A face recognition system based on big data

Abstract

The present application discloses a face recognition system based on big data, which relates to the field of face recognition technology, and includes obtaining a face recognition request instruction of a user, pre-processing a face image based on the face recognition request instruction, and confirming the front face information; comparing with the pre-stored face information, confirming the user information corresponding to the front face information, generating relevant sentence information based on the user information, and confirming the preset lip change curve; obtaining the audio data and video data of the user's response to the relevant sentence information, and obtaining the corresponding voice information and the lip opening information corresponding to the video data based on the audio data; drawing the lip opening size-sentence change curve based on the lip opening information, comparing the lip opening size-sentence change curve with the preset lip change curve, and confirming the lip opening fit, and passing the face recognition request when the lip opening fit is greater than the preset fit threshold. The present application has the effect of improving the accuracy of face recognition.

Description

Technical Field

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

[0002] At present, the face recognition system is an emerging biometric technology with face recognition technology as its core, and is a high-tech technology that is being tackled in the international scientific and technological field today. Faces are popular because they are non-replicable, easy to collect, and do not require the cooperation of the person being photographed. Facial recognition systems have a wide range of applications: eliminating voting fraud, identity verification for cash withdrawals, computer security, etc. However, with the promotion and application of face recognition systems, more and more attack methods of forging portraits have also emerged.

[0003] In the related art, in some household face recognition systems, the door opening and closing are controlled by simple face recognition. Obviously, this current control method has the following shortcomings:

[0004] When facial recognition authentication is performed by forging a real-life photo of the user, there is a possibility that the authentication will be successful, which makes the facial recognition system vulnerable to attacks and has low security, resulting in reduced facial recognition accuracy;

[0005] When the face recognition system receives a face recognition request, it randomly generates a sentence and uses the sentence to determine whether the face recognition is a fake photo of the user's real life. However, the randomly generated sentence cannot be customized according to the user's information, which makes it impossible for children or the elderly to read some sentences normally, thus bringing adverse effects. There is room for improvement. Summary of the invention

[0006] In view of the deficiencies in the prior art, the present invention provides a face recognition system based on big data.

[0007] In a first aspect, the present application provides a face recognition method based on big data, comprising the following steps:.

[0008] Obtaining a face recognition request instruction from a user, preprocessing a face image based on the face recognition request instruction, and confirming frontal face information;

[0009] According to the frontal face information and the pre-stored face information, confirm the user information corresponding to the frontal face information, generate relevant sentence information according to the user information, and confirm the preset lip change curve based on the relevant sentence information;

[0010] Acquire audio data and video data of a user's response to the relevant sentence information, and acquire corresponding voice information based on the audio data and lip opening information corresponding to the video data;

[0011] A lip opening size-sentence change curve is drawn based on the lip opening information, and the lip opening size-sentence change curve is compared with the preset lip change curve to confirm the lip opening fit. When the lip opening fit is greater than the preset fit threshold, the face recognition request is passed.

[0012] Preferably, the preprocessing of the facial image based on the facial recognition request instruction to confirm the frontal facial information specifically includes:

[0013] Detecting a face image, and marking and locating feature points of the detected face, wherein the feature points include a left eye center point, a right eye center point, and a lip center point of the face, to obtain a first connecting line and a second connecting line, wherein the first connecting line is set as a connecting line formed by the left eye center point and the right eye center point, and the second connecting line is set as a connecting line formed by the lip center point perpendicular to the first connecting line, and respectively confirming a first spacing of the first connecting line and a second spacing of the second connecting line;

[0014] The preset relationship between the first spacing, the second spacing and the face deflection angle is obtained, the face deflection angle is confirmed, and the face feature points are reconstructed based on the face deflection angle to confirm the front face information.

[0015] Preferably, the confirming of user information corresponding to the frontal face information, generating relevant sentence information according to the user information, and confirming a preset lip change curve based on the relevant sentence information specifically includes:

[0016] confirming the user's age information based on the user information, and dividing the user into different age groups according to age, including a first age group, a second age group, and a third age group;

[0017] When it is determined based on the user information that the user is in the first age group, the poem information learned by the user is confirmed based on the user age information, and a first sentence is generated based on the poem information; when the user is in the second age group, sentence information is randomly generated and set as a second sentence; when the user is in the third age group, the information of the user's relatives, friends and children is obtained, and a third sentence is generated based on the information of the relatives, friends and children; wherein the relevant sentence information includes a first sentence, a second sentence and a third sentence, the second sentence is composed of text, English words or pinyin, and the first sentence, the second sentence and the third sentence are automatically updated after a preset time interval;

[0018] The corresponding lip opening size is confirmed based on the first statement, the second statement and the third statement, and then the preset lip change curve is confirmed based on the lip opening size.

