A security detection method and system, and a storage medium

Abstract

The application relates to the technical field of security detection, in particular to a security detection method and system and a storage medium, a security detection method comprising the following steps: obtaining a to-be-detected face model; obtaining a corresponding interval duration according to the to-be-detected face model; if the interval duration corresponding to the to-be-detected face model is greater than a preset duration, determining whether a corresponding filed medical record can be obtained in a preset database according to the to-be-detected face model; if the corresponding filed medical record is not obtained in the preset database according to the to-be-detected face model, matching in a preset registration database according to the to-be-detected face model; if the to-be-detected face model is not matched in the preset registration database, informing a security personnel to perform real-time monitoring on a person corresponding to the to-be-detected face model. The application has the effects that when a fraud behavior is found, the fraud behavior can be timely stopped, the patients and the patient families can be prevented from being cheated, and the security detection efficiency of the security personnel on the whole hospital is improved.

Description

Technical Field

[0001] This application relates to the technical field of security detection, and in particular to a security detection method, system and storage medium. Background Technology

[0002] Hospitals are bustling with people, including patients, their family members, and medical staff. However, there are always some criminals who impersonate medical staff to defraud patients and their families, causing many patients to lose their life-saving money and thus preventing their conditions from being treated in a timely manner.

[0003] The hospital currently has cameras installed at each exit, and security personnel monitor the area covered by these cameras.

[0004] However, due to the wide scope of the hospital and the limited number of security personnel in the monitoring room, and the inability of security personnel to screen out suspicious persons entering and leaving the hospital from multiple camera coverage areas, some patients or their families may still be deceived. Summary of the Invention

[0005] To improve security personnel's monitoring of suspicious individuals entering and exiting hospitals, this application provides a security detection method, system, and storage medium.

[0006] Firstly, this application provides a security detection method, which adopts the following technical solution: A security detection method includes the following steps: Obtain the face model of the subject; Based on the face model to be tested and the preset database, the initial recording time and the filtered recording time corresponding to the face model to be tested are obtained; The interval duration corresponding to the face model under test is obtained based on the initial recording time and the screening recording time of the face model under test. The face model under test and the interval duration are in one-to-one correspondence. Compare the interval duration of the corresponding face model with the preset duration; If the interval length corresponding to the face model to be tested is longer than the preset length, then it is determined whether the face model to be tested can be matched with the corresponding medical record in the preset database. If the face model to be tested does not match the corresponding medical record in the preset database, then it is determined whether the face model to be tested can be matched with the corresponding face model in the preset registration database. If a face model corresponding to the face model to be tested is matched in the preset registration database, the face model to be tested is recorded in the security personnel list. If no matching face model is found in the preset registration database, security personnel will be notified to monitor the person corresponding to the face model in real time.

[0007] In some of these embodiments, personal medical records are obtained from a preset database based on the face model to be tested. Each personal medical record in the preset database has a matching face model to be tested, and the personal medical record includes the medical record update time. The update time of the individual medical record is compared with the preset medical record time; If the update time of the personal medical record is not earlier than the preset medical record time, it is determined that the personal medical record is the most recently created, and the obtained personal medical record is used as the corresponding archived medical record. If the update time of the personal medical record is earlier than the preset medical record time, the corresponding medical record is obtained based on the face model to be tested, and the corresponding medical record is represented as the family medical record of the face model to be tested. Compare the update time of the corresponding medical record with the preset medical record time; If the update time of the corresponding medical record is not earlier than the preset medical record time, it is determined that the corresponding medical record is the most recently created, and the obtained corresponding medical record is used as the corresponding archived medical record.

[0008] In some embodiments, the preset database also includes multiple social security photos, and the corresponding medical records are obtained based on the face model to be tested, including the following steps: Based on the face model to be tested, the corresponding social security photo is obtained from the preset database; Based on the corresponding social security photo, obtain the corresponding medical record.

[0009] In some embodiments, obtaining the initial recording time and the filtered recording time of the face model under test based on the face model under test and a preset database includes the following steps: Determine whether a face model corresponding to the face model to be tested can be matched in the preset database; If no matching face model is found in the preset database, the face model to be tested is recorded in the preset database and the recording time is set as the initial recording time. If a face model corresponding to the face model to be tested is matched in the preset database, the face model to be tested is recorded in the preset database, and the recording time is set as the screening recording time.

[0010] In some embodiments, after the step of not finding a matching face model for the face model to be tested in a preset registration database, the following steps are included: The face model to be tested is matched against a preset blacklist database, and it is determined whether a corresponding face model is matched against the preset blacklist database. The preset blacklist database represents the face information of the fraudster. If a corresponding face model is matched in the preset blacklist database, then the face model to be tested is matched in the preset database and it is determined whether the corresponding medical record is obtained. If the corresponding medical record is not found in the preset database, security personnel will be notified to monitor the person corresponding to the face model to be tested in real time.

[0011] In some embodiments, the preset database includes the number of fraud attempts, each with a corresponding facial model. If no corresponding medical record is found in the preset database, the following steps are included: Based on the face model to be tested, the corresponding face model is obtained from the preset database; Based on the corresponding face model, the number of fraud attempts is obtained, and the number of fraud attempts is compared with the preset number of attempts. If the number of fraud attempts exceeds the preset number, the corresponding number of fraud attempts will be incremented by 1, and the face model to be tested and the number of fraud attempts will be recorded in the preset blacklist database.

