A smart campus access control system

Abstract

The present invention relates to the technical field of smart campus, and specifically to a door access control system for a smart campus. The system comprises a server and a door access control management device, wherein the door access control management device comprises an information identification module, and the information identification module comprises identification terminals of at least two information identification modes; a traffic sensing module is arranged in front of an electronic fence to sense the passing personnel, and the server comprises a site identification module for obtaining site information of the door access management device, and obtaining the site personnel density and personnel behavior within the preset range of the electronic fence through a camera arranged just above the electronic fence; a behavior identification module for estimating the number of personnel who will pass through the door access management device within a preset period of time according to the personnel behavior; and an identification selection module, wherein the site information is preset with a verification level, and the higher the verification level, the higher the verification accuracy requirement, and the module is used to select the information identification mode to be started according to the verification level, the estimated number of personnel, and the length of the site personnel density.

Description

Technical Field

[0001] The present invention relates to the technical field of smart campuses, and in particular to an access control system for a smart campus. Background Art

[0002] Smart campus is the top-level design of national education informatization. With the rapid development and rapid marketization of advanced information technologies such as big data, cloud computing and the Internet of Things, the concept of smart campus has been put on the construction agenda of the education system along with the popularity of smart concepts such as "smart earth" and "smart city". The concept of smart campus is to realize rapid information exchange and management of various applications in the campus, including people, finance, and materials, and to improve the fun and orderly development of daily teaching, learning, research, management and other business activities on campus. Campus safety, library management, asset management and other aspects.

[0003] The smart campus includes an intelligent access control system, which can use Internet technologies such as smart cards, identity tags, face recognition, fingerprint recognition and other personnel information identification methods to achieve scientific management of the entry and exit of personnel in various venues. It is commonly used in libraries, dormitories, school gates and other places on campus to replace the previous manual registration method.

[0004] However, this intelligent identification method has certain defects. For example, although the identification speed is fast through identity tags and face recognition, it is greatly affected by the number of people. When the number of people is large, some people's faces in the recognition screen are blocked and cannot be recognized. Also, when using identity tags for identification in an environment with a high density of people, the human body and other obstacles may block or interfere with the signal transmission between the RFID tag and the reader. This interference may cause problems such as signal attenuation and multipath effects, thereby affecting the recognition accuracy and reliability of the RFID system. Although the recognition speed is high through fingerprints or electronic cards, only one person can be recognized at a time, and the recognition efficiency is low. Summary of the invention

[0005] The technical problem solved by the present invention is to provide an access control system for a smart campus, which can select the most appropriate identification method according to the site conditions.

[0006] The basic solution provided by the present invention is: a smart campus access control system, including a server and an access control management device, the access control management device includes an information recognition module, a traffic sensing module and an electronic fence, the information recognition module includes an identification terminal of at least two information recognition modes, the traffic sensing module is arranged in front of the electronic fence to sense the passing personnel, the server includes a site recognition module, a behavior recognition module, and an identification selection module;

[0007] The site identification module is used to obtain site information of the access control management device and obtain the site personnel density and personnel behavior within the preset range of the electronic fence through the camera located just above the electronic fence;

[0008] The behavior recognition module is used to estimate the number of people who will pass through the access control management device within a preset period of time based on their behavior;

[0009] The identification selection module, the site information is preset with a verification level, the higher the verification level, the higher the verification accuracy requirement, is used to select the information identification method to be started according to the verification level, the estimated number of personnel and the length of the densely populated area of ​​the site, start the corresponding identification terminal and obtain the personnel information collected by it;

[0010] The unlocking management module is used to obtain the number of passers-by m through the pass sensing module, and calculate the recognition accuracy based on the number of passers-by m and the number of personnel information obtained n. When the verification accuracy is higher than the verification accuracy, the electronic fence is controlled to open. When the verification accuracy is lower than or equal to the verification accuracy, the electronic fence is controlled to close until If the accuracy rate is higher than the verification rate, the electronic fence will be controlled to open.