[0019] Preferably, the step of obtaining audio data and video data of a user's response to the relevant sentence information, and obtaining corresponding voice information based on the audio data and lip opening information corresponding to the video data, further comprises:

[0020] Determining the sentence relevance between the voice information and the relevant sentence information;

[0021] When the sentence relevance is lower than a preset sentence relevance threshold, it is prompted that the face recognition request has not been passed;

[0022] When the sentence relevance is higher than a preset sentence relevance threshold, the lip opening fit is confirmed based on the lip opening information, and it is determined whether the lip opening fit is greater than a preset fit threshold.

[0023] Preferably, the step of drawing a lip opening size-sentence change curve based on the lip opening information specifically further includes:

[0024] Confirming the lip opening error of each word in the relevant sentence information based on the lip opening size-sentence change curve and the preset lip change curve, and drawing the lip opening error-word change curve;

[0025] Obtaining the slope value of the lip opening error-word change curve and comparing it with a preset curve slope standard value to determine whether the lip change value corresponding to each word in the face recognition system needs to be adjusted;

[0026] When the slope value of the lip opening error-word change curve is less than the preset curve slope standard value, it is determined that the lip change values ​​corresponding to each word in the current face recognition system do not need to be adjusted;

[0027] When the slope value of the lip opening error-word change curve is greater than the preset curve slope standard value, it is determined that the lip change values ​​corresponding to each word in the current face recognition system need to be adjusted.

[0028] Preferably, when the slope value of the lip opening error-word change curve is greater than the preset curve slope standard value, it is determined that the lip change value corresponding to each word in the current face recognition system needs to be adjusted, specifically including:

[0029] Based on the preset detection time and frontal face information, the lip feature points of each face image are obtained when reading each word in the relevant sentence information, and the lip change value compensation coefficient corresponding to each word in the current face recognition system is obtained;

[0030] The current face recognition system is adjusted based on the lip change value compensation coefficient corresponding to each word to ensure the accuracy of the lip change value of each word in the face recognition system.

[0031] Preferably, it also includes:

[0032] When the face recognition system obtains a face recognition request instruction but fails to recognize a face image corresponding to the face recognition request instruction, it is determined whether the door is passively opened. When the door is passively opened, the face image corresponding to the face recognition request instruction is marked as a related face image and stored in a first face image set. When a user in the first face image set outputs a face recognition request instruction again, related sentence information is generated based on the user information, and related audio data and video data are collected, and then a judgment is made as to whether the face recognition is passed based on the related audio data and video data.

[0033] In a second aspect, the present application provides a face recognition system based on big data, comprising:

[0034] A response request instruction module is used to obtain a user's face recognition request instruction, pre-process the face image based on the face recognition request instruction, and confirm the front face information;

[0035] A related sentence generating module, used for confirming the user information corresponding to the front face information according to the front face information and the pre-stored face information, generating related sentence information according to the user information, and confirming the preset lip change curve based on the related sentence information;

[0036] A data acquisition module, used to acquire audio data and video data of a user's response to the relevant sentence information, and acquire corresponding voice information and lip opening information corresponding to the video data based on the audio data;

[0037] The recognition and judgment module is used to draw a lip opening size-sentence change curve based on the lip opening information, and compare the lip opening size-sentence change curve with the preset lip change curve to confirm the lip opening fit. When the lip opening fit is greater than a preset fit threshold, the face recognition request is passed.

[0038] In a third aspect, the present application provides a computer-readable storage medium storing instructions, which, when executed on a computer, enables the computer to execute any one of the above-mentioned face recognition methods based on big data.

