A special identification method and system for AI intelligent platform

Abstract

The present invention discloses a special recognition method and system for an AI intelligent platform. The method includes collecting a user's face image, preprocessing the face image, extracting the geometric features and shape features of the preprocessed face image; identifying the first emotional features of the preprocessed face image according to a neural network model; generating a first text vector from the geometric features, the geometric features, and the emotional features, and compressing the first text vector to generate a QR code; during recognition, collecting a real-time face image and an image of a user holding a work card containing the QR code; identifying the QR code and parsing it into a first text vector, extracting and identifying the real-time face image geometric features and shape features, and the second emotional features, and generating a second text vector. The face features in the image and the features in the QR code are identified by image acquisition on the local end, and user authentication can be achieved by comparing the similarity between the two, thereby reducing the load of cloud transmission.

Description

Technical Field

[0001] The present invention belongs to the field of computer recognition, and specifically relates to a special recognition system and method for an AI intelligent platform. Background Art

[0002] In recent years, QR codes were initially used in product tracking, object identification, file management, etc. In recent years, with the rapid development of mobile Internet, QR codes have quickly become one of the main ways of information exchange, and using mobile phones to scan them anytime and anywhere has become a habit. Scanning QR codes to surf the Internet, scan QR codes to shop, scan QR codes to pay, etc. have become part of people's daily lives.

[0003] How to apply QR codes to face recognition to improve the efficiency of face recognition has become a hot issue. Summary of the invention

[0004] In view of the existing problems, the present invention provides an AI intelligent platform-specific identification method, which includes the following steps:

[0005] S1. Collecting a user's facial image, preprocessing the facial image, and extracting geometric features and shape features of the preprocessed facial image; the facial image is stored in the cloud;

[0006] S2, identifying the first emotion feature of the preprocessed facial image according to the neural network model;

[0007] S3, generating a first text vector from geometric features, shape features, and emotional features, and compressing the first text vector to generate a QR code;

[0008] S4. During recognition, a real-time facial image and an image of the user holding a work card containing the QR code are collected;

[0009] S5, identifying the two-dimensional code and parsing it into a first text vector, extracting and identifying the geometric features and shape features of the real-time face image and the second emotional features, and generating a second text vector; the neural network model used for identifying the second emotional features is different from that used for identifying the first emotional features;

[0010] S6, calculate the similarity between the first text vector and the second text vector, if the similarity is greater than or equal to the threshold, the recognition is successful, if the similarity is less than the threshold; split the first text vector into a third text vector composed of geometric features and shape features, and a fourth text vector composed of emotional features; split the second text vector into a fifth text vector composed of geometric features and shape features, and a sixth text vector composed of emotional features; calculate the similarity between the third text vector and the fifth text vector, if the similarity is greater than or equal to the threshold, prompt the recognition is successful, and send the sixth text vector to the cloud; the cloud sends a prompt message to the user's mobile terminal; the prompt message is a text or voice link of emotion enhancement; otherwise, the recognition is unsuccessful;

[0011] S7. When the QR code is blurred or the face collection image is unclear, the QR code or face collection image is compared with the face image features stored in the cloud by an offset value. When the offset value is less than a preset value, the recognition is successful.

[0012] The facial images stored in the cloud are periodically replaced.

[0013] Furthermore, the geometric features include the relative distances between eyes, nose and mouth; and the shape features include the width of the outline and forehead.

[0014] Furthermore, the first emotional characteristic is happiness.

[0015] Further, generating a first text vector from the geometric features, the shape features, and the emotional features includes:

[0016] S301, connecting geometric features, shape features, and emotional features to generate an intermediate vector;

[0017] S302: compress the intermediate vector to generate a first text vector.

