Face recognition system and method for data security based on door lock

Abstract

The application relates to the field of face recognition applications, in particular to a face recognition system and method based on a door lock and ensuring data security, which comprises a collection end S1, a camera module S11, a face algorithm module S12 and a network transmission module S13; the camera module S11 comprises a plurality of cameras in different directions; the face algorithm module S12 is used for image scanning and judgment of local image information in all directions and extraction of local feature data; and the network transmission module S13 is used for encryption, packaging, labeling and transmission of all local feature data; the application provides a face recognition system and method based on a door lock and ensuring data security, different parts of a face are recognized through a plurality of cameras contained in the collection end, thereby controlling extraction of feature quantity, distinguishing and protecting features, preventing face data from being read, and realizing safety control of face data in face recognition of a door lock.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of face recognition applications, in particular to a face recognition system and method based on door locks for data security. BACKGROUND

[0002] Face recognition technology is a kind of biometric technology based on facial feature information. By collecting facial data information, comparison and recognition are performed. With the continuous development of face recognition technology, various face recognition devices have been widely applied. Face recognition technology is also widely used in the intelligent door lock industry. The intelligent door lock collects the face through the front camera and cooperates with the server to identify and authenticate. In the actual application process, it is usually necessary to record a large amount of identity information of the identified person and collect face pictures. At present, in public places such as hotels and apartments, most of the solutions for the use of intelligent door locks in various scenarios are to collect user pictures through a mobile app or a program, and upload the user pictures to the server for registration, so that the terminal can compare the collected picture data uploaded to the server and then perform comparison and recognition, and finally the server feeds back the recognition result to the identification terminal. Since the collected picture data is uploaded to the terminal server, there is a risk that the user's face picture information will be retained or leaked during the identification and feedback process. At the same time, the processing capacity of the intelligent door lock is slow, and it needs to process the recognition result returned by the server network, which results in a relatively long recognition time and a poor user experience. The current face recognition system used by the intelligent door lock is a conventional encryption transmission of data information, and personnel in the relevant field can steal information by methods such as dictionary attack and encryption cracking, so the security level is not high. Face data is personal sensitive information, and once it is leaked, it will cause great security problems and even property losses to the user. SUMMARY

[0003] The purpose of the present application is to solve the problem of insufficient protection of face picture data security in the face recognition scene of the door lock in the current face recognition technology application scene, and to provide a face recognition system and method based on door locks for data security.

[0004] In order to achieve the above purpose, the technical scheme adopted by the present application is as follows:

[0005] A face recognition system based on door locks for data security, comprising:

[0006] The collection end S1 includes a camera module S11, a face algorithm module S12, and a network transmission module S13. The camera module S11 includes several cameras in different directions, which independently collect effective local image information in their own directions. The face algorithm module S12 is used for image scanning and judgment of local image information in all directions, extraction of local feature data. The network transmission module S13 is used for encryption, packaging, labeling, and transmission of all local feature data.

[0007] The transfer end S2 is used for transmitting encrypted data.

[0008] The client end S3 is used for data reception, storage, and comparison of related data.

[0009] Preferably, the camera module S11 includes a left camera S111, a main camera S112, and a right camera S113, which are arranged horizontally on an arc surface. The camera module S11 collects different face image information from three field angles, respectively. The left camera S111 collects left image data of the face, the main camera S112 collects front image data of the face, and the right camera S113 collects right image data of the face.

[0010] Preferably, the collection end S1 only dynamically generates local face information in different directions.

[0011] Preferably, the face algorithm module S12 is used for feature extraction of image data, including first feature data J1 for identifying left face contour and left ear, second feature data J2 for identifying eyes, nose, lips, and chin, and third feature data J3 for identifying right face contour and right ear.

[0012] Preferably, the network transmission module S13 is used for adding and packaging output of the first feature data J1, the second feature data J2, and the third feature data J3.

[0013] The application discloses a face recognition method based on door lock data security, which comprises the following steps: collecting local image information of a user's face in different directions by a collection terminal S1 to generate local face feature data Jn in the corresponding direction, encrypting and packing the local face feature data Jn in different directions by position, and adding a serial number or a label; forwarding the packed data to a client S3 by a transfer terminal S2 in a predetermined data transmission mode; decrypting the packed data by the client S3, restoring the local face feature data Jn in different directions, and then fusing the local face feature data Jn in all directions to generate overall face feature data F1 by an algorithm, and storing the overall face feature data F1 in the client; collecting face information F2 on site when the user opens the lock in the client, comparing the face information F2 with the overall face feature data F1, and determining whether to perform the opening action according to the comparison result.

[0014] Preferably, the local face feature data Jn is transmitted by twice encryption or a self-defined data format.

[0015] Preferably, the collection terminal S1, the transfer terminal S2 and the client S3 are arranged in a local area network.

