emergency and surgery, any delay in identity verification will directly affect the efficiency of patient treatment or rescue, and may even have serious consequences. Summary of the invention

[0005] The present invention aims to solve at least one of the technical problems in the related art to a certain extent. To this end, the first purpose of the present invention is to propose a safety control method for a treatment cabinet based on face recognition, which can improve the adaptability and recognition accuracy of medical personnel wearing protective gear at the expense of some safety by introducing a mimetic fuzzy repair algorithm, and effectively improve the working efficiency of face recognition in medical emergency scenarios;

[0006] The second object of the present invention is to provide a safety control system of a treatment cabinet based on face recognition.

[0007] To achieve the above-mentioned purpose, a first aspect of the present invention provides a safety control method for a treatment cabinet based on face recognition, the method comprising the following steps:

[0008] S100, irradiating the face to be recognized with a near-infrared light source to obtain a near-infrared light source image;

[0009] S200, graying and meshing the near-infrared light source image and obtaining grayscale information to obtain a near-infrared light source grayscale image;

[0010] S300, calculating the coverage brightness range difference according to the grayscale information of the near-infrared light source grayscale image, and obtaining the facial obstruction area through the coverage brightness range difference;

[0011] S400, performing image segmentation on the near-infrared light source image according to the facial occlusion area, obtaining an uncovered facial image area, and performing mimicry blur repair on the uncovered facial image area;

[0012] S500, identifying the uncovered facial image area after the mimicry blur repair and the corresponding uncovered facial image area in the stored near-infrared light source image information of the user, and opening the treatment cabinet if the identification is successful.

[0013] According to the security control method of the embodiment of the present invention, the adaptability and recognition accuracy of medical personnel when wearing protective equipment can be improved at the expense of some safety by introducing a mimetic fuzzy repair algorithm, thereby effectively improving the work efficiency of face recognition in medical emergency scenarios.

[0014] Furthermore, in step S100, irradiating the face to be identified by a near-infrared light source, and obtaining a near-infrared light source image includes:

[0015] The face to be identified is illuminated by a near-infrared light source, where the near-infrared light source includes a near-infrared sensor and an image sensor. The near-infrared sensor uses a detector based on CMOS (complementary metal oxide semiconductor); the image sensor images the face in real time, captures the intensity of the reflected near-infrared light, and forms a near-infrared light source picture.

[0016] Since treatment cabinets are often placed in medical care rooms at night or in low light conditions, and nurses wear masks and glasses, visible light-based face recognition systems often fail to recognize patients or have low recognition accuracy in such environments. Although facial recognition technology based on near-infrared light sources can overcome the problem of insufficient light, when nurses wear masks, the area blocked by the masks may still affect the accuracy of the facial recognition system. When nurses wear glasses, especially glasses with larger frames, they may block a large part of the eye area. Although near-infrared light sources can penetrate the surface of the eyes and capture deep information under the eyes, reflection or occlusion from the glasses may interfere with the reflection of light, causing the recognition system to be unable to obtain complete eye information and unable to find facial features that can be captured.

[0017] In order to solve the above problem, in step S200, the near-infrared light source image is grayed and meshed and grayscale information is obtained. Obtaining the near-infrared light source grayscale image includes:

[0018] The near-infrared light source image is gridded and grayed out using a grid division algorithm. The grid size is one thousandth of the near-infrared light source image. The near-infrared light source image is divided into K grids, where K = 1000, to obtain a near-infrared light source grayscale image.

[0019] In step S300, calculating the coverage brightness range value according to the grayscale information of the near-infrared light source grayscale image includes:

[0020] Let s(i) represent the grayscale value of the i-th grid of the near-infrared light source grayscale image, the value of i is [1, K], K is the number of grids after the near-infrared light source grayscale image is divided; obtain the average value of the grayscale values ​​in each grid of the near-infrared light source grayscale image and record it as ZM; let SL(i) represent the average value of the grayscale values ​​of the adjacent grids of the i-th grid.