[0018] Furthermore, compressing the first text vector to generate a QR code includes:

[0019] S310, analyzing a first text vector to be embedded, and determining an encoding method for encoding the text vector;

[0020] S311, encoding the first text vector into a data stream according to the encoding format, and adding a supplementary code at the end of the data stream to supplement the data stream to an appropriate length;

[0021] S312, dividing the data stream into different byte segments, and calculating the verification code for each byte segment;

[0022] S313, cross-arranging the data stream and the check code;

[0023] S314, drawing the data stream and the functional area into the QR code to obtain an intermediate QR code;

[0024] S315, respectively use multiple mask patterns to perform mask operations on the intermediate two-dimensional code, and select the two-dimensional code with the best balanced black mode as the final result; fill corresponding data in the positions corresponding to the function area version information and the mask code, and output the two-dimensional code pattern.

[0025] Further, identifying the QR code includes:

[0026] S51, performing a preprocessing operation on the candidate area containing the QR code to eliminate the uneven illumination and noise introduced by the shooting environment and image acquisition;

[0027] S52, detecting the position of the QR code and detecting the corner points of the graphics as alignment key points, and performing QR code correction;

[0028] S53, decoding the processed QR code image and outputting the result.

[0029] An AI intelligent platform dedicated recognition system, the system comprising:

[0030] The sensor module is used to collect the user's face image when creating a file and to collect the real-time face image and the image of the user holding the work card containing the QR code when identifying;

[0031] An image processing module is used to extract geometric features and shape features of the preprocessed face image after preprocessing the face image; and to identify the first emotion feature of the preprocessed face image according to the neural network model;

[0032] A cloud storage module for storing facial images;

[0033] A QR code generation module, used to generate a first text vector from geometric features, shape features, and emotional features, and to compress the first text vector to generate a QR code;

[0034] An identity verification module is used to identify the two-dimensional code and parse it into a first text vector, extract and identify the geometric features and shape features of the real-time facial image, and the second emotional features, and generate a second text vector; the neural network model used for identifying the second emotional feature is different from that used for identifying the first emotional feature; calculate the similarity between the first text vector and the second text vector, and if the similarity is greater than or equal to a threshold, the recognition is successful; if the similarity is less than the threshold; then split the first text vector into a third text vector composed of geometric features and shape features, and a fourth text vector composed of emotional features; split the second text vector into a fifth text vector composed of geometric features and shape features, and a sixth text vector composed of emotional features; calculate the similarity between the third text vector and the fifth text vector, and if the similarity is greater than or equal to a threshold, then prompt that the recognition is successful, and send the sixth text vector to the cloud; the cloud sends a prompt message to the user's mobile terminal; the prompt message is a text or voice connection for emotional enhancement; otherwise, the recognition is unsuccessful;

[0035] The fuzzy verification module is used to compare the offset value of the QR code or face collection image with the face image features stored in the cloud when the QR code is fuzzy or the face collection image is unclear. When the offset value is less than the preset value, the recognition is successful;

[0036] Information prompt module, used to send emotion-enhancing text or voice links.

[0037] An electronic device, characterized in that it includes a processor, a memory, a communication interface and a program, wherein the one or more programs are stored in the memory and configured to be executed by the processor, and the program includes instructions for executing the steps in the AI ​​intelligent platform-specific identification method.

[0038] A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement instructions for the steps of an AI intelligent platform-specific identification method.

[0039] The beneficial effects of the present invention are as follows:

[0040] In the face recognition process, the present invention uses local image acquisition to identify face features in the image and features in the QR code, and can achieve user authentication by comparing the similarity between the two, thereby reducing the load of cloud transmission.

[0041] By identifying the user's emotions, sensing the user's current state, and sending information prompts through the mobile terminal, the user can actively adjust his or her emotions to achieve the effect of stabilizing emotions. At the same time, it can identify specific emotions and provide guidance to restore the user's mental health.

