Face recognition system based on information technology

Abstract

The present invention relates to the field of face recognition technology, and in particular to a face recognition system based on information technology, comprising: a control terminal, which is the main control end of the system and is used to issue an execution command; a storage module, which is used to store user facial data; a camera module, which is used to collect user facial images in real time; a search module, which is used to receive user facial images collected by the camera module in real time, and use the collected user facial images as search targets to search for user facial data with the highest similarity in the storage module; the present invention provides user facial verification conditions by storing user facial data and collecting real-time user facial images, and in the process of user facial verification, by bidirectional segmentation, selection and further configuration of user facial images and user facial data, the user facial verification process is no longer dependent on the completed user facial images, thereby making the user facial verification process faster.

Description

Technical Field

[0001] The present invention relates to the field of face recognition technology, and in particular to a face recognition system based on information technology. Background Art

[0002] Face recognition is a biometric technology that identifies people based on their facial features. It uses a camera or camcorder to collect images or video streams containing faces, automatically detects and tracks faces in images, and then performs facial recognition on the detected faces. It is also commonly called portrait recognition or facial recognition.

[0003] However, the current usage scenarios of facial recognition technology as a method of access control gate verification often use the global information contained in the user's face for identification and verification. This method requires collecting the user's global facial image. When the user's face is blocked or unclear due to various reasons, this facial recognition technology will not be able to recognize and verify the user's face. Summary of the invention

[0004] In view of the above-mentioned shortcomings of the prior art, the present invention provides a face recognition system based on information technology, which solves the technical problems raised in the above-mentioned background technology.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0006] Face recognition system based on information technology, including:

[0007] The control terminal is the main control terminal of the system and is used to issue execution commands;

[0008] A storage module, used for storing user's facial data;

[0009] A camera module is used to collect the user's facial image in real time;

[0010] A search module is used to receive the user face image collected by the camera module in real time, use the collected user face image as the search target, and search for the user face data with the highest similarity in the storage module;

[0011] A selection module, used to select an image area in the user face image collected by the camera module, and a face area corresponding to the image area in the user face data searched by the search module;

[0012] The determination module is used to receive the image area in the user facial image and the facial area in the user facial data selected in the selection module, analyze the similarity between the image area and its corresponding facial area, calculate the mean similarity, and based on the mean similarity, determine whether the user corresponding to the user facial image currently collected by the camera module has been verified.

[0013] Furthermore, the storage module is provided with submodules at the lower level, including:

[0014] An uploading unit, used to upload the user's facial data;

[0015] A management unit, used to record the number of times the user face data stored in the storage module is called;

[0016] Among them, the system end user manually uploads the user's facial data through the uploading unit, the storage module receives the user facial data uploaded in the uploading unit, and stores the user facial data. A recording period is set in the management unit. The management unit records the number of times the user facial data stored in the storage module is called based on the recording period. After each recording period, an activity determination threshold is set in the management unit. The management unit calculates the user facial data retrieval frequency based on the recorded number of times the user facial data is called, and the calculated user facial data retrieval frequency is compared with the activity determination threshold. If the user facial data retrieval frequency is outside the activity determination threshold, the user facial data corresponding to the user facial data retrieval frequency outside the activity determination threshold is deleted in the storage module. Otherwise, it jumps to the next comparison stage of the user facial data retrieval frequency and the activity determination threshold.

[0017] Furthermore, the user facial data stored in the storage module is a three-dimensional user facial model.

[0018] Furthermore, the camera module is deployed on the surface of the barrier and the access control system. Each time the search module finds the user facial data with the highest similarity to the user facial image collected in real time by the camera module in the storage module, the search module synchronously feeds back the searched user facial data to the management unit, and the management unit accumulates the number of times the searched user facial data is called.

[0019] Furthermore, after the camera module collects the user's facial image, it simultaneously optimizes the user's facial image, and after the optimization is completed, it feeds back the user's facial image to the search module. The user's facial image optimization processing logic is expressed as:

[0020]

[0021] Where: F(u,v) is the optimized output user face image; I(x,y) is the original user face image; j is the imaginary unit; M and N are the width and height of the image; u and v are frequency domain coordinates, indicating the horizontal and vertical frequencies of the image; x and y are spatial domain coordinates, indicating the horizontal and vertical pixel positions of the image; d is the differential.