[0012] In some embodiments, the preset database includes fraud estimates, each of which has a corresponding facial model. If no corresponding medical record is found in the preset database, the following steps are also included: Based on the face model to be tested, the corresponding face model is obtained from the preset database; Based on the corresponding facial model, multiple contacts are obtained, and the number of contacts is compared with the preset number. If the number of people contacted exceeds the preset number, the fraud valuation corresponding to the face model under test will be updated.

[0013] In some embodiments, before obtaining the initial recording time and the filtered recording time of the face model under test based on the face model under test and a preset database, the following steps are included: The determination is based on whether the face model under test has obtained a specific marker, which is the identifier of medical personnel; If a specific marker is obtained based on the face model to be tested, the face model to be tested is matched in a preset whitelist database, and it is determined whether the face model to be tested is matched in the preset whitelist database. If the face model to be tested is not matched in the preset whitelist database, the security personnel will be reminded to monitor the face model in real time and update the fraud valuation corresponding to the face model.

[0014] Secondly, this application provides a security detection system, which adopts the following technical solution: A security detection system, comprising: The data generation module is used to acquire the face model of the subject. The data calculation module is used to obtain the initial recording time and the filtering recording time of the face model under test based on the face model under test and the preset database. The data calculation module is also used to obtain the interval duration corresponding to the face model under test based on the initial recording time and the filtering recording time corresponding to the face model under test. The face model under test and the interval duration are in one-to-one correspondence. The data analysis module is used to compare the interval duration of the corresponding face model with the preset duration. The data processing module is configured to: if the interval duration corresponding to the face model under test exceeds a preset duration, determine whether a corresponding medical record can be matched in a preset database based on the face model under test; if no corresponding medical record is matched in the preset database based on the face model under test, determine whether a corresponding face model can be matched in a preset registration database based on the face model under test; if a face model corresponding to the face model under test is matched in the preset registration database, record the face model under test in the security personnel list; if no face model corresponding to the face model under test is matched in the preset registration database, notify security personnel to conduct real-time monitoring of the personnel corresponding to the face model under test.

[0015] Thirdly, this application provides a readable storage medium, which adopts the following technical solution: A readable storage medium storing a computer program that can be loaded by a processor and executed as described in the security detection method.

[0016] The security detection method, system, and storage medium provided in this application embodiment can obtain the corresponding initial recording time and screening recording time from the acquired face model in a preset database. Suspicious individuals can then be screened in advance based on the corresponding initial recording time and screening recording time. Security personnel can then be alerted to monitor these individuals closely. If fraudulent behavior is discovered, it can be stopped in time, thereby reducing the chances of patients and their families being deceived and improving the security detection efficiency of security personnel throughout the hospital. Attached Figure Description

[0017] Figure 1 This is a flowchart of the steps involved in a security detection method.

[0018] Figure 2 This is a flowchart showing the steps for obtaining the initial recording time and the filtering of recorded times.

[0019] Figure 3 This is a flowchart of the steps for obtaining medical records for record keeping.

[0020] Figure 4 This is a flowchart of the preset blacklist database filtering steps.

[0021] Figure 5 This is a flowchart of the whitelist database filtering steps. Detailed Implementation

[0022] To better understand the purpose, technical solutions, and advantages of this application, it has been described and illustrated below with reference to the accompanying drawings and embodiments. However, those skilled in the art should understand that this application can be implemented without these details. In some cases, to avoid unnecessary descriptions that obscure various aspects of this application, well-known methods, processes, and systems already described at a higher level will not be elaborated upon. It will be apparent to those skilled in the art that various modifications can be made to the embodiments disclosed in this application, and the general principles defined in this application can be applied to other embodiments and application scenarios without departing from the principles and scope of this application. Therefore, this application is not limited to the illustrated embodiments, but conforms to the broadest scope consistent with the scope of protection claimed in this application.

[0023] Hospitals are bustling with people, and many families lose their hard-earned savings for life-saving treatments due to scams by criminals. To enhance hospital security, cameras are installed at all exits, and security personnel monitor these cameras to conduct security checks. However, due to the multiple exits and corners of a hospital, and the limited number of security personnel in the monitoring room, and the possibility that some security personnel may have face blindness, they may not be able to accurately identify every suspicious person in the monitoring room. Therefore, to enable more accurate and rapid identification of suspicious persons in hospital security checks and improve the monitoring accuracy of security personnel in the monitoring room, this application discloses a security detection method.

[0024] This application discloses a security detection method, referring to... Figure 1 This includes the following steps: S100, acquire the face model of the subject to be tested.

[0025] The face model to be tested is represented by the face model captured by the camera. Specific face models to be tested can be selected from photos taken by cameras placed at various exits and corners of the hospital. It should be noted that after each camera takes a photo, it is transmitted to the central server for data processing. It is particularly important to note that the cameras should be placed in areas with high pedestrian traffic within the hospital.

[0026] S200: Based on the face model to be tested and the preset database, obtain the initial recording time and the screening recording time of the face model to be tested.