[0016] Preferably, after receiving all the encrypted data, the client S3 deletes the valid local image information Jn in all directions collected by the collection terminal S1, and deletes all the encrypted data cached by the transfer terminal S2.

[0017] Preferably, the collection terminal S1 adopts a low storage scheme, and only generates the local face information Jn in different directions dynamically, without storing face picture information or face feature data.

[0018] Beneficial effects: the application provides a face recognition system and method based on door lock data security, which collects different parts of a face by a collection terminal and multiple cameras contained in the collection terminal, controls the extraction of feature quantity, distinguishes and protects the features, prevents the face data from being read, realizes the safety control of face data in door lock face recognition, provides safety protection for the face data in the whole data link process from face data collection to transmission link and face data verification and recognition, avoids illegal acquisition, use and tampering of face picture information and face data, and guarantees the data safety of the face recognition use scene. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 is a flow chart of the face recognition system and method based on door lock data security in the embodiment of the application;

[0020] Figure 2This is a schematic diagram of the workflow of the face data acquisition terminal in an embodiment of the present invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and key points of this invention clearer, the following will describe the invention in further detail with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0022] like Figure 1 As shown in the embodiment of this solution, a face recognition system based on door lock data security is provided, which consists of a collection terminal S1, a relay terminal S2 and a client S3. The system collects local image information of the user's face from different directions through the collection terminal S1, generates local face feature data Jn in the corresponding directions, and performs encrypted transmission.

[0023] The acquisition terminal S1 is equipped with a camera module S11, which scans faces. To ensure the protection of user facial data, the camera module S11 uses three sets of cameras: a left camera S111, a main camera S112, and a right camera S113. The left camera S111, the main camera S112, and the right camera S113 are horizontally arranged on the same arc surface with an included angle of 150 degrees, so that the left camera S111, the main camera S112, and the right camera S113 form an enclosing trend, constituting a triangular image acquisition area on the arc surface. During the face scanning process, facial information is recorded and scanned from three different angles.

[0024] The collection end S1 is not only provided with the camera module S11, but also includes a face algorithm module S12. The face algorithm module S12 cooperates with the scanned and recorded information to simplify and adjust, and performs deep scanning on the data collected by the camera module S11 to form feature data. The left camera S111 collects left face image data and forms feature data of the left face contour and the left ear under the calculation of the face algorithm module S12. As described above, the main camera S112 cooperates with the face algorithm module S12 to collect front face image data to form feature data of the eyes, nose, lips and chin of the face. The right camera S113 cooperates with the face algorithm module S12 to collect right face image data to form feature data of the right face contour and the right ear. At least three times of image collection are performed during the collection process, and multiple scanning and comparison are performed to collect the effective face data of the user as accurately as possible, so as to ensure the authenticity and effectiveness of the data and facilitate the client to quickly identify and compare. The face algorithm module S12 finally forms three effective feature data, which are respectively a first feature data J1 for identifying the left face contour and the left ear, a second feature data J2 for identifying the eyes, nose, lips and chin, and a third feature data J3 for identifying the right face contour and the right ear.

[0025] To ensure safety, the collection end adopts a low storage scheme, and only different local face information is dynamically generated. After transmission, the face information is deleted, and no related feature data is retained, so as to ensure the safety of the face data of the user.

[0026] After the feature data is generated, the corresponding data needs to be encrypted to prevent data theft during transmission. Therefore, the collection end S1 further includes a network transmission module S13. The network transmission module S13 is used to uniformly encrypt the local feature data Jn generated by the face algorithm module S12. Each feature data is separately packaged and encrypted, and then different labels are allocated as identifiers according to the predetermined rules of the cameras S111, S112 and S113. After the encrypted data of each local feature is added with an identifier, it is packaged and output to the transfer end S2.

[0027] The face recognition system based on the data security of the door lock further includes a transfer end S2. The transfer end S2 is an information transmission device, which further transmits the data received by the network transmission module S13.

[0028] The face recognition system based on the data security of the door lock further includes a client S3 for decryption and identification. The client S3 is in communication connection with the transfer end S2. The client S3 receives all the packaged encrypted data through the agreed data transmission mode. The transfer end S2 performs secondary encryption or an agreed private transmission protocol, so as to ensure safety and ensure the safety of the transmission between the transfer end S2 and the client S3.

[0029] The client S3 receives all packaged data, decrypts the package, restores the local facial feature data Jn of different orientations, and then generates the overall facial feature data F1 by fusing all orientations of the local facial feature data Jn using an algorithm, and stores the overall facial feature data F1 in the client S3; when the user unlocks the client S3, the facial feature information F2 is collected on site, compared with the overall facial feature data F1 that has been stored and issued, and whether the unlocking action needs to be performed is determined through the comparison result; when the comparison work is performed, the facial feature data can be fused into a complete face F1 after the partial facial data J1, J2 and J3 collected by the cameras S111, S112 and S113 are fused, and then output; the complete facial feature data F1 after fusion is used for comparison when the client S3 is compared, which can reduce the data fusion time during comparison and improve the comparison speed.