[0022] Furthermore, the selection module is internally provided with submodules, including:

[0023] A logic unit, used to set the selection logic when selecting the image area and the face area during the operation phase of the selection module;

[0024] Among them, the selection logic of the image area and the face area in the selection module includes: selecting the position and selecting the quantity;

[0025] The selection module is provided with submodules at the lower level, including:

[0026] A segmentation unit, used for segmenting the user face image and the user face data, so as to segment them into a plurality of image blocks of equal number and size, and in one-to-one correspondence;

[0027] The image blocks obtained by segmenting the user facial image and the user facial data are the image area and the facial area.

[0028] Furthermore, before segmenting the user facial data, based on the user facial image acquisition perspective, a user facial image of the same perspective is obtained on the user facial data, and the user facial image obtained on the user facial data and the user facial image collected in the camera module are used as processing targets for the selection module and the segmentation unit to run and process.

[0029] Furthermore, in the logic for selecting the image area and the facial area in the logic unit, when the image area and the facial area are selected twice in a row for the same set of user facial images and user facial data, the selected image area and the facial area are different, and the number of selections is subject to the following logic:

[0030]

[0031] Where: m is the number of image areas and facial areas selected; g is the base number of selections; t is the number of times the user's facial data is called; k is the current verification pass rate of the user's facial data; f is the balance of the number of times the user's facial data is called based on the recording period; λ is the correction;

[0032] The value of the number of selected image areas and facial areas, m, is rounded to the nearest integer, and the value of g is 1 or 2.

[0033] Furthermore, the similarity mean in the determination module is obtained by the following formula:

[0034]

[0035] Where: is the similarity between image region a and its corresponding facial region b; n is the total number of image regions and facial regions selected by the selection module; p a ∩p bis the number of pixels with the same feature vector in the intersection of the set of feature vectors of each pixel in image region a and the set of feature vectors of each pixel in image region b; p a ∪p b is the total number of pixels with different feature vectors in the union of the set of feature vectors of each pixel in image region a and the set of feature vectors of each pixel in image region b;

[0036] in, The judgment module determines that the verification is passed, otherwise, the verification is not passed. During the operation phase of the search module and the judgment module, when the image similarity is obtained, the similarity of the two groups of images can be expressed as

[0037] Furthermore, the control terminal is electrically connected to a storage module via a medium, the storage module is electrically connected to an upload unit and a management unit at a lower level via a medium, the storage module is electrically connected to a camera module and a search module via a medium, the search module is electrically connected to the management unit via a medium, the search module is electrically connected to a selection module and a determination module via a medium, and the interior and lower level of the selection module are electrically connected to a logic unit and a segmentation unit via a medium.

[0038] Compared with the known public technology, the technical solution provided by the present invention has the following beneficial effects:

[0039] 1. The present invention provides a face recognition system based on information technology. During operation, the system provides user face verification conditions by storing user face data and collecting real-time user face images. In the process of user face verification, the system performs two-way segmentation, selection and further configuration of user face images and user face data, so that the user face verification process no longer relies on the completed user face images, thereby making the user face verification process faster.

[0040] 2. During the operation of the system of the present invention, during the segmentation and selection stage of user facial images and user facial data, the system can adaptively adapt to the user's facial verification habits through the set segmentation and selection logic, thereby ensuring that the user facial verification process is fast and making the user facial verification results more accurate and reliable.

[0041] 3. During operation, the system of the present invention can also optimize the collected user facial images, so as to perform user facial verification with the optimized user facial images, ensuring that the user facial images collected by the system will not fail the user facial verification due to unclear images as much as possible. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0043] Figure 1 It is a structural diagram of a face recognition system based on information technology;

[0044] The numbers in the figure represent: 1. control terminal; 2. storage module; 21. upload unit; 22. management unit; 3. camera module; 4. search module; 5. selection module; 51. logic unit; 52. segmentation unit; 6. determination module. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are 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.

[0046] The present invention will be further described below in conjunction with the embodiments.

[0047] Example 1

[0048] The face recognition system based on information technology in this embodiment is as follows: Figure 1 As shown, including:

[0049] Control terminal 1 is the main control terminal of the system and is used to issue execution commands;

[0050] Storage module 2, used to store user's facial data;

[0051] The camera module 3 is used to collect the user's facial image in real time;

[0052] The search module 4 is used to receive the user face image collected by the camera module 3 in real time, and use the collected user face image as the search target to search for the user face data with the highest similarity in the storage module 2;

[0053] A selection module 5, used for selecting an image area in the user face image collected by the camera module 3, and a face area corresponding to the image area in the user face data searched by the search module 4;