[0027] The preset database is the database processed by the backend server. It includes multiple sets of face models, initial recording times, and selection / recording times. Each initial recording time and selection / recording time corresponds one-to-one with a face model. The initial recording time represents the time the corresponding face model was entered into the hospital database, and the selection / recording time represents the time the face model left the hospital.

[0028] To more comprehensively identify the initial recording time and the selected recording time for the face model under test, the methods for obtaining the initial recording time and the selected recording time of the face model under test are combined with... Figure 2 Specifically, it includes the following steps: S210, determine whether a face model corresponding to the face model to be tested can be matched in the preset database.

[0029] S220, if no face model corresponding to the face model to be tested is found in the preset database, the face model to be tested is recorded in the preset database and the recording time is set as the initial recording time.

[0030] S230, if a face model corresponding to the face model to be tested is matched in the preset database, the face model to be tested is recorded in the preset database, and the recording time is set as the filtering recording time.

[0031] Furthermore, when a face model corresponding to the test face model can be matched in the preset database, the initial recording time of the test face model in the preset database will change within a fixed period, updating each time the person corresponding to the test face model enters the hospital. The screening recording time is continuously updated each time the test face model leaves the hospital. In order to make data screening more efficient, in this embodiment, the fixed period is set to one week, and the data in the preset database will be automatically updated.

[0032] It's important to explain how to determine whether a person corresponding to the face model being tested is leaving or entering the hospital. The specific steps involve installing two sets of cameras at each of the hospital's exits. The first set of cameras faces inwards, while the second set faces away from the hospital. By comparing the images captured by the two sets of cameras, it's possible to determine whether the person corresponding to the face model is leaving or entering the hospital.

[0033] In this embodiment, the analysis of photos taken by the two sets of cameras mainly focuses on photos representing the same person. The specific analysis process for determining whether the person is leaving or entering the hospital is as follows: If the first set of cameras captures a frontal view of the person, while the second set captures a back view, and the first set of cameras captured the image earlier than the second set, then the person corresponding to the tested face model is determined to have left the hospital. Conversely, if the first set of cameras captures a back view of the person, while the second set captures a frontal view, and the second set captures the image earlier than the first set, then the person corresponding to the tested face model is determined to have entered the hospital.

[0034] It's important to note that when comparing the shooting times of photos taken by the two sets of cameras, the time interval between the two sets of photos needs to be within a preset time interval to ensure more accurate matching data. This preset time interval can be set based on the distance between the two sets of cameras and the time interval required for the person to walk past both sets of cameras. It's necessary to consider situations where some patients have mobility issues or are carrying children and walking slowly; therefore, the time interval can be set based on the slowest possible time it would take to pass between the two sets of cameras.

[0035] Furthermore, only when the time interval between the photos taken by the two sets of cameras is less than a preset time interval can the system determine whether the person corresponding to the tested facial model has left or entered the hospital. If the time interval between the photos taken by the two sets of cameras is not less than the preset time interval, it means that the two sets of photos cannot accurately determine whether the person corresponding to the tested facial model has left or entered the hospital.

[0036] S300 obtains the interval time corresponding to the face model under test based on the initial recording time and the screening recording time.

[0037] The interval duration represents the duration of the face model under test in the hospital under a fixed period, and there is a one-to-one correspondence between the face model under test and the interval duration.

[0038] It should be noted here that for the interval duration under a fixed period, the specific steps include: S310: Based on the face model to be tested, obtain the initial recording time and the filtered recording time of each day from the preset database.

[0039] S320 obtains the daily time interval in the hospital based on the initial recording time and the selected recording time.

[0040] S330 sums up all the daily time intervals of a fixed period to obtain the interval duration corresponding to the face model under test.

[0041] Since the individuals corresponding to the face models being tested are not active in the hospital at all times, it is necessary to record the daily time intervals corresponding to the face models being tested. Each daily time interval represents the cumulative total of all time the face models were active in the hospital that day. The daily time intervals within a fixed period are then summed together to determine the interval duration.

[0042] It's important to note that when obtaining the interval duration, the initial recording time and the filtered recording time must first be checked for consistency. If they are the same, they are considered to be on the same day. In this case, simply subtract the corresponding times from the initial recording time to obtain the interval duration. If they are different, they are considered to be on different days. In this case, subtract the corresponding dates from the initial recording time to obtain the interval in days, and then subtract the corresponding times from the initial recording time to obtain the temporary interval duration. Finally, multiply the interval in days by 24 and add it to the temporary interval duration to obtain the actual interval duration. S400: Compare the interval duration of the corresponding face model with the preset duration.

[0043] The preset duration represents the minimum time a suspicious person can spend in the hospital, and it is a cumulative time. In this embodiment, the preset duration is set to a cumulative one-week time interval, specifically 148 hours. When the suspected person's activity time in the hospital exceeds the preset duration, further screening is required when the person corresponding to the face model being tested exceeds the preset duration, since hospitalized patients and their caregivers also need to spend a long time in the hospital.

[0044] S500: If the interval length corresponding to the face model to be tested is longer than a preset time length, it is determined whether the face model to be tested can be matched with a corresponding medical record in a preset database. The preset database includes medical records, which represent the patient's medical files established at the hospital. Each medical record in the preset database has a corresponding face model. When the camera captures a face model as the face model to be tested, the face model to be tested is matched against the preset database. If a matching face model is found, it is further determined whether the face model to be tested has visited the hospital, based on whether the corresponding face model can be matched with a corresponding medical record in the preset database.