[0039] In summary, the present application includes at least one of the following beneficial technical effects:

[0040] The present invention provides a face recognition method based on big data, which confirms the front face information of the user through a face recognition request instruction, and then confirms the user information, and generates a first sentence, a second sentence and a third sentence based on the user information, and confirms the preset lip change curve corresponding to each sentence according to the lip opening size of each word in the first sentence, the second sentence and the third sentence, and then obtains the lip opening information based on the audio data and the video data, and confirms the lip opening size-sentence change curve, so as to compare the lip opening size-sentence change curve with the preset lip change curve, and confirm whether the face recognition request is passed, thereby effectively improving the accuracy of face recognition;

[0041] By dividing users into different age groups according to their age information and generating different relevant sentence information according to different age groups, some users can read the relevant sentence information normally, so that the face recognition system can judge the authenticity of the face through the sentence, thereby effectively improving the efficiency of the face recognition system;

[0042] By reading the up and down changes in the lips and the front and back changes in the lip feature points of each word in the corresponding sentence information of each user, the lip change value compensation coefficient corresponding to each word in the relevant sentence information of each user is confirmed, and the preset lip change curve in the face recognition system is adjusted, thereby effectively improving the accuracy of face recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a flow chart of the method for face recognition based on big data in an embodiment of the present application.

[0044] Figure 2 It is a system diagram of face recognition based on big data in an embodiment of the present application. DETAILED DESCRIPTION

[0045] The following is combined with Figure 1-2 This application is described in further detail.

[0046] Example 1

[0047] The embodiment of the present application discloses a face recognition method based on big data.

[0048] Reference Figure 1 , a face recognition method based on big data, comprising the following steps:

[0049] S1. Obtain the user's face recognition request instruction, pre-process the face image based on the face recognition request instruction, and confirm the front face information;

[0050] S2. According to the front face information and the pre-stored face information, confirm the user information corresponding to the front face information, generate relevant sentence information according to the user information, and confirm the preset lip change curve based on the relevant sentence information;

[0051] S3. Obtaining audio data and video data of the user's response to the relevant sentence information, and obtaining the corresponding voice information and lip opening information corresponding to the video data based on the audio data;

[0052] S4. Draw the lip opening size-sentence change curve based on the lip opening information, and compare the lip opening size-sentence change curve with the preset lip change curve to confirm the lip opening fit. When the lip opening fit is greater than the preset fit threshold, pass the face recognition request.

[0053] Among them, in step S1, the face image is pre-processed based on the face recognition request instruction to confirm the front face information, which specifically includes:

[0054] Detecting a face image, and marking and locating feature points of the detected face, wherein the feature points include the center point of the left eye, the center point of the right eye, and the center point of the lips of the face, to obtain a first connecting line and a second connecting line, wherein the first connecting line is set as a connecting line formed by the center point of the left eye and the center point of the right eye, and the second connecting line is set as a connecting line formed by the center point of the lips perpendicular to the first connecting line, and the first spacing of the first connecting line and the second spacing of the second connecting line are respectively obtained;

[0055] The preset relationship between the first spacing, the second spacing and the face deflection angle is obtained, the face deflection angle is confirmed, and the face feature points are reconstructed based on the face deflection angle to confirm the front face information.

[0056] Specifically, the preset relationship between the first spacing, the second spacing and the face deflection angle is set as:

[0057] Based on the formula , confirm the face deflection angle ,in, Represented as the first spacing, It is indicated as the second spacing;

[0058] After the face deflection angle is obtained, the feature points marked on the face are rotated based on the face deflection angle, and then the face feature points are reconstructed to confirm the frontal face information.

[0059] Among them, in step S2, the user information corresponding to the front face information is confirmed, and relevant sentence information is generated according to the user information, and the preset lip change curve is confirmed based on the relevant sentence information, which specifically includes:

[0060] The user's age information is confirmed based on the user information, and the users are divided into different age groups according to their ages, including a first age group, a second age group, and a third age group; wherein, the first age group can be divided into 6-12 years old, the second age group can be divided into 12-60 years old, and the third age group can be divided into 60-70 years old. In this embodiment of the application, children under 6 years old and people over 70 years old are deemed to be accompanied by relatives and friends.

[0061] When it is determined based on the user information that the user is in the first age group, the information of the poem learned by the user is confirmed based on the user age information, and a first sentence is generated based on the poem information; when the user is in the second age group, sentence information is randomly generated and set as the second sentence; when the user is in the third age group, the information of the user's relatives, friends and children is obtained, and a third sentence is generated based on the information of the relatives, friends and children; wherein the relevant sentence information includes the first sentence, the second sentence and the third sentence, the second sentence is composed of text, English words or pinyin, and the first sentence, the second sentence and the third sentence are automatically updated after a preset time interval;

[0062] The corresponding lip opening size is confirmed based on the first sentence, the second sentence and the third sentence, and then the preset lip change curve is confirmed based on the lip opening size.