[0042] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented in accordance with the contents of the specification. In order to make the above description and other purposes, features and advantages of the present invention more obvious and easy to understand, the preferred embodiments are specifically cited and described in detail as follows. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0044] Figure 1 Structural diagram of the AI ​​intelligent platform dedicated recognition system DETAILED DESCRIPTION

[0045] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0046] In the description of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be connected, detachably connected, or integrated; it can be mechanically connected or electrically connected; it can be directly connected or indirectly connected through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. Example 1

[0047] An AI intelligent platform-specific identification method, the method comprising the following steps:

[0048] S1. Collecting a user's facial image, preprocessing the facial image, and extracting geometric features and shape features of the preprocessed facial image; the facial image is stored in the cloud;

[0049] S2, identifying the first emotion feature of the preprocessed facial image according to the neural network model;

[0050] S3, generating a first text vector from geometric features, shape features, and emotional features, and compressing the first text vector to generate a QR code;

[0051] S4. During recognition, a real-time facial image and an image of the user holding a work card containing the QR code are collected;

[0052] S5, identifying the two-dimensional code and parsing it into a first text vector, extracting and identifying the geometric features and shape features of the real-time face image and the second emotional features, and generating a second text vector; the neural network model used for identifying the second emotional features is different from that used for identifying the first emotional features;

[0053] S6, calculate the similarity between the first text vector and the second text vector, if the similarity is greater than or equal to the threshold, the recognition is successful, if the similarity is less than the threshold; split the first text vector into a third text vector composed of geometric features and shape features, and a fourth text vector composed of emotional features; split the second text vector into a fifth text vector composed of geometric features and shape features, and a sixth text vector composed of emotional features; calculate the similarity between the third text vector and the fifth text vector, if the similarity is greater than or equal to the threshold, prompt the recognition is successful, and send the sixth text vector to the cloud; the cloud sends a prompt message to the user's mobile terminal; the prompt message is a text or voice link of emotion enhancement; otherwise, the recognition is unsuccessful;

[0054] S7. When the QR code is blurred or the face collection image is unclear, the QR code or face collection image is compared with the face image features stored in the cloud by an offset value. When the offset value is less than a preset value, the recognition is successful.

[0055] The facial images stored in the cloud are periodically replaced.

[0056] Furthermore, the geometric features include the relative distances between eyes, nose and mouth; and the shape features include the width of the outline and forehead.

[0057] Furthermore, the first emotional characteristic is happiness.

[0058] Further, generating a first text vector from the geometric features, the shape features, and the emotional features includes:

[0059] S301, connecting geometric features, shape features, and emotional features to generate an intermediate vector;

[0060] S302: compress the intermediate vector to generate a first text vector.

[0061] Furthermore, compressing the first text vector to generate a QR code includes:

[0062] S310, analyzing a first text vector to be embedded, and determining an encoding method for encoding the text vector;

[0063] S311, encoding the first text vector into a data stream according to the encoding format, and adding a supplementary code at the end of the data stream to supplement the data stream to an appropriate length;

[0064] S312, dividing the data stream into different byte segments, and calculating the verification code for each byte segment;

[0065] S313, cross-arranging the data stream and the check code;

[0066] S314, drawing the data stream and the functional area into the QR code to obtain an intermediate QR code;

[0067] S315, respectively use multiple mask patterns to perform mask operations on the intermediate two-dimensional code, and select the two-dimensional code with the best balanced black mode as the final result; fill corresponding data in the positions corresponding to the function area version information and the mask code, and output the two-dimensional code pattern.

[0068] Further, identifying the QR code includes:

[0069] S51, performing a preprocessing operation on the candidate area containing the QR code to eliminate the uneven illumination and noise introduced by the shooting environment and image acquisition;

[0070] S52, detecting the position of the QR code and detecting the corner points of the graphics as alignment key points, and performing QR code correction;

[0071] S53, decoding the processed QR code image and outputting the result.

[0072] Furthermore, when the application scenario is a school, the first emotion can also be anger or loss; the corresponding prompt information of the mobile terminal is a piece of white noise music and text information; the text information includes but is not limited to ancient poems and famous quotes, etc. At the same time, the user status is sent to the cloud; when the user's mood is abnormal for many consecutive days, the user status is sent to the psychological teacher, who will intervene.

[0073] Furthermore, the similarity calculation is weighted cosine distance calculation.

[0074]

[0075] Where A and B are the first text vector and the second text vector respectively;

[0076] Wi is the weight of the i-th dimension;

[0077] n is the dimension of the text vector;

[0078] An electronic device, characterized in that it includes a processor, a memory, a communication interface and a program, wherein the one or more programs are stored in the memory and configured to be executed by the processor, and the program includes instructions for executing the steps in the AI ​​intelligent platform-specific identification method.