[0045] It should be noted that if the interval time corresponding to the face model under test is not greater than the preset time, it means that the person corresponding to the face model under test is normally active in the hospital. In this case, there is no need to remind the security personnel to conduct real-time detection, which reduces the energy of the security personnel and thus improves the security personnel's ability to monitor suspicious persons.

[0046] However, it is possible that the face model to be tested is a family member accompanying a patient to the hospital. Therefore, in some embodiments, the face model to be tested is matched with the corresponding medical record in the preset database. At this time, it can be determined that the face model to be tested is a patient or a family member. Therefore, it can be further determined whether the medical record is a personal medical record or a family medical record.

[0047] The determination is based on whether the face model under test can obtain the corresponding medical record in the preset database, combined with... Figure 3 This includes the following steps: S510: Based on the face model to be tested, retrieve personal medical records from the preset database. Each personal medical record in the preset database has a matching face model to be tested, and the personal medical record includes the medical record update time.

[0048] S520 compares the update time of the individual medical record with the preset medical record time.

[0049] S530, if the update time of a personal medical record is not earlier than the preset medical record time, then it is determined that the personal medical record was recently created, and the obtained personal medical record is used as the corresponding archived medical record.

[0050] S540, if the update time of the personal medical record is earlier than the preset medical record time, the corresponding medical record is obtained based on the face model to be tested, and the corresponding medical record is represented as the family medical record of the face model to be tested.

[0051] S550 compares the update time of the corresponding medical record with the preset medical record time.

[0052] S560, if the update time of the corresponding medical record is not earlier than the preset medical record time, then the corresponding medical record is determined to be the most recently created, and the obtained corresponding medical record is used as the corresponding archived medical record.

[0053] It should be added that if the update time of the corresponding medical record is earlier than the preset medical record time, it is determined that the corresponding medical record is not recently created, and therefore the corresponding medical record cannot be used as the corresponding archived medical record.

[0054] In this database, the personal medical record refers to the medical record file of the person corresponding to the face model to be tested, and the medical record update time refers to the update time of the personal medical record. The preset medical record time refers to the lowest time standard at which the medical record file is determined to be recently established. The family medical record refers to the medical record files of family members associated with the face model to be tested. Due to the large number of people in the hospital, not everyone will have a personal medical record file established at the hospital, but those who have established personal medical record files all have a matching face model to be tested in the preset database. In this embodiment, there may be multiple family medical records corresponding to one face model to be tested, while there is only one personal medical record. Both the personal medical record and the family medical record are used as the archived medical records corresponding to the face model to be tested.

[0055] Furthermore, the boundaries of family medical records are limited to those with direct kinship. In this embodiment, those within three generations above and below are considered direct relatives and can be included in family medical records.

[0056] In this embodiment, the specific steps for obtaining the preset medical record time are as follows: Obtain the current day's time for the face model to be tested, where the current day's time represents the time when the face model was acquired. Subtract the current day's time corresponding to the face model to be tested from the preset time period to obtain the preset medical record time. The preset time period can be set to one week.

[0057] For example, assuming the date of the obtained face model is March 8, 2022, subtracting the preset time period from the date corresponding to the face model can determine the preset medical record date as March 1, 2022. However, if the update date of the medical record corresponding to the face model is March 4, 2022, then the update date of the medical record corresponding to the face model is later than the preset medical record date, indicating that the medical record is the most recently updated.

[0058] It's important to note that if the medical record corresponding to the face model being tested is updated earlier than the preset medical record date, the corresponding treatment period needs to be obtained based on that medical record. The treatment period represents the treatment cycle for the patient corresponding to that medical record, thus determining whether the face model needs to remain in the hospital for a longer period. If the treatment period is longer than the preset time period, it indicates that the face model needs to remain in the hospital for an extended period, and in this case, the treatment period is used as the preset time period. If the treatment period is not longer than the preset time period, it indicates that the face model does not need to remain in the hospital for an extended period, suggesting that the medical record is not recently updated, and therefore the person corresponding to the face model does not need to seek medical treatment or care for family members at the hospital.

[0059] For example, suppose the date the face model to be tested is obtained is March 8, 2022. Subtracting a preset time period from the date corresponding to the face model gives a preset medical record date of March 1, 2022. However, the update date of the medical record corresponding to the face model is February 20, 2022. Since the update date of the medical record is earlier than the preset medical record date, it indicates that the medical record may not be recently updated. Therefore, it is necessary to obtain the patient's treatment period of 3 weeks. Upon analysis, it is found that the patient's treatment period is longer than the preset time period. Therefore, the treatment period of the patient corresponding to the medical record needs to be used as the preset time period. The corresponding preset medical record date is then the date corresponding to the face model to be tested minus the new preset time period, specifically February 15, 2022. The update date of the medical record corresponding to the face model is later than the preset time period, indicating that the medical record is also recently updated.

[0060] In summary, determining whether the medical record corresponding to the face model under test is the most recently updated only requires comparing the record's update time with a preset record time. If the update time of the medical record corresponding to the face model under test is later than the preset record time, it indicates that the medical record is the most recently updated. If the update time of the medical record corresponding to the face model under test is earlier than the preset record time, it indicates that the medical record is not the most recently updated.