[0063] Specifically, different sentence information is generated for users of different age groups. The generation settings of sentence information are more in line with the living habits of different age groups. This is to facilitate each user to read the relevant voice information output by the face recognition system more effectively and quickly, and then effectively conduct judgment and analysis.

[0064] Exemplarily, when the user is 7 years old, the user is judged to be in the first age group, and the user's grade information is obtained. When the user's grade information is second grade, a sentence is set as a poem in the second grade Chinese textbook. This setting can not only make it easier for the user to read the first sentence but also examine the user's learning situation. When the user is 65 years old, the user is judged to be in the third age group, and the user's pre-stored children's information is further obtained, so as to generate a third sentence based on the children's information, so that the user can read the third sentence more easily, thereby effectively improving the efficiency of face recognition.

[0065] Among them, after step S3, it also includes:

[0066] Determine the sentence relevance between the speech information and the related sentence information;

[0067] When the sentence relevance is lower than the preset sentence relevance threshold, it will prompt that the face recognition request has not been passed;

[0068] When the sentence relevance is higher than the preset sentence relevance threshold, the lip opening fit is confirmed based on the lip opening information, and it is determined whether the lip opening fit is greater than the preset fit threshold.

[0069] In step S4, a lip opening size-sentence change curve is drawn based on the lip opening information, which specifically includes:

[0070] Based on the lip opening size-sentence change curve and the preset lip change curve, the lip opening error of each word in the relevant sentence information is confirmed, and the lip opening error-word change curve is drawn;

[0071] Obtain the slope value of the lip opening error-word change curve and compare it with the preset curve slope standard value to determine whether the lip change value corresponding to each word in the face recognition system needs to be adjusted;

[0072] When the slope value of the lip opening error-word change curve is less than the preset curve slope standard value, it is determined that the lip change values ​​corresponding to each word in the current face recognition system do not need to be adjusted;

[0073] When the slope value of the lip opening error-word change curve is greater than the preset curve slope standard value, it is determined that the lip change values ​​corresponding to each word in the current face recognition system need to be adjusted.

[0074] Further, when the slope value of the lip opening error-word change curve is greater than the preset curve slope standard value, it is determined that the lip change value corresponding to each word in the current face recognition system needs to be adjusted, specifically including:

[0075] Based on the preset detection time and frontal face information, the lip feature points of each face image are obtained when reading out the lip up and down changes and the lip front and back changes of each word in the related sentence information, and then the lip change value compensation coefficient corresponding to each word in the current face recognition system is obtained;

[0076] The current face recognition system is adjusted based on the lip change value compensation coefficient corresponding to each word to ensure the accuracy of the lip change value of each word in the face recognition system.

[0077] Specifically, the up and down changes in the lips and the front and back changes in the lips of each word when each user reads the corresponding relevant sentences are confirmed, and then the lip change value compensation coefficient of each user for each word in different sentences is confirmed, and the relevant sentences in the database are updated based on the lip change value compensation coefficient, thereby effectively improving the accuracy of face recognition, wherein the relevant sentence information is automatically updated after a preset time interval.

[0078] Furthermore, it also includes:

[0079] When the face recognition system obtains a face recognition request instruction but fails to recognize a face image corresponding to the face recognition request instruction, it determines whether the door is passively opened. When the door is passively opened, the face image corresponding to the face recognition request instruction is marked as a related face image and stored in the first face image set. When the user in the first face image set outputs a face recognition request instruction again, related sentence information is generated based on the user information, and related audio data and video data are collected, and then a judgment is made as to whether the face recognition is passed based on the related audio data and video data.

[0080] Example 2

[0081] The embodiment of the present application also discloses a face recognition system based on big data.

[0082] Reference Figure 2 , a face recognition system based on big data, comprising:

[0083] The response request instruction module 1 is used to obtain the user's face recognition request instruction, pre-process the face image based on the face recognition request instruction, and confirm the front face information;

[0084] A related sentence generating module 2 is used to confirm the user information corresponding to the front face information according to the front face information and the pre-stored face information, generate related sentence information according to the user information, and confirm the preset lip change curve based on the related sentence information;

[0085] Data acquisition module 3, used to acquire audio data and video data of the user's response to the relevant sentence information, and acquire the corresponding voice information and lip opening information corresponding to the video data based on the audio data;

[0086] The recognition and judgment module 4 is used to draw a lip opening size-sentence change curve based on the lip opening information, and compare the lip opening size-sentence change curve with the preset lip change curve to confirm the lip opening fit. When the lip opening fit is greater than the preset fit threshold, the face recognition request is passed.