[0021] Furthermore, the adjacent grid refers to a grid that shares a border with the current grid, wherein the grayscale value of the grid is the average of the grayscale values ​​of all pixels in the grid, wherein the average grayscale value SL(i) of the adjacent grids of the i-th grid is the grayscale value of the sum of the grayscale values ​​of all adjacent grids of the i-th grid divided by the number of adjacent grids;

[0022] Perform facial occlusion analysis on near-infrared light source grayscale images:

[0023] S301, calculating the coverage brightness range difference of the near-infrared light source grayscale image;

[0024] Calculate the coverage brightness range difference: let the absolute value of the difference between SL(i) and SL(i+1) be ZG(j), the absolute value of the difference between s(i) and s(i+1) be SG(j), the value of j is [1, K-1], add all ZG(j) and all SG(j) and divide by 2j to get the coverage brightness range difference HDC;

[0025] The coverage brightness difference HDC is used to quantify the change and unevenness of the occluded area, and is used as a standard to determine whether the current grid is a reflection point or a blurred point of the occluder.

[0026] S302, define an integer variable k, set the initial value to 1, and create a blank sequence Z1;

[0027] S303, performing facial occlusion analysis, wherein the facial occlusion analysis is: comparing the values ​​of s(k) and ZM, and comparing the values ​​of [ZG(j)+SG(j)] and HDC, if s(k) is less than ZM and [ZG(k)+SG(k)] is less than HDC, then adding s(k) to the sequence Z1;

[0028] S304, judging whether the facial occlusion analysis is completed, the specific judging method is: if the current variable k is less than K, then k is increased by 1, and the process returns to step S303 to continue the facial occlusion analysis; if the current variable k is equal to K, then it means that the facial occlusion analysis has been processed and the process goes to step S305;

[0029] S305 , record the grids corresponding to all elements in the sequence Z1 as face occlusion grids, and record the area consisting of all face occlusion grids and the grids adjacent to the face occlusion grids as the face occlusion area.

[0030] The beneficial effect of this step is: through facial occlusion analysis based on local changes in image grayscale values, combined with the grayscale difference of adjacent grids and the coverage brightness range difference (HDC), the occlusion area in the image can be judged. Through facial occlusion analysis, the real facial feature area and the interference area caused by occlusion, noise or blur can be effectively distinguished. By effectively analyzing and identifying the occluded area, the system can remove these interfering parts from the image and focus on identifying the real facial features to avoid affecting the accuracy of the facial recognition system when medical staff or other users wear masks, glasses, goggles and other occluders.

[0031] Further, in step S400, the near-infrared light source image is segmented according to the facial occlusion area, the uncovered facial image area is obtained, and the mimicry blur repair is performed on the uncovered facial image area, including:

[0032] Furthermore, since medical staff wear goggles during work, the eye imaging after the near-infrared light source passes through the goggles causes tearing and noise, and the separation during the facial occlusion analysis produces edge tearing and local blurring, so it is necessary to perform mimic blur repair on the uncovered facial image area;

[0033] The process of mimic fuzzy repair is as follows:

[0034] S401, obtaining the grayscale value of the grid without the facial image area and the average value of the grayscale values ​​of the adjacent grids of the grid without the facial image area;

[0035] Specifically, u(y) represents the gray value of the yth grid in the uncovered facial image area, the value of y is [1, G], and G is the number of grids in the uncovered facial image area; the median of the gray values ​​in each grid in the uncovered facial image area is obtained and recorded as PD, and the average value of the gray values ​​in each grid in u(y) is obtained and recorded as PM; SOG(y) represents the average value of the gray values ​​of the adjacent grids of the yth grid in the uncovered facial image area;

[0036] S402, calculate the difference fluctuation gray value KJ(y) of each grid in the uncovered facial image area; wherein KJ(y) is the sum of Du(y) and DS(y); wherein Du(y) is the absolute value of 2u(y) minus the sum of u(y-1) and u(y+1); and Ds(y) is the absolute value of 2SOG(y) minus the sum of SOG(y-1) and SOG(y+1);

[0037] S403, calculating the difference fluctuation grayscale range; wherein the difference fluctuation grayscale range KFW is the sum of the absolute values ​​of the differences between all KJ(y) and the coverage brightness range difference HDC divided by G;

[0038] S404, correcting the grayscale value of the grid with abnormal grayscale fluctuation according to the differential fluctuation grayscale value and the differential fluctuation grayscale range of each grid;

[0039] Compare the absolute value of the difference between KJ(y) and HDC with the size of KFW. When the absolute value of the difference between KJ(y) and HDC is greater than KFW, the grayscale value of u(y) is corrected to the sum of u(y-1), u(y+1), PD, PM, SOG(y), SOG(y-1) and SOG(y+1) divided by 7;

