Multi-access control management system based on image recognition

Abstract

The invention discloses a multi-access control management system based on image recognition, and relates to the technical field of access control. Comprising an identification recognition module and a face recognition module, the identification recognition module is used for collecting a to-be-recognized image in a recognition area N, matching the to-be-recognized image with a template image to generate a matching degree image, comparing the matching degree with a threshold value to judge the number of identifications, and then sending a recognition result to an access control module; the face recognition module is used for matching the recognized face information with the face information of the first storage unit, and feeding back a third recognition result to the access control module after successful matching; and the access control module controls the opening or closing of the Nth access control according to the identification result, and can judge the distribution condition of the to-be-passed personnel in the identification area N according to different combinations of the identification results, so that the opening time of the Nth access control is adjusted more flexibly, and the personnel in the front half part of the identification area N are prevented from following to pass through the access control.

Description

Technical Field

[0001] The present invention relates to the field of access control technology, and in particular to a multi-access control management system based on image recognition. Background Art

[0002] The access control system based on image recognition generally refers to the process of determining whether to release the lock restrictions through technical means such as face capture, face recognition comparison, face modeling and retrieval. It is done by presetting the face information in the comparison database. After the access control recognizes the face, it is compared with the face information in the comparison database. If the comparison results are consistent, the lock restrictions are released, which greatly reduces labor costs.

[0003] In the prior art, during the facial information collection process of the access control system, there is often a situation where there are multiple people waiting to pass through in front of the access control at the same time. When there are multiple people waiting to pass, their facial information is often entered into the system for facial recognition. On the one hand, this increases the workload of facial recognition feature comparison, and the simultaneous facial recognition of multiple people waiting to pass will cause great congestion at the access control. Whether the distribution positions of people waiting to pass at the access control recognition point can be restricted to a certain extent will become the key to solving the congestion problem at the access control; on the other hand, when multiple people waiting to pass undergo facial recognition at the same time, the distance between them is relatively close, and their positions close to the access control entrance are also roughly the same. When the facial recognition of one of the people waiting to pass is successful, it is easy for other people to immediately follow and pass through the access control. Therefore, it is necessary to design an access control system that ensures that multiple people waiting to pass undergo facial recognition one by one to achieve orderly passage through the access control. Summary of the invention

[0004] In view of the above technical deficiencies, the present invention provides a multi-access control management system based on image recognition to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention is implemented through the following technical solutions: A multi-access control management system based on image recognition is used to control n access controls in an area. An identification area N is set in front of the Nth access control, where 1≤N≤n. The Nth access control includes: an identification recognition module whose identification direction faces the identification area N and a face recognition module whose identification direction faces the front of the Nth access control. The identification area N is provided with m identifications evenly distributed along the access control entry and exit directions. The corresponding serial numbers of the identifications are recorded as M from back to front. 1 、M 2 , ..., M m; The identification recognition module is used to collect the image to be recognized in the recognition area N, and match it with the template image to generate a matching degree image, compare the matching degree with the threshold to determine the number of identifications, and then send the recognition result to the access control module; the face recognition module is used to match the recognized face information with the face information of the first storage unit, and after the match is successful, the third recognition result is fed back to the access control module; the access control module is used to receive the recognition result and the remote control instruction sent by the management end, and control the electronic lock according to the recognition result or the remote control instruction; the management end is used to view the recognition results of the identification recognition module and the face recognition module, and can remotely control the electronic locks of n access control in the control area; wherein the identification recognition module compares the image to be recognized with the template image N in sequence 11 and template image N 12 Match and generate matching degree images N respectively 13 And the matching image N 14 , the matching degree image N 13 The matching degrees of the m / 2 marks in the corresponding areas are compared with the threshold in turn. If the matching degree is greater than the threshold, it is determined that there is a mark in the corresponding area of ​​the image to be identified. The m / 2 judgment results are integrated. If there are (m / 2-1) marks, it is determined that there is only one person to pass behind the identification area N, and the first recognition result is sent to the access control module. At the same time, the matching image N is also sent. 14 The matching degrees of the m / 2 identification areas corresponding to the m / 2 identifications in the image are compared with the threshold value in turn. If the matching degree is greater than the threshold value, it is determined that there is an identification in the corresponding area of ​​the image to be identified, and the m / 2 judgment results are integrated. If there are m / 2 identifications, it is determined that there is no following person in front of the identification area N, and the second identification result is sent to the access control module; when the access control module receives the first identification result, the second identification result and the third identification result at the same time, the electronic lock is controlled to open to unlock the Nth access control, and the Nth access control is closed after time T1; when the access control module only receives the first identification result and the third identification result, the electronic lock is controlled to open to unlock the Nth access control, and the Nth access control is closed after time T1 / 2.