[0011] The principles and advantages of the present invention are as follows: First, the site information is acquired. Different sites have different requirements for the progress of personnel identification. For example, the personnel information identification of libraries, school gates, laboratories, and dormitories requires a higher recognition accuracy rate because it is necessary to count the specific people entering and leaving. For places such as gymnasiums, cafeterias, and teaching buildings, it is only necessary to count the approximate number of people there, so the recognition accuracy rate is lower. Thereafter, the density of people and the behavior of people within the preset range of the electronic fence are obtained. Specifically, in the present application, identification can be performed by means of monitoring images, by setting a monitoring camera at the electronic fence, and obtaining the image near the electronic fence through the camera, and sending it to the server. The site identification module obtains the density of people and the behavior of people within the preset range of the electronic fence based on the image recognition.

[0012] The behavior recognition module estimates the number of people who will pass through the access control management device within the preset time period based on their behavior. Estimate the number of people who will pass through. Not all people who appear in the preset range in the image are people who will pass through the electronic fence. Some of them are passing nearby or have just passed through the electronic fence in the opposite direction of traffic. However, if these people are near the electronic fence, it may have a certain impact on the recognition progress. Therefore, it is necessary to identify the density of people separately and estimate the number of people who will pass through the access control management device based on their behavior.

[0013] After that, according to the verification level preset by the venue information, the higher the verification level, the higher the verification accuracy requirement. Here, the verification accuracy refers to the percentage of the number of people who successfully verify the personnel information among the total number of people required to pass through each venue. For example, for libraries and laboratories, it is necessary to count the specific personnel situation, and the verification accuracy recognition needs to be set to 100%, that is, if 100 people pass, the personnel information of these 100 people must be verified. For teaching buildings and gymnasiums, only the general personnel situation needs to be verified, and it can be set to 80%. Assuming that there are 100 people passing, when the personnel information of more than 80 people is verified, the verification accuracy is met, and if only the personnel information of less than 80 people is verified, it is not met. The canteen has lower requirements and can be set to 70%. The recognition selection module selects the verification method to be enabled according to the verification level, the density of people, and the estimated number of people who will pass. For example, when the density of people is high and the estimated number of people passing is low, if the verification level is high, fingerprint recognition is enabled, and if the verification level is low, face recognition or RFID recognition can be used. If the density of people is low, facial recognition or RFID recognition can be used directly regardless of the verification level, thus realizing the selection of different verification methods according to different situations.

[0014] After that, the management module recognizes that the number of people passing through the traffic sensing module can be an infrared sensor, which is set in front of the electronic fence to obtain the number of people passing through, and then according to the ratio of the number of people passing through m and the number of people information obtained n, , the recognition gets the current recognition accuracy. If it is not reached, the electronic fence will be closed. If face recognition is used, since the people in the picture are always moving, after closing the electronic fence, the picture changes and more faces are recognized. If RFID recognition is used, it is also because of the movement of people that the signal interference disappears, and more people are verified. Therefore, the ratio It will continue to improve, and when the verification accuracy is reached, it will continue to be released.

[0015] The technical solution of the present application achieves a more flexible verification compared to the single verification method used in the prior art. Different verification methods are used according to different situations, which avoids the situation where face or RFID cannot be recognized when there are dense crowds of people. It is also possible to choose to enable a faster verification method when there are fewer people.

[0016] In addition, in this application, the density of people and the estimated people who will pass through are identified separately, and the verification accuracy requirements of the venue are combined to adapt to the selection of verification methods in various complex environments. For example, for places such as libraries where the verification accuracy requirements are high, fingerprint verification or smart card recognition can be selected when the population is relatively dense to ensure the accuracy of the verification of each person entering and leaving, and face recognition or RFID recognition can be selected when there are fewer people to achieve the effect of fast passage.