[0054] The determination module 6 is used to receive the image area in the user facial image and the facial area in the user facial data selected by the selection module 5, analyze the similarity between the image area and the corresponding facial area, obtain the mean similarity, and determine whether the user corresponding to the user facial image currently collected by the camera module 3 has passed the verification based on the mean similarity;

[0055] The storage module 2 is provided with submodules at the lower level, including:

[0056] An uploading unit 21, used for uploading the user's facial data;

[0057] The management unit 22 is used to record the number of times the user face data stored in the storage module 2 is called;

[0058] Among them, the system end user manually uploads the user's facial data through the uploading unit 21, the storage module 2 receives the user facial data uploaded in the uploading unit 21, and stores the user facial data. The management unit 22 is provided with a recording cycle. The management unit 22 records the number of times the user facial data stored in the storage module 2 is called based on the recording cycle. After each recording cycle, the management unit 22 is provided with an activity determination threshold. The management unit 22 obtains the user facial data retrieval frequency based on the recorded number of times the user facial data is called, and the obtained user facial data retrieval frequency is compared with the activity determination threshold. If the user facial data retrieval frequency is outside the activity determination threshold, the user facial data corresponding to the user facial data retrieval frequency outside the activity determination threshold is deleted in the storage module 2. Otherwise, it jumps to the next comparison stage of the user facial data retrieval frequency and the activity determination threshold.

[0059] The user face data stored in the storage module 2 is a three-dimensional user face model;

[0060] The selection module 5 is internally provided with submodules, including:

[0061] The logic unit 51 is used to set the selection logic when the selection module 5 selects the image area and the face area during the operation stage;

[0062] Among them, the selection logic of the image area and the face area in the selection module 5 includes: selecting the position and selecting the quantity;

[0063] The selection module 5 is provided with submodules at the lower level, including:

[0064] A segmentation unit 52, used to segment the user face image and the user face data into a plurality of image blocks of equal number and size, and in one-to-one correspondence;

[0065] The image blocks obtained by segmenting the user's facial image and the user's facial data are the image area and the facial area;

[0066] The similarity mean in the determination module 6 is obtained by the following formula:

[0067]

[0068] Where: is the similarity between the image region a and its corresponding facial region b; n is the total number of image regions and facial regions selected by the selection module 5; p a ∩p b is the number of pixels with the same feature vector in the intersection of the set of feature vectors of each pixel in image region a and the set of feature vectors of each pixel in image region b; p a ∪p b is the total number of pixels with different feature vectors in the union of the set of feature vectors of each pixel in image region a and the set of feature vectors of each pixel in image region b;

[0069] in, The judgment result of the judgment module 6 is that the verification is passed, otherwise, the verification is not passed. During the operation phase of the search module 4 and the judgment module 6, when the image similarity is obtained, the similarity of the two groups of images can be expressed as

[0070] The control terminal 1 is electrically connected to the storage module 2 through a medium, the storage module 2 is electrically connected to the upload unit 21 and the management unit 22 through a medium, the storage module 2 is electrically connected to the camera module 3 and the search module 4 through a medium, the search module 4 is electrically connected to the management unit 22 through a medium, the search module 4 is electrically connected to the selection module 5 and the determination module 6 through a medium, and the interior and the lower level of the selection module 5 are electrically connected to the logic unit 51 and the segmentation unit 52 through a medium.

[0071] In this embodiment, the control terminal 1 controls the storage module 2 to operate and store the facial data of the user, the camera module 3 collects the facial image of the user in real time, the search module 4 synchronously receives the facial image of the user collected by the camera module 3 in real time, takes the collected facial image of the user as the search target, searches for the facial data of the user with the highest similarity in the storage module 2, the selection module 5 is post-operated to select the image area in the facial image of the user collected by the camera module 3, and the facial area corresponding to the image area in the facial data of the user searched in the search module 4, and finally the image area in the facial image of the user selected in the selection module 5 and the facial area in the facial data of the user are received by the determination module 6, the similarity between the image area and the corresponding facial area is analyzed, the similarity mean is obtained, and based on the similarity mean, it is determined whether the user corresponding to the facial image of the user currently collected by the camera module 3 has been verified;

[0072] Furthermore, the submodule set at the lower level of the storage module 2 provides the storage module with further user facial data management logic when storing user facial data, thereby providing necessary operation data support for the operation of subsequent modules in the system.

[0073] Example 2

[0074] In terms of specific implementation, based on Example 1, this example refers to Figure 1 The face recognition system based on information technology in Example 1 is further described in detail:

[0075] The camera module 3 is deployed on the surface of the gate and the access control. Every time the search module 4 finds the user facial data with the highest similarity to the user facial image collected in real time by the camera module 3 in the storage module 2, the search module 4 synchronously feeds back the found user facial data to the management unit 22, and the management unit 22 accumulates and records the number of times the found user facial data is called.