[0006] Further, the identification recognition module is also connected to an unidentified identification detection unit, the unidentified identification detection unit is connected to an unidentified identification serial number comparison unit, the unidentified identification serial number comparison unit is connected to the electronic lock, the unidentified identification detection unit is used to compare the preset number of identifications with the number of identifications actually recognized to determine whether there are unidentified identifications, and send the unidentified identification serial numbers to the unidentified identification serial number comparison unit, the unidentified identification serial number comparison unit is used to compare the interval [M 1 , M m / 2 ] in the unrecognized identification number and interval (M m / 2 , M m] are subtracted in sequence, and then the difference with the smallest value is obtained, and the time T1 for opening the Nth access control is adjusted according to the size of the difference with the smallest value.

[0007] Furthermore, the identification module is also connected to a template updating unit, which is connected to the management terminal. The template updating unit is used to receive a new template image sent by the management terminal to implement the template image N in the identification module. 11 and template image N 12 Updates.

[0008] Furthermore, the identification recognition module includes: an image acquisition unit, a preprocessing unit, a feature extraction unit, and a second storage unit. The image acquisition unit is used to acquire the image to be identified in the identification area N and send it to the preprocessing unit. The preprocessing unit preprocesses the image to be identified to improve the accuracy of subsequent matching. The feature extraction unit is used to extract relevant features of m identifications from the preprocessed image for template matching. The second storage unit is used to store the template image N corresponding to the identification area N. 11 and template image N 12 , for the template matching unit to read during matching.

[0009] Furthermore, the identification recognition module further includes: a template matching unit, a threshold judgment unit and a recognition result transmission unit. The template matching unit sequentially matches the image to be recognized with the template image N by a normalized square difference matching method. 11 and template image N 12 Match and generate matching degree images N respectively 13 And the matching image N 14 The threshold judgment unit is used to compare the matching degree with the preset threshold to determine whether there is an identification in the corresponding area of ​​the image to be identified, and then integrate the judgment results, and generate or not generate an identification result based on the number of identifications existing in the image to be identified. The identification result transmission unit is used to send the first identification result or the second identification result generated by the threshold judgment unit to the access control module.

[0010] Furthermore, the Nth access control system further includes: a password input unit, each of which is connected to the access control module, and the password input unit is used to provide a method for unlocking the Nth access control system for an outside visitor.

[0011] Furthermore, the password input unit is also connected to an early warning unit, the early warning unit is connected to a wireless communication unit, the wireless communication unit is connected to a management end, the early warning unit is used to generate an early warning message after the number of unlocking errors exceeds a preset number, and send the early warning message to the wireless communication unit, the wireless communication unit is used to receive the early warning message sent by the early warning unit, and further send it to the management end to remind the management personnel.

[0012] Furthermore, the Nth access control further comprises: a voice unit, the voice unit being connected to the unrecognized identification detection unit, the voice unit being used to detect the unrecognized identification detection unit in the interval [M 1 , M m / 2 ] When the number of unrecognized identification marks detected in the area is greater than 1, a preset reminder voice is played to remind multiple people waiting to pass through in the identification area N to keep a certain distance.

[0013] Furthermore, the management end includes: a remote control unit and a display unit, the remote control unit is used to send remote control instructions to one or more of the access control modules corresponding to n access control systems, thereby controlling the electronic lock to open to unlock the Nth access control system, and the display unit is used to display the number of successful unlocking, the number of failed unlocking and the corresponding unlocking time of the n access control systems in the form of a list.

[0014] The beneficial effects of the present invention are: The access control module controls the opening or closing of the Nth access control according to the recognition results sent by the identification recognition module and the face recognition module, and can judge the distribution of people to be passed through in the recognition area N according to different combinations of recognition results, thereby more flexibly adjusting the opening time of the Nth access control to avoid people in the first half of the recognition area N from following through the access control; at the same time, when there are multiple people to be passed through in the second half of the recognition area N, the identification recognition module does not send the recognition result to the access control module. At this time, the access control module still does not unlock the Nth access control after receiving the recognition result of successful face recognition, thereby avoiding congestion caused by multiple people to be passed through performing face recognition at the same time, and ensuring that multiple people to be passed through pass through the Nth access control in an orderly manner; in addition, setting m identifications to cover the recognition area N as much as possible can increase the identification recognition module's perception of the recognition area N, thereby improving the identification recognition module's recognition accuracy for whether there are people to be passed through in the recognition area N. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a structural diagram of a multi-access control management system.