[0040] Among them, PD represents the median of the grayscale value of the current grid, which can effectively reflect the grayscale distribution of the local area and reduce noise interference; PM represents the average value of the grayscale value of the current grid, which can provide the grayscale center trend of the local area; the difference fluctuation grayscale value KJ(y) represents the grayscale change of the current grid and the change of its adjacent area; Du(y) is used to measure the intensity of the grayscale change of the current grid. Ds(y) is used to measure the grayscale change of the adjacent area of ​​the current grid. KJ(y) represents the grayscale fluctuation degree of the current grid and its adjacent grids, reflecting the degree of change of the image in this area. The difference fluctuation grayscale range KFW is the mean of the difference between the difference fluctuation grayscale value KJ(y) of all grids and the coverage brightness range difference HDC, which represents the average level of the fluctuation difference of all grids and is used to measure the overall fluctuation range of the image.

[0041] Furthermore, mimetic blur repair determines whether repair is needed by grayscale difference fluctuations and the average grayscale values ​​of adjacent grids. If the difference fluctuation is large and exceeds the normal fluctuation range KFW, it means that the grid may be blurred and its grayscale value needs to be corrected by the grayscale values ​​of the surrounding grids.

[0042] The beneficial effect of this step is: by repairing part of the grayscale information, it solves the problem of tearing and noise in the eye imaging caused by the near-infrared light source passing through the goggles due to medical staff wearing goggles or glasses, which affects the collection of eye features and greatly reduces the success rate of the recognition process and the flexibility of use. The problem of being unable to quickly perform facial recognition in high-pressure environments such as emergency and surgery leads to the inability to open the medical cabinet in time.

[0043] S500, identifying the uncovered facial image area after the mimicry blur repair and the corresponding uncovered facial image area in the stored near-infrared light source image information of the user, and opening the treatment cabinet if the identification is successful.

[0044] Feature extraction is performed on the uncovered facial image area, and the feature extraction method includes texture-based feature, shape feature or key point feature extraction.

[0045] Specifically, texture features are obtained through methods such as gray-level co-occurrence matrix (GLCM), shape features are described by the relative positions of facial feature points, and deep learning methods extract higher-level facial features through networks such as convolutional neural network (CNN).

[0046] Furthermore, the stored near-infrared light source image information of the user has been pre-processed and features have been extracted for use as recognition templates. These templates also contain texture, shape, and key point information for matching with the features of the current facial image.

[0047] Furthermore, the features of the uncovered facial area of ​​the current user are compared with the stored templates, and a similarity metric is calculated, and a deep learning model is used for feature extraction, and the similarity is calculated through the output of the neural network to obtain a matching score. Among them, the matching score can measure the similarity between the current face and the stored template.

[0048] Furthermore, during the comparison process, if the similarity between the current facial features and the template exceeds the preset matching score, the identity of the currently identified user is verified, and the treatment cabinet is opened by controlling the hardware device (such as a relay, smart socket, etc.). If the matching degree is lower than the set matching score, it means that the recognition has failed, and it is judged that the user's facial information does not match due to occlusion, image quality and other problems, and the user is required to re-perform facial recognition or a corresponding error prompt is given.

[0049] The beneficial effects of the present invention are as follows: the present invention sacrifices a certain degree of security through mimetic blur repair and separation of occluded areas, but improves the robustness and adaptability of the facial recognition system, especially in complex environments where medical staff wear masks, goggles and other occluders. The assumption and processing of part of the image information in the image repair process sacrifices security, but by introducing the mimetic blur repair algorithm, the blur, noise and tearing caused by the occluders can be effectively eliminated, ensuring that sufficiently accurate facial features can be extracted even when wearing goggles or masks. In practical applications, the presence of facial occluders, especially the wearing of medical protective equipment such as masks and glasses, seriously affects the recognition accuracy and stability of traditional face recognition systems. The security requirements of traditional face recognition methods usually focus on minimizing the risk of misidentification, while ignoring that the work rhythm of medical staff is usually very tight, especially in high-pressure environments such as emergency and surgery. Any delay will affect the treatment or rescue efficiency of patients. Therefore, facial recognition technology must not only have high accuracy, but also respond quickly to ensure that identity authentication can be completed in a short time and related equipment can be enabled. The present invention makes appropriate technical compromises and sacrifices some security in exchange for a higher recognition success rate and flexibility of use, so that the recognition system can still reliably perform identity authentication in medical and nursing environments, especially in situations where lighting is insufficient and personnel are wearing protective gear. In addition, although some grayscale information is repaired, the repair and difference adjustment of the image during the repair process is essentially a certain degree of estimation and correction of the original data, but the repair process will strictly follow the overall structural characteristics of the image and maintain the consistency and naturalness of the facial features. This method ensures that the recognition system has strong fault tolerance and adaptability while ensuring accuracy; therefore, the present invention focuses more on improving the scope of application and recognition effect of the system, especially in medical and emergency scenarios, it can quickly and accurately identify identities, reduce tedious operations in medical work, and improve work efficiency.