[0076] Through the above settings, the deployment environment of the camera module 3 is further limited.

[0077] like Figure 1 As shown, after the camera module 3 collects the user's facial image, it simultaneously optimizes the user's facial image. After the optimization is completed, the user's facial image is fed back to the search module 4. The user's facial image optimization processing logic is expressed as:

[0078]

[0079] Where: F(u,v) is the optimized output user face image; I(x,y) is the original user face image; j is the imaginary unit; M and N are the width and height of the image; u and v are frequency domain coordinates, indicating the horizontal and vertical frequencies of the image; x and y are spatial domain coordinates, indicating the horizontal and vertical pixel positions of the image; d is the differential.

[0080] By calculating through the above formula, the user's facial image is optimized, thereby providing more accurate operation data support for the operation of subsequent modules in the system, thereby making the final operation output result of the system more accurate.

[0081] like Figure 1 As shown, before segmenting the user facial data, based on the user facial image acquisition perspective, a user facial image of the same perspective is obtained on the user facial data, and the user facial image obtained on the user facial data and the user facial image collected in the camera module 3 are used as processing targets for the selection module 5 and the segmentation unit 22 to run and process.

[0082] Through the above settings, the user facial image used in the user facial data is further limited when the user facial data is segmented and selected synchronously with the user facial image.

[0083] Example 3

[0084] In terms of specific implementation, based on Example 1, this example refers to Figure 1 The face recognition system based on information technology in Example 1 is further described in detail:

[0085] In the logic for selecting the image area and the facial area in the logic unit 51, when the image area and the facial area are selected twice in a row for the same set of user facial images and user facial data, the selected image area and the facial area are different, and the number of selections is subject to the following logic:

[0086]

[0087] Where: m is the number of image areas and facial areas selected; g is the base number of selections; t is the number of times the user's facial data is called; k is the current verification pass rate of the user's facial data; f is the balance of the number of times the user's facial data is called based on the recording period; λ is the correction;

[0088] The value of the number of selected image areas and facial areas, m, is rounded to the nearest integer, and the value of g is 1 or 2.

[0089] Through the calculation of the above formula, a specified selection logic is further provided to ensure that a specified number of image areas and facial areas are selected, so that the selected specified number of image areas and facial areas are then applied to complete the user face verification operation.

[0090] In summary, during the operation of the system in the above embodiment, the user facial verification conditions are provided by storing the user facial data and collecting the real-time user facial images, and in the process of user facial verification, the user facial verification process is no longer dependent on the completed user facial images through the two-way segmentation, selection and further configuration of the user facial images and the user facial data, thereby making the user facial verification process faster; and during the operation of the system, in the segmentation and selection stages of the user facial images and the user facial data, the set segmentation and selection logic can be used to adaptively adapt to the user's facial verification habits, while ensuring the user facial verification process is fast, making the user facial verification results more accurate and reliable; at the same time, during the operation of the system, the collected user facial images can also be optimized, so as to perform user facial verification with the optimized user facial images, ensuring that the user facial images collected by the system will not fail the user facial verification due to unclear images as much as possible.

[0091] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A face recognition system based on information technology, characterized in that: include: The control terminal (1) is the main control terminal of the system and is used to issue execution commands; A storage module (2), used for storing user facial data; A camera module (3) is used to collect a user's facial image in real time; A search module (4) is used to receive a user facial image collected in real time by the camera module (3), use the collected user facial image as a search target, and search for the user facial data with the highest similarity in the storage module (2); A selection module (5) is used to select an image area in the user facial image collected by the camera module (3) and a facial area corresponding to the image area in the user facial data searched by the search module (4); A determination module (6) is used to receive the image area in the user facial image and the facial area in the user facial data selected in the selection module (5), analyze the similarity between the image area and the corresponding facial area, calculate the average similarity, and determine whether the user corresponding to the user facial image currently collected by the camera module (3) has passed the verification based on the average similarity; The selection module (5) is internally provided with submodules, including: A logic unit (51), used to set the selection logic when the selection module (5) selects the image area and the facial area during the operation phase; The selection logic of the image area and the facial area in the selection module (5) includes: selecting a position and selecting a quantity; The selection module (5) is provided with submodules at the lower level, including: A segmentation unit (52) is used to segment the user facial image and the user facial data so as to segment them into a plurality of groups of image blocks of equal number and size and in one-to-one correspondence; The image blocks obtained by segmenting the user's facial image and the user's facial data are the image area and the facial area; In the logic for selecting the image area and the facial area in the logic unit (51), when the image area and the facial area are selected twice in a row for the same set of user facial images and user facial data, the selected image area and the facial area are different, and the number of selections is subject to the following logic: ； Where: The number of choices for the image area and the face area; To select the quantity base; The number of times the user's facial data is called; The current verification pass rate of user facial data; The number of times the user's facial data is called is balanced based on the recording period; To amend; Among them, the number of selections of image area and face area The value of is rounded to the nearest integer. The value is 1 or 2. .