[0016] Figure 2 This is a schematic diagram of the management side structure.

[0017] Figure 3It is a structural diagram of the identification module.

[0018] In the figure, 1. identification recognition module; 2. face recognition module; 3. access control module; 4. first storage unit; 5. management terminal; 6. electronic lock; 7. unidentified identification detection unit; 8. unidentified identification serial number comparison unit; 9. template update unit; 10. image acquisition unit; 11. preprocessing unit; 12. feature extraction unit; 13. second storage unit; 14. template matching unit; 15. threshold judgment unit; 16. recognition result transmission unit; 17. password input unit; 18. early warning unit; 19. wireless communication unit; 20. voice unit; 21. remote control unit; 22. display unit. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] The present invention provides Figures 1 to 3 In combination with the multi-access control management system based on image recognition shown, it is used to control n access controls in an area, and an identification area N is set in front of the Nth access control, wherein 1≤N≤n, and the Nth access control includes: an identification recognition module 1 whose identification direction faces the identification area N and a face recognition module 2 whose identification direction faces the front of the Nth access control, and the identification area N is provided with m identifications evenly distributed along the access control entry and exit direction, and the corresponding serial numbers of the identifications are recorded as M from back to front. 1 、M 2 , ..., M m ; The identification recognition module 1 is used to collect the image to be recognized in the recognition area N, and match it with the template image to generate a matching degree image, compare the matching degree with the threshold to determine the number of identifications, and then send the recognition result to the access control module 3; the face recognition module 2 is used to match the recognized face information with the face information of the first storage unit 4, and after the match is successful, the third recognition result is fed back to the access control module 3; the access control module 3 is used to receive the recognition result and the remote control instruction sent by the management terminal 5, and control the electronic lock 6 according to the recognition result or the remote control instruction; the management terminal 5 is used to view the recognition results of the identification recognition module 1 and the face recognition module 2, and can remotely control the electronic locks 6 of n access control in the control area; wherein, the identification recognition module 1 compares the image to be recognized with the template image N in sequence 11 and template image N 12 Match and generate matching degree images N respectively 13And the matching image N 14 , the matching degree image N 13 The matching degrees of the m / 2 marks corresponding to the areas in the image are compared with the threshold value in turn. If the matching degree is greater than the threshold value, it is determined that there is a mark in the corresponding area of ​​the image to be identified. The m / 2 judgment results are integrated. If there are (m / 2-1) marks, it is determined that there is only one person to pass behind the identification area N, and the first recognition result is sent to the access control module 3. At the same time, the matching degree image N is also sent to the access control module 3. 14 The matching degrees of the m / 2 identification areas corresponding to the images are compared with the thresholds in turn. If the matching degree is greater than the threshold, it is determined that there are identifications in the corresponding areas of the image to be identified. The m / 2 judgment results are integrated. If there are m / 2 identifications, it is determined that there are no following persons in front of the identification area N, and the second identification result is sent to the access control module 3; when the access control module 3 receives the first identification result, the second identification result and the third identification result at the same time, the electronic lock 6 is controlled to open to unlock the Nth access control, and the Nth access control is closed after a time T1; when the access control module 3 only receives the first identification result and the third identification result, the electronic lock 6 is controlled to open to unlock the Nth access control, and the Nth access control is closed after a time T1 / 2.

[0021] Among them, the multi-access control management system is mainly used to uniformly manage n access control at n entrances and exits of a community. For the convenience of explanation, the Nth access control among the n access control is taken as an example for explanation, and the remaining (n-1) access control are the same, and will not be repeated below; for the convenience of description, the location of the Nth access control is taken as the rear, and the location of the area outside the side of the Nth access control before entering the community is taken as the front, and this area is the identification area N corresponding to the Nth access control. The identification area N is rectangular and its length direction is parallel to the entry and exit direction of the Nth access control. The middle part of the left side of the Nth access control is set There is a face recognition module 2, and the top of the Nth access control is connected to the identification recognition module 1 through a horizontally arranged connecting rod, so that the identification recognition module 1 is located directly above the identification area N corresponding to the Nth access control, so that the identification recognition module 1 has a recognition direction that is vertically downward from directly above the identification area N, which is conducive to the identification recognition module 1 collecting the image to be recognized in the identification area N, and to a certain extent improves the recognition accuracy of the identification recognition module 1; m identifications are arranged in the rectangular identification area N, evenly distributed along the entry and exit direction of the Nth access control, and the corresponding serial numbers of the identifications are recorded as M from the back to the front. 1 、M 2 , ..., M m , that is, taking the location of the Nth access control as a reference, the identification number closest to the Nth access control is M 1 , the identification number farthest from the Nth access control is M m , it can be seen that in the interval [M 1 , M m / 2The marks in ] are evenly distributed in the second half of the recognition area N, located in the interval (M m / 2 , M m ] are evenly distributed in the first half of the recognition area N, and m is an even number, so that the number of logos in the second half of the recognition area N is equal to the number of logos in the first half of the recognition area N, so that the logo recognition module 1 can subsequently compare the image to be recognized with the template image N in sequence. 11 and template image N 12 Matching; by setting multiple identifications to cover the recognition area N as much as possible, the identification recognition module 1 can increase the perception ability of the recognition area N, thereby improving the identification recognition module 1 to identify whether there is a person in the recognition area N recognition accuracy.