[0050] To achieve the above-mentioned purpose, the second aspect embodiment of the present invention also proposes a safety control system of a treatment cabinet based on facial recognition, the safety control system of the treatment cabinet based on facial recognition includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, the processor implements the steps in a safety control method of a treatment cabinet based on facial recognition when executing the computer program, and the safety control system of the treatment cabinet based on facial recognition runs on desktop computers, notebooks, PDAs and computing devices in cloud data centers.

[0051] By implementing a safety control method for a treatment cabinet based on face recognition through a safety control system of a treatment cabinet based on face recognition, the adaptability and recognition accuracy of medical staff wearing protective equipment can be improved at the expense of some safety by introducing a mimetic fuzzy repair algorithm, thereby effectively improving the work efficiency of face recognition in medical emergency scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 Shown is a flow chart of a safety control method for a treatment cabinet based on face recognition;

[0053] Figure 2 Shown is the structural diagram of the safety control system of the treatment cabinet based on face recognition. DETAILED DESCRIPTION

[0054] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and should not be construed as limiting the present invention.

[0055] Figure 1 Shown is a flow chart of a safety control method for a treatment cabinet based on face recognition.

[0056] Reference Figure 1 The present invention proposes a safety control method for a treatment cabinet based on face recognition, the method comprising the following steps:

[0057] S100, irradiating the face to be recognized with a near-infrared light source to obtain a near-infrared light source image;

[0058] S200, graying and meshing the near-infrared light source image and obtaining grayscale information to obtain a near-infrared light source grayscale image;

[0059] S300, calculating the coverage brightness range difference according to the grayscale information of the near-infrared light source grayscale image, and obtaining the facial obstruction area through the coverage brightness range difference;

[0060] S400, performing image segmentation on the near-infrared light source image according to the facial occlusion area, obtaining an uncovered facial image area, and performing mimicry blur repair on the uncovered facial image area;

[0061] S500, identifying the uncovered facial image area after the mimicry blur repair and the corresponding uncovered facial image area in the stored near-infrared light source image information of the user, and opening the treatment cabinet if the identification is successful.

[0062] According to the security control method of the embodiment of the present invention, the adaptability and recognition accuracy of medical personnel when wearing protective equipment can be improved at the expense of some safety by introducing a mimetic fuzzy repair algorithm, thereby effectively improving the work efficiency of face recognition in medical emergency scenarios.

[0063] Furthermore, in step S100, irradiating the face to be identified by a near-infrared light source, and obtaining a near-infrared light source image includes:

[0064] The face to be identified is illuminated by a near-infrared light source, where the near-infrared light source includes a near-infrared sensor and an image sensor. The near-infrared sensor uses a detector based on CMOS (complementary metal oxide semiconductor); the image sensor images the face in real time, captures the intensity of the reflected near-infrared light, and forms a near-infrared light source picture.

[0065] Since treatment cabinets are often placed in medical care rooms at night or in low light conditions, and nurses wear masks and glasses, visible light-based face recognition systems often fail to recognize patients or have low recognition accuracy in such environments. Although facial recognition technology based on near-infrared light sources can overcome the problem of insufficient light, when nurses wear masks, the area blocked by the masks may still affect the accuracy of the facial recognition system. When nurses wear glasses, especially glasses with larger frames, they may block a large part of the eye area. Although near-infrared light sources can penetrate the surface of the eyes and capture deep information under the eyes, reflection or occlusion from the glasses may interfere with the reflection of light, causing the recognition system to be unable to obtain complete eye information and unable to find facial features that can be captured.

[0066] In order to solve the above problem, in step S200, the near-infrared light source image is grayed and meshed and grayscale information is obtained. Obtaining the near-infrared light source grayscale image includes:

[0067] The near-infrared light source image is gridded and grayed out using a grid division algorithm. The grid size is one thousandth of the near-infrared light source image. The near-infrared light source image is divided into K grids, where K = 1000, to obtain a near-infrared light source grayscale image.

