Reminding method and device based on intelligent face analysis

Abstract

The invention discloses a reminding method and device based on intelligent face analysis, and relates to the technical field of face analysis, and the method comprises the steps: obtaining the image data of a target object through a camera, carrying out the face image detection, determining whether a face exists in the image data or not, and positioning the key points of the face through a face key point positioning algorithm, and based on the key points of the positioned face, using a feature description algorithm to extract the features of the face image, and representing the face features as mathematical vectors, the mathematical vectors including the feature information of the face image. Through face image detection, key point positioning and feature extraction algorithms, the identity of the target object can be accurately judged, and compared with traditional identity authentication modes such as passwords and fingerprints, face recognition has higher accuracy and safety.

Description

Technical Field

[0001] The present invention relates to the technical field of face analysis, and in particular to a method and device based on intelligent face analysis reminder. Background Art

[0002] The method and device based on intelligent face analysis reminder can be used in multiple fields. Intelligent face analysis can be used in access control systems and security monitoring. By real-time recognition and verification of face information, it ensures that only authorized personnel can enter specific areas, and timely reminds and alarms illegal intrusions. Identity authentication based on face features can be applied to various scenarios, such as mobile phone unlocking, electronic payment, online banking, etc. It provides a convenient, fast and secure identity authentication method, reducing the use of traditional passwords and PIN codes.

[0003] Traditional methods usually rely on manual analysis and processing of data, requiring personnel to manually input, calculate and judge. This method is inefficient and prone to errors, especially in the case of large-scale data analysis. In traditional methods, monitoring and analysis are often performed offline, resulting in possible delays in analysis results and easy omission of key information, which may lead to the failure to timely discover important anomalies or events. In addition, due to technical and resource limitations, traditional methods are often unable to handle large-scale and complex data sets, making it difficult to discover hidden patterns, associations and anomalies. Summary of the invention

[0004] The purpose of the present invention is to provide a method and device based on intelligent face analysis reminder to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solution: a method based on intelligent face analysis reminder, comprising:

[0006] Acquire the image data of the target object through the camera, perform face image detection, determine whether there is a face in the image data, and use the face key point positioning algorithm to locate the key points of the face;

[0007] Based on locating key points of the face, extracting features of the face image using a feature description algorithm, and representing the face features as a mathematical vector, wherein the mathematical vector includes feature information of the face image;

[0008] According to the stored target object feature library, the extracted facial features are matched with the facial features in the target object feature library, and the identity of the target object is determined by comparing the similarity between the facial features;

[0009] If the face match of the target object is successful, then reading the associated information corresponding to the identity of the target object in the associated information database according to the stored associated information database, wherein the associated information includes basic information of the target object;

[0010] If the target object face matching fails, a first reminder message is displayed on the user terminal to inform the user that the identity authentication has failed. If the target object face matching fails more than a preset number of times, an early warning mechanism is triggered;

[0011] After reading the associated information corresponding to the target object in the associated information library, the target object inputs the basic information of the target object in the user terminal, compares the basic information with the associated information, and determines whether the basic information and the associated information completely match;

[0012] If the basic information and the associated information completely match each other, a second reminder message indicating successful authentication is generated, and the generated second reminder message indicating successful authentication is transmitted to the user terminal to remind the target object that the identity authentication is successful;

[0013] If the basic information fails to completely match the associated information, a third reminder message of verification failure is generated, and the generated third reminder message of verification failure is transmitted to the user terminal to remind the target object to re-match the associated information.

[0014] Preferably, the step of acquiring image data of the target object through a camera, performing face image detection, determining whether a face exists in the image data, and locating key points of the face using a face key point location algorithm comprises:

[0015] Use a camera to obtain a static image of the target object;

[0016] The image is processed by the face detection algorithm to detect whether there is a face in it. The face detection algorithm is as follows:

[0017]

[0018] Among them, G inner represents the intra-class distance and m i represents the mean of each cluster after clustering, m represents the mean of the entire sample, represents the jth object in the i-th sample, and k represents the number of clusters;

[0019] If a face is detected, the face key point positioning algorithm is used to locate the face key points and determine the position of the face key points. The face key point positioning algorithm is as follows:

[0020]

[0021] in, represents the heat map of two-dimensional key points, k represents the key point, p represents the probability that the key point exists at the pixel point, X n,k represents the coordinates of the key points of the face, σ 2 Represents the variance of the pixel;

[0022] According to the position information of the key points, the key points are visualized on the face image.