[0079] A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement instructions for the steps of an AI intelligent platform-specific identification method. Example 2

[0080] An AI intelligent platform dedicated recognition system, the system comprising:

[0081] The sensor module is used to collect the user's face image when creating a file and to collect the real-time face image and the image of the user holding the work card containing the QR code when identifying;

[0082] An image processing module is used to extract geometric features and shape features of the preprocessed face image after preprocessing the face image; and to identify the first emotion feature of the preprocessed face image according to the neural network model;

[0083] A cloud storage module for storing facial images;

[0084] A QR code generation module, used to generate a first text vector from geometric features, shape features, and emotional features, and to compress the first text vector to generate a QR code;

[0085] An identity verification module is used to identify the two-dimensional code and parse it into a first text vector, extract and identify the geometric features and shape features of the real-time facial image, and the second emotional features, and generate a second text vector; the neural network model used for identifying the second emotional feature is different from that used for identifying the first emotional feature; calculate the similarity between the first text vector and the second text vector, and if the similarity is greater than or equal to a threshold, the recognition is successful; if the similarity is less than the threshold; then split the first text vector into a third text vector composed of geometric features and shape features, and a fourth text vector composed of emotional features; split the second text vector into a fifth text vector composed of geometric features and shape features, and a sixth text vector composed of emotional features; calculate the similarity between the third text vector and the fifth text vector, and if the similarity is greater than or equal to a threshold, then prompt that the recognition is successful, and send the sixth text vector to the cloud; the cloud sends a prompt message to the user's mobile terminal; the prompt message is a text or voice connection for emotional enhancement; otherwise, the recognition is unsuccessful;

[0086] The fuzzy verification module is used to compare the offset value of the QR code or face collection image with the face image features stored in the cloud when the QR code is fuzzy or the face collection image is unclear. When the offset value is less than the preset value, the recognition is successful;

[0087] Information prompt module, used to send emotion-enhancing text or voice links.

[0088] The advantages of the present invention are as follows:

[0089] In the face recognition process, the present invention uses local image acquisition to identify face features in the image and features in the QR code, and can achieve user authentication by comparing the similarity between the two, thereby reducing the load of cloud transmission.

[0090] By identifying the user's emotions, sensing the user's current state, and sending information prompts through the mobile terminal, the user can actively adjust his or her emotions to achieve the effect of stabilizing emotions. At the same time, it can identify specific emotions and provide guidance to restore the user's mental health.

[0091] In the several embodiments provided in the present application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of the units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the device or unit can be electrical or other forms.

[0092] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0093] Each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above integrated unit may be implemented in the form of hardware or in the form of software program modules.

[0094] If the integrated unit is implemented in the form of a software program module and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a memory, including a number of instructions to enable a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned memory includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or CD-ROM and other media that can store program codes.

[0095] A person skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable memory, and the memory can include: a flash drive, a read-only memory (English: Read-Only Memory, abbreviated as: ROM), a random access memory (English: Random Access Memory, abbreviated as: RAM), a magnetic disk or an optical disk, etc.

[0096] The embodiments of the present application are introduced in detail above. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and core idea of ​​the present application. At the same time, for general technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A special identification method for an AI intelligent platform, characterized in that: The method comprises the following steps: S1. Collecting a user's facial image, preprocessing the facial image, and extracting geometric features and shape features of the preprocessed facial image; the facial image is stored in the cloud; S2, identifying the first emotion feature of the preprocessed facial image according to the neural network model; S3, generating a first text vector from geometric features, shape features, and emotional features, and compressing the first text vector to generate a QR code; S4. During recognition, a real-time facial image and an image of the user holding a work card containing the QR code are collected; S5, identifying the two-dimensional code and parsing it into a first text vector, extracting and identifying the geometric features and shape features of the real-time face image and the second emotional features, and generating a second text vector; the neural network model used for identifying the second emotional features is different from that used for identifying the first emotional features; S6, calculate the similarity between the first text vector and the second text vector, if the similarity is greater than or equal to a threshold, the recognition is successful; if the similarity is less than the threshold; split the first text vector into a third text vector composed of geometric features and shape features, and a fourth text vector composed of emotional features; split the second text vector into a fifth text vector composed of geometric features and shape features, and a sixth text vector composed of emotional features; calculate the similarity between the third text vector and the fifth text vector, if the similarity is greater than or equal to a threshold, prompt the recognition is successful, and send the sixth text vector to the cloud; the cloud sends a prompt message to the user's mobile terminal; the prompt message is a text or voice link of emotion enhancement; otherwise, the recognition is unsuccessful; S7. When the QR code is blurred or the face collection image is unclear, the QR code or face collection image is compared with the face image features stored in the cloud by an offset value. When the offset value is less than a preset value, the recognition is successful. The facial images stored in the cloud are periodically replaced.