[0061] In order to more quickly match the medical records corresponding to the face model to be tested, in some embodiments, the corresponding medical records are obtained based on the face model to be tested, specifically including the following steps: S541, based on the face model to be tested, obtain the corresponding social security photo from the preset database.

[0062] S542, obtain the corresponding medical record based on the corresponding social security photo.

[0063] The preset database also includes multiple social security photos, with a one-to-one correspondence between the facial models in the database and the social security photos. Each social security photo represents a facial image matched by the facial model in the preset database, and the corresponding medical record for the facial model under test is obtained through the facial image information.

[0064] Furthermore, for face models without corresponding social security photos, individuals traveling with the face model can be identified as related persons based on photos taken by the camera. Related persons are defined as those who have a relationship with the face model, and these related persons must have traveled with the face model more than a preset number of times. Traveling persons are selected from photos taken by the camera, and the preset number of travels is the minimum number of times a traveler is identified as a related person for that face model. The preset number of travels is recorded in a preset database, and each face model in the database corresponds to a set of multiple traveling persons, each with a certain number of travels with that face model. The preset number of travels can be set according to specific circumstances; in this embodiment, it is set to 3 times.

[0065] It should be noted here that the relevant personnel have corresponding social security photos or medical records at the hospital.

[0066] For example, a camera captures a face model to be tested. If no matching photo is found in a pre-set database, the camera can then match the face model with a group of people traveling with it. From this group, the person with the highest number of travel companions is selected, and their number of travel companions is compared to a pre-set limit. If the number of travel companions is less than the pre-set limit, this person is not considered a related party of the face model. Security personnel should then monitor this person normally, typically through routine questioning. If the number of travel companions is not less than the pre-set limit, this person is considered a related party of the face model.

[0067] For some staff members who are stationed at the hospital, who may be taking care of patients for a long time, if the corresponding medical records are not found in the preset database based on the face model to be tested, it is necessary to further match the face model to be tested in the preset level database.

[0068] S600: If the face model to be tested does not match the corresponding medical record in the preset database, then determine whether the face model to be tested can be matched in the preset registration database.

[0069] The preset registration database is represented by the back-end database of the inpatient department. The preset registration database includes the social security photo corresponding to the face model to be tested and the name of the corresponding patient who needs care.

[0070] S700: If a face model corresponding to the face model to be tested is matched in the preset registration database, the face model to be tested is recorded in the security personnel list.

[0071] The "safe personnel list" represents a list of individuals who have legitimate reasons for prolonged activity within the hospital. This list includes the social security photo and name of the person corresponding to the tested facial model. Since the person corresponding to the tested facial model needs to be present in the hospital for extended periods due to caring for patients, they can be recorded in the safe personnel list.

[0072] It should be noted that if a person in the security personnel list is found to be engaging in fraudulent activities, their information will be removed from the security personnel list.

[0073] S800: If no matching face model is found in the preset registration database, the security personnel will be notified to monitor the person corresponding to the face model in real time.

[0074] For suspicious individuals who linger in the hospital for extended periods, even if they are not there for medical treatment or have patients requiring care, security personnel should be notified to conduct special surveillance. This will reduce the number of times patients and their families are deceived and improve the overall security of the hospital.

[0075] It should be noted that real-time monitoring can be achieved by controlling cameras to monitor the suspicious person's movements in real time. The images captured by the cameras along the suspicious person's path are processed in the background to monitor whether fraudulent behavior occurs. Manual monitoring is also possible.

[0076] To improve efficiency during each data processing step, if no matching face model is found in the pre-defined registration database for the face model to be tested, then... Figure 4 This includes the following steps: S810 matches the face model to be tested against a preset blacklist database and determines whether the corresponding face model is matched in the preset blacklist database.

[0077] S820: If a corresponding face model is matched in the preset blacklist database, then the face model to be tested is matched in the preset database and it is determined whether the corresponding medical record is obtained.

[0078] S830: If the corresponding medical record is not found in the preset database, the security personnel will be notified to monitor the person corresponding to the face model to be tested in real time.

[0079] The preset blacklist database represents the facial information of fraudsters. When a corresponding facial model is matched in the preset blacklist database, further screening is needed to more accurately determine whether the person is suspicious. This is done by checking if a corresponding medical record is found in the preset database to determine if the person corresponding to the facial model is visiting a hospital or caring for a patient. If a corresponding medical record is found in the preset database, it indicates that the person is visiting a hospital or caring for a patient, and therefore, no further surveillance is required. If no corresponding medical record is found in the preset database, it indicates that the person corresponding to the facial model is suspicious, and therefore, security personnel need to be notified to monitor the person corresponding to the facial model in real time.

[0080] It should be noted that if no matching face model is found in the preset blacklist database, the S100-S800 operation will then be performed to filter out suspicious persons.

[0081] To ensure more accurate data processing, the preset blacklist database needs to be updated in real time during data processing. Therefore, in some embodiments, after failing to retrieve the corresponding medical record from the preset database, the following steps are included: S831, obtain the corresponding face model from the preset database based on the face model to be tested.

[0082] S832: Based on the corresponding face model, obtain the corresponding number of fraud attempts and compare the corresponding number of fraud attempts with the preset number of attempts.