[0023] Preferably, based on locating the key points of the face, a face recognition algorithm is used to extract the features of the face image, and the face features are represented by a mathematical vector, wherein the mathematical vector includes feature information of the face image, including:

[0024] Based on the location information of the key points, the feature vector is extracted using the feature description algorithm. The feature description algorithm is as follows:

[0025]

[0026] Among them, L x and L y represents the first-order nonlinear filter response, L xx , L xy and L yy represents the second-order nonlinear filter response, L ww represents the second-order canonical derivative;

[0027] The extracted feature vector is encoded and converted into a mathematical vector representation.

[0028] Preferably, the step of matching the extracted facial features with the facial features in the target object feature library according to the stored target object feature library, and determining the identity of the target object by comparing the similarity between the facial features, includes:

[0029] Compare the feature vector of the face to be compared with the features in the target object feature library, and set a threshold when comparing the similarity to determine whether the match is successful;

[0030] Based on the similarity obtained by comparison, a judgment is made through a set threshold. If the similarity exceeds the threshold, the face to be compared successfully matches the face in the target object feature library, and the identity of the target object is determined.

[0031] Preferably, if the target object fails to match, a first reminder message is displayed on the user terminal to inform the user that the identity authentication fails. If the target object fails to match more than a preset number of times, an early warning mechanism is triggered, including:

[0032] After the target object face matching fails, a first reminder message is displayed to the user terminal to inform the user that the identity authentication fails;

[0033] Every time a match fails, the counter is incremented by 1;

[0034] If the counter reaches the preset value, the early warning mechanism is triggered and the administrator is notified for processing.

[0035] Preferably, after reading the associated information corresponding to the target object in the associated information library, the target object inputs basic information of the target object in a user terminal, compares the basic information with the associated information, and determines whether the basic information completely matches the associated information, including:

[0036] The target object inputs the basic information of the target object in the user terminal;

[0037] According to the basic information input by the target object, the associated information corresponding to the identity of the target object is searched in the associated information database. If the corresponding record is found, it is read out. Otherwise, it indicates that the associated information of the target object cannot be found, and a fourth reminder message is issued and transmitted to the user terminal;

[0038] The input basic information is compared with the associated information to determine whether the basic information and the associated information completely match. If all information completely matches, it is determined that the input basic information matches the associated information; otherwise, it is determined that the input basic information does not match the associated information.

[0039] The present invention also provides the following technical solution: a device based on intelligent face analysis and reminder, comprising an associated information library module, a matching and verification module, a reminder message module and an early warning mechanism module, and also comprising;

[0040] A camera module, wherein the camera module is used to obtain image data of a target object and collect a static image of the target object through a camera;

[0041] A face detection module is used to process the image collected by the camera to detect whether there is a face therein, and if there is a face, locate the key points of the face and extract its features;

[0042] A facial key point positioning module, the facial key point positioning module is used to determine the key point positions of the face according to the detected face using a facial key point positioning algorithm;

[0043] A feature extraction module, wherein the feature extraction module is used to extract features of a face image based on position information of key points of the face using a feature description algorithm, and to represent the features of the face image as a mathematical vector;

[0044] A target object feature library module is used to store facial features of registered target objects.

[0045] Preferably, the association information library module is used to store target object association information for comparison with the input basic information;

[0046] The matching and verification module is used to match the extracted facial features with the facial features in the target object feature library, calculate the similarity and determine whether the identity is successfully matched, and compare the input basic information with the associated information to determine whether it is a complete match;

[0047] The reminder message module is used to generate corresponding reminder messages according to the matching and verification results, and transmit them to the user terminal. The reminder messages include a first reminder message, a second reminder message, a third reminder message and a fourth reminder message. The first reminder message is used to inform the target object that the face verification failed, the second reminder message is used to remind the target object that the identity authentication was successful, the third reminder message is used to inform the target object that the associated information verification failed, and the fourth reminder message is used to remind the target object that the associated information of the target object cannot be found;

[0048] The early warning mechanism module is used to trigger the early warning mechanism, and when the number of target object matching failures exceeds a preset value, the administrator is notified to handle the problem.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] 1. The present invention can accurately determine the identity of the target object through face image detection, key point positioning and feature extraction algorithms. Compared with traditional identity authentication methods such as passwords and fingerprints, face recognition has higher accuracy and security. Once the face match of the target object fails or exceeds the preset number of times, it can remind the user of identity authentication failure in real time and trigger an early warning mechanism to notify the administrator, which helps to protect the security of the user terminal and the system and prevent unauthorized access. In addition, the face recognition is performed by a camera without the need for additional hardware equipment, which is convenient and flexible to use. By displaying a reminder message on the user terminal, the user is informed of the result of identity authentication in a timely manner, thereby improving user experience and satisfaction.