[0022] Furthermore, when a person who is about to pass through walks from outside the community to the face recognition module 2 for face recognition, he or she will also be located in the second half of the recognition area N. At this time, the body of the person who is about to pass through will block the area in the interval [M 1 , M m / 2 ], the identification module 1 collects the image to be identified in the identification direction vertically downward from the top of the identification area N, then the image to be identified collected at this time has a total of (m-1) identifications; specifically, the identification module 1 uses the normalized square difference matching method to sequentially match the image to be identified collected at this time with the template image N 11 and template image N 12 Match and generate matching degree images N respectively 13 And the matching image N 14 , firstly, the matching image N 13 The matching degree of the m / 2 marks corresponding to the area in the image is compared with the threshold in turn. If the matching degree is greater than the threshold, it is judged that there is a mark in the second half of the corresponding identification area N in the image to be identified. 1 , M m / 2 ], so the matching degree corresponding to the identifier will be less than the threshold, then it is judged that the identifier does not exist in the second half of the corresponding identification area N in the image to be identified, and m / 2 judgment results are integrated. At this time, there will be (m / 2-1) identifiers in the second half of the corresponding identification area N in the image to be identified. Therefore, it is judged that there is only one person to pass behind the identification area N, and the first recognition result is sent to the access control module 3; at the same time, the matching degree image N 14 The matching degree of the m / 2 marks corresponding to the area in the image is compared with the threshold in turn. If the matching degree is greater than the threshold, it is judged that there is a mark in the first half of the corresponding identification area N in the image to be identified. If there is no following person blocking the interval (M m / 2 , M m Any symbol in the interval (M m / 2 , Mm ] The matching degree corresponding to the identification in the image to be identified will be greater than the threshold, and m / 2 judgment results will be integrated. At this time, there will be m / 2 identifications in the front half of the corresponding identification area N in the image to be identified. It is judged that there is no following person in front of the identification area N, and the second recognition result is sent to the access control module 3; when the access control module 3 receives the first recognition result and the second recognition result at the same time, it means that when a person to be passed is performing face recognition in the second half of the identification area N, there is no following person in the front half of the identification area N. If the access control module 3 also receives the third recognition result sent by the face recognition module 2 after successful recognition, the access control module 3 controls the electronic lock 6 to open to unlock the Nth access control, and after time T1 Close the Nth access control; when the access control control module 3 only receives the first recognition result and the third recognition result, it means that a person to be passed through performs face recognition in the second half of the recognition area N and the face recognition is successful, but there is a follower in the first half of the recognition area N. When there is a follower and the follower is far away from the Nth access control entrance, the electronic lock 6 is still controlled to open to unlock the Nth access control, but the Nth access control is immediately closed after the time T1 / 2 has passed. By shortening the opening time of the Nth access control, it is avoided that the follower in the first half of the recognition area N follows through the Nth access control, thereby improving the security of the access control; it should be noted that the template image in this embodiment includes the template image N 11 and template image N 12 By setting two template images so that the identification recognition module 1 can recognize the identification in the second half and the first half of the identification area N respectively, when the management personnel need to update the template image later, the template image corresponding to the second half or the first half of the identification area N can be updated, which can effectively reduce the workload of the management personnel in updating the template image later. 11 The preset identification is based on the interval [M 1 , M m / 2 The arrangement order of the m / 2 marks in ] is set accordingly and the areas other than the marks are blank areas. The template image N 12 The preset identification is based on the interval (M m / 2 , M m ] are arranged in a corresponding order and other areas except the logos are blank areas. The logos preset in the template image are arranged in a corresponding order according to the arrangement order of the logos in the interval, which is conducive to accurate recognition by the logo recognition module 1. The areas except the logos in the template image are also set as blank areas, thereby reducing the matching degree between the areas except the logos in the template image and the corresponding areas of the image to be recognized, thereby avoiding the logo recognition module 1 from misidentifying other objects in the image to be recognized as logos, thereby improving the recognition accuracy of the logo recognition module 1 to a certain extent.