[0068] In step S300, calculating the coverage brightness range value according to the grayscale information of the near-infrared light source grayscale image includes:

[0069] Let s(i) represent the grayscale value of the i-th grid of the near-infrared light source grayscale image, the value of i is [1, K], K is the number of grids after the near-infrared light source grayscale image is divided; obtain the average value of the grayscale values ​​in each grid of the near-infrared light source grayscale image and record it as ZM; let SL(i) represent the average value of the grayscale values ​​of the adjacent grids of the i-th grid.

[0070] Furthermore, the adjacent grid refers to a grid that shares a border with the current grid, wherein the grayscale value of the grid is the average of the grayscale values ​​of all pixels in the grid, wherein the average grayscale value SL(i) of the adjacent grids of the i-th grid is the grayscale value of the sum of the grayscale values ​​of all adjacent grids of the i-th grid divided by the number of adjacent grids;

[0071] Perform facial occlusion analysis on near-infrared light source grayscale images:

[0072] S301, calculating the coverage brightness range difference of the near-infrared light source grayscale image;

[0073] Calculate the coverage brightness range difference: let the absolute value of the difference between SL(i) and SL(i+1) be ZG(j), the absolute value of the difference between s(i) and s(i+1) be SG(j), the value of j is [1, K-1], add all ZG(j) and all SG(j) and divide by 2j to get the coverage brightness range difference HDC;

[0074] The coverage brightness difference HDC is used to quantify the change and unevenness of the occluded area, and is used as a standard to determine whether the current grid is a reflection point or a blurred point of the occluder.

[0075] S302, define an integer variable k, set the initial value to 1, and create a blank sequence Z1;

[0076] S303, performing facial occlusion analysis, wherein the facial occlusion analysis is: comparing the values ​​of s(k) and ZM, and comparing the values ​​of [ZG(j)+SG(j)] and HDC, if s(k) is less than ZM and [ZG(k)+SG(k)] is less than HDC, then adding s(k) to the sequence Z1;

[0077] S304, judging whether the facial occlusion analysis is completed, the specific judging method is: if the current variable k is less than K, then k is increased by 1, and the process returns to step S303 to continue the facial occlusion analysis; if the current variable k is equal to K, then it means that the facial occlusion analysis has been processed and the process goes to step S305;

[0078] S305 , record the grids corresponding to all elements in the sequence Z1 as face occlusion grids, and record the area consisting of all face occlusion grids and the grids adjacent to the face occlusion grids as the face occlusion area.

[0079] The beneficial effect of this step is: through facial occlusion analysis based on local changes in image grayscale values, combined with the grayscale difference of adjacent grids and the coverage brightness range difference (HDC), the occlusion area in the image can be judged. Through facial occlusion analysis, the real facial feature area and the interference area caused by occlusion, noise or blur can be effectively distinguished. By effectively analyzing and identifying the occluded area, the system can remove these interfering parts from the image and focus on identifying the real facial features to avoid affecting the accuracy of the facial recognition system when medical staff or other users wear masks, glasses, goggles and other occluders.

[0080] Further, in step S400, the near-infrared light source image is segmented according to the facial occlusion area, the uncovered facial image area is obtained, and the mimicry blur repair is performed on the uncovered facial image area, including:

[0081] Furthermore, since medical staff wear goggles during work, the eye imaging after the near-infrared light source passes through the goggles causes tearing and noise, and the separation during the facial occlusion analysis produces edge tearing and local blurring, so it is necessary to perform mimic blur repair on the uncovered facial image area;

[0082] The process of mimic fuzzy repair is as follows:

[0083] S401, obtaining the grayscale value of the grid without the facial image area and the average value of the grayscale values ​​of the adjacent grids of the grid without the facial image area;

[0084] Specifically, u(y) represents the gray value of the yth grid in the uncovered facial image area, the value of y is [1, G], and G is the number of grids in the uncovered facial image area; the median of the gray values ​​in each grid in the uncovered facial image area is obtained and recorded as PD, and the average value of the gray values ​​in each grid in u(y) is obtained and recorded as PM; SOG(y) represents the average value of the gray values ​​of the adjacent grids of the yth grid in the uncovered facial image area;

[0085] S402, calculate the difference fluctuation gray value KJ(y) of each grid in the uncovered facial image area; wherein KJ(y) is the sum of Du(y) and DS(y); wherein Du(y) is the absolute value of 2u(y) minus the sum of u(y-1) and u(y+1); and Ds(y) is the absolute value of 2SOG(y) minus the sum of SOG(y-1) and SOG(y+1);