[0017] Further, the behavior recognition module includes an image acquisition module and an image recognition module;

[0018] An image acquisition module is used to acquire the image captured by the camera;

[0019] The image recognition module includes a density recognition module and a personnel recognition module;

[0020] The density recognition module is used to recognize the number of people in the image and determine the density of people based on the number of people;

[0021] The personnel identification module is used to identify the movement trajectory of the personnel in the image and the completeness of the face, calculate the possibility score, determine whether the possibility score is higher than the preset score threshold, mark the personnel above the score threshold as the personnel who will pass through the electronic fence, and determine the number of personnel who will pass through the access control management device within the preset time period based on the marking.

[0022] The camera placed on the electronic fence is used to obtain video images, identify the image, identify the number of people in the image, and determine the density of people. Specifically, the density of people is calculated using the following formula:

[0023]

[0024] Where K is the density of people, S is the area within the preset range of the electronic fence in the video image where people can stand, and S is the number of people in the venue within the pre-screening range of the electronic fence.

[0025] After that, the possibility score is calculated based on the movement trajectory and the completeness of the face. The person appearing in the image may be passing by, staying nearby, or just passing through the passage in the opposite direction. According to the movement trajectory, the person in action is identified. If the movement trajectory is toward the electronic fence, the possibility that the person is about to pass is high. And the completeness of the face is identified. Since the electronic camera in this solution is located directly above the electronic fence, if a complete face appears in the image, it means that the person is facing the electronic fence and is likely to pass through. If the complete face does not appear, it means that the person is facing away from the electronic fence or is diagonally facing the electronic fence, and the possibility that he is about to pass through is low.

[0026] Further, the personnel identification module includes a trajectory scoring module, a face scoring module, a possibility scoring module and a scoring judgment module;

[0027] The trajectory scoring module is used to identify the angle between the movement trajectory of the person and the shooting direction of the camera. The angle range is 0-180°. The trajectory score is generated according to the angle. The smaller the angle, the greater the trajectory score.

[0028] The face scoring module is used to generate a face score according to the face completeness. The higher the face completeness, the higher the face score.

[0029] The possibility scoring module calculates the possibility score based on the trajectory score and face score:

[0030]

[0031] Where P is the possibility score, G is the trajectory score, and R is the face score. , These are the preset weights for trajectory scoring and face scoring respectively;

[0032] The scoring judgment module compares the possibility score of each person with the score threshold, and marks the person whose possibility score is higher than the score threshold as the person who will pass through the electronic fence. The more people there are in the image, the lower the score threshold.

[0033] According to the action and the completeness of the face, the possibility of a person passing through the electronic fence is scored. When the person's action trajectory is closer to 0 degrees, it means that the person is moving towards the electronic fence, so the trajectory score is higher. The closer it is to 180 degrees, the person is moving away from the electronic fence, so the trajectory score is lower. At the same time, according to the completeness of the face, the higher the face completeness, the closer the person is to the electronic fence, and the lower the face completeness, the farther away from the electronic fence. According to the action trajectory and the completeness of the face, the trajectory score and the face score are obtained respectively, and then the possibility score is calculated based on the preset weights. Finally, whether the possibility score is higher than the score threshold is determined to determine whether the person is likely to pass through the electronic fence. In addition, the more people there are in the image, the lower the score threshold. Since there are more people, the possibility of people being blocked in the shooting picture is higher, so there may be some people who are not recognized. Since this solution estimates the number of people who may pass through rather than which specific person passes, the score threshold is lowered to increase the estimated number of people.

[0034] Furthermore, the identification methods include RFID identification and face recognition in descending order according to the influence of the density of people on the identification accuracy, and fingerprint recognition is not affected by the density of people;

[0035] The identification and selection module includes an interval management module, a density judgment module, a first selection module, a second selection module and a third selection module;

[0036] The interval management module pre-stores the interval range of personnel density applicable to RFID and face recognition, as well as the changing trend of recognition accuracy after exceeding the maximum interval range;

[0037] The density judgment module is used to determine the verification accuracy required by the site based on the site information, and to determine whether it is within the range of population density applicable to RFID identification and face recognition based on the population density;