[0087] Among them, the response request instruction module 1 also includes a feature point marking unit and a deflection angle calculation unit, the feature point marking unit is used to mark the left eye center point, the right eye center point and the lip center point of the face to obtain a first connecting line and a second connecting line, wherein the first connecting line is set to a connecting line formed by the left eye center point and the right eye center point, and the second connecting line is set to a connecting line formed by the lip center point perpendicular to the first connecting line, and the first spacing of the first connecting line and the second spacing of the second connecting line are respectively confirmed; the deflection angle calculation unit is used to confirm the face deflection angle based on a preset relationship between the first spacing, the second spacing and the face deflection angle.

[0088] The relevant sentence generation module 2 includes an age division unit, a sentence generation unit and a preset lip change curve confirmation unit; the age division unit is used to divide users into different age groups based on user age information, including a first age group, a second age group and a third age group; the sentence generation unit is used to generate different sentences based on different age groups. When the user is in the first age group, the poem information learned by the user is confirmed based on the user age information, and a first sentence is generated based on the poem information; when the user is in the second age group, sentence information is randomly generated and set as a second sentence; when the user is in the third age group, the user's relatives and children information is obtained, and a third sentence is generated based on the relatives and children information; the preset lip change curve confirmation unit is used to confirm the lip opening size according to the corresponding relevant sentence information, and confirm the preset lip change curve based on the lip opening size.

[0089] The recognition and judgment module 4 includes a threshold judgment unit and an error adjustment unit. The threshold judgment unit is used to compare and judge the sentence relevance of the voice information and the relevant sentence information with the preset sentence relevance threshold, and to compare and judge the lip opening fit with the preset fit threshold; the error adjustment unit is used to obtain the lip feature points of each face image when reading each word in the relevant sentence information, and then obtain the lip change value compensation coefficient corresponding to each word in the current face recognition system; and adjust the current face recognition system based on the lip change value compensation coefficient corresponding to each word to ensure the accuracy of the lip change value of each word in the face recognition system.

[0090] Example 3

[0091] According to an exemplary embodiment, a computer-readable storage medium is shown, on which a rewritable computer program is stored;

[0092] When the computer program runs on a computer device, the computer device executes the above-mentioned face recognition method based on big data.

[0093] The above contents are merely examples and explanations of the concept of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the claims, they should all fall within the protection scope of the present invention.