2. The method according to claim 1, characterized in that: The geometric features include the relative distances between eyes, nose and mouth; the shape features include the width of the outline and forehead.

3. The method according to claim 2, characterized in that: The first emotion characteristic is happiness.

4. The method according to claim 1, characterized in that: Generating the first text vector from the geometric features, shape features, and emotional features includes: S301, connecting geometric features, shape features, and emotional features to generate an intermediate vector; S302: compress the intermediate vector to generate a first text vector.

5. The method according to claim 1, characterized in that: Compressing the first text vector to generate a QR code includes: S310, analyzing a first text vector to be embedded, and determining an encoding method for encoding the text vector; S311, encoding the first text vector into a data stream according to the encoding format, and adding a supplementary code at the end of the data stream to supplement the data stream to an appropriate length; S312, dividing the data stream into different byte segments, and calculating the verification code for each byte segment; S313, cross-arranging the data stream and the check code; S314, drawing the data stream and the functional area into the QR code to obtain an intermediate QR code; S315, respectively use multiple mask patterns to perform mask operations on the intermediate two-dimensional code, and select the two-dimensional code with the best balanced black mode as the final result; fill corresponding data in the positions corresponding to the function area version information and the mask code, and output the two-dimensional code pattern.

6. The method according to claim 1, characterized in that: Identifying the QR code includes: S51, performing a preprocessing operation on the candidate area containing the QR code to eliminate the uneven illumination and noise introduced by the shooting environment and image acquisition; S52, detecting the position of the QR code and detecting the corner points of the graphics as alignment key points, and performing QR code correction; S53, decoding the processed QR code image and outputting the result.

7. An AI intelligent platform dedicated identification system, used to execute the AI ​​intelligent platform dedicated identification method according to any one of claims 1 to 6, characterized in that: The system includes: The sensor module is used to collect the user's face image when creating a file and to collect the real-time face image and the image of the user holding the work card containing the QR code when identifying; An image processing module is used to extract geometric features and shape features of the preprocessed face image after preprocessing the face image; and to identify the first emotion feature of the preprocessed face image according to the neural network model; A cloud storage module for storing facial images; A QR code generation module, used to generate a first text vector from geometric features, shape features, and emotional features, and to compress the first text vector to generate a QR code; The identity verification module is used to identify the two-dimensional code and parse it into a first text vector, extract and identify the geometric features and shape features of the real-time face image, and the second emotional features, and generate a second text vector; the neural network model used for identifying the second emotional feature is different from that used for identifying the first emotional feature; calculate the similarity between the first text vector and the second text vector, and if the similarity is greater than or equal to a threshold, the recognition is successful; The fuzzy verification module is used to compare the offset value of the QR code or face collection image with the face image features stored in the cloud when the QR code is fuzzy or the face collection image is unclear. When the offset value is less than the preset value, the recognition is successful; Information prompt module, used to send emotion-enhancing text or voice links.

8. An electronic device, characterized in that: The method comprises a processor, a memory, a communication interface and a program, wherein the one or more programs are stored in the memory and configured to be executed by the processor, and the program comprises instructions for executing the steps in any one of the methods of claims 1-6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the method according to any one of claims 1 to 6.

Images