[0038] The first selection module is used for, when the population density is within the range of population density applicable to RFID identification and face recognition, if the ratio of the number of people who can pass through to the number of people identified is higher than a preset ratio, RFID identification is used, and if it is lower than the preset ratio, face recognition is used;

[0039] The second selection module is used to select face recognition when the density of people is only within the range applicable to face recognition;

[0040] The third selection module is used to judge whether the face recognition method under the current population density meets the recognition accuracy requirements of the site when the population density is not within the population density range applicable to RFID recognition and face recognition, based on the changing trend of the recognition accuracy after face recognition exceeds the maximum interval range. If so, face recognition is used; if not, fingerprint recognition is used.

[0041] In crowded environments, both RFID and face recognition technologies face certain challenges, but RFID is more significantly affected. This is because RFID technology relies on radio signals for contactless identification, and when there are dense crowds, the ultra-high frequency band RFID system will absorb electromagnetic waves and affect the recognition distance of the tag due to human body occlusion and proximity, thus greatly reducing the accuracy and efficiency of recognition. In contrast, although face recognition technology is also affected to a certain extent, it performs well in multi-target recognition and can simultaneously identify multiple target objects in crowded places, improving the accuracy and efficiency of positioning.

[0042] In this method, the range of population density applicable to RFID recognition and face recognition is pre-set. After the population density is identified, it is first determined whether the population density is within the range of population density applicable to RFID recognition and face recognition. If both are within the range, the selection of RFID recognition or face recognition is determined based on the ratio of the number of people who may pass through to the number of all people. When the ratio is large, the predicted proportion of people who may pass through is high. At this time, face recognition is used. At this time, the number of people facing the camera is more, and the face recognition effect is worse than RFID, so RFID recognition is used. When the ratio is low, the number of people facing the camera is smaller, and the accuracy of face recognition will be higher. When it is only within the range of face recognition, face recognition is selected. When neither is within the range, whether face recognition can meet the verification accuracy required by the site at this time is determined. If the verification accuracy requirement of the site is low and face recognition can meet it, face recognition can be used. If the requirement is high and face recognition cannot meet it, fingerprint recognition is used.

[0043] Furthermore, the recognition selection module is used to use the face data collected by the personnel recognition module when the selected recognition method is face recognition.

[0044] Further, the unlocking management module includes a time acquisition module;

[0045] Time acquisition module, used for The closing time of the electronic fence is controlled when the verification accuracy is lower than or equal to the verification accuracy;

[0046] The recognition selection module is also used to select a recognition method in which the density of people has less influence on the recognition accuracy when the shutdown time is higher than a preset time threshold. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 This is a schematic diagram of an implementation example of a smart campus access control system of the present invention. DETAILED DESCRIPTION

[0048] The following is further described in detail through specific implementation methods:

[0049] A smart campus access control system includes a server and an access control management device, wherein the access control management device includes an information recognition module, a traffic sensing module and an electronic fence, wherein the information recognition module includes an identification terminal of at least two information recognition modes, the traffic sensing module is arranged in front of the electronic fence to sense people passing by, and the server includes a site recognition module, a behavior recognition module and an identification selection module;

[0050] The site identification module is used to obtain site information of the access control management device and obtain the site personnel density and personnel behavior within the preset range of the electronic fence through the camera located just above the electronic fence;

[0051] The behavior recognition module is used to estimate the number of people who will pass through the access control management device within a preset period of time based on their behavior;

[0052] The identification selection module, the site information is preset with a verification level, the higher the verification level, the higher the verification accuracy requirement, is used to select the information identification method to be started according to the verification level, the estimated number of personnel and the length of the densely populated area of ​​the site, start the corresponding identification terminal and obtain the personnel information collected by it;

[0053] The unlocking management module is used to obtain the number of passers-by m through the pass sensing module, and calculate the recognition accuracy based on the number of passers-by m and the number of personnel information obtained n. When the verification accuracy is higher than the verification accuracy, the electronic fence is controlled to open. When the verification accuracy is lower than or equal to the verification accuracy, the electronic fence is controlled to close until If the accuracy rate is higher than the verification rate, the electronic fence will be controlled to open.