[0023] Furthermore, the above mainly describes two situations in which the access control module 3 controls the electronic lock 6 to open, that is, when the access control module 3 receives the first recognition result, the second recognition result and the third recognition result at the same time or only receives the first recognition result and the third recognition result, it controls the electronic lock 6 to open to unlock the Nth access control, and the Nth access control is not unlocked in all other situations; to avoid redundancy, some situations in which the Nth access control is not unlocked are briefly described here. Specifically, when the access control module 3 does not receive the first recognition result, it means that the number of identifications recognized by the identification recognition module 1 in the second half of the corresponding identification area N in the image to be recognized is greater than or less than (m / 2-1), indicating that there is no person to be passed in the second half of the identification area N or there are There are multiple people to pass through, and the second half of the identification area N is closer to the Nth access control entrance. When there are multiple people to pass through, opening the Nth access control will greatly increase the risk of the people to pass through following through the access control. It can also avoid the congestion problem caused by the simultaneous face recognition of multiple people to pass through, and ensure that multiple people to pass through pass through the Nth access control in an orderly manner. Therefore, the access control control module 3 does not unlock the Nth access control when the first recognition result is not received; when the access control module 3 does not receive the third recognition result, it means that the face recognition module 2 fails to match the face information of the person to pass with the face information of the first storage unit 4, indicating that the person to pass is not an internal member of the community, so the access control module 3 does not unlock the Nth access control when the third recognition result is not received.

[0024] In addition, the above mainly describes the situation where the person to be passed through passes through the Nth gate from the outside of the community to enter the community. Here, a brief description is given of the situation where the person to be passed through passes through the Nth gate from the inside of the community to reach the outside of the community. Specifically, another face recognition module 2 is arranged in the middle of the right side of the Nth gate. The recognition direction of the face recognition module 2 located in the middle of the right side of the Nth gate is facing the rear of the Nth gate. When the people inside the community need to go out, face recognition is performed at the face recognition module 2 facing the rear of the Nth gate in the recognition direction. After the face recognition is successful, the face recognition module 2 directly controls the electronic lock 6 to open through the access control module 3 to unlock the Nth gate, thereby ensuring the normal travel of the people inside the community.

[0025] In an optional embodiment, if Figure 1 As shown, the identification recognition module 1 is also connected to the unidentified identification detection unit 7, the unidentified identification detection unit 7 is connected to the unidentified identification serial number comparison unit 8, the unidentified identification serial number comparison unit 8 is connected to the electronic lock 6, the unidentified identification detection unit 7 is used to compare the preset identification number and the actual identification number to determine whether there is an unidentified identification, and send the unidentified identification serial number to the unidentified identification serial number comparison unit 8, the unidentified identification serial number comparison unit 8 is used to compare the interval [M1 , M m / 2 ] in the unrecognized identification number and interval (M m / 2 , M m ] are subtracted in sequence, and then the difference with the smallest value is obtained, and the access control opening time T1 is adjusted according to the size of the smallest difference.

[0026] When there are (m / 2-1) marks in the second half of the corresponding identification area N in the image to be identified, the mark identification module 1 determines that there is only one person to pass behind the identification area N. At this time, the unidentified mark number comparison unit 8 will set the interval [M 1 , M m / 2 ] is the only unrecognized identification number and interval (M m / 2 , M m ] are subtracted in sequence, thereby obtaining the difference in distance between a person to be passed in the second half of the identification area N and one or more persons to be passed in the first half of the identification area N, and then the difference with the smallest value is obtained, which represents the distance between a person to be passed in the second half of the identification area N and a person to be passed closest to the person to be passed, and the opening time T1 of the Nth access control is flexibly adjusted according to the size of the smallest difference; specifically, when the smallest difference is larger, it means that the distance between the two is far, and the opening time T1 of the Nth access control can be appropriately increased to enhance the flexibility and intelligence of the access control management system; when the smallest difference is smaller, it means that the distance between the two is close, and the opening time T1 of the Nth access control can be appropriately reduced to avoid the person to be passed in the first half of the identification area N following the person to be passed in the second half of the identification area N through the Nth access control, thereby improving the security of the Nth access control to a certain extent.

[0027] In an optional embodiment, if Figure 1 As shown, the identification recognition module 1 is also connected to the template updating unit 9, and the template updating unit 9 is connected to the management terminal 5. The template updating unit 9 is used to receive the new template image sent by the administrator through the management terminal 5 to implement the template image N in the identification recognition module 1. 11 and template image N 12 Updates.