[0086] S403, calculating the difference fluctuation grayscale range; wherein the difference fluctuation grayscale range KFW is the sum of the absolute values ​​of the differences between all KJ(y) and the coverage brightness range difference HDC divided by G;

[0087] S404, correcting the grayscale value of the grid with abnormal grayscale fluctuation according to the differential fluctuation grayscale value and the differential fluctuation grayscale range of each grid;

[0088] Compare the absolute value of the difference between KJ(y) and HDC with the size of KFW. When the absolute value of the difference between KJ(y) and HDC is greater than KFW, the grayscale value of u(y) is corrected to the sum of u(y-1), u(y+1), PD, PM, SOG(y), SOG(y-1) and SOG(y+1) divided by 7;

[0089] Among them, PD represents the median of the grayscale value of the current grid, which can effectively reflect the grayscale distribution of the local area and reduce noise interference; PM represents the average value of the grayscale value of the current grid, which can provide the grayscale center trend of the local area; the difference fluctuation grayscale value KJ(y) represents the grayscale change of the current grid and the change of its adjacent area; Du(y) is used to measure the intensity of the grayscale change of the current grid. Ds(y) is used to measure the grayscale change of the adjacent area of ​​the current grid. KJ(y) represents the grayscale fluctuation degree of the current grid and its adjacent grids, reflecting the degree of change of the image in this area. The difference fluctuation grayscale range KFW is the mean of the difference between the difference fluctuation grayscale value KJ(y) of all grids and the coverage brightness range difference HDC, which represents the average level of the fluctuation difference of all grids and is used to measure the overall fluctuation range of the image.

[0090] Furthermore, mimetic blur repair determines whether repair is needed by grayscale difference fluctuations and the average grayscale values ​​of adjacent grids. If the difference fluctuation is large and exceeds the normal fluctuation range KFW, it means that the grid may be blurred and its grayscale value needs to be corrected by the grayscale values ​​of the surrounding grids.

[0091] The beneficial effect of this step is: by repairing part of the grayscale information, it solves the problem of tearing and noise in the eye imaging caused by the near-infrared light source passing through the goggles due to medical staff wearing goggles or glasses, which affects the collection of eye features and greatly reduces the success rate of the recognition process and the flexibility of use. The problem of being unable to quickly perform facial recognition in high-pressure environments such as emergency and surgery leads to the inability to open the medical cabinet in time.

[0092] S500, identifying the uncovered facial image area after the mimicry blur repair and the corresponding uncovered facial image area in the stored near-infrared light source image information of the user, and opening the treatment cabinet if the identification is successful.

[0093] Feature extraction is performed on the uncovered facial image area, and the feature extraction method includes texture-based feature, shape feature or key point feature extraction.

[0094] Specifically, texture features are obtained through methods such as gray-level co-occurrence matrix (GLCM), shape features are described by the relative positions of facial feature points, and deep learning methods extract higher-level facial features through networks such as convolutional neural network (CNN).

[0095] Furthermore, the stored near-infrared light source image information of the user has been pre-processed and features have been extracted for use as recognition templates. These templates also contain texture, shape, and key point information for matching with the features of the current facial image.

[0096] Furthermore, the features of the uncovered facial area of ​​the current user are compared with the stored templates, and a similarity metric is calculated, and a deep learning model is used for feature extraction, and the similarity is calculated through the output of the neural network to obtain a matching score. Among them, the matching score can measure the similarity between the current face and the stored template.

[0097] Furthermore, during the comparison process, if the similarity between the current facial features and the template exceeds the preset matching score, the identity of the currently identified user is verified, and the treatment cabinet is opened by controlling the hardware device (such as a relay, smart socket, etc.). If the matching degree is lower than the set matching score, it means that the recognition has failed, and it is judged that the user's facial information does not match due to occlusion, image quality and other problems, and the user is required to re-perform facial recognition or a corresponding error prompt is given.