[0054] First, the venue information is acquired. Different venues have different requirements for personnel recognition progress. For example, for personnel information recognition in libraries, school gates, laboratories, and dormitories, since it is necessary to count the specific people entering and exiting, a higher recognition accuracy rate is required. For venues such as gymnasiums, canteens, and teaching buildings, it is only necessary to count the approximate number of people, so the recognition accuracy rate is lower. Thereafter, the density of people and the behavior of people within the preset range of the electronic fence are obtained. Specifically, in the present application, identification can be performed by means of monitoring images, by setting a monitoring camera at the electronic fence, and obtaining the image near the electronic fence through the camera, and sending it to the server. The venue recognition module obtains the density of people and the behavior of people within the preset range of the electronic fence based on the image recognition.

[0055] The behavior recognition module estimates the number of people who will pass through the access control management device within the preset time period based on their behavior. Estimate the number of people who will pass through. Not all people who appear in the preset range in the image are people who will pass through the electronic fence. Some of them are passing nearby or have just passed through the electronic fence in the opposite direction of traffic. However, if these people are near the electronic fence, it may have a certain impact on the recognition progress. Therefore, it is necessary to identify the density of people separately and estimate the number of people who will pass through the access control management device based on their behavior.

[0056] After that, according to the verification level preset by the venue information, the higher the verification level, the higher the verification accuracy requirement. Here, the verification accuracy refers to the percentage of the number of people who successfully verify the personnel information among the total number of people required to pass through each venue. For example, for libraries and laboratories, it is necessary to count the specific personnel situation, and the verification accuracy recognition needs to be set to 100%, that is, if 100 people pass, the personnel information of these 100 people must be verified. For teaching buildings and gymnasiums, only the general personnel situation needs to be verified, and it can be set to 80%. Assuming that there are 100 people passing, when the personnel information of more than 80 people is verified, the verification accuracy is met, and if only the personnel information of less than 80 people is verified, it is not met. The canteen has lower requirements and can be set to 70%. The recognition selection module selects the verification method to be enabled according to the verification level, the density of people, and the estimated number of people who will pass. For example, when the density of people is high and the estimated number of people passing is low, if the verification level is high, fingerprint recognition is enabled, and if the verification level is low, face recognition or RFID recognition can be used. If the density of people is low, facial recognition or RFID recognition can be used directly regardless of the verification level, thus realizing the selection of different verification methods according to different situations.

[0057] After that, the management module recognizes that the number of people passing through the traffic sensing module can be an infrared sensor, which is set in front of the electronic fence to obtain the number of people passing through, and then according to the ratio of the number of people passing through m and the number of people information obtained n, , the recognition gets the current recognition accuracy. If it is not reached, the electronic fence will be closed. If face recognition is used, since the people in the picture are always moving, after closing the electronic fence, the picture changes and more faces are recognized. If RFID recognition is used, it is also because of the movement of people that the signal interference disappears, and more people are verified. Therefore, the ratio It will continue to improve, and when the verification accuracy is reached, it will continue to be released.

[0058] The technical solution of the present application achieves a more flexible verification compared to the single verification method used in the prior art. Different verification methods are used according to different situations, which avoids the situation where face or RFID cannot be recognized when there are dense crowds of people. It is also possible to choose to enable a faster verification method when there are fewer people.

[0059] In addition, in this application, the density of people and the estimated people who will pass through are identified separately, and the verification accuracy requirements of the venue are combined to adapt to the selection of verification methods in various complex environments. For example, for places such as libraries where the verification accuracy requirements are high, fingerprint verification or smart card recognition can be selected when the population is relatively dense to ensure the accuracy of the verification of each person entering and leaving, and face recognition or RFID recognition can be selected when there are fewer people to achieve the effect of fast passage.