[0030] Further, to ensure the security of transmission, the network communication of the collection end S1, the transfer end S2 and the client S3 is placed in the internal local area network of hotels, apartments and the like, reducing the communication connection with the external Internet, and ensuring the security of data transmission and storage.

[0031] The present application provides a kind of based on door lock's data security face recognition system and method, realize the security control of face data in door lock face recognition, in the face recognition solution and use scene based on door lock, from the collection of face data, to the transmission link of face data, and face data verification identification, entire data link process all provide the security guarantee of face data.Avoid face picture information and face data are illegally acquired, used and tampered with, guarantee the data security of face recognition use scene.

[0032] The above only describes the preferred embodiments of the present application, and it should be pointed out that for ordinary skilled in the art, without departing from the principles of the present application, a number of improvements can be made, and these improvements should be considered as the protection scope of the present application.

Claims

1. A face recognition system based on door lock data security, characterized in that: The application relates to a face recognition system, which comprises a collection end S1, a transfer end S2 and a client end S3. The collection end S1 comprises a camera module S11, a face algorithm module S12 and a network transmission module S13, the camera module S11 comprises a plurality of cameras in different directions, the cameras independently collect effective local image information in their own directions, the face algorithm module S12 is used for image scanning and judgment of the local image information in all directions, extraction of local feature data, the network transmission module S13 is used for encryption, packaging, labeling and transmission of all local feature data, after the client end S3 receives all the packaged data, the client end S3 performs decryption and restores local face feature data Jn in different directions, then the client end S3 generates overall face feature data F1 by fusing the local face feature data Jn in all directions through an algorithm, and stores the overall face feature data F1 in the client end; when a user opens a lock in the client end, on-site face information F2 is collected, and the on-site face information F2 is compared with the overall face feature data F1 which has been stored and issued, and whether the opening action needs to be performed is judged according to the comparison result; the local face feature data Jn is transmitted through secondary encryption or self-defined data format; the transfer end S2 is used for transmission of encrypted data; and the client end S3 is used for data reception, storage and comparison of related data.

2. The face recognition system based on door lock for data security according to claim 1, characterized in that: The camera module S11 comprises a left camera S111, a main camera S112 and a right camera S113, the three cameras are horizontally arranged on an arc surface, the camera module S11 collects different face image information from three field angles, the left camera S111 collects left face image data, the main camera S112 collects front face image data, and the right camera S113 collects right face image data.

3. The face recognition system based on door lock for data security according to claim 2, characterized in that: The collection end S1 only dynamically generates local face information in different directions.

4. The face recognition system based on door lock for data security according to claim 2, characterized in that: The face algorithm module S12 is used for feature extraction of image data, and comprises first feature data J1, second feature data J2 and third feature data J3, the first feature data J1 is used for recognizing a left face contour and ears, the second feature data J2 is used for recognizing eyes, a nose, lips and a chin, and the third feature data J3 is used for recognizing a right face contour and ears.

5. The face recognition system based on door lock for data security according to claim 4, characterized in that: The network transmission module S13 is used for adding and packaging output of the first feature data J1, the second feature data J2 and the third feature data J3.

6. A face recognition method based on door lock data security, characterized in that: The face recognition system for the door lock-based data security according to any one of claims 1 to 5, by the collection end S1, the face of the user is separately collected in different orientations, the local face feature data Jn in the corresponding orientation is generated, the local face feature data Jn in different orientations is encrypted and packed with serial numbers or labels, the data transmission mode is agreed, the packaged data is forwarded to the client S3, after the client S3 receives all the packaged data, the decryption is carried out, the local face feature data Jn in different orientations is restored, then the local face feature data Jn in all orientations is fused to generate the overall face feature data F1 by using the algorithm, and is stored in the client; when the user opens the lock in the client, the face information F2 is collected on site, is compared with the overall face feature data F1 which has been stored and issued, and whether the opening action is needed is judged by the comparison result; the local face feature data Jn transmission adopts secondary encryption or custom data format for transmission.

7. The face recognition method based on door lock data security according to claim 6, characterized in that: The collection end S1, the transfer end S2 and the client S3 are all deployed in the local area network.

8. The face recognition method based on door lock data security according to claim 6, characterized in that: After the client S3 receives all the encrypted data after packing, the effective local image information Jn data in all orientations which has been collected in the collection end S1 is deleted, and all the encrypted data cached in the transfer end S2 is deleted.

Images