[0083] S833, if the number of fraudulent acts is greater than the preset number, then the number of fraudulent acts is incremented by 1, and the face model to be tested and the number of fraudulent acts are recorded in the preset blacklist database.

[0084] The preset database includes the number of fraudulent acts, each with a corresponding facial model. The number of fraudulent acts represents the number of times a facial model under test has engaged in fraudulent behavior. The preset number of acts represents the minimum number of fraudulent acts that a facial model under test must have to be listed in a preset blacklist database. In this embodiment, the preset number of acts is 3.

[0085] For example, a camera captures a corresponding face model to be tested. This face model has been matched with a pre-set database three times for fraudulent activity. At this point, the person corresponding to this face model does not meet the minimum criteria for fraud in the pre-set blacklist database, therefore, this face model is not matched in the pre-set blacklist database. When, through data processing, this face model commits fraud again, the fraud count for this face model is incremented by 1, and the photo of the person corresponding to this face model, along with the corresponding fraud count, is recorded in the pre-set blacklist database for future data comparison.

[0086] In some embodiments, the preset database includes fraud estimates, each with a corresponding facial model. If no corresponding medical record is found in the preset database, the following steps are also included: S834: Obtain the corresponding face model from the preset database based on the face model to be tested.

[0087] S835 obtains multiple contacts based on the corresponding face model and compares the number of contacts with the preset number.

[0088] S836, if the number of people contacted is greater than the preset number, it is determined that the person corresponding to the face model under test has the potential to commit fraud, and the fraud valuation corresponding to the face model under test is updated.

[0089] The preset number of people represents the minimum number of people a suspected individual would come into contact with. The fraud estimate represents the likelihood of fraud corresponding to the tested facial model, with a value ranging from 1 to 100. A higher fraud estimate indicates a greater likelihood that the person corresponding to the tested facial model is likely to commit fraud. The fraud estimate is primarily based on the predicted behavior of the person corresponding to the tested facial model.

[0090] It should be noted that the definition of contact personnel is when the communication time reaches a preset communication time. The preset communication time can be set according to the actual situation. In this embodiment, the preset communication time is defined as 3 hours, and the preset communication time is a time interval.

[0091] For example, in this embodiment, the preset valuation is set to 50. When the camera captures a face model to be tested, and the number of contacts corresponding to that face model is detected to be 30, the probability of that face model being fraudulent is very small, so the fraud valuation for that face model is not updated. When 100 contacts corresponding to that face model are detected, the probability of that person being fraudulent is very high, and the fraud valuation can be set according to the ratio of contacts to the preset valuation. In this case, the fraud valuation increases by 50. When 80 contacts corresponding to that face model are detected, the corresponding fraud valuation increases by 62.

[0092] Some suspicious individuals may impersonate medical staff to deceive patients and their families. Therefore, in some embodiments, before obtaining the initial recording time and screening recording time of the face model to be tested based on the face model to be tested and a preset database, the following steps are taken: Figure 5 This includes the following steps: S910, determines whether a specific marker has been obtained based on the face model of the subject.

[0093] S920, if a specific marker is obtained based on the face model to be tested, the face model to be tested is matched in the preset whitelist database, and it is determined whether the face model to be tested is matched in the preset whitelist database.

[0094] S930: If the face model to be tested is not matched in the preset whitelist database, the security personnel will be reminded to monitor the face model in real time and update the fraud valuation corresponding to the face model to be tested.

[0095] Among these, specific identifiers are used to identify medical personnel. Examples include the distinctive shape of a medical worker's hat and white coat. However, this is not a limitation; any identifier that can represent the identity of a medical worker can be used as a specific identifier. A pre-defined whitelist database represents information on resident medical personnel.

[0096] For example, if a camera captures a face model posing as a medical worker, but no match is found in the preset whitelist database, the face model's behavior can be deemed suspicious, and its fraudulent valuation can be increased by 20. When the fraudulent valuation of the face model exceeds the preset valuation, security personnel are alerted to monitor it in real time. In this embodiment, the preset valuation is set to 100. When suspicious behavior is detected in the person corresponding to the face model, the fraudulent valuation is updated accordingly.

[0097] The implementation principle of a security detection method according to an embodiment of this application is as follows: After the camera captures a face model to be tested, the face model needs to be matched against a preset blacklist database. If a matching face model is found in the preset blacklist database, the face model is then matched against the preset database again to determine whether a corresponding medical record can be obtained. If no corresponding medical record is found in the preset database, security personnel are notified to monitor the person corresponding to the face model in real time. If no matching face model is found in the preset blacklist database, the interval duration of the face model within a fixed period is obtained from the preset database, and the interval duration of the face model is compared with a preset duration. If the interval duration of the face model is longer than the preset duration, further screening is performed by determining whether a corresponding medical record can be obtained from the preset database for the face model. If no corresponding medical record is found in the preset database based on the face model being tested, it indicates that the person corresponding to the face model may be there to care for or visit a patient. Therefore, it is necessary to match the face model with the preset registration database to check if the person corresponding to the face model is registered there. If the face model is matched in the preset registration database, it indicates that the person is there to care for a patient long-term, and therefore, the face model is recorded in the security personnel list. If the face model is not matched in the preset registration database, it indicates that the person corresponding to the face model does not have a patient to care for or visit, and is staying at the hospital for an extended period, making them a suspicious person. Therefore, security personnel are notified to monitor the person corresponding to the face model in real time.