[0094] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0095] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A face recognition method based on big data, characterized in that: The following steps are involved: Obtaining a face recognition request instruction from a user, preprocessing a face image based on the face recognition request instruction, and confirming frontal face information; The preprocessing of the face image based on the face recognition request instruction to confirm the front face information specifically includes: Detecting a face image, and marking and locating feature points of the detected face, wherein the feature points include a left eye center point, a right eye center point, and a lip center point of the face, to obtain a first connecting line and a second connecting line, wherein the first connecting line is set as a connecting line formed by the left eye center point and the right eye center point, and the second connecting line is set as a connecting line formed by the lip center point perpendicular to the first connecting line, and respectively confirming a first spacing of the first connecting line and a second spacing of the second connecting line; Acquire a preset relationship between the first spacing, the second spacing, and a face deflection angle, confirm the face deflection angle, and reconstruct facial feature points based on the face deflection angle, thereby confirming frontal face information; According to the frontal face information and the pre-stored face information, confirm the user information corresponding to the frontal face information, generate relevant sentence information according to the user information, and confirm the preset lip change curve based on the relevant sentence information; The confirming the user information corresponding to the front face information, generating relevant sentence information according to the user information, and confirming the preset lip change curve based on the relevant sentence information specifically includes: confirming the user's age information based on the user information, and dividing the user into different age groups according to age, including a first age group, a second age group, and a third age group; When it is determined based on the user information that the user is in the first age group, the poem information learned by the user is confirmed based on the user age information, and a first sentence is generated based on the poem information; when the user is in the second age group, sentence information is randomly generated and set as a second sentence; when the user is in the third age group, the information of the user's relatives, friends and children is obtained, and a third sentence is generated based on the information of the relatives, friends and children; wherein the relevant sentence information includes a first sentence, a second sentence and a third sentence, the second sentence is composed of text, English words or pinyin, and the first sentence, the second sentence and the third sentence are automatically updated after a preset time interval; Determine the corresponding lip opening size based on the first sentence, the second sentence and the third sentence, and then determine the preset lip change curve based on the lip opening size; Acquire audio data and video data of a user's response to the relevant sentence information, and acquire corresponding voice information based on the audio data and lip opening information corresponding to the video data; Draw a lip opening size-sentence change curve based on the lip opening information, and compare the lip opening size-sentence change curve with the preset lip change curve to confirm the lip opening fit, and when the lip opening fit is greater than the preset fit threshold, pass the face recognition request; The step of drawing a lip opening size-sentence change curve based on the lip opening information specifically includes: Confirming the lip opening error of each word in the relevant sentence information based on the lip opening size-sentence change curve and the preset lip change curve, and drawing the lip opening error-word change curve; Obtaining the slope value of the lip opening error-word change curve and comparing it with a preset curve slope standard value to determine whether the lip change value corresponding to each word in the face recognition system needs to be adjusted; When the slope value of the lip opening error-word change curve is less than the preset curve slope standard value, it is determined that the lip change values ​​corresponding to each word in the current face recognition system do not need to be adjusted; When the slope value of the lip opening error-word change curve is greater than the preset curve slope standard value, it is determined that the lip change values ​​corresponding to each word in the current face recognition system need to be adjusted; When the slope value of the lip opening error-word change curve is greater than the preset curve slope standard value, it is determined that the lip change values ​​corresponding to each word in the current face recognition system need to be adjusted, specifically including: Based on the preset detection time and frontal face information, the lip feature points of each face image are obtained when reading each word in the relevant sentence information, and the lip change value compensation coefficient corresponding to each word in the current face recognition system is obtained; The current face recognition system is adjusted based on the lip change value compensation coefficient corresponding to each word to ensure the accuracy of the lip change value of each word in the face recognition system.

2. The face recognition method based on big data according to claim 1, characterized in that: The step of obtaining audio data and video data of a user's response to the relevant sentence information, and obtaining corresponding voice information and lip opening information corresponding to the video data based on the audio data, further includes: Determining the sentence relevance between the voice information and the relevant sentence information; When the sentence relevance is lower than a preset sentence relevance threshold, it is prompted that the face recognition request has not been passed; When the sentence relevance is higher than a preset sentence relevance threshold, the lip opening fit is confirmed based on the lip opening information, and it is determined whether the lip opening fit is greater than a preset fit threshold.

3. The face recognition method based on big data according to claim 2, characterized in that: Also includes: When the face recognition system obtains a face recognition request instruction but fails to recognize a face image corresponding to the face recognition request instruction, it is determined whether the door is passively opened. When the door is passively opened, the face image corresponding to the face recognition request instruction is marked as a related face image and stored in a first face image set. When a user in the first face image set outputs a face recognition request instruction again, relevant sentence information is generated based on the user information, and relevant audio data and video data are collected, and then a judgment is made as to whether the face recognition is passed based on the relevant audio data and video data.

4. A face recognition system based on big data, applied to a face recognition method based on big data as described in any one of claims 1 to 3, characterized in that: include: A response request instruction module (1) is used to obtain a user's face recognition request instruction, pre-process the face image based on the face recognition request instruction, and confirm the front face information; A related sentence generating module (2) is used to confirm the user information corresponding to the front face information according to the front face information and the pre-stored face information, generate related sentence information according to the user information, and confirm the preset lip change curve based on the related sentence information; A data acquisition module (3) is used to acquire audio data and video data of a user's response to the relevant sentence information, and acquire corresponding voice information and lip opening information corresponding to the video data based on the audio data; The recognition and judgment module (4) is used to draw a lip opening size-sentence change curve based on the lip opening information, and compare the lip opening size-sentence change curve with the preset lip change curve to confirm the lip opening fit. When the lip opening fit is greater than a preset fit threshold, the face recognition request is passed.

5. A computer-readable storage medium, characterized in that: Instructions are stored, and when the instructions are executed on a computer, the computer executes a face recognition method based on big data as described in any one of claims 1 to 3.

Images