[0028] The environment of the Nth access control station may change over time, for example, it may be affected by factors such as light, angle, and occlusion. In this embodiment, a template updating unit 9 is provided to update the template image N of the Nth access control station. 11 and template image N 12The template image N is updated so that the multi-access control management system can better adapt to the above-mentioned environmental changes and improve the stability and accuracy of recognition. At the same time, updating the template image can enable the multi-access control management system to recognize new identification types, thereby maintaining the advancement and practicality of the system. In addition, by continuously updating the template image N 11 and template image N 12 , the multi-access control management system can learn and adapt to new recognition patterns, thereby improving its intelligence and adaptability, enabling it to better meet user needs.

[0029] To sum up, setting the template update unit 9 to update the template image is one of the important means to maintain the performance of the multi-access control management system and continuously improve its functions. Through regular updates, the multi-access control management system can keep pace with environmental and technological changes, thereby ensuring its long-term stable operation and providing the best service.

[0030] In an optional embodiment, if Figure 1 and Figure 3 As shown in combination, the identification recognition module 1 includes: an image acquisition unit 10, a preprocessing unit 11, a feature extraction unit 12 and a second storage unit 13. The image acquisition unit 10 is used to acquire the image to be identified in the identification area N and send it to the preprocessing unit 11. The preprocessing unit 11 preprocesses the image to be identified to improve the accuracy of subsequent matching. The feature extraction unit 12 is used to extract relevant features of m identifications from the preprocessed image for template matching. The second storage unit 13 is used to store the template image N corresponding to the identification area N. 11 and template image N 12 , for the template matching unit 14 to read during matching.

[0031] Among them, in actual application, the preprocessing unit 11 can selectively perform one or more of the preprocessing steps such as grayscale processing, size standardization, noise removal, edge detection and image enhancement according to actual needs, thereby improving the accuracy and stability of the subsequent template matching unit 14.

[0032] In an optional embodiment, if Figure 1 and Figure 3 As shown in the figure, the identification recognition module 1 also includes: a template matching unit 14, a threshold judgment unit 15 and a recognition result transmission unit 16. The template matching unit 14 uses a normalized square difference matching method to sequentially match the image to be recognized with the template image N. 11 and template image N 12 Match and generate matching degree images N respectively 13 And the matching image N 14The threshold judgment unit 15 is used to compare the matching degree with the preset threshold to determine whether there is an identification in the corresponding area of ​​the image to be identified, and then integrate the judgment results, and generate or not generate an identification result in combination with the number of identifications existing in the image to be identified. The identification result transmission unit 16 is used to send the first identification result or the second identification result generated by the threshold judgment unit 15 to the access control module 3.

[0033] The template matching unit 14 uses the normalized square difference matching method to sequentially match the collected image to be identified with the template image N 11 and template image N 12 Match and generate matching degree images N respectively 13 And the matching image N 14 The threshold judgment unit 15 first matches the image N 13 The matching degrees of the m / 2 marks corresponding to the areas in the image are compared with the threshold value in turn. If the matching degree is greater than the threshold value, it is determined that there is a mark in the second half of the corresponding identification area N in the image to be identified. The m / 2 judgment results are integrated. If there are (m / 2-1) marks in the second half of the corresponding identification area N in the image to be identified, it is determined that there is only one person to pass behind the identification area N and a first identification result is generated. The identification result transmission unit 16 sends the first identification result to the access control module 3. The threshold judgment unit 15 also sends the matching degree image N to the access control module 3. 14 The matching degrees of the m / 2 identification areas corresponding to the image are compared with the thresholds in turn. If the matching degree is greater than the threshold, it is determined that there is an identification in the front half of the corresponding identification area N in the image to be identified. The m / 2 judgment results are integrated. If there are m / 2 identifications in the front half of the corresponding identification area N in the image to be identified, it is determined that there is no following person in front of the identification area N and a second identification result is generated. The identification result transmission unit 16 sends the second identification result to the access control module 3.

[0034] In an optional embodiment, if Figure 1 As shown, the Nth access control further includes: a password input unit 17, and the password input unit 17 is connected to the access control module 3 respectively, and the password input unit 17 is used to provide a way for external visitors to unlock the Nth access control.

[0035] Among them, the password input unit 17 is used to input a temporary digital password. When the temporary digital password is entered correctly, the access control module 3 controls the electronic lock 6 to open to unlock the Nth access control, thereby providing certain convenience for external visitors who have access rights but have not recorded facial recognition information to enter and exit the Nth access control.