[0098] This application also discloses a security detection system that executes a security detection method. The system includes a data generation module, a data calculation module, a data analysis module, and a data processing module. Specifically, the data generation module acquires a face model to be tested. The data calculation module obtains the initial recording time and the filtered recording time of the face model to be tested based on the face model and a preset database. The data calculation module also obtains the interval duration corresponding to the face model to be tested based on the initial recording time and the filtered recording time. There is a one-to-one correspondence between the face model to be tested and the interval duration. The data analysis module compares the interval duration of the corresponding face model with the preset duration.

[0099] The data processing module is used to determine whether a corresponding medical record can be matched in a preset database if the interval length corresponding to the face model under test is longer than a preset time. The data processing module is also used to determine whether a corresponding face model can be matched in a preset registration database if no corresponding medical record is found in the preset database. If a face model corresponding to the face model under test is matched in the preset registration database, the data processing module records the face model under test in the security personnel list. If a face model corresponding to the face model under test is not matched in the preset registration database, the data processing module notifies security personnel to conduct real-time monitoring of the personnel corresponding to the face model under test.

[0100] It should be noted that the data calculation module is also used to retrieve personal medical records from a preset database based on the face model to be tested. The data analysis module is also used to compare the update time of personal medical records with a preset medical record time, and to compare the update time of corresponding medical records with the preset medical record time. The data processing module is also used to determine if a personal medical record's update time is not earlier than the preset medical record time, and to use the retrieved personal medical record as the corresponding archived medical record. The data processing module is also used to determine if a personal medical record's update time is earlier than the preset medical record time, and to retrieve the corresponding medical record based on the face model to be tested. The corresponding medical record is represented as the family medical record of the face model to be tested. The data processing module is also used to determine if a corresponding medical record's update time is not earlier than the preset medical record time, and to use the retrieved corresponding medical record as the corresponding archived medical record.

[0101] It should be noted that the data analysis module is also used to retrieve the corresponding social security photo from a preset database based on the face model to be tested. The data processing module is also used to retrieve the corresponding medical record based on the social security photo.

[0102] It should be noted that the data analysis module is also used to determine whether a corresponding face model can be matched in the preset database. The data processing module is also used to handle situations where no matching face model is found in the preset database; in this case, the matching face model is recorded in the preset database, and the recording time is set as the initial recording time. The data processing module is also used to handle situations where a matching face model is found in the preset database; in this case, the matching face model is recorded in the preset database, and the recording time is set as the filtering recording time.

[0103] It should be noted that the data analysis module is also used to match the face model under test against a pre-defined blacklist database. The data processing module is also used to process data where a matching face model is found in the pre-defined blacklist database, and then to match the face model under test against the pre-defined database and determine if a corresponding medical record is found. The data processing module is also used to process data where no corresponding medical record is found in the pre-defined database, and then to notify security personnel to conduct real-time monitoring of the person corresponding to the face model under test.

[0104] It should be noted that the data calculation module is also used to retrieve the corresponding face model from the preset database based on the face model to be tested, and to retrieve the corresponding number of fraud attempts based on the corresponding face model. The data analysis module is also used to compare the corresponding number of fraud attempts with the preset number. The data processing module is also used to update the corresponding number of fraud attempts if the corresponding number of fraud attempts is greater than the preset number, and to record the face model to be tested and the number of fraud attempts in the preset blacklist database.

[0105] It should be noted that the data generation module is also used to obtain the corresponding face model from a preset database based on the face model to be tested. The data calculation module is also used to obtain multiple contacts based on the corresponding face model. The data analysis module is also used to compare the number of contacts with a preset number. The data processing module is also used to determine if the number of contacts exceeds the preset number, thus determining that the person corresponding to the face model to be tested has the potential to commit fraud, and updating the fraud valuation corresponding to the face model to be tested.

[0106] It should be noted that the data generation module is also used to determine whether a specific marker has been obtained based on the face model under test. The data calculation module is also used to match the face model under test against a preset whitelist database if a specific marker has been obtained. The data analysis module is also used to determine whether the face model under test has been matched against the preset whitelist database. The data processing module is also used to handle cases where the face model under test has not been matched against the preset whitelist database, by alerting security personnel to monitor the face model under test in real time and updating the fraud assessment value corresponding to the face model under test.

[0107] This application also discloses a readable storage medium storing a computer program that can be loaded and executed by a processor to provide a security detection method and a security detection system.

[0108] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A security detection method, characterized in that, Includes the following steps: Obtain the face model to be tested; Based on the face model to be tested and the preset database, obtain the initial recording time and the filtered recording time corresponding to the face model to be tested; Based on the initial recording time and the screening recording time of the face model under test, the interval duration corresponding to the face model under test is obtained, and there is a one-to-one correspondence between the face model under test and the interval duration. The interval duration of the corresponding face model is compared with the preset duration. If the interval duration corresponding to the face model under test is greater than the preset duration, it is determined whether the face model under test can be matched with a corresponding medical record in a preset database. If the face model under test is not matched with a corresponding medical record in the preset database, it is determined whether the face model under test can be matched with a corresponding face model in a preset registration database. If the face model corresponding to the face model under test is matched in the preset registration database, the face model under test is recorded in the security personnel list. If no matching face model is found in the preset registration database, security personnel will be notified to monitor the person corresponding to the face model in real time. If no matching face model is found in the preset registration database, the process includes the following steps: matching the face model to be tested in a preset blacklist database, and determining whether a matching face model is found in the preset blacklist database, where the preset blacklist database represents the facial information of fraudsters; if a matching face model is found in the preset blacklist database, then matching the face model to be tested in the preset database again and determining whether a corresponding medical record is obtained; If the corresponding medical record is not found in the preset database, security personnel will be notified to monitor the person corresponding to the face model to be tested in real time.