[0051] 2. Based on the stored associated information library, the present invention can compare the basic information input by the target object with the associated information to determine whether they are completely matched, which helps to ensure the accuracy and credibility of the identity of the target object. In addition, the method based on intelligent face analysis can automatically perform operations such as face detection, feature extraction and identity matching, reducing the need for manual operation and improving reaction efficiency and response speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 A flow chart of the overall method provided by an embodiment of the present invention;

[0053] Figure 2A flowchart of step S1 provided in an embodiment of the present invention;

[0054] Figure 3 A flowchart of step S2 provided in an embodiment of the present invention;

[0055] Figure 4 A flowchart of step S3 provided in an embodiment of the present invention;

[0056] Figure 5 A flowchart of step S5 provided in an embodiment of the present invention;

[0057] Figure 6 A flowchart of step S6 provided in an embodiment of the present invention;

[0058] Figure 7 A flowchart of an overall device provided by an embodiment of the present invention.

[0059] In the figure: 1. Camera module; 2. Face detection module; 3. Face key point positioning module; 4. Feature extraction module; 5. Target object feature library module; 6. Related information library module; 7. Matching and verification module; 8. Reminder message module; 9. Early warning mechanism module. DETAILED DESCRIPTION

[0060] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0061] See also Figure 1-6 The present invention provides a technical solution: a method based on intelligent face analysis and reminder, comprising the following steps:

[0062] S1. Acquire image data of a target object through a camera, perform face image detection, determine whether there is a face in the image data, and locate the key points of the face using a face key point location algorithm;

[0063] S2. Based on the key points of the human face, the features of the human face image are extracted using a feature description algorithm, and the facial features are represented as a mathematical vector, which includes feature information of the human face image;

[0064] S3, according to the stored target object feature library, the extracted facial features are matched with the facial features in the target object feature library, and the identity of the target object is determined by comparing the similarity between the facial features;

[0065] S4. If the face match of the target object is successful, then read the associated information corresponding to the identity of the target object in the associated information database according to the stored associated information database, where the associated information includes the basic information of the target object;

[0066] S5. If the target object face matching fails, a first reminder message is displayed on the user terminal to inform the user that the identity authentication has failed. If the target object face matching fails more than a preset number of times, an early warning mechanism is triggered.

[0067] S6. After reading the associated information corresponding to the target object in the associated information database, the target object inputs the basic information of the target object in the user terminal, compares the basic information with the associated information, and determines whether the basic information and the associated information completely match;

[0068] S7. If the basic information and the associated information completely match each other, a second reminder message indicating successful authentication is generated, and the generated second reminder message indicating successful authentication is transmitted to the user terminal to remind the target object that the identity authentication is successful;

[0069] S8. If the basic information fails to completely match the associated information, a third reminder message of verification failure is generated, and the generated third reminder message of verification failure is transmitted to the user terminal to remind the target object to re-match the associated information.

[0070] The image data of the target object is obtained through a camera, and face image detection is performed to determine whether there is a face in the image data, and the key points of the face are located using a face key point location algorithm, including the following steps:

[0071] S101, using a camera to obtain a static image of a target object;

[0072] S102: Process the image using a face detection algorithm to detect whether there is a face in the image. The face detection algorithm is specifically:

[0073]

[0074] Among them, G inner represents the intra-class distance and m i represents the mean of each cluster after clustering, m represents the mean of the entire sample, represents the jth object in the i-th sample, and k represents the number of clusters;

[0075] S103: If a face is detected, locate the key points of the face using a face key point location algorithm to determine the positions of the key points of the face. The face key point location algorithm is specifically as follows:

[0076]

[0077] in, represents the heat map of two-dimensional key points, k represents the key point, p represents the probability that the key point exists at the pixel point, X n,k represents the coordinates of the key points of the face, σ 2 Represents the variance of the pixel;

[0078] S104, visualizing the key points on the face image according to the position information of the key points;

[0079] Based on the key points of the human face, the features of the human face image are extracted using a face recognition algorithm, and the facial features are represented as mathematical vectors, which include the feature information of the human face image, including the following steps:

[0080] S201, based on the position information of the key points, a feature description algorithm is used to extract a feature vector, and the feature description algorithm is specifically:

[0081]

[0082] Among them, L x and L y represents the first-order nonlinear filter response, L xx , L xy and L yy represents the second-order nonlinear filter response, L ww represents the second-order canonical derivative;

[0083] S202, encoding the extracted feature vector and converting it into a mathematical vector representation;

[0084] According to the stored target object feature library, the extracted facial features are matched with the facial features in the target object feature library, and the identity of the target object is determined by comparing the similarity between the facial features, including the following steps:

[0085] S301, comparing the feature vector of the face to be compared with the features in the target object feature library, and setting a threshold when comparing the similarity to determine whether the match is successful;

[0086] S302, judging by a set threshold value based on the similarity obtained by comparison, if the similarity exceeds the threshold value, the face to be compared successfully matches the face in the target object feature library, and the identity of the target object is determined;

[0087] If the target object fails to match, a first reminder message is displayed on the user terminal to inform the user that the identity authentication has failed. If the target object fails to match more than a preset number of times, an early warning mechanism is triggered, including the following steps:

[0088] S501, after the target object face matching fails, displaying a first reminder message to the user terminal to inform the user that the identity authentication fails;

[0089] S502, every time a match fails, the counter is incremented by 1;

[0090] S503: If the counter reaches a preset value, the early warning mechanism is triggered to notify the administrator to handle the problem;

[0091] After reading the associated information corresponding to the target object in the associated information database, the target object inputs the basic information of the target object in the user terminal, compares the basic information with the associated information, and determines whether the basic information and the associated information completely match, including the following steps:

[0092] S601, the target object inputs basic information of the target object in the user terminal;

[0093] S602: searching for association information corresponding to the identity of the target object in the association information database according to the basic information input by the target object; if a corresponding record is found, reading it out; otherwise, indicating that the association information of the target object cannot be found, a fourth reminder message is issued and transmitted to the user terminal;

[0094] S603: Compare the input basic information with the associated information to determine whether the basic information and the associated information completely match. If all the information completely matches, it is determined that the input basic information matches the associated information. Otherwise, it is determined that the input basic information does not match the associated information.

[0095] See also Figure 7 ,The present invention also provides a technical solution: a device based on intelligent face analysis and reminder, including an associated information library module 6, a matching and verification module 7, a reminder message module 8 and an early warning mechanism module 9, and also includes;

[0096] The camera module 1 is used to obtain image data of the target object and collect a static image of the target object through the camera;

[0097] Face detection module 2, which is used to process the image collected by the camera to detect whether there is a face. If there is a face, it will locate the key points of the face and extract its features;

[0098] The face key point positioning module 3 is used to determine the key point positions of the face according to the detected face using the face key point positioning algorithm;

[0099] Feature extraction module 4, the feature extraction module 4 is used to extract the features of the face image based on the position information of the key points of the face using a feature description algorithm, and represent the features of the face image as a mathematical vector;

[0100] The target object feature library module 5 is used to store the facial features of the registered target objects.

[0101] The associated information library module 6 is used to store the target object associated information for comparison with the input basic information;

[0102] The matching and verification module 7 is used to match the extracted facial features with the facial features in the target object feature library, calculate the similarity and determine whether the identity is successfully matched, and compare the input basic information with the associated information to determine whether it is a complete match;

[0103] The reminder message module 8 is used to generate corresponding reminder messages according to the matching and verification results, and transmit them to the user terminal. The reminder messages include a first reminder message, a second reminder message, a third reminder message and a fourth reminder message. The first reminder message is used to inform the target object that the face verification failed, the second reminder message is used to remind the target object that the identity authentication was successful, the third reminder message is used to inform the target object that the associated information verification failed, and the fourth reminder message is used to remind the target object that the associated information of the target object cannot be found;

[0104] The early warning mechanism module 9 is used to trigger the early warning mechanism. When the number of target object matching failures exceeds a preset value, the administrator is notified to handle the problem.

[0105] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0106] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method based on intelligent face analysis and reminder, characterized in that: include: Acquire the image data of the target object through the camera, perform face image detection, determine whether there is a face in the image data, and use the face key point positioning algorithm to locate the key points of the face; Based on locating key points of the face, extracting features of the face image using a feature description algorithm, and representing the face features as a mathematical vector, wherein the mathematical vector includes feature information of the face image; According to the stored target object feature library, the extracted facial features are matched with the facial features in the target object feature library, and the identity of the target object is determined by comparing the similarity between the facial features; If the face match of the target object is successful, then reading the associated information corresponding to the identity of the target object in the associated information database according to the stored associated information database, wherein the associated information includes basic information of the target object; If the target object face matching fails, a first reminder message is displayed on the user terminal to inform the user that the identity authentication has failed. If the target object face matching fails more than a preset number of times, an early warning mechanism is triggered; After reading the associated information corresponding to the target object in the associated information library, the target object inputs the basic information of the target object in the user terminal, compares the basic information with the associated information, and determines whether the basic information and the associated information completely match; If the basic information and the associated information completely match each other, a second reminder message indicating successful authentication is generated, and the generated second reminder message indicating successful authentication is transmitted to the user terminal to remind the target object that the identity authentication is successful; If the basic information fails to completely match the associated information, a third reminder message of verification failure is generated, and the generated third reminder message of verification failure is transmitted to the user terminal to remind the target object to re-match the associated information.