[0036] In an optional embodiment, if Figure 1As shown, the password input unit 17 is also connected to the early warning unit 18, the early warning unit 18 is connected to the wireless communication unit 19, the wireless communication unit 19 is connected to the management terminal 5, the early warning unit 18 is used to generate an early warning message after the number of unlocking errors exceeds a preset number, and send the early warning message to the wireless communication unit 19, the wireless communication unit 19 is used to receive the early warning message sent by the early warning unit 18, and further send it to the management terminal 5 to remind the management personnel.

[0037] Among them, when the number of times the temporary digital password is entered exceeds the preset number of times, the early warning unit 18 determines that an outsider who does not have the right to enter and exit has attempted to unlock the Nth access control by entering the temporary digital password multiple times. At this time, the early warning unit 18 generates an early warning message and sends it to the management terminal 5 through the wireless communication unit 19, so that the management personnel can understand the status of the Nth access control in time.

[0038] In an optional embodiment, if Figure 1 As shown, the Nth access control further includes: a speech unit 20, the speech unit 20 is connected to the unrecognized identification detection unit 7, and the speech unit 20 is used to detect the unrecognized identification detection unit 7 in the interval [M 1 , M m / 2 ] When the number of unrecognized identification marks detected in the area is greater than 1, a preset reminder voice is played to remind multiple people waiting to pass through in the identification area N to keep a certain distance.

[0039] Among them, the unrecognized mark detection unit 7 is in the interval [M 1 , M m / 2 ] when the number of unrecognized identification marks detected is greater than 1, it indicates that there are multiple people waiting to pass through in the second half of the recognition area N. At this time, the voice unit 20 plays a preset reminder voice to remind multiple people waiting to pass to queue up while keeping a certain distance, thereby avoiding congestion caused by multiple people waiting to pass through simultaneously performing face recognition, and ensuring that multiple people waiting to pass through pass through the Nth access control in an orderly manner.

[0040] In an optional embodiment, if Figure 2 As shown, the management terminal 5 includes: a remote control unit 21 and a display unit 22. The remote control unit 21 is used to send remote control instructions to one or more of the access control modules 3 corresponding to n access control systems, thereby controlling the electronic lock 6 to open to unlock the Nth access control system. The display unit 22 is used to display the number of successful unlocking, the number of failed unlocking and the corresponding unlocking time of the n access control systems in the form of a list.

[0041] Among them, the community is generally equipped with multiple entrances and exits, and multiple entrances and exits will correspond to the access control system. The management terminal 5 in this embodiment is equipped with a remote control unit 21, so that the management personnel can remotely and uniformly manage the access control modules 3 corresponding to n access control systems through the management terminal 5, so as to facilitate the management personnel to manage the community with multiple access control systems; in addition, the display unit 22 is used to display the number of successful unlocking, the number of failed unlocking and the corresponding unlocking time of n access control systems in the form of a list, so as to facilitate the management personnel to check the unlocking failure situation and improve the management efficiency of the management personnel.

[0042] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.

Claims

1. A multi-access control management system based on image recognition, used to control n access control in an area, It is characterized in that There is an identification area N in front of the Nth access control, where 1≤N≤n. The Nth access control includes: an identification recognition module whose identification direction faces the identification area N and a face recognition module whose identification direction faces the front of the Nth access control. The identification area N is provided with m identifications evenly distributed along the access control entry and exit direction. The corresponding serial numbers of the identifications are recorded as M from the back to the front. 1 、M 2 , ..., M m ; The identification recognition module is used to collect the image to be recognized in the recognition area N, match it with the template image to generate a matching degree image, compare the matching degree with the threshold to determine the number of identifications, and then send the recognition result to the access control module; The face recognition module is used to match the recognized face information with the face information of the first storage unit, and feed back the third recognition result to the access control module after the match is successful; The access control module is used to receive the recognition result and the remote control command sent by the management terminal, and control the electronic lock according to the recognition result or the remote control command; The management terminal is used to view the recognition results of the identification recognition module and the face recognition module, and can remotely control the electronic locks of n access control doors in the control area; The identification module compares the image to be identified with the template image N in sequence. 11 and template image N 12 Match and generate matching degree images N respectively 13 And the matching image N 14 , the matching degree image N 13 The matching degrees of the m / 2 marks in the corresponding areas are compared with the threshold in turn. If the matching degree is greater than the threshold, it is determined that there is a mark in the corresponding area of ​​the image to be identified. The m / 2 judgment results are integrated. If there are (m / 2-1) marks, it is determined that there is only one person to pass behind the identification area N, and the first recognition result is sent to the access control module. At the same time, the matching image N is also sent. 14 The matching degrees of the areas corresponding to the m / 2 identifications in the image are compared with the threshold value in turn. If the matching degree is greater than the threshold value, it is determined that there is an identification in the area corresponding to the image to be identified, and the m / 2 judgment results are integrated. If there are m / 2 identifications, it is determined that there is no following person in front of the identification area N, and the second identification result is sent to the access control module; when the access control module receives the first identification result, the second identification result and the third identification result at the same time, the electronic lock is controlled to open to unlock the Nth access control, and the Nth access control is closed after time T1; when the access control module only receives the first identification result and the third identification result, the electronic lock is controlled to open to unlock the Nth access control, and the Nth access control is closed after time T1 / 2.