2. The security detection method according to claim 1, characterized in that, The step of determining whether to retrieve the corresponding medical record from the preset database based on the face model to be tested includes the following steps: Retrieving a personal medical record from the preset database based on the face model to be tested, wherein each personal medical record in the preset database has a matching face model to be tested, and the personal medical record includes a medical record update time; Comparing the update time of the personal medical record with a preset medical record time; If the update time of the personal medical record is not earlier than the preset medical record time, then it is determined that the personal medical record is recently created, and the retrieved personal medical record is used as the corresponding medical record for filing; If the update time of the personal medical record is earlier than the preset medical record time, then a corresponding medical record is retrieved based on the face model to be tested, and the corresponding medical record is represented as a family member's medical record of the face model to be tested; Comparing the update time of the corresponding medical record with the preset medical record time; If the update time of the corresponding medical record is not earlier than the preset medical record time, then it is determined that the corresponding medical record is recently created, and the retrieved corresponding medical record is used as the corresponding medical record for filing.

3. The security detection method according to claim 2, characterized in that, The preset database also includes multiple social security photos. The process of obtaining the corresponding medical record based on the face model to be tested includes the following steps: obtaining the corresponding social security photo from the preset database based on the face model to be tested; obtaining the corresponding medical record based on the social security photo.

4. The security detection method according to claim 1, characterized in that, The step of obtaining the initial recording time and the filtered recording time of the face model under test based on the face model under test and the preset database includes the following steps: determining whether a face model corresponding to the face model under test can be matched in the preset database; if no face model corresponding to the face model under test is matched in the preset database, the face model under test is recorded in the preset database, and the recording time is set as the initial recording time; if a face model corresponding to the face model under test is matched in the preset database, the face model under test is recorded in the preset database, and the recording time is set as the filtered recording time.

5. The security detection method according to claim 1, characterized in that, The preset database includes the number of fraud attempts, each with a corresponding facial model. If no corresponding medical record is found in the preset database, the process includes the following steps: obtaining the corresponding facial model from the preset database based on the facial model to be tested; obtaining the corresponding number of fraud attempts based on the corresponding facial model, and comparing the corresponding number of fraud attempts with a preset number; if the corresponding number of fraud attempts is greater than the preset number, incrementing the corresponding number of fraud attempts by 1, and recording the facial model to be tested and the number of fraud attempts in a preset blacklist database.

6. The security detection method according to claim 5, characterized in that, The preset database includes fraud estimates, each with a corresponding facial model. If no corresponding medical record is found in the preset database, the process further includes the following steps: Obtaining the corresponding facial model from the preset database based on the facial model to be tested; Obtaining multiple contacts based on the corresponding facial model and comparing the number of contacts with a preset number; If the number of contacts is greater than the preset number, updating the fraud estimate corresponding to the facial model to be tested.

7. The security detection method according to claim 1, characterized in that, Before obtaining the initial recording time and the filtered recording time of the face model under test based on the face model under test and the preset database, the following steps are included: determining whether a specific marker is obtained based on the face model under test, wherein the specific marker is the identifier of medical personnel; if the specific marker is obtained based on the face model under test, then the face model under test is matched in the preset whitelist database, and it is determined whether the face model under test is matched in the preset whitelist database; if the face model under test is not matched in the preset whitelist database, then security personnel are reminded to monitor the face model under test in real time and update the fraud valuation corresponding to the face model under test.

8. A security detection system, characterized in that, A security detection system for implementing the security detection method according to any one of claims 1-7, comprising: a data generation module for acquiring a face model to be tested; a data calculation module for acquiring the initial recording time and the filtering recording time of the face model to be tested based on the face model to be tested and a preset database, the data calculation module further being used to acquire the interval duration corresponding to the face model to be tested based on the initial recording time and the filtering recording time corresponding to the face model to be tested, wherein the face model to be tested and the interval duration are in one-to-one correspondence; and a data analysis module for comparing the interval duration of the corresponding face model with the preset duration. The data processing module is configured to: if the interval duration corresponding to the face model under test exceeds a preset duration, determine whether a corresponding medical record can be matched in a preset database based on the face model under test; if no corresponding medical record is matched in the preset database based on the face model under test, determine whether a corresponding face model can be matched in a preset registration database based on the face model under test; if a face model corresponding to the face model under test is matched in the preset registration database, record the face model under test in the security personnel list; if no face model corresponding to the face model under test is matched in the preset registration database, notify security personnel to conduct real-time monitoring of the personnel corresponding to the face model under test.

9. A readable storage medium, characterized in that, The computer program is stored and can be loaded by a processor and executed as described in any one of claims 1 to 7.

Images