[0098] The beneficial effects of the present invention are as follows: the present invention sacrifices a certain degree of security through mimetic blur repair and separation of occluded areas, but improves the robustness and adaptability of the facial recognition system, especially in complex environments where medical staff wear masks, goggles and other occluders. The assumption and processing of part of the image information in the image repair process sacrifices security, but by introducing the mimetic blur repair algorithm, the blur, noise and tearing caused by the occluders can be effectively eliminated, ensuring that sufficiently accurate facial features can be extracted even when wearing goggles or masks. In practical applications, the presence of facial occluders, especially the wearing of medical protective equipment such as masks and glasses, seriously affects the recognition accuracy and stability of traditional face recognition systems. The security requirements of traditional face recognition methods usually focus on minimizing the risk of misidentification, while ignoring that the work rhythm of medical staff is usually very tight, especially in high-pressure environments such as emergency and surgery. Any delay will affect the treatment or rescue efficiency of patients. Therefore, facial recognition technology must not only have high accuracy, but also respond quickly to ensure that identity authentication can be completed in a short time and related equipment can be enabled. The present invention makes appropriate technical compromises and sacrifices some security in exchange for a higher recognition success rate and flexibility of use, so that the recognition system can still reliably perform identity authentication in medical and nursing environments, especially in situations where lighting is insufficient and personnel are wearing protective gear. In addition, although some grayscale information is repaired, the repair and difference adjustment of the image during the repair process is essentially a certain degree of estimation and correction of the original data, but the repair process will strictly follow the overall structural characteristics of the image and maintain the consistency and naturalness of the facial features. This method ensures that the recognition system has strong fault tolerance and adaptability while ensuring accuracy; therefore, the present invention focuses more on improving the scope of application and recognition effect of the system, especially in medical and emergency scenarios, it can quickly and accurately identify identities, reduce tedious operations in medical work, and improve work efficiency.

[0099] Figure 2 Shown is the structural diagram of the safety control system of the treatment cabinet based on face recognition.

[0100] Reference Figure 2 The present invention also proposes a safety control system 20 for a treatment cabinet based on face recognition. The safety control system 20 for a treatment cabinet based on face recognition includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps in a safety control method for a treatment cabinet based on face recognition are implemented. The safety control system 20 for a treatment cabinet based on face recognition runs on a desktop computer, a notebook, a PDA, and a computing device in a cloud data center.

[0101] The safety control system comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to run in the following units of the safety control system:

[0102] The acquisition unit 21 is used to illuminate the face to be identified through a near-infrared light source to obtain a near-infrared light source image;

[0103] A conversion unit 22 is used to grayscale and grid the near-infrared light source image and obtain grayscale information to obtain a near-infrared light source grayscale image;

[0104] A calculation unit 23 is used to calculate the coverage brightness range difference according to the grayscale information of the near-infrared light source grayscale image, and obtain the facial occlusion area through the coverage brightness range difference;

[0105] The management unit 24 is used to perform image segmentation on the near-infrared light source image according to the facial occlusion area, obtain the uncovered facial image area, and perform mimicry blur repair on the uncovered facial image area;

[0106] The recognition unit 25 is used to recognize the uncovered facial image area after the mimicry blur restoration and the corresponding uncovered facial image area in the stored near-infrared light source image information of the user, and open the treatment cabinet if the recognition is successful.

[0107] The safety control system of the treatment cabinet based on face recognition can be run in computing devices such as desktop computers, notebooks, PDAs and cloud servers. The safety control system of the treatment cabinet based on face recognition, the operational safety control system may include, but is not limited to, a processor and a memory. Those skilled in the art will understand that the example is only an example of a safety control system 20 of a treatment cabinet based on face recognition, and does not constitute a limitation on a safety control system 20 of a treatment cabinet based on face recognition, and may include more or fewer components than the example, or a combination of certain components, or different components, for example, the safety control system of the treatment cabinet based on face recognition may also include input and output devices, network access devices, buses, etc.

[0108] By executing the safety control method of the treatment cabinet based on face recognition through the safety control system 20 of the treatment cabinet based on face recognition, the adaptability and recognition accuracy of medical personnel when wearing protective equipment can be improved at the expense of some safety by introducing a mimetic fuzzy repair algorithm, thereby effectively improving the work efficiency of face recognition in medical emergency scenarios.

[0109] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection portion with one or more wirings (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0110] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0111] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0112] In the description of the present invention, it is to be understood that the terms “center”, “longitudinal”, “lateral”, “length”, “width”, “thickness”, “up”, “down”, “front”, “back”, “left”, “right”, “vertical”, “horizontal”, “top”, “bottom”, “inside”, “outside”, “clockwise”, “counterclockwise”, “axial”, “radial”, “circumferential”, etc., indicating orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0113] In addition, the terms "first", "second", etc. used in the embodiments of the present invention are only used for descriptive purposes and should not be understood as indicating or implying relative importance, or implicitly indicating the number of technical features indicated in the present embodiment. Therefore, the features defined by the terms "first", "second", etc. in the embodiments of the present invention can explicitly or implicitly indicate that the embodiment includes at least one of the features. In the description of the present invention, the word "multiple" means at least two or two or more, such as two, three, four, etc., unless otherwise clearly and specifically defined in the embodiments.