2. According to the multi-access control management system based on image recognition according to claim 1, It is characterized in that The identification module is also connected to an unidentified identification detection unit, which is connected to an unidentified identification serial number comparison unit, which is connected to the electronic lock, and the unidentified identification detection unit is used to compare the preset number of identifications with the number of identifications actually identified to determine whether there are unidentified identifications, and send the unidentified identification serial numbers to the unidentified identification serial number comparison unit, and the unidentified identification serial number comparison unit is used to compare the interval [M 1 , M m / 2 ] in the unrecognized identification number and interval (M m / 2 , M m ] are subtracted in sequence, and then the difference with the smallest value is obtained, and the time T1 for opening the Nth access control is adjusted according to the size of the difference with the smallest value.

3. According to the multi-access control management system based on image recognition according to claim 2, It is characterized in that The identification module is also connected to a template updating unit, which is connected to the management terminal. The template updating unit is used to receive a new template image sent by the management terminal to implement the template image N in the identification module. 11 and template image N 12 Updates.

4. According to the multi-access control management system based on image recognition according to claim 1, It is characterized in that The identification recognition module includes: an image acquisition unit, a preprocessing unit, a feature extraction unit and a second storage unit. The image acquisition unit is used to acquire the image to be identified in the identification area N and send it to the preprocessing unit. The preprocessing unit preprocesses the image to be identified to improve the accuracy of subsequent matching. The feature extraction unit is used to extract relevant features of m identifications from the preprocessed image for template matching. The second storage unit is used to store the template image N corresponding to the identification area N. 11 and template image N 12 , for the template matching unit to read during matching.

5. According to the multi-access control management system based on image recognition according to claim 4, It is characterized in that The identification recognition module also includes: a template matching unit, a threshold judgment unit and a recognition result transmission unit. The template matching unit sequentially matches the image to be recognized with the template image N by a normalized square difference matching method. 11 and template image N 12 Match and generate matching degree images N respectively 13 And the matching image N 14 The threshold judgment unit is used to compare the matching degree with the preset threshold to determine whether there is an identification in the corresponding area of ​​the image to be identified, and then integrate the judgment results, and generate or not generate an identification result based on the number of identifications existing in the image to be identified. The identification result transmission unit is used to send the first identification result or the second identification result generated by the threshold judgment unit to the access control module.

6. According to the multi-access control management system based on image recognition according to claim 1, It is characterized in that The Nth access control also includes: a password input unit, which is connected to the access control module respectively, and is used to provide a way for external visitors to unlock the Nth access control.

7. According to claim 6, a multi-access control management system based on image recognition, It is characterized in that The password input unit is also connected to the early warning unit, the early warning unit is connected to the wireless communication unit, the wireless communication unit is connected to the management end, the early warning unit is used to generate early warning information after the number of unlocking errors exceeds a preset number, and send the early warning information to the wireless communication unit, the wireless communication unit is used to receive the early warning information sent by the early warning unit, and further send it to the management end to remind the management personnel.

8. According to the multi-access control management system based on image recognition as claimed in claim 1, It is characterized in that The Nth access control further includes: a voice unit, the voice unit being connected to the unrecognized identification detection unit, the voice unit being used to detect the unrecognized identification in the interval [M 1 , M m / 2 ] When the number of unrecognized identification marks detected in the area is greater than 1, a preset reminder voice is played to remind multiple people waiting to pass through in the identification area N to keep a certain distance.

9. The multi-access control management system based on image recognition according to claim 1, It is characterized in that The management end includes: a remote control unit and a display unit, the remote control unit is used to send remote control instructions to one or more of the access control modules corresponding to n access control systems, thereby controlling the electronic lock to open to unlock the Nth access control system, and the display unit is used to display the number of successful unlocking, the number of failed unlocking and the corresponding unlocking time of the n access control systems in the form of a list.

Images