[0060] The behavior recognition module includes an image acquisition module and an image recognition module;

[0061] An image acquisition module is used to acquire the image captured by the camera;

[0062] The image recognition module includes a density recognition module and a personnel recognition module;

[0063] The density recognition module is used to recognize the number of people in the image and determine the density of people based on the number of people;

[0064] The personnel identification module is used to identify the movement trajectory of the personnel in the image and the completeness of the face, calculate the possibility score, determine whether the possibility score is higher than the preset score threshold, mark the personnel above the score threshold as the personnel who will pass through the electronic fence, and determine the number of personnel who will pass through the access control management device within the preset time period based on the marking.

[0065] The camera placed on the electronic fence is used to obtain video images, identify the image, identify the number of people in the image, and determine the density of people. Specifically, the density of people is calculated using the following formula:

[0066]

[0067] Where K is the density of people, S is the area within the preset range of the electronic fence in the video image where people can stand, and S is the number of people in the venue within the pre-screening range of the electronic fence.

[0068] After that, the possibility score is calculated based on the movement trajectory and the completeness of the face. The person appearing in the image may be passing by, staying nearby, or just passing through the passage in the opposite direction. According to the movement trajectory, the person in action is identified. If the movement trajectory is toward the electronic fence, the possibility that the person is about to pass is high. And the completeness of the face is identified. Since the electronic camera in this solution is located directly above the electronic fence, if a complete face appears in the image, it means that the person is facing the electronic fence and is likely to pass through. If the complete face does not appear, it means that the person is facing away from the electronic fence or is diagonally facing the electronic fence, and the possibility that he is about to pass through is low.

[0069] The personnel identification module includes a trajectory scoring module, a face scoring module, a possibility scoring module and a scoring judgment module;

[0070] The trajectory scoring module is used to identify the angle between the movement trajectory of the person and the shooting direction of the camera. The angle range is 0-180°. The trajectory score is generated according to the angle. The smaller the angle, the greater the trajectory score.

[0071] The face scoring module is used to generate a face score according to the face completeness. The higher the face completeness, the higher the face score.

[0072] The possibility scoring module calculates the possibility score based on the trajectory score and face score:

[0073]

[0074] Where P is the possibility score, G is the trajectory score, and R is the face score. , These are the preset weights for trajectory scoring and face scoring respectively;

[0075] The scoring judgment module compares the possibility score of each person with the score threshold, and marks the person whose possibility score is higher than the score threshold as the person who will pass through the electronic fence. The more people there are in the image, the lower the score threshold.

[0076] According to the action and the completeness of the face, the possibility of a person passing through the electronic fence is scored. When the person's action trajectory is closer to 0 degrees, it means that the person is moving towards the electronic fence, so the trajectory score is higher. The closer it is to 180 degrees, the person is moving away from the electronic fence, so the trajectory score is lower. At the same time, according to the completeness of the face, the higher the face completeness, the closer the person is to the electronic fence, and the lower the face completeness, the farther away from the electronic fence. According to the action trajectory and the completeness of the face, the trajectory score and the face score are obtained respectively, and then the possibility score is calculated based on the preset weights. Finally, whether the possibility score is higher than the score threshold is determined to determine whether the person is likely to pass through the electronic fence. In addition, the more people there are in the image, the lower the score threshold. Since there are more people, the possibility of people being blocked in the shooting picture is higher, so there may be some people who are not recognized. Since this solution estimates the number of people who may pass through rather than which specific person passes, the score threshold is lowered to increase the estimated number of people.