2. The face recognition system based on information technology according to claim 1 is characterized in that: The storage module (2) is provided with submodules at the lower level, including: An uploading unit (21), used for uploading the user's facial data; A management unit (22), used for recording the number of times the user facial data stored in the storage module (2) is called; In which, a system end user manually uploads the user's facial data through the uploading unit (21), the storage module (2) receives the user's facial data uploaded in the uploading unit (21), and stores the user's facial data. A recording cycle is set in the management unit (22), and the management unit (22) records the number of times the user's facial data stored in the storage module (2) is called based on the recording cycle. After each recording cycle ends, an activity determination threshold is set in the management unit (22), and the management unit (22) obtains the user's facial data retrieval frequency based on the recorded number of times the user's facial data is called, and compares the obtained user's facial data retrieval frequency with the activity determination threshold. If the user's facial data retrieval frequency is outside the activity determination threshold, the user's facial data corresponding to the user's facial data retrieval frequency outside the activity determination threshold is deleted in the storage module (2), otherwise, it jumps to the next user facial data retrieval frequency and activity determination threshold comparison stage.

3. The face recognition system based on information technology according to claim 1 is characterized in that: The user facial data stored in the storage module (2) is a three-dimensional user facial model.

4. The face recognition system based on information technology according to claim 1 is characterized in that: The camera module (3) is deployed on the surface of the gate or the access control system. Each time the search module (4) finds in the storage module (2) the user facial data with the highest similarity to the user facial image collected in real time by the camera module (3), the search module (4) synchronously feeds back the found user facial data to the management unit (22). The management unit (22) accumulates and records the number of times the found user facial data is called.

5. The face recognition system based on information technology according to claim 1 is characterized in that: After the camera module (3) acquires the user's facial image, it simultaneously optimizes the user's facial image. After the optimization is completed, the user's facial image is fed back to the search module (4). The user's facial image optimization logic is expressed as follows: ； Where: The user's facial image is output after optimization; is the original user's facial image; is an imaginary unit; , is the width and height of the image; , is the frequency domain coordinate, indicating the frequency of the image in the horizontal and vertical directions; , is the spatial coordinate, indicating the pixel position in the horizontal and vertical directions of the image; is a differential.

6. The face recognition system based on information technology according to claim 5 is characterized in that: Before segmenting the user facial data, based on the user facial image acquisition angle of view, a user facial image of the same angle of view is acquired on the user facial data, and the user facial image acquired on the user facial data and the user facial image acquired in the camera module (3) are used as processing targets for the selection module (5) and the segmentation unit (22) to run and process.

7. The face recognition system based on information technology according to claim 1 is characterized in that: The similarity mean in the determination module (6) is calculated by the following formula: ； Where: is the similarity between image region a and its corresponding facial region b; Selecting the total amount of image area and facial area for the selection module (5); is the number of pixels with the same feature vector in the intersection of the set of feature vectors of each pixel in image region a and the set of feature vectors of each pixel in image region b; is the total number of pixels with different feature vectors in the union of the set of feature vectors of each pixel in image region a and the set of feature vectors of each pixel in image region b; in, ≥95%, then the judgment module (6) judges that the verification is passed, otherwise, the verification is not passed. During the operation phase of the search module (4) and the judgment module (6), when the image similarity is calculated, the similarity of the two groups of images can be expressed as .

8. The face recognition system based on information technology according to claim 1 is characterized in that: The control terminal (1) is electrically connected to a storage module (2) via a medium; the storage module (2) is electrically connected to an upload unit (21) and a management unit (22) via a medium; the storage module (2) is electrically connected to a camera module (3) and a search module (4) via a medium; the search module (4) is electrically connected to the management unit (22) via a medium; the search module (4) is electrically connected to a selection module (5) and a determination module (6) via a medium; the selection module (5) is electrically connected to a logic unit (51) and a segmentation unit (52) inside and at a lower level via a medium.

Images