[0114] In the present invention, unless otherwise clearly specified or limited in the embodiments, the terms "installed", "connected", "connected" and "fixed" etc. in the embodiments should be understood in a broad sense. For example, the connection can be a fixed connection, a detachable connection, or an integrated connection. It can be understood that it can also be a mechanical connection, an electrical connection, etc.; of course, it can also be a direct connection, or an indirect connection through an intermediate medium, or it can be the internal connection of two elements, or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to the specific implementation situation.

[0115] In the present invention, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, a first feature being "above", "above" or "above" a second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature. A first feature being "below", "below" or "below" a second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is lower in level than the second feature.

[0116] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A safety control method for a treatment cabinet based on face recognition, characterized in that: The method comprises the following steps: S100, irradiating the face to be recognized with a near-infrared light source to obtain a near-infrared light source image; S200, graying and meshing the near-infrared light source image and obtaining grayscale information to obtain a near-infrared light source grayscale image; S300, calculating the coverage brightness range difference according to the grayscale information of the near-infrared light source grayscale image, and obtaining the facial obstruction area through the coverage brightness range difference; S400, performing image segmentation on the near-infrared light source image according to the facial occlusion area, obtaining an uncovered facial image area, and performing mimicry blur repair on the uncovered facial image area; S500, identifying the uncovered facial image area after the mimicry blur repair and the corresponding uncovered facial image area in the stored near-infrared light source image information of the user, and opening the treatment cabinet if the identification is successful.

2. The safety control method of a treatment cabinet based on face recognition according to claim 1 is characterized in that: In step S100, irradiating the face to be identified by a near-infrared light source, and obtaining a near-infrared light source image includes: The face to be identified is illuminated by a near-infrared light source, wherein the near-infrared light source includes a near-infrared sensor and an image sensor, wherein the near-infrared sensor adopts a CMOS-based detector; the image sensor images the face in real time, captures the intensity of the reflected near-infrared light, and forms a near-infrared light source picture.

3. The safety control method of a treatment cabinet based on face recognition according to claim 1 is characterized in that: In step S200, grayscale and grid division are performed on the near-infrared light source image and grayscale information is obtained. Obtaining the near-infrared light source grayscale image includes: The near-infrared light source image is gridded and grayed out using a grid division algorithm. The grid size is one thousandth of the near-infrared light source image. The near-infrared light source image is divided into K grids, where K = 1000, to obtain a near-infrared light source grayscale image.

4. The safety control method of a treatment cabinet based on face recognition according to claim 1 is characterized in that: In step S300, calculating the coverage brightness range value according to the grayscale information of the near-infrared light source grayscale image includes: S301, calculating the coverage brightness range difference of the near-infrared light source grayscale image; S302, define an integer variable k, set the initial value to 1, and create a blank sequence Z1; S303, performing facial occlusion analysis; S304, determine whether the facial occlusion analysis is completed; if the current variable k is less than G, increase k by 1, and return to step S303 to continue the facial occlusion analysis; if the current variable k is equal to K, it means that the facial occlusion analysis has been processed and go to step S305; S305 , record the grids corresponding to all elements in the sequence Z1 as face occlusion grids, and record the area consisting of all face occlusion grids and the grids adjacent to the face occlusion grids as the face occlusion area.

5. The safety control method of a treatment cabinet based on face recognition according to claim 1 is characterized in that: In step S400, the near-infrared light source image is segmented according to the facial occlusion area, the uncovered facial image area is obtained, and the mimicry blur repair is performed on the uncovered facial image area, including: S401, obtaining the grayscale value of the grid without the facial image area and the average value of the grayscale values ​​of the adjacent grids of the grid without the facial image area; S402, calculating the difference fluctuation gray value of each grid in the uncovered facial image area; S403, calculating the difference fluctuation grayscale range; S404, correcting the grayscale value of the grid with abnormal grayscale fluctuation according to the differential fluctuation grayscale value and the differential fluctuation grayscale range of each grid.

6. A safety control system for a treatment cabinet based on face recognition, characterized in that: The safety control system of a treatment cabinet based on facial recognition comprises: a processor, a memory, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the steps of the safety control method of the treatment cabinet based on facial recognition described in any one of claims 1 to 5 are implemented.

Images