[0077] The identification methods include RFID identification and face recognition in descending order according to the influence of population density on the identification accuracy, and fingerprint recognition is not affected by population density;

[0078] The identification and selection module includes an interval management module, a density judgment module, a first selection module, a second selection module and a third selection module;

[0079] The interval management module pre-stores the interval range of personnel density applicable to RFID and face recognition, as well as the changing trend of recognition accuracy after exceeding the maximum interval range;

[0080] The density judgment module is used to determine the verification accuracy required by the site based on the site information, and to determine whether it is within the range of population density applicable to RFID identification and face recognition based on the population density;

[0081] The first selection module is used for, when the population density is within the range of population density applicable to RFID identification and face recognition, if the ratio of the number of people who can pass through to the number of people identified is higher than a preset ratio, RFID identification is used, and if it is lower than the preset ratio, face recognition is used;

[0082] The second selection module is used to select face recognition when the density of people is only within the range applicable to face recognition;

[0083] The third selection module is used to judge whether the face recognition method under the current population density meets the recognition accuracy requirements of the site when the population density is not within the population density range applicable to RFID recognition and face recognition, based on the changing trend of the recognition accuracy after face recognition exceeds the maximum interval range. If so, face recognition is used; if not, fingerprint recognition is used.

[0084] In crowded environments, both RFID and face recognition technologies face certain challenges, but RFID is more significantly affected. This is because RFID technology relies on radio signals for contactless identification, and when there are dense crowds, the ultra-high frequency band RFID system will absorb electromagnetic waves and affect the recognition distance of the tag due to human body occlusion and proximity, thus greatly reducing the accuracy and efficiency of recognition. In contrast, although face recognition technology is also affected to a certain extent, it performs well in multi-target recognition and can simultaneously identify multiple target objects in crowded places, improving the accuracy and efficiency of positioning.

[0085] In this method, the range of population density applicable to RFID recognition and face recognition is pre-set. After the population density is identified, it is first determined whether the population density is within the range of population density applicable to RFID recognition and face recognition. If both are within the range, the selection of RFID recognition or face recognition is determined based on the ratio of the number of people who may pass through to the number of all people. When the ratio is large, the predicted proportion of people who may pass through is high. At this time, face recognition is used. At this time, the number of people facing the camera is more, and the face recognition effect is worse than RFID, so RFID recognition is used. When the ratio is low, the number of people facing the camera is smaller, and the accuracy of face recognition will be higher. When it is only within the range of face recognition, face recognition is selected. When neither is within the range, whether face recognition can meet the verification accuracy required by the site at this time is determined. If the verification accuracy requirement of the site is low and face recognition can meet it, face recognition can be used. If the requirement is high and face recognition cannot meet it, fingerprint recognition is used.

[0086] The unlocking management module includes a time acquisition module;

[0087] Time acquisition module, used for The closing time of the electronic fence is controlled when the verification accuracy is lower than or equal to the verification accuracy;

[0088] The recognition selection module is also used to select a recognition method in which the density of people has less influence on the recognition accuracy when the shutdown time is higher than a preset time threshold.