2. The method according to claim 1, characterized in that: The method of acquiring image data of the target object through a camera and performing face image detection to determine whether a face exists in the image data and locating key points of the face using a face key point positioning algorithm includes: Use a camera to obtain a static image of the target object; Process the image using a face detection algorithm to detect whether there is a face in it; If a face is detected, the face key points are located by using the face key point location algorithm to determine the position of the key points of the face; According to the position information of the key points, the key points are visualized on the face image.

3. The method of intelligent face analysis and reminder according to claim 1, characterized in that: Based on the key points of the human face, the features of the human face image are extracted using a face recognition algorithm, and the facial features are represented as mathematical vectors, wherein the mathematical vectors include feature information of the human face image, including: Based on the location information of key points, feature vectors are extracted using feature description algorithms; The extracted feature vector is encoded and converted into a mathematical vector representation.

4. The method of claim 3 based on intelligent face analysis and reminder, characterized in that: The step of matching the extracted facial features with the facial features in the target object feature library according to the stored target object feature library, and determining the identity of the target object by comparing the similarity between the facial features, includes: Compare the feature vector of the face to be compared with the features in the target object feature library, and set a threshold when comparing the similarity to determine whether the match is successful; Based on the similarity obtained by comparison, a judgment is made through a set threshold. If the similarity exceeds the threshold, the face to be compared successfully matches the face in the target object feature library, and the identity of the target object is determined.

5. The method of intelligent face analysis and reminder according to claim 1, characterized in that: If the target object fails to match, a first reminder message is displayed on the user terminal to inform the user that the identity authentication fails. If the target object fails to match more than a preset number of times, an early warning mechanism is triggered, including: After the target object face matching fails, a first reminder message is displayed to the user terminal to inform the user that the identity authentication fails; Every time a match fails, the counter is incremented by 1; If the counter reaches the preset value, the early warning mechanism is triggered and the administrator is notified for processing.

6. The method of intelligent face analysis and reminder according to claim 1, characterized in that: After reading the associated information corresponding to the target object in the associated information library, the target object inputs the basic information of the target object in the user terminal, compares the basic information with the associated information, and determines whether the basic information and the associated information completely match, including: The target object inputs the basic information of the target object in the user terminal; According to the basic information input by the target object, the associated information corresponding to the identity of the target object is searched in the associated information database. If the corresponding record is found, it is read out. Otherwise, it indicates that the associated information of the target object cannot be found, and a fourth reminder message is issued and transmitted to the user terminal; The input basic information is compared with the associated information to determine whether the basic information and the associated information completely match. If all information completely matches, it is determined that the input basic information matches the associated information; otherwise, it is determined that the input basic information does not match the associated information.

7. A device based on intelligent face analysis and reminder according to any one of claims 1 to 6, comprising an associated information library module, a matching and verification module, a reminder message module and an early warning mechanism module, characterized in that: A camera module, wherein the camera module is used to obtain image data of a target object and collect a static image of the target object through a camera; A face detection module is used to process the image collected by the camera to detect whether there is a face therein, and if there is a face, locate the key points of the face and extract its features; A facial key point positioning module, the facial key point positioning module is used to determine the key point positions of the face according to the detected face using a facial key point positioning algorithm; A feature extraction module, wherein the feature extraction module is used to extract features of a face image based on position information of key points of the face using a feature description algorithm, and to represent the features of the face image as a mathematical vector; A target object feature library module is used to store facial features of registered target objects.

8. The device based on intelligent face analysis and reminder according to claim 7, characterized in that: The associated information library module is used to store target object associated information for comparison with input basic information; The matching and verification module is used to match the extracted facial features with the facial features in the target object feature library, calculate the similarity and determine whether the identity is successfully matched, and compare the input basic information with the associated information to determine whether it is a complete match; The reminder message module is used to generate corresponding reminder messages according to the matching and verification results, and transmit them to the user terminal. The reminder messages include a first reminder message, a second reminder message, a third reminder message and a fourth reminder message. The first reminder message is used to inform the target object that the face verification failed, the second reminder message is used to remind the target object that the identity authentication was successful, the third reminder message is used to inform the target object that the associated information verification failed, and the fourth reminder message is used to remind the target object that the associated information of the target object cannot be found; The early warning mechanism module is used to trigger the early warning mechanism, and when the number of target object matching failures exceeds a preset value, the administrator is notified to handle the problem.

Images