[0089] The above are only embodiments of the present invention. Common knowledge such as the known specific structures and characteristics in the scheme is not described in detail here. Ordinary technicians in the relevant field know all the common technical knowledge in the technical field to which the invention belongs before the application date or priority date, can obtain all the existing technologies in the field, and have the ability to apply conventional experimental means before that date. Ordinary technicians in the relevant field can improve and implement this scheme in combination with their own abilities under the enlightenment given by this application. Some typical known structures or known methods should not become obstacles for ordinary technicians in the relevant field to implement this application. It should be pointed out that for those skilled in the art, without departing from the structure of the present invention, several deformations and improvements can be made, which should also be regarded as the scope of protection of the present invention, and these will not affect the effect of the implementation of the present invention and the practicality of the patent. The scope of protection required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. A smart campus access control system, characterized by: It includes a server and an access control management device, the access control management device includes an information recognition module, a traffic sensing module and an electronic fence, the information recognition module includes an identification terminal of at least two information recognition modes, the traffic sensing module is arranged in front of the electronic fence to sense the passing personnel, and the server includes a site recognition module, a behavior recognition module, and an identification selection module; The site identification module is used to obtain site information of the access control management device, and obtain the site personnel density and personnel behavior within the preset range of the electronic fence through the camera located just above the electronic fence; The behavior recognition module is used to estimate the number of people who will pass through the access control management device within a preset period of time based on their behavior; The identification selection module, the site information is preset with a verification level, the higher the verification level, the higher the verification accuracy requirement, and is used to select the information identification method to be started according to the verification level, the estimated number of personnel and the personnel density of the site, start the corresponding identification terminal and obtain the personnel information collected by it; The unlocking management module is used to obtain the number of passers-by m through the pass sensing module, and calculate whether the recognition accuracy reaches the verification accuracy based on the number of passers-by m and the number of personnel information obtained n. When the verification accuracy is higher than the verification accuracy, the electronic fence is controlled to open. When the verification accuracy is lower than or equal to the verification accuracy, the electronic fence is controlled to close until If the accuracy rate is higher than the verification rate, the electronic fence will be controlled to open; The identification methods include RFID identification and face recognition in descending order according to the influence of population density on the identification accuracy, and fingerprint recognition is not affected by population density; The identification and selection module includes an interval management module, a density judgment module, a first selection module, a second selection module and a third selection module; The interval management module pre-stores the interval range of personnel density applicable to RFID and face recognition, as well as the changing trend of recognition accuracy after exceeding the maximum interval range; The density judgment module is used to determine the verification accuracy required by the site based on the site information, and to determine whether it is within the range of population density applicable to RFID identification and face recognition based on the population density; The first selection module is used for, when the population density is within the range of population density applicable to RFID identification and face recognition, if the ratio of the number of people who can pass through to the number of people identified is higher than a preset ratio, RFID identification is used, and if it is lower than the preset ratio, face recognition is used; The second selection module is used to select face recognition when the density of people is only within the range applicable to face recognition; The third selection module is used to judge whether the face recognition method under the current population density meets the recognition accuracy requirements of the site when the population density is not within the population density range applicable to RFID recognition and face recognition, based on the changing trend of the recognition accuracy after face recognition exceeds the maximum interval range. If so, face recognition is used; if not, fingerprint recognition is used.

2. According to claim 1, the access control system for a smart campus is characterized by: The behavior recognition module includes an image acquisition module and an image recognition module; An image acquisition module is used to acquire the image captured by the camera; The image recognition module includes a density recognition module and a personnel recognition module; The density recognition module is used to recognize the number of people in the image and determine the density of people based on the number of people; The personnel identification module is used to identify the movement trajectory of the personnel in the image and the completeness of the face, calculate the possibility score, determine whether the possibility score is higher than the preset score threshold, mark the personnel above the score threshold as the personnel who will pass through the electronic fence, and determine the number of personnel who will pass through the access control management device within the preset time period based on the marking.

3. According to claim 2, the access control system for a smart campus is characterized in that: The personnel identification module includes a trajectory scoring module, a face scoring module, a possibility scoring module and a scoring judgment module; The trajectory scoring module is used to identify the angle between the movement trajectory of the person and the shooting direction of the camera. The angle range is 0-180°. The trajectory score is generated according to the angle. The smaller the angle, the greater the trajectory score. The face scoring module is used to generate a face score according to the face completeness. The higher the face completeness, the higher the face score. The possibility scoring module calculates the possibility score based on the trajectory score and face score: Where P is the possibility score, G is the trajectory score, and R is the face score. , These are the preset weights for trajectory scoring and face scoring respectively; The scoring judgment module compares the possibility score of each person with the score threshold, and marks the person whose possibility score is higher than the score threshold as the person who will pass through the electronic fence. The more people there are in the image, the lower the score threshold.

4. According to claim 3, the access control system for a smart campus is characterized in that: The recognition selection module is used to use the face data collected by the personnel recognition module when the selected recognition method is face recognition.

5. According to claim 4, the access control system for a smart campus is characterized in that: The unlocking management module includes a time acquisition module; Time acquisition module, used for The closing time of the electronic fence is controlled when the verification accuracy is lower than or equal to the verification accuracy; The recognition selection module is also used to select a recognition method in which the density of people has less influence on the recognition accuracy when the shutdown time is higher than a preset